archive/issues_020499.json:
{
"assignees": [],
"body": "<div id=\"comment:0\"></div>\n\nThe patchbot is currently installing in a very not kosher way, as explained in\n\nhttps://github.com/robertwb/sage-patchbot/issues/79\n\nThis ticket will change completely the installation mode. The patchbot will become a python package, to be installed using pip in\nthe system python as explained below.\n\nThis ticket also updates the patchbot, in particular to cope with the new https trac.\n\nNew tarball for patchbot: http://www-irma.u-strasbg.fr/~chapoton/sage-patchbot-2.6.1.tar.gz\n\ninstallation (two equivalent methods at your disposal):\n1. Download the tarball and then\n\n```\npip install --user sage-patchbot-2.6.1.tar.gz\n```\n\n2. (recommended)\n\n```\npip install --user git+https://github.com/sagemath/[email protected]\n```\n\nThere are now two possible ways to launch the patchbot:\n1.\n\n```\npython -m sage_patchbot.patchbot --sage-root=XXXX\n```\nor\n2. Checkout the branch here and run\n\n```\nmake && ./sage --patchbot\n```\n\nDepends on #22473\n\nCC: @embray @jdemeyer\n\nComponent: **packages: optional**\n\nKeywords: **patchbot**\n\nAuthor: **Fr\u00e9d\u00e9ric Chapoton**\n\nBranch/Commit: **[`c9c4101`](https://github.com/sagemath/sagetrac-mirror/commit/c9c4101b961c454b2acac23508a74ccfbcffb7c4)**\n\nReviewer: **Jeroen Demeyer**\n\n_Issue created by migration from https://trac.sagemath.org/ticket/20736_\n\n",
"closed_at": "2017-03-05T15:13:53Z",
"created_at": "2016-05-31T18:31:53Z",
"labels": [
"https://github.com/sagemath/sage/labels/c%3A%20packages%3A%20optional",
"https://github.com/sagemath/sage/labels/p%3A%20major%20/%203",
"https://github.com/sagemath/sage/labels/enhancement"
],
"milestone": "https://github.com/sagemath/sage/milestones/sage-7.6",
"reactions": [],
"repository": "https://github.com/sagemath/sage",
"title": "Upgrade patchbot to 2.6.1 as a system package",
"type": "issue",
"updated_at": "2017-03-05T15:13:53Z",
"url": "https://github.com/sagemath/sage/issues/20736",
"user": "https://github.com/fchapoton"
}
The patchbot is currently installing in a very not kosher way, as explained in
This ticket will change completely the installation mode. The patchbot will become a python package, to be installed using pip in the system python as explained below.
This ticket also updates the patchbot, in particular to cope with the new https trac.
New tarball for patchbot: http://www-irma.u-strasbg.fr/~chapoton/sage-patchbot-2.6.1.tar.gz
installation (two equivalent methods at your disposal):
- Download the tarball and then
pip install --user sage-patchbot-2.6.1.tar.gz
- (recommended)
pip install --user git+https://github.com/sagemath/[email protected]
There are now two possible ways to launch the patchbot: 1.
python -m sage_patchbot.patchbot --sage-root=XXXX
or 2. Checkout the branch here and run
make && ./sage --patchbot
Depends on #22473
CC: @embray @jdemeyer
Component: packages: optional
Keywords: patchbot
Author: Frédéric Chapoton
Branch/Commit: c9c4101
Reviewer: Jeroen Demeyer
Issue created by migration from https://trac.sagemath.org/ticket/20736
archive/issue_events_289385.json:
{
"actor": "https://github.com/fchapoton",
"created_at": "2016-05-31T18:31:53Z",
"event": "milestoned",
"issue": "https://github.com/sagemath/sage/issues/20736",
"milestone_number": null,
"milestone_title": "sage-7.3",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289385"
}
archive/issue_events_289386.json:
{
"actor": "https://github.com/fchapoton",
"created_at": "2016-05-31T18:31:53Z",
"event": "labeled",
"issue": "https://github.com/sagemath/sage/issues/20736",
"label": "https://github.com/sagemath/sage/labels/c%3A%20packages%3A%20optional",
"label_color": "0000b0",
"label_name": "c: packages: optional",
"label_text_color": "ffffff",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289386"
}
archive/issue_events_289387.json:
{
"actor": "https://github.com/fchapoton",
"created_at": "2016-05-31T18:31:53Z",
"event": "labeled",
"issue": "https://github.com/sagemath/sage/issues/20736",
"label": "https://github.com/sagemath/sage/labels/p%3A%20major%20/%203",
"label_color": "ffbb00",
"label_name": "p: major / 3",
"label_text_color": "ffffff",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289387"
}
archive/issue_events_289388.json:
{
"actor": "https://github.com/fchapoton",
"created_at": "2016-05-31T18:31:53Z",
"event": "labeled",
"issue": "https://github.com/sagemath/sage/issues/20736",
"label": "https://github.com/sagemath/sage/labels/enhancement",
"label_color": "696969",
"label_name": "enhancement",
"label_text_color": "ffffff",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289388"
}
archive/issue_comments_300294.json:
{
"body": "Description changed:\n``````diff\n--- \n+++ \n@@ -3,3 +3,5 @@\n https://github.com/robertwb/sage-patchbot/issues/79\n \n Let us correct that.\n+\n+Tarball: \u200b\u200b\u200b\u200bhttp://www-irma.u-strasbg.fr/~chapoton/patchbot-2.5.8.tar.bz2\n``````\n",
"created_at": "2016-05-31T18:36:31Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300294",
"user": "https://github.com/fchapoton"
}
Description changed:
---
+++
@@ -3,3 +3,5 @@
https://github.com/robertwb/sage-patchbot/issues/79
Let us correct that.
+
+Tarball: http://www-irma.u-strasbg.fr/~chapoton/patchbot-2.5.8.tar.bz2
archive/issue_comments_300295.json:
{
"body": "<div id=\"comment:2\" align=\"right\">comment:2</div>\n\nI tried and failed so far. Help is required, please.\n\n---\nNew commits:\n<table><tr><td><a href=\"https://github.com/sagemath/sagetrac-mirror/commit/a827087d9b4335f351605f2d2a119b53d0c3bd25\"><code>a827087</code></a></td><td><code>trac 20736 first try, failing</code></td></tr></table>\n",
"created_at": "2016-05-31T18:57:25Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300295",
"user": "https://github.com/fchapoton"
}
I tried and failed so far. Help is required, please.
New commits:
a827087 | trac 20736 first try, failing |
archive/issue_comments_300296.json:
{
"body": "Branch: **[public/20736](https://github.com/sagemath/sagetrac-mirror/tree/public/20736)**",
"created_at": "2016-05-31T18:57:25Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300296",
"user": "https://github.com/fchapoton"
}
Branch: public/20736
archive/issue_comments_300297.json:
{
"body": "Commit: **[`a827087`](https://github.com/sagemath/sagetrac-mirror/commit/a827087d9b4335f351605f2d2a119b53d0c3bd25)**",
"created_at": "2016-05-31T18:57:25Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300297",
"user": "https://github.com/fchapoton"
}
Commit: a827087
archive/issue_comments_300298.json:
{
"body": "<div id=\"comment:3\" align=\"right\">comment:3</div>\n\nNot related to installation problem, but in the `dependencies` file you should put\n\n```\n$(PYTHON) | setuptools\n```\n\nThat is, if `setuptools` is reinstalled it's not necessary to reinstall a Python package that happens to use it. But it is a prerequisite to installing.",
"created_at": "2016-06-01T11:31:35Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300298",
"user": "https://github.com/embray"
}
Not related to installation problem, but in the dependencies
file you should put
$(PYTHON) | setuptools
That is, if setuptools
is reinstalled it's not necessary to reinstall a Python package that happens to use it. But it is a prerequisite to installing.
archive/issue_comments_300299.json:
{
"body": "<div id=\"comment:4\"></div>\n\nBranch pushed to git repo; I updated commit sha1. New commits:\n<table><tr><td><a href=\"https://github.com/sagemath/sagetrac-mirror/commit/c9907efed9c7f02a928af40a2514a28f2ed7e356\"><code>c9907ef</code></a></td><td><code>trac 20736 change in dependencies</code></td></tr></table>\n",
"created_at": "2016-06-01T11:34:00Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300299",
"user": "https://github.com/sagetrac-git"
}
Branch pushed to git repo; I updated commit sha1. New commits:
c9907ef | trac 20736 change in dependencies |
archive/issue_comments_300300.json:
{
"body": "Changed commit from **[`a827087`](https://github.com/sagemath/sagetrac-mirror/commit/a827087d9b4335f351605f2d2a119b53d0c3bd25)** to **[`c9907ef`](https://github.com/sagemath/sagetrac-mirror/commit/c9907efed9c7f02a928af40a2514a28f2ed7e356)**",
"created_at": "2016-06-01T11:34:00Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300300",
"user": "https://github.com/sagetrac-git"
}
Changed commit from a827087
to c9907ef
archive/issue_comments_300301.json:
{
"body": "<div id=\"comment:5\" align=\"right\">comment:5</div>\n\nBTW, when building a source package of a Python project, use `./setup.py sdist` to make the tarball (also make sure the tarball contains all the files it should contain, and none that it shouldn't contain--you might have to edit a `MANIFEST.in` file to adjust it a la https://docs.python.org/2/distutils/sourcedist.html#specifying-the-files-to-distribute)",
"created_at": "2016-06-01T11:36:56Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300301",
"user": "https://github.com/embray"
}
BTW, when building a source package of a Python project, use ./setup.py sdist
to make the tarball (also make sure the tarball contains all the files it should contain, and none that it shouldn't contain--you might have to edit a MANIFEST.in
file to adjust it a la https://docs.python.org/2/distutils/sourcedist.html#specifying-the-files-to-distribute)
archive/issue_comments_300302.json:
{
"body": "Description changed:\n``````diff\n--- \n+++ \n@@ -4,4 +4,6 @@\n \n Let us correct that.\n \n-Tarball: \u200b\u200b\u200b\u200bhttp://www-irma.u-strasbg.fr/~chapoton/patchbot-2.5.8.tar.bz2\n+Tarball: \u200b\u200b\u200b\u200bhttp://www-irma.u-strasbg.fr/~chapoton/sage-patchbot-2.5.9.tar.gz\n+\n+(needs to be renamed patchbot-2.5.9.tar.gz)\n``````\n",
"created_at": "2016-06-01T15:11:50Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300302",
"user": "https://github.com/fchapoton"
}
Description changed:
---
+++
@@ -4,4 +4,6 @@
Let us correct that.
-Tarball: http://www-irma.u-strasbg.fr/~chapoton/patchbot-2.5.8.tar.bz2
+Tarball: http://www-irma.u-strasbg.fr/~chapoton/sage-patchbot-2.5.9.tar.gz
+
+(needs to be renamed patchbot-2.5.9.tar.gz)
archive/issue_comments_300303.json:
{
"body": "Changed commit from **[`c9907ef`](https://github.com/sagemath/sagetrac-mirror/commit/c9907efed9c7f02a928af40a2514a28f2ed7e356)** to **[`f3839fd`](https://github.com/sagemath/sagetrac-mirror/commit/f3839fd6c0fb2562fb7731e0a62f73a2d3b89124)**",
"created_at": "2016-06-01T15:17:01Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300303",
"user": "https://github.com/sagetrac-git"
}
Changed commit from c9907ef
to f3839fd
archive/issue_comments_300304.json:
{
"body": "<div id=\"comment:7\"></div>\n\nBranch pushed to git repo; I updated commit sha1. New commits:\n<table><tr><td><a href=\"https://github.com/sagemath/sagetrac-mirror/commit/f3839fd6c0fb2562fb7731e0a62f73a2d3b89124\"><code>f3839fd</code></a></td><td><code>another try, failing the same way</code></td></tr></table>\n",
"created_at": "2016-06-01T15:17:01Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300304",
"user": "https://github.com/sagetrac-git"
}
Branch pushed to git repo; I updated commit sha1. New commits:
f3839fd | another try, failing the same way |
archive/issue_comments_300305.json:
{
"body": "<div id=\"comment:8\" align=\"right\">comment:8</div>\n\nI am making some slow progress.",
"created_at": "2016-06-01T15:45:32Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300305",
"user": "https://github.com/fchapoton"
}
I am making some slow progress.
archive/issue_comments_300306.json:
{
"body": "Description changed:\n``````diff\n--- \n+++ \n@@ -4,6 +4,5 @@\n \n Let us correct that.\n \n-Tarball: \u200b\u200b\u200b\u200bhttp://www-irma.u-strasbg.fr/~chapoton/sage-patchbot-2.5.9.tar.gz\n+Tarball: \u200b\u200b\u200b\u200bhttp://www-irma.u-strasbg.fr/~chapoton/patchbot-2.5.9.tar.gz\n \n-(needs to be renamed patchbot-2.5.9.tar.gz)\n``````\n",
"created_at": "2016-06-01T15:46:03Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300306",
"user": "https://github.com/fchapoton"
}
Description changed:
---
+++
@@ -4,6 +4,5 @@
Let us correct that.
-Tarball: http://www-irma.u-strasbg.fr/~chapoton/sage-patchbot-2.5.9.tar.gz
+Tarball: http://www-irma.u-strasbg.fr/~chapoton/patchbot-2.5.9.tar.gz
-(needs to be renamed patchbot-2.5.9.tar.gz)
archive/issue_comments_300307.json:
{
"body": "<div id=\"comment:10\" align=\"right\">comment:10</div>\n\nThere is a fundamental problem here, mentioned also at https://github.com/robertwb/sage-patchbot/issues/79\n\nWe cannot use the Sage Python for the patchbot, since the patchbot might reinstall Python. So I don't know what the best solution is, but installing it \"the usual way\" is **not** the right thing to do here.",
"created_at": "2016-06-01T15:47:10Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300307",
"user": "https://github.com/jdemeyer"
}
There is a fundamental problem here, mentioned also at sagemath/sage-patchbot#79
We cannot use the Sage Python for the patchbot, since the patchbot might reinstall Python. So I don't know what the best solution is, but installing it "the usual way" is not the right thing to do here.
archive/issue_comments_300308.json:
{
"body": "<div id=\"comment:11\"></div>\n\nBranch pushed to git repo; I updated commit sha1. New commits:\n<table><tr><td><a href=\"https://github.com/sagemath/sagetrac-mirror/commit/160332c495872a01047d109a8584aca955d98252\"><code>160332c</code></a></td><td><code>fixing the spkg-install script</code></td></tr></table>\n",
"created_at": "2016-06-01T15:50:27Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300308",
"user": "https://github.com/sagetrac-git"
}
Branch pushed to git repo; I updated commit sha1. New commits:
160332c | fixing the spkg-install script |
archive/issue_comments_300309.json:
{
"body": "Changed commit from **[`f3839fd`](https://github.com/sagemath/sagetrac-mirror/commit/f3839fd6c0fb2562fb7731e0a62f73a2d3b89124)** to **[`160332c`](https://github.com/sagemath/sagetrac-mirror/commit/160332c495872a01047d109a8584aca955d98252)**",
"created_at": "2016-06-01T15:50:27Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300309",
"user": "https://github.com/sagetrac-git"
}
Changed commit from f3839fd
to 160332c
archive/issue_comments_300310.json:
{
"body": "<div id=\"comment:12\" align=\"right\">comment:12</div>\n\nSo it's a build dep for Sage and should be installed \"the usual way\" using the system Python.",
"created_at": "2016-06-02T09:12:40Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300310",
"user": "https://github.com/embray"
}
So it's a build dep for Sage and should be installed "the usual way" using the system Python.
archive/issue_comments_300311.json:
{
"body": "<div id=\"comment:13\" align=\"right\">comment:13</div>\n\nReplying to [@embray](#comment%3A12):\n> So it's a build dep for Sage and should be installed \"the usual way\" using the system Python.\n\nI don't get why you say a \"build dep\", but I agree with the second part.",
"created_at": "2016-06-02T09:40:38Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300311",
"user": "https://github.com/jdemeyer"
}
Replying to @embray:
So it's a build dep for Sage and should be installed "the usual way" using the system Python.
I don't get why you say a "build dep", but I agree with the second part.
archive/issue_comments_300312.json:
{
"body": "<div id=\"comment:14\" align=\"right\">comment:14</div>\n\nI changed the name of the package back to sage-patchbot. But I am little bit stuck here.\n\nIt kind of worked with the package name \"patchbot\", but it was installed in SAGE_LOCAL.",
"created_at": "2016-06-02T09:54:48Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300312",
"user": "https://github.com/fchapoton"
}
I changed the name of the package back to sage-patchbot. But I am little bit stuck here.
It kind of worked with the package name "patchbot", but it was installed in SAGE_LOCAL.
archive/issue_comments_300313.json:
{
"body": "<div id=\"comment:15\" align=\"right\">comment:15</div>\n\nReplying to [@jdemeyer](#comment%3A13):\n> Replying to [@embray](#comment%3A12):\n> > So it's a build dep for Sage and should be installed \"the usual way\" using the system Python.\n> \n> \n> I don't get why you say a \"build dep\", but I agree with the second part.\n\n\"build dep\" is probably the wrong description--but I just meant a developer tool for development of Sage, outside the sage distribution.",
"created_at": "2016-06-02T13:20:43Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300313",
"user": "https://github.com/embray"
}
Replying to @jdemeyer:
Replying to @embray:
So it's a build dep for Sage and should be installed "the usual way" using the system Python.
I don't get why you say a "build dep", but I agree with the second part.
"build dep" is probably the wrong description--but I just meant a developer tool for development of Sage, outside the sage distribution.
archive/issue_comments_300314.json:
{
"body": "<div id=\"comment:16\" align=\"right\">comment:16</div>\n\nReplying to [@fchapoton](#comment%3A14):\n> I changed the name of the package back to sage-patchbot. But I am little bit stuck here.\n> \n> It kind of worked with the package name \"patchbot\", but it was installed in SAGE_LOCAL.\n\nTo be clear, this is a case where the term \"package\" is often overloaded in Python:\n\n* Here the package, as in the name of what you import in Python is `patchbot`\n\n* The name in the `setup.py` is different from the package name, and as sometimes referred to as the \"distribution name\" or the \"PyPI name\" to reduce confusion. Often the package name and the distribution name are the same, but they don't have to be. Either way it shouldn't be relevant here.\n\nI think the main takeaway here is that the patchbot should not be a Sage spkg at all. It's not a part of sage, the distribution. It's a tool for testing sage the distribution and as such should be entirely external to it.\n\nSo to install the patchbot on say a Linux system one would use `sudo pip install <path/to/tarball>` or even just `sudo pip install sage-patchbot` if the latest version is uploaded to PyPI.\n\nAlternatively this can be done within a virtualenv if we prefer to avoid the system install.",
"created_at": "2016-06-02T13:23:27Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300314",
"user": "https://github.com/embray"
}
Replying to @fchapoton:
I changed the name of the package back to sage-patchbot. But I am little bit stuck here.
It kind of worked with the package name "patchbot", but it was installed in SAGE_LOCAL.
To be clear, this is a case where the term "package" is often overloaded in Python:
-
Here the package, as in the name of what you import in Python is
patchbot
-
The name in the
setup.py
is different from the package name, and as sometimes referred to as the "distribution name" or the "PyPI name" to reduce confusion. Often the package name and the distribution name are the same, but they don't have to be. Either way it shouldn't be relevant here.
I think the main takeaway here is that the patchbot should not be a Sage spkg at all. It's not a part of sage, the distribution. It's a tool for testing sage the distribution and as such should be entirely external to it.
So to install the patchbot on say a Linux system one would use sudo pip install <path/to/tarball>
or even just sudo pip install sage-patchbot
if the latest version is uploaded to PyPI.
Alternatively this can be done within a virtualenv if we prefer to avoid the system install.
archive/issue_comments_300315.json:
{
"body": "<div id=\"comment:17\"></div>\n\nNew commits:\n<table><tr><td><a href=\"https://github.com/sagemath/sagetrac-mirror/commit/eed116f97f2e467ccc445b24807588a802038e29\"><code>eed116f</code></a></td><td><code>Merge branch 'public/20736' into 7.3.b4</code></td></tr><tr><td><a href=\"https://github.com/sagemath/sagetrac-mirror/commit/228842b1aad446925645dcadb72f9c768358fe0d\"><code>228842b</code></a></td><td><code>update of patchbot to 2.6.0</code></td></tr></table>\n",
"created_at": "2016-06-22T18:38:57Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300315",
"user": "https://github.com/fchapoton"
}
New commits:
eed116f | Merge branch 'public/20736' into 7.3.b4 |
228842b | update of patchbot to 2.6.0 |
archive/issue_events_289389.json:
{
"actor": "https://github.com/fchapoton",
"created_at": "2016-06-22T18:38:57Z",
"event": "renamed",
"issue": "https://github.com/sagemath/sage/issues/20736",
"title_is": "upgrade patchbot to 2.6.0 (compatible with https trac)",
"title_was": "make patchbot a correct python package",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289389"
}
archive/issue_comments_300316.json:
{
"body": "Changed commit from **[`160332c`](https://github.com/sagemath/sagetrac-mirror/commit/160332c495872a01047d109a8584aca955d98252)** to **[`228842b`](https://github.com/sagemath/sagetrac-mirror/commit/228842b1aad446925645dcadb72f9c768358fe0d)**",
"created_at": "2016-06-22T18:38:57Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300316",
"user": "https://github.com/fchapoton"
}
Changed commit from 160332c
to 228842b
archive/issue_comments_300317.json:
{
"body": "Description changed:\n``````diff\n--- \n+++ \n@@ -2,7 +2,8 @@\n \n https://github.com/robertwb/sage-patchbot/issues/79\n \n-Let us correct that.\n+This ticket does not solve that, but just updates the patchbot,\n+in particular to cope with the new https trac\n \n-Tarball: \u200b\u200b\u200b\u200bhttp://www-irma.u-strasbg.fr/~chapoton/patchbot-2.5.9.tar.gz\n+Tarball: \u200b\u200b\u200b\u200bhttp://www-irma.u-strasbg.fr/~chapoton/patchbot-2.6.0.tar.gz\n \n``````\n",
"created_at": "2016-06-22T18:38:57Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300317",
"user": "https://github.com/fchapoton"
}
Description changed:
---
+++
@@ -2,7 +2,8 @@
https://github.com/robertwb/sage-patchbot/issues/79
-Let us correct that.
+This ticket does not solve that, but just updates the patchbot,
+in particular to cope with the new https trac
-Tarball: http://www-irma.u-strasbg.fr/~chapoton/patchbot-2.5.9.tar.gz
+Tarball: http://www-irma.u-strasbg.fr/~chapoton/patchbot-2.6.0.tar.gz
archive/issue_comments_300318.json:
{
"body": "<div id=\"comment:18\" align=\"right\">comment:18</div>\n\nThis is becoming urgent, to cope with https trac.\n\nI tried to go back to the previous way but this is not working. It complains about\n\n```\nsage -patchbot --ticket=20862\nTraceback (most recent call last):\n File \"sage/local/bin/patchbot/patchbot.py\", line 63, in <module>\n from .trac import scrape, pull_from_trac, TracServer, Config\nValueError: Attempted relative import in non-package\n```\nI need help. Should I just remove all relative imports newly introduced ?? should I keep the strange source dir of patchbot inside the local/bin directory ??",
"created_at": "2016-06-22T18:46:32Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300318",
"user": "https://github.com/fchapoton"
}
This is becoming urgent, to cope with https trac.
I tried to go back to the previous way but this is not working. It complains about
sage -patchbot --ticket=20862
Traceback (most recent call last):
File "sage/local/bin/patchbot/patchbot.py", line 63, in <module>
from .trac import scrape, pull_from_trac, TracServer, Config
ValueError: Attempted relative import in non-package
I need help. Should I just remove all relative imports newly introduced ?? should I keep the strange source dir of patchbot inside the local/bin directory ??
archive/issue_comments_300319.json:
{
"body": "Description changed:\n``````diff\n--- \n+++ \n@@ -5,5 +5,5 @@\n This ticket does not solve that, but just updates the patchbot,\n in particular to cope with the new https trac\n \n-Tarball: \u200b\u200b\u200b\u200bhttp://www-irma.u-strasbg.fr/~chapoton/patchbot-2.6.0.tar.gz\n+Tarball: \u200b\u200b\u200b\u200bhttp://www-irma.u-strasbg.fr/~chapoton/patchbot-2.6.0.tar.bz2\n \n``````\n",
"created_at": "2016-06-22T19:14:46Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300319",
"user": "https://github.com/fchapoton"
}
Description changed:
---
+++
@@ -5,5 +5,5 @@
This ticket does not solve that, but just updates the patchbot,
in particular to cope with the new https trac
-Tarball: http://www-irma.u-strasbg.fr/~chapoton/patchbot-2.6.0.tar.gz
+Tarball: http://www-irma.u-strasbg.fr/~chapoton/patchbot-2.6.0.tar.bz2
archive/issue_comments_300320.json:
{
"body": "Description changed:\n``````diff\n--- \n+++ \n@@ -2,8 +2,11 @@\n \n https://github.com/robertwb/sage-patchbot/issues/79\n \n-This ticket does not solve that, but just updates the patchbot,\n-in particular to cope with the new https trac\n+This also updates the patchbot, in particular to cope with the new https trac.\n \n-Tarball: \u200b\u200b\u200b\u200bhttp://www-irma.u-strasbg.fr/~chapoton/patchbot-2.6.0.tar.bz2\n+Broken Tarball for spkg: \u200b\u200b\u200b\u200bhttp://www-irma.u-strasbg.fr/~chapoton/patchbot-2.6.0.tar.bz2\n \n+New tarball for pip: sage-patchbot-2.6.0.tar.gz\n+\n+usage: sudo pip install sage-patchbot-2.6.0.tar.gz\n+\n``````\n",
"created_at": "2016-06-22T20:18:10Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300320",
"user": "https://github.com/fchapoton"
}
Description changed:
---
+++
@@ -2,8 +2,11 @@
https://github.com/robertwb/sage-patchbot/issues/79
-This ticket does not solve that, but just updates the patchbot,
-in particular to cope with the new https trac
+This also updates the patchbot, in particular to cope with the new https trac.
-Tarball: http://www-irma.u-strasbg.fr/~chapoton/patchbot-2.6.0.tar.bz2
+Broken Tarball for spkg: http://www-irma.u-strasbg.fr/~chapoton/patchbot-2.6.0.tar.bz2
+New tarball for pip: sage-patchbot-2.6.0.tar.gz
+
+usage: sudo pip install sage-patchbot-2.6.0.tar.gz
+
archive/issue_comments_300321.json:
{
"body": "<div id=\"comment:20\" align=\"right\">comment:20</div>\n\nI am trying a new way:\n\n* have a tarball made with \"python ./setup.py sdist\"\n* install it using \"pip install sage-patchbot-tarball\"\n* run it as usual by \"sage -patchbot\" (with sage on the current branch)",
"created_at": "2016-06-22T20:18:10Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300321",
"user": "https://github.com/fchapoton"
}
I am trying a new way:
- have a tarball made with "python ./setup.py sdist"
- install it using "pip install sage-patchbot-tarball"
- run it as usual by "sage -patchbot" (with sage on the current branch)
archive/issue_comments_300322.json:
{
"body": "Description changed:\n``````diff\n--- \n+++ \n@@ -6,7 +6,7 @@\n \n Broken Tarball for spkg: \u200b\u200b\u200b\u200bhttp://www-irma.u-strasbg.fr/~chapoton/patchbot-2.6.0.tar.bz2\n \n-New tarball for pip: sage-patchbot-2.6.0.tar.gz\n+New tarball for pip: http://www-irma.u-strasbg.fr/~chapoton/sage-patchbot-2.6.0.tar.gz\n \n usage: sudo pip install sage-patchbot-2.6.0.tar.gz\n \n``````\n",
"created_at": "2016-06-22T20:18:50Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300322",
"user": "https://github.com/fchapoton"
}
Description changed:
---
+++
@@ -6,7 +6,7 @@
Broken Tarball for spkg: http://www-irma.u-strasbg.fr/~chapoton/patchbot-2.6.0.tar.bz2
-New tarball for pip: sage-patchbot-2.6.0.tar.gz
+New tarball for pip: http://www-irma.u-strasbg.fr/~chapoton/sage-patchbot-2.6.0.tar.gz
usage: sudo pip install sage-patchbot-2.6.0.tar.gz
archive/issue_comments_300323.json:
{
"body": "Changed commit from **[`228842b`](https://github.com/sagemath/sagetrac-mirror/commit/228842b1aad446925645dcadb72f9c768358fe0d)** to **[`91fc492`](https://github.com/sagemath/sagetrac-mirror/commit/91fc4927d5115ffd9af20b45b9cc5d05a4b284ef)**",
"created_at": "2016-06-23T08:09:55Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300323",
"user": "https://github.com/jdemeyer"
}
Changed commit from 228842b
to 91fc492
archive/issue_comments_300324.json:
{
"body": "Replying to [@fchapoton](https://github.com/sagemath/sage/issues/20736#comment:0):\n> usage: sudo pip install sage-patchbot-2.6.0.tar.gz\n\n**VERY** -1 to this!\n\n---\nNew commits:\n<table><tr><td><a href=\"https://github.com/sagemath/sagetrac-mirror/commit/91fc4927d5115ffd9af20b45b9cc5d05a4b284ef\"><code>91fc492</code></a></td><td><code>patchbot : trying to use pip install</code></td></tr></table>\n",
"created_at": "2016-06-23T08:09:55Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300324",
"user": "https://github.com/jdemeyer"
}
Replying to @fchapoton:
usage: sudo pip install sage-patchbot-2.6.0.tar.gz
VERY -1 to this!
New commits:
91fc492 | patchbot : trying to use pip install |
archive/issue_comments_300325.json:
{
"body": "<div id=\"comment:23\" align=\"right\">comment:23</div>\n\nCan you please update the patchbot to *only* deal with the https thing and not change the whole install procedure?",
"created_at": "2016-06-23T08:12:31Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300325",
"user": "https://github.com/jdemeyer"
}
Can you please update the patchbot to only deal with the https thing and not change the whole install procedure?
archive/issue_comments_300326.json:
{
"body": "<div id=\"comment:24\" align=\"right\">comment:24</div>\n\nI am doing what I can, but this is a bit too complicated for me to handle easily.\n\nI changed all the imports in such a way that patchbot must now be used a a python package. The simpplest way (and it works) is what I suggest above.\n\nWhy do you so strongly object to pip installation in the system python ?\n\nNow if you have better suggestions, I would appreciate help.",
"created_at": "2016-06-23T08:15:06Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300326",
"user": "https://github.com/fchapoton"
}
I am doing what I can, but this is a bit too complicated for me to handle easily.
I changed all the imports in such a way that patchbot must now be used a a python package. The simpplest way (and it works) is what I suggest above.
Why do you so strongly object to pip installation in the system python ?
Now if you have better suggestions, I would appreciate help.
archive/issue_comments_300327.json:
{
"body": "<div id=\"comment:25\" align=\"right\">comment:25</div>\n\nReplying to [@fchapoton](#comment%3A24):\n> I changed all the imports in such a way that patchbot must now be used a a python package.\n\nI don't understand how the imports should be related to how the package is installed.\n\n> Why do you so strongly object to pip installation in the system python ?\n\nIt should be possible to install and run the patchbot without `sudo` access.",
"created_at": "2016-06-23T08:21:36Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300327",
"user": "https://github.com/jdemeyer"
}
Replying to @fchapoton:
I changed all the imports in such a way that patchbot must now be used a a python package.
I don't understand how the imports should be related to how the package is installed.
Why do you so strongly object to pip installation in the system python ?
It should be possible to install and run the patchbot without sudo
access.
archive/issue_comments_300328.json:
{
"body": "<div id=\"comment:26\" align=\"right\">comment:26</div>\n\nReplying to [@jdemeyer](#comment%3A25):\n> Replying to [@fchapoton](#comment%3A24):\n> > I changed all the imports in such a way that patchbot must now be used a a python package.\n> \n> \n> I don't understand how the imports should be related to how the package is installed.\n\nRelative imports are not allowed when a python file is run as a script, it seems.\n\n> > Why do you so strongly object to pip installation in the system python ?\n> \n> \n> It should be possible to install and run the patchbot without `sudo` access.\n\nHow is it compatible with not installing in the sage python ?\n\nIs there any way to install somewhere inside sage a python package using the system python ? Is there any existing case ? Where should it go ? How should I do that ?",
"created_at": "2016-06-23T08:24:08Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300328",
"user": "https://github.com/fchapoton"
}
Replying to @jdemeyer:
Replying to @fchapoton:
I changed all the imports in such a way that patchbot must now be used a a python package.
I don't understand how the imports should be related to how the package is installed.
Relative imports are not allowed when a python file is run as a script, it seems.
Why do you so strongly object to pip installation in the system python ?
It should be possible to install and run the patchbot without
sudo
access.
How is it compatible with not installing in the sage python ?
Is there any way to install somewhere inside sage a python package using the system python ? Is there any existing case ? Where should it go ? How should I do that ?
archive/issue_comments_300329.json:
{
"body": "Description changed:\n``````diff\n--- \n+++ \n@@ -4,9 +4,11 @@\n \n This also updates the patchbot, in particular to cope with the new https trac.\n \n-Broken Tarball for spkg: \u200b\u200b\u200b\u200bhttp://www-irma.u-strasbg.fr/~chapoton/patchbot-2.6.0.tar.bz2\n-\n New tarball for pip: http://www-irma.u-strasbg.fr/~chapoton/sage-patchbot-2.6.0.tar.gz\n \n-usage: sudo pip install sage-patchbot-2.6.0.tar.gz\n+usage:\n \n+```\n+pip install --user sage-patchbot-2.6.0.tar.gz\n+```\n+\n``````\n",
"created_at": "2016-06-23T08:29:57Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300329",
"user": "https://github.com/jdemeyer"
}
Description changed:
---
+++
@@ -4,9 +4,11 @@
This also updates the patchbot, in particular to cope with the new https trac.
-Broken Tarball for spkg: http://www-irma.u-strasbg.fr/~chapoton/patchbot-2.6.0.tar.bz2
-
New tarball for pip: http://www-irma.u-strasbg.fr/~chapoton/sage-patchbot-2.6.0.tar.gz
-usage: sudo pip install sage-patchbot-2.6.0.tar.gz
+usage:
+```
+pip install --user sage-patchbot-2.6.0.tar.gz
+```
+
archive/issue_events_289390.json:
{
"actor": "https://github.com/jdemeyer",
"created_at": "2016-06-23T08:29:57Z",
"event": "renamed",
"issue": "https://github.com/sagemath/sage/issues/20736",
"title_is": "Upgrade patchbot to 2.6.0 to system package",
"title_was": "upgrade patchbot to 2.6.0 (compatible with https trac)",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289390"
}
archive/issue_comments_300330.json:
{
"body": "<div id=\"comment:28\" align=\"right\">comment:28</div>\n\nReplying to [@fchapoton](#comment%3A26):\n> Relative imports are not allowed when a python file is run as a script, it seems.\n\nThat can't be true. You're probably just missing some `__init__.py` file.",
"created_at": "2016-06-23T08:30:56Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300330",
"user": "https://github.com/jdemeyer"
}
Replying to @fchapoton:
Relative imports are not allowed when a python file is run as a script, it seems.
That can't be true. You're probably just missing some __init__.py
file.
archive/issue_comments_300331.json:
{
"body": "<div id=\"comment:29\" align=\"right\">comment:29</div>\n\nReplying to [@fchapoton](#comment%3A26):\n> Is there any way to install somewhere inside sage a python package using the system python ? Is there any existing case ? Where should it go ? How should I do that ?\n\nUser installs are possible with `pip`, I edited the ticket description.",
"created_at": "2016-06-23T08:31:31Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300331",
"user": "https://github.com/jdemeyer"
}
Replying to @fchapoton:
Is there any way to install somewhere inside sage a python package using the system python ? Is there any existing case ? Where should it go ? How should I do that ?
User installs are possible with pip
, I edited the ticket description.
archive/issue_comments_300332.json:
{
"body": "<div id=\"comment:30\" align=\"right\">comment:30</div>\n\nIf the `patchbot` becomes a system package, it should be removed as Sage package. I.e. remove the whole directory `build/pkgs/patchbot`.",
"created_at": "2016-06-23T08:32:29Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300332",
"user": "https://github.com/jdemeyer"
}
If the patchbot
becomes a system package, it should be removed as Sage package. I.e. remove the whole directory build/pkgs/patchbot
.
archive/issue_comments_300333.json:
{
"body": "<div id=\"comment:31\" align=\"right\">comment:31</div>\n\nReplying to [@jdemeyer](#comment%3A30):\n> If the `patchbot` becomes a system package, it should be removed as Sage package. I.e. remove the whole directory `build/pkgs/patchbot`.\n\nyes, I agree.\n\nAnd there is an `__init__.py` in the package.",
"created_at": "2016-06-23T08:34:50Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300333",
"user": "https://github.com/fchapoton"
}
Replying to @jdemeyer:
If the
patchbot
becomes a system package, it should be removed as Sage package. I.e. remove the whole directorybuild/pkgs/patchbot
.
yes, I agree.
And there is an __init__.py
in the package.
archive/issue_comments_300334.json:
{
"body": "Changed commit from **[`91fc492`](https://github.com/sagemath/sagetrac-mirror/commit/91fc4927d5115ffd9af20b45b9cc5d05a4b284ef)** to **[`897f5b2`](https://github.com/sagemath/sagetrac-mirror/commit/897f5b220c7eb63a994853092be376d62e40fe39)**",
"created_at": "2016-06-23T08:42:02Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300334",
"user": "https://github.com/jdemeyer"
}
Changed commit from 91fc492
to 897f5b2
archive/issue_comments_300335.json:
{
"body": "<div id=\"comment:32\" align=\"right\">comment:32</div>\n\nReplying to [@fchapoton](#comment%3A31):\n> And there is an `__init__.py` in the package.\n\nWell, if you want me to have a look, I need more details. But if it works as intended outside of Sage with a *user* install, that's good enough.\n\n---\nNew commits:\n<table><tr><td><a href=\"https://github.com/sagemath/sagetrac-mirror/commit/897f5b220c7eb63a994853092be376d62e40fe39\"><code>897f5b2</code></a></td><td><code>trac 20736 removing patchbot as spkg</code></td></tr></table>\n",
"created_at": "2016-06-23T08:42:02Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300335",
"user": "https://github.com/jdemeyer"
}
Replying to @fchapoton:
And there is an
__init__.py
in the package.
Well, if you want me to have a look, I need more details. But if it works as intended outside of Sage with a user install, that's good enough.
New commits:
897f5b2 | trac 20736 removing patchbot as spkg |
archive/issue_comments_300336.json:
{
"body": "<div id=\"comment:33\" align=\"right\">comment:33</div>\n\nIf installed correctly there should be no `sage/local/bin/patchbot/patchbot.py`\n\nThe exception you're getting should be unrelated to the new patchbot package. Instead, the old patchbot spkg should *not* be getting installed.",
"created_at": "2016-06-23T09:34:51Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300336",
"user": "https://github.com/embray"
}
If installed correctly there should be no sage/local/bin/patchbot/patchbot.py
The exception you're getting should be unrelated to the new patchbot package. Instead, the old patchbot spkg should not be getting installed.
archive/issue_comments_300337.json:
{
"body": "<div id=\"comment:34\" align=\"right\">comment:34</div>\n\nReplying to [@jdemeyer](#comment%3A22):\n> Replying to [@fchapoton](https://github.com/sagemath/sage/issues/20736#comment:0):\n> > usage: sudo pip install sage-patchbot-2.6.0.tar.gz\n> \n> \n> **VERY** -1 to this!\n> \n> ---\n> New commits:\n<table><tr><td><a href=\"https://github.com/sagemath/sagetrac-mirror/commit/91fc4927d5115ffd9af20b45b9cc5d05a4b284ef\"><code>91fc492</code></a></td><td><code>patchbot : trying to use pip install</code></td></tr></table>\n\n+1 to this. This is an appropriate way to install a Python package that is not (and will likely never be) provided by the system packager. Debian, at least, handles this well. \n\nThe alternative would be some kind of `~/.local` install--need to ensure that the `PYTHONPATH` is correct when running the patchbot though.\n\nAs I've written before on this thread the patchbot should not be a Sage spkg.",
"created_at": "2016-06-23T09:39:09Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300337",
"user": "https://github.com/embray"
}
Replying to @jdemeyer:
Replying to @fchapoton:
usage: sudo pip install sage-patchbot-2.6.0.tar.gz
VERY -1 to this!
New commits:
91fc492 | patchbot : trying to use pip install |
+1 to this. This is an appropriate way to install a Python package that is not (and will likely never be) provided by the system packager. Debian, at least, handles this well.
The alternative would be some kind of ~/.local
install--need to ensure that the PYTHONPATH
is correct when running the patchbot though.
As I've written before on this thread the patchbot should not be a Sage spkg.
archive/issue_events_289391.json:
{
"actor": "https://github.com/fchapoton",
"created_at": "2016-06-24T08:32:47Z",
"event": "labeled",
"issue": "https://github.com/sagemath/sage/issues/20736",
"label": "https://github.com/sagemath/sage/labels/needs%20review",
"label_color": "7fff00",
"label_name": "needs review",
"label_text_color": "ffffff",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289391"
}
archive/issue_comments_300338.json:
{
"body": "<div id=\"comment:35\" align=\"right\">comment:35</div>\n\nThe 2.6.0 seems to work well enough on sage4, although at some point it tested closed tickets.\n\nOn the other hand, arando is uselessly running 2.5.7.",
"created_at": "2016-06-24T08:32:47Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300338",
"user": "https://github.com/fchapoton"
}
The 2.6.0 seems to work well enough on sage4, although at some point it tested closed tickets.
On the other hand, arando is uselessly running 2.5.7.
archive/issue_comments_300339.json:
{
"body": "<div id=\"comment:36\" align=\"right\">comment:36</div>\n\nReplying to [@fchapoton](#comment%3A35):\n> On the other hand, arando is uselessly running 2.5.7.\n\nI know, I have not upgraded it yet.",
"created_at": "2016-06-24T08:55:19Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300339",
"user": "https://github.com/jdemeyer"
}
Replying to @fchapoton:
On the other hand, arando is uselessly running 2.5.7.
I know, I have not upgraded it yet.
archive/issue_events_289392.json:
{
"actor": "https://github.com/jdemeyer",
"created_at": "2016-06-24T08:58:50Z",
"event": "unlabeled",
"issue": "https://github.com/sagemath/sage/issues/20736",
"label": "https://github.com/sagemath/sage/labels/needs%20review",
"label_color": "7fff00",
"label_name": "needs review",
"label_text_color": "ffffff",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289392"
}
archive/issue_events_289393.json:
{
"actor": "https://github.com/jdemeyer",
"created_at": "2016-06-24T08:58:50Z",
"event": "labeled",
"issue": "https://github.com/sagemath/sage/issues/20736",
"label": "https://github.com/sagemath/sage/labels/needs%20work",
"label_color": "ffff00",
"label_name": "needs work",
"label_text_color": "ffffff",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289393"
}
archive/issue_comments_300340.json:
{
"body": "Reviewer: **Jeroen Demeyer**",
"created_at": "2016-06-24T08:58:50Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300340",
"user": "https://github.com/jdemeyer"
}
Reviewer: Jeroen Demeyer
archive/issue_comments_300341.json:
{
"body": "<div id=\"comment:37\" align=\"right\">comment:37</div>\n\nReplying to [@fchapoton](#comment%3A35):\n> The 2.6.0 seems to work well enough on sage4, although at some point it tested closed tickets.\n\nI did have to manually change `PYTHONPATH` for it to work in a user install.\n\nI think you should be able to do\n\n```\n./sage --pip install --user patchbot\n```\n\nHowever, since Sage sets `PYTHONUSERBASE` (no idea why), this patchbot is installed in a directory different from what the system Python expects. To get it working, I had to set `PYTHONPATH` to the `site-packages` directory in `~/.sage/local`. Because of confusion between `lib` and `lib64`, just setting `PYTHONUSERBASE` system-wide was not sufficient.\n\nFor this reason, I believe that `./sage --patchbot` should set `PYTHONPATH` before running the patchbot.",
"created_at": "2016-06-24T08:58:50Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300341",
"user": "https://github.com/jdemeyer"
}
Replying to @fchapoton:
The 2.6.0 seems to work well enough on sage4, although at some point it tested closed tickets.
I did have to manually change PYTHONPATH
for it to work in a user install.
I think you should be able to do
./sage --pip install --user patchbot
However, since Sage sets PYTHONUSERBASE
(no idea why), this patchbot is installed in a directory different from what the system Python expects. To get it working, I had to set PYTHONPATH
to the site-packages
directory in ~/.sage/local
. Because of confusion between lib
and lib64
, just setting PYTHONUSERBASE
system-wide was not sufficient.
For this reason, I believe that ./sage --patchbot
should set PYTHONPATH
before running the patchbot.
archive/issue_comments_300342.json:
{
"body": "<div id=\"comment:38\" align=\"right\">comment:38</div>\n\nI don't know why you would run `./sage --patchbot` at all, as opposed to just `patchbot`. Why is it necessary to use sage to run the thing that's supposed to be testing sage?",
"created_at": "2016-06-24T10:56:15Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300342",
"user": "https://github.com/embray"
}
I don't know why you would run ./sage --patchbot
at all, as opposed to just patchbot
. Why is it necessary to use sage to run the thing that's supposed to be testing sage?
archive/issue_comments_300343.json:
{
"body": "<div id=\"comment:39\" align=\"right\">comment:39</div>\n\nI propose the following \"modus operandi\" for patchbot installation and use\n\n* get the latest tarball.tar.gz from somewhere (git or my web page?)\n* pip install --user tarball.tar.gz\n* sage -patchbot --options (1) OR python -m sage_patchbot.patchbot --options (2)\n\nEverything (pip and python) should take place in the system python. I do not see any reason to involve the inner python of sage.\n\nI am not completely sure if this already works with this branch, but it was my vision of things. I think that I did (1) or (2) on my own patchbot client, which seems to be running well.\n\nNow Jeroen and Eric do not seem to agree on what to do. Please consider that I have not a lot of time to spend on that.",
"created_at": "2016-06-24T15:00:27Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300343",
"user": "https://github.com/fchapoton"
}
I propose the following "modus operandi" for patchbot installation and use
- get the latest tarball.tar.gz from somewhere (git or my web page?)
- pip install --user tarball.tar.gz
- sage -patchbot --options (1) OR python -m sage_patchbot.patchbot --options (2)
Everything (pip and python) should take place in the system python. I do not see any reason to involve the inner python of sage.
I am not completely sure if this already works with this branch, but it was my vision of things. I think that I did (1) or (2) on my own patchbot client, which seems to be running well.
Now Jeroen and Eric do not seem to agree on what to do. Please consider that I have not a lot of time to spend on that.
archive/issue_comments_300344.json:
{
"body": "<div id=\"comment:40\" align=\"right\">comment:40</div>\n\nmuch more annoying is the fact that the patchbot persist to test closed tickets..",
"created_at": "2016-06-24T15:03:10Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300344",
"user": "https://github.com/fchapoton"
}
much more annoying is the fact that the patchbot persist to test closed tickets..
archive/issue_comments_300345.json:
{
"body": "<div id=\"comment:41\" align=\"right\">comment:41</div>\n\n> get the latest tarball.tar.gz from somewhere (git or my web page?)\n\nWhy not put it on PyPI, and put out some proper releases?\n\nAlternatively, pip allows download direct from a git repository, including specific tags.\n\n> pip install --user tarball.tar.gz\n\nWhatever; people can `pip install` however they're comfortable with.\n\n> Everything (pip and python) should take place in the system python. I do not see any reason to involve the inner python of sage.\n\n+1\n\n> I am not completely sure if this already works with this branch, but it was my vision of things. I think that I did (1) or (2) on my own patchbot client, which seems to be running well.\n\nI don't know. Is there anything in the patchbot that tries to import `sage`? I don't *think* there is--as far as I can tell the patchbot can run entirely outside of the sage environment, as it should.\n\n> Now Jeroen and Eric do not seem to agree on what to do. Please consider that I have not a lot of time to spend on that.\n\nAFAICT the only disagreement is over whether or not running `sudo pip install` is ever appropriate. My point is that it doesn't really matter--as long as the user is installing into a Python outside of the Sage distribution (be it a virtualenv, a `--user` install, or whatever) it should work fine. It doesn't really matter and the most appropriate choice depends on the context.\n\nFor that matter, there's probably nothing wrong with installing the patchbot package into Sage's Python too, again just using `pip install`. But in general it should not be necessary to.",
"created_at": "2016-06-27T08:26:33Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300345",
"user": "https://github.com/embray"
}
get the latest tarball.tar.gz from somewhere (git or my web page?)
Why not put it on PyPI, and put out some proper releases?
Alternatively, pip allows download direct from a git repository, including specific tags.
pip install --user tarball.tar.gz
Whatever; people can pip install
however they're comfortable with.
Everything (pip and python) should take place in the system python. I do not see any reason to involve the inner python of sage.
+1
I am not completely sure if this already works with this branch, but it was my vision of things. I think that I did (1) or (2) on my own patchbot client, which seems to be running well.
I don't know. Is there anything in the patchbot that tries to import sage
? I don't think there is--as far as I can tell the patchbot can run entirely outside of the sage environment, as it should.
Now Jeroen and Eric do not seem to agree on what to do. Please consider that I have not a lot of time to spend on that.
AFAICT the only disagreement is over whether or not running sudo pip install
is ever appropriate. My point is that it doesn't really matter--as long as the user is installing into a Python outside of the Sage distribution (be it a virtualenv, a --user
install, or whatever) it should work fine. It doesn't really matter and the most appropriate choice depends on the context.
For that matter, there's probably nothing wrong with installing the patchbot package into Sage's Python too, again just using pip install
. But in general it should not be necessary to.
archive/issue_comments_300346.json:
{
"body": "<div id=\"comment:42\" align=\"right\">comment:42</div>\n\nLet us try to move forward, please. This works, as far as I can tell.\n\nThe patchbot package is installed in the system python, not as root but as user.\n\nI would prefer not to use pypi for the moment.\n\nCould somebody try to follow exactly the instructions and see if everything runs smoothly ?",
"created_at": "2016-07-12T14:46:00Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300346",
"user": "https://github.com/fchapoton"
}
Let us try to move forward, please. This works, as far as I can tell.
The patchbot package is installed in the system python, not as root but as user.
I would prefer not to use pypi for the moment.
Could somebody try to follow exactly the instructions and see if everything runs smoothly ?
archive/issue_events_289394.json:
{
"actor": "https://github.com/fchapoton",
"created_at": "2016-07-12T14:46:00Z",
"event": "unlabeled",
"issue": "https://github.com/sagemath/sage/issues/20736",
"label": "https://github.com/sagemath/sage/labels/needs%20work",
"label_color": "ffff00",
"label_name": "needs work",
"label_text_color": "ffffff",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289394"
}
archive/issue_events_289395.json:
{
"actor": "https://github.com/fchapoton",
"created_at": "2016-07-12T14:46:00Z",
"event": "labeled",
"issue": "https://github.com/sagemath/sage/issues/20736",
"label": "https://github.com/sagemath/sage/labels/needs%20review",
"label_color": "7fff00",
"label_name": "needs review",
"label_text_color": "ffffff",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289395"
}
archive/issue_events_289396.json:
{
"actor": "https://github.com/fchapoton",
"created_at": "2016-07-12T14:46:00Z",
"event": "renamed",
"issue": "https://github.com/sagemath/sage/issues/20736",
"title_is": "Upgrade patchbot to 2.6.1 as a system package",
"title_was": "Upgrade patchbot to 2.6.0 to system package",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289396"
}
archive/issue_comments_300347.json:
{
"body": "Description changed:\n``````diff\n--- \n+++ \n@@ -2,13 +2,19 @@\n \n https://github.com/robertwb/sage-patchbot/issues/79\n \n-This also updates the patchbot, in particular to cope with the new https trac.\n+This ticket also updates the patchbot, in particular to cope with the new https trac.\n \n-New tarball for pip: http://www-irma.u-strasbg.fr/~chapoton/sage-patchbot-2.6.0.tar.gz\n+New tarball for pip: http://www-irma.u-strasbg.fr/~chapoton/sage-patchbot-2.6.1.tar.gz\n \n-usage:\n+installation:\n \n ```\n pip install --user sage-patchbot-2.6.0.tar.gz\n ```\n \n+usage (starting on the branch here):\n+\n+```\n+./sage -patchbot\n+```\n+\n``````\n",
"created_at": "2016-07-12T14:46:00Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300347",
"user": "https://github.com/fchapoton"
}
Description changed:
---
+++
@@ -2,13 +2,19 @@
https://github.com/robertwb/sage-patchbot/issues/79
-This also updates the patchbot, in particular to cope with the new https trac.
+This ticket also updates the patchbot, in particular to cope with the new https trac.
-New tarball for pip: http://www-irma.u-strasbg.fr/~chapoton/sage-patchbot-2.6.0.tar.gz
+New tarball for pip: http://www-irma.u-strasbg.fr/~chapoton/sage-patchbot-2.6.1.tar.gz
-usage:
+installation:
```
pip install --user sage-patchbot-2.6.0.tar.gz
```
+usage (starting on the branch here):
+
+```
+./sage -patchbot
+```
+
archive/issue_comments_300348.json:
{
"body": "<div id=\"comment:43\" align=\"right\">comment:43</div>\n\nIf I install the package on my system Python, then doing\n\n```\n$ python -m sage_patchbot.patchbot\n```\nI got\n\n```\nTraceback (most recent call last):\n File \"/usr/lib/python2.7/runpy.py\", line 162, in _run_module_as_main\n \"__main__\", fname, loader, pkg_name)\n File \"/usr/lib/python2.7/runpy.py\", line 72, in _run_code\n exec code in run_globals\n File \"/nfs4/home6/vdelecro/.local/lib/python2.7/site-packages/sage_patchbot/patchbot.py\", line 1471, in <module>\n main(args)\n File \"/nfs4/home6/vdelecro/.local/lib/python2.7/site-packages/sage_patchbot/patchbot.py\", line 1401, in main\n patchbot = Patchbot(os.path.abspath(options.sage_root),\n File \"/usr/lib/python2.7/posixpath.py\", line 367, in abspath\n if not isabs(path):\n File \"/usr/lib/python2.7/posixpath.py\", line 61, in isabs\n return s.startswith('/')\nAttributeError: 'NoneType' object has no attribute 'startswith'\n```",
"created_at": "2016-07-12T22:16:15Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300348",
"user": "https://github.com/videlec"
}
If I install the package on my system Python, then doing
$ python -m sage_patchbot.patchbot
I got
Traceback (most recent call last):
File "/usr/lib/python2.7/runpy.py", line 162, in _run_module_as_main
"__main__", fname, loader, pkg_name)
File "/usr/lib/python2.7/runpy.py", line 72, in _run_code
exec code in run_globals
File "/nfs4/home6/vdelecro/.local/lib/python2.7/site-packages/sage_patchbot/patchbot.py", line 1471, in <module>
main(args)
File "/nfs4/home6/vdelecro/.local/lib/python2.7/site-packages/sage_patchbot/patchbot.py", line 1401, in main
patchbot = Patchbot(os.path.abspath(options.sage_root),
File "/usr/lib/python2.7/posixpath.py", line 367, in abspath
if not isabs(path):
File "/usr/lib/python2.7/posixpath.py", line 61, in isabs
return s.startswith('/')
AttributeError: 'NoneType' object has no attribute 'startswith'
archive/issue_comments_300349.json:
{
"body": "<div id=\"comment:44\" align=\"right\">comment:44</div>\n\nAll right, the command *must* specify the sage root as\n\n```\n$ python -m sage_patchbot.patchbot --sage-root=WHATEVER\n```\n(or setting `SAGE_ROOT` environment variable). I think there should be a more specific error message than what I got in [comment:43](#comment%3A43). Note that if the package is installed in Sage python then it will work out of the box since `SAGE_ROOT` is defined.\n\nI don't understand why you do not want to put it on PyPI. I can do it myself if you do not want to investigate. But it really takes less than a minute and greatly simplify the installation, version upgrade, online documentation, ...",
"created_at": "2016-07-12T22:22:55Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300349",
"user": "https://github.com/videlec"
}
All right, the command must specify the sage root as
$ python -m sage_patchbot.patchbot --sage-root=WHATEVER
(or setting SAGE_ROOT
environment variable). I think there should be a more specific error message than what I got in comment:43. Note that if the package is installed in Sage python then it will work out of the box since SAGE_ROOT
is defined.
I don't understand why you do not want to put it on PyPI. I can do it myself if you do not want to investigate. But it really takes less than a minute and greatly simplify the installation, version upgrade, online documentation, ...
archive/issue_comments_300350.json:
{
"body": "<div id=\"comment:45\" align=\"right\">comment:45</div>\n\n(I launch the package version of the patchbot on librae)",
"created_at": "2016-07-12T22:23:16Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300350",
"user": "https://github.com/videlec"
}
(I launch the package version of the patchbot on librae)
archive/issue_comments_300351.json:
{
"body": "<div id=\"comment:46\" align=\"right\">comment:46</div>\n\nI guess that the log should not go under `SAGE_ROOT/logs/patchbot` anymore now. The path should be part of the configuration file. What do you think?",
"created_at": "2016-07-12T22:26:11Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300351",
"user": "https://github.com/videlec"
}
I guess that the log should not go under SAGE_ROOT/logs/patchbot
anymore now. The path should be part of the configuration file. What do you think?
archive/issue_comments_300352.json:
{
"body": "<div id=\"comment:47\" align=\"right\">comment:47</div>\n\nWhat about switching to [logging Python module](https://docs.python.org/2/library/logging.html)?",
"created_at": "2016-07-12T22:29:51Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300352",
"user": "https://github.com/videlec"
}
What about switching to logging Python module?
archive/issue_comments_300353.json:
{
"body": "<div id=\"comment:48\" align=\"right\">comment:48</div>\n\nReplying to [@videlec](#comment%3A43):\n> If I install the package on my system Python, then doing\n> \n> ```\n> $ python -m sage_patchbot.patchbot\n> ```\n\nThanks for the feedback.\n\nThese are not the instructions on how to run the patchbot given in the ticket description.\n\nBut this will be another possible way.\n\nLibrae is meeting a whole bunch of doctest failure with lrslib. Maybe it should be checked that the bot is not causing them.\n\nConcerning pypi, it may be simpler once done, but I am not really ready to add yet another layer of complexity.\n\nConcerning the log, I would prefer to postpone the problem, once the new setting is in place.",
"created_at": "2016-07-13T06:45:18Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300353",
"user": "https://github.com/fchapoton"
}
Replying to @videlec:
If I install the package on my system Python, then doing
$ python -m sage_patchbot.patchbot
Thanks for the feedback.
These are not the instructions on how to run the patchbot given in the ticket description.
But this will be another possible way.
Librae is meeting a whole bunch of doctest failure with lrslib. Maybe it should be checked that the bot is not causing them.
Concerning pypi, it may be simpler once done, but I am not really ready to add yet another layer of complexity.
Concerning the log, I would prefer to postpone the problem, once the new setting is in place.
archive/issue_comments_300354.json:
{
"body": "<div id=\"comment:49\" align=\"right\">comment:49</div>\n\nReplying to [@videlec](#comment%3A44):\n> I don't understand why you do not want to put it on PyPI. I can do it myself if you do not want to investigate. But it really takes less than a minute and greatly simplify the installation, version upgrade, online documentation, ...\n\nI don't think it's deeply necessary--`pip` can install fine from a git repository for example (including specific tags). But it's not a big deal either way.\n\nAnd like I also wrote earlier it shouldn't matter where `pathbot` is installed either as long as it's installed correctly and running `patchbot` from the command-line works.",
"created_at": "2016-07-13T10:23:19Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300354",
"user": "https://github.com/embray"
}
Replying to @videlec:
I don't understand why you do not want to put it on PyPI. I can do it myself if you do not want to investigate. But it really takes less than a minute and greatly simplify the installation, version upgrade, online documentation, ...
I don't think it's deeply necessary--pip
can install fine from a git repository for example (including specific tags). But it's not a big deal either way.
And like I also wrote earlier it shouldn't matter where pathbot
is installed either as long as it's installed correctly and running patchbot
from the command-line works.
archive/issue_comments_300355.json:
{
"body": "<div id=\"comment:50\" align=\"right\">comment:50</div>\n\nReplying to [@fchapoton](#comment%3A48):\n> Concerning pypi, it may be simpler once done, but I am not really ready to add yet another layer of complexity.\n\nYou've already fixed up the patchbot to use distutils properly, so this is a matter of making an account on PyPI, setting up your credentials, and running `./setup.py register upload`. There's no additional layer of complexity.",
"created_at": "2016-07-13T10:25:28Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300355",
"user": "https://github.com/embray"
}
Replying to @fchapoton:
Concerning pypi, it may be simpler once done, but I am not really ready to add yet another layer of complexity.
You've already fixed up the patchbot to use distutils properly, so this is a matter of making an account on PyPI, setting up your credentials, and running ./setup.py register upload
. There's no additional layer of complexity.
archive/issue_comments_300356.json:
{
"body": "<div id=\"comment:51\" align=\"right\">comment:51</div>\n\nReplying to [@fchapoton](#comment%3A48):\n> Replying to [@videlec](#comment%3A43):\n> > If I install the package on my system Python, then doing\n> > \n> > ```\n> > $ python -m sage_patchbot.patchbot\n> > ```\n> \n> \n> Thanks for the feedback.\n> \n> These are not the instructions on how to run the patchbot given in the ticket description.\n>\n>\n> But this will be another possible way.\n\nIndeed. If the patchbot is a Python package it is much more natural to use it without the Sage environment variables set.\n\n> Librae is meeting a whole bunch of doctest failure with lrslib. Maybe it should be checked that the bot is not causing them.\n\nNot the bot: #21012\n\n> Concerning pypi, it may be simpler once done, but I am not really ready to add yet another layer of complexity.\n\nIt is not a new layer. It is simply a way to distribute the module. Instead of having a tarball on your webpage there is a tarball on PyPI. It is just simpler because for installation you would just have to do `pip install sage-patchbot --user` (downloading would be automatic). And for upgrade `pip install sage-patchbot --upgrade`.\n \n> Concerning the log, I would prefer to postpone the problem, once the new setting is in place.\n\nOf course this was not intended to be for this ticket. Still it is a question.",
"created_at": "2016-07-13T13:51:44Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300356",
"user": "https://github.com/videlec"
}
Replying to @fchapoton:
Replying to @videlec:
If I install the package on my system Python, then doing
$ python -m sage_patchbot.patchbot
Thanks for the feedback.
These are not the instructions on how to run the patchbot given in the ticket description.
But this will be another possible way.
Indeed. If the patchbot is a Python package it is much more natural to use it without the Sage environment variables set.
Librae is meeting a whole bunch of doctest failure with lrslib. Maybe it should be checked that the bot is not causing them.
Not the bot: #21012
Concerning pypi, it may be simpler once done, but I am not really ready to add yet another layer of complexity.
It is not a new layer. It is simply a way to distribute the module. Instead of having a tarball on your webpage there is a tarball on PyPI. It is just simpler because for installation you would just have to do pip install sage-patchbot --user
(downloading would be automatic). And for upgrade pip install sage-patchbot --upgrade
.
Concerning the log, I would prefer to postpone the problem, once the new setting is in place.
Of course this was not intended to be for this ticket. Still it is a question.
archive/issue_comments_300357.json:
{
"body": "<div id=\"comment:52\" align=\"right\">comment:52</div>\n\nThe patchbot librae is hanging on #19092 since more than a day\n\nThe current date is `Fri Jul 22 21:25:47 CEST 2016`. The end of `logs/patchbot/patchbot.log`\n\n```\n[2016-07-21 23:38:46] Getting ticket list...\n[2016-07-21 23:39:22] testing found ticket #19092\n[2016-07-21 23:39:22] #19092: init phase\n```\nlast lines in `logs/install.log`\n\n```\n[dochtml] [tutorial ] copying extra files... done\n[dochtml] [tutorial ] dumping search index in German (code: de) ... done\n[dochtml] [tutorial ] dumping object inventory... done\n[dochtml] [tutorial ] build succeeded.\n[dochtml] Build finished. The built documents can be found in /home/vincent/sage_patchbot/local/share/doc/sage/html/de/tutorial\n```\n\nAfter a `Ctrl-C` I had the following traceback (appearing only in `logs/install.log`)\n\n```\n[dochtml] Process PoolWorker-173:\n[dochtml] Process PoolWorker-178:\n[dochtml] Process PoolWorker-174:\n[dochtml] Process PoolWorker-177:\n[dochtml] Traceback (most recent call last):\n[dochtml] Traceback (most recent call last):\n[dochtml] Traceback (most recent call last):\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py\", line 258, in _bootstrap\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py\", line 258, in _bootstrap\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py\", line 258, in _bootstrap\n[dochtml] Process PoolWorker-176:\n[dochtml] Process PoolWorker-175:\n[dochtml] Traceback (most recent call last):\n[dochtml] Traceback (most recent call last):\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py\", line 258, in _bootstrap\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py\", line 258, in _bootstrap\n[dochtml] Traceback (most recent call last):\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py\", line 258, in _bootstrap\n[dochtml] self.run()\n[dochtml] self.run()\n[dochtml] self.run()\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py\", line 114, in run\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py\", line 114, in run\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py\", line 114, in run\n[dochtml] self.run()\n[dochtml] self._target(*self._args, **self._kwargs)\n[dochtml] self._target(*self._args, **self._kwargs)\n[dochtml] self._target(*self._args, **self._kwargs)\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py\", line 114, in run\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/pool.py\", line 102, in worker\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/pool.py\", line 102, in worker\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/pool.py\", line 102, in worker\n[dochtml] self.run()\n[dochtml] self._target(*self._args, **self._kwargs)\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py\", line 114, in run\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/pool.py\", line 102, in worker\n[dochtml] self._target(*self._args, **self._kwargs)\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/pool.py\", line 102, in worker\n[dochtml] self.run()\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py\", line 114, in run\n[dochtml] self._target(*self._args, **self._kwargs)\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/pool.py\", line 102, in worker\n[dochtml] task = get()\n[dochtml] task = get()\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/queues.py\", line 376, in get\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/queues.py\", line 376, in get\n[dochtml] task = get()\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/queues.py\", line 376, in get\n[dochtml] task = get()\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/queues.py\", line 376, in get\n[dochtml] task = get()\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/queues.py\", line 376, in get\n[dochtml] task = get()\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/queues.py\", line 378, in get\n[dochtml] return recv()\n[dochtml] File \"src/cysignals/signals.pyx\", line 225, in cysignals.signals.python_check_interrupt (build/src/cysignals/signals.c:2429)\n[dochtml] racquire()\n[dochtml] racquire()\n[dochtml] File \"src/cysignals/signals.pyx\", line 225, in cysignals.signals.python_check_interrupt (build/src/cysignals/signals.c:2429)\n[dochtml] File \"src/cysignals/signals.pyx\", line 225, in cysignals.signals.python_check_interrupt (build/src/cysignals/signals.c:2429)\n[dochtml] racquire()\n[dochtml] racquire()\n[dochtml] File \"src/cysignals/signals.pyx\", line 225, in cysignals.signals.python_check_interrupt (build/src/cysignals/signals.c:2429)\n[dochtml] File \"src/cysignals/signals.pyx\", line 225, in cysignals.signals.python_check_interrupt (build/src/cysignals/signals.c:2429)\n[dochtml] racquire()\n[dochtml] File \"src/cysignals/signals.pyx\", line 225, in cysignals.signals.python_check_interrupt (build/src/cysignals/signals.c:2429)\n[dochtml] File \"src/cysignals/signals.pyx\", line 96, in cysignals.signals.sig_raise_exception (build/src/cysignals/signals.c:1116)\n[dochtml] File \"src/cysignals/signals.pyx\", line 96, in cysignals.signals.sig_raise_exception (build/src/cysignals/signals.c:1116)\n[dochtml] File \"src/cysignals/signals.pyx\", line 96, in cysignals.signals.sig_raise_exception (build/src/cysignals/signals.c:1116)\n[dochtml] File \"src/cysignals/signals.pyx\", line 96, in cysignals.signals.sig_raise_exception (build/src/cysignals/signals.c:1116)\n[dochtml] File \"src/cysignals/signals.pyx\", line 96, in cysignals.signals.sig_raise_exception (build/src/cysignals/signals.c:1116)\n[dochtml] KeyboardInterrupt\n[dochtml] KeyboardInterrupt\n[dochtml] File \"src/cysignals/signals.pyx\", line 96, in cysignals.signals.sig_raise_exception (build/src/cysignals/signals.c:1116)\n[dochtml] KeyboardInterrupt\n[dochtml] KeyboardInterrupt\n[dochtml] KeyboardInterrupt\n[dochtml] KeyboardInterrupt\n[dochtml] Error building the documentation.\n[dochtml] Traceback (most recent call last):\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/runpy.py\", line 162, in _run_module_as_main\n[dochtml] \"__main__\", fname, loader, pkg_name)\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/runpy.py\", line 72, in _run_code\n[dochtml] exec code in run_globals\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python2.7/site-packages/sage_setup/docbuild/__main__.py\", line 2, in <module>\n[dochtml] main()\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python2.7/site-packages/sage_setup/docbuild/__init__.py\", line 1630, in main\n[dochtml] builder()\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python2.7/site-packages/sage_setup/docbuild/__init__.py\", line 298, in _wrapper\n[dochtml] x.get(99999)\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/pool.py\", line 561, in get\n[dochtml] self.wait(timeout)\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/multiprocessing/pool.py\", line 556, in wait\n[dochtml] self._cond.wait(timeout)\n[dochtml] File \"/home/vincent/sage_patchbot/local/lib/python/threading.py\", line 359, in wait\n[dochtml] _sleep(delay)\n[dochtml] File \"src/cysignals/signals.pyx\", line 225, in cysignals.signals.python_check_interrupt (build/src/cysignals/signals.c:2429)\n[dochtml] File \"src/cysignals/signals.pyx\", line 96, in cysignals.signals.sig_raise_exception (build/src/cysignals/signals.c:1116)\n[dochtml] KeyboardInterrupt\nmake[1]: *** [doc-html] Error 1\n```",
"created_at": "2016-07-22T19:31:27Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300357",
"user": "https://github.com/videlec"
}
The patchbot librae is hanging on #19092 since more than a day
The current date is Fri Jul 22 21:25:47 CEST 2016
. The end of logs/patchbot/patchbot.log
[2016-07-21 23:38:46] Getting ticket list...
[2016-07-21 23:39:22] testing found ticket #19092
[2016-07-21 23:39:22] #19092: init phase
last lines in logs/install.log
[dochtml] [tutorial ] copying extra files... done
[dochtml] [tutorial ] dumping search index in German (code: de) ... done
[dochtml] [tutorial ] dumping object inventory... done
[dochtml] [tutorial ] build succeeded.
[dochtml] Build finished. The built documents can be found in /home/vincent/sage_patchbot/local/share/doc/sage/html/de/tutorial
After a Ctrl-C
I had the following traceback (appearing only in logs/install.log
)
[dochtml] Process PoolWorker-173:
[dochtml] Process PoolWorker-178:
[dochtml] Process PoolWorker-174:
[dochtml] Process PoolWorker-177:
[dochtml] Traceback (most recent call last):
[dochtml] Traceback (most recent call last):
[dochtml] Traceback (most recent call last):
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py", line 258, in _bootstrap
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py", line 258, in _bootstrap
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py", line 258, in _bootstrap
[dochtml] Process PoolWorker-176:
[dochtml] Process PoolWorker-175:
[dochtml] Traceback (most recent call last):
[dochtml] Traceback (most recent call last):
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py", line 258, in _bootstrap
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py", line 258, in _bootstrap
[dochtml] Traceback (most recent call last):
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py", line 258, in _bootstrap
[dochtml] self.run()
[dochtml] self.run()
[dochtml] self.run()
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py", line 114, in run
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py", line 114, in run
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py", line 114, in run
[dochtml] self.run()
[dochtml] self._target(*self._args, **self._kwargs)
[dochtml] self._target(*self._args, **self._kwargs)
[dochtml] self._target(*self._args, **self._kwargs)
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py", line 114, in run
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/pool.py", line 102, in worker
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/pool.py", line 102, in worker
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/pool.py", line 102, in worker
[dochtml] self.run()
[dochtml] self._target(*self._args, **self._kwargs)
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py", line 114, in run
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/pool.py", line 102, in worker
[dochtml] self._target(*self._args, **self._kwargs)
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/pool.py", line 102, in worker
[dochtml] self.run()
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/process.py", line 114, in run
[dochtml] self._target(*self._args, **self._kwargs)
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/pool.py", line 102, in worker
[dochtml] task = get()
[dochtml] task = get()
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/queues.py", line 376, in get
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/queues.py", line 376, in get
[dochtml] task = get()
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/queues.py", line 376, in get
[dochtml] task = get()
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/queues.py", line 376, in get
[dochtml] task = get()
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/queues.py", line 376, in get
[dochtml] task = get()
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/queues.py", line 378, in get
[dochtml] return recv()
[dochtml] File "src/cysignals/signals.pyx", line 225, in cysignals.signals.python_check_interrupt (build/src/cysignals/signals.c:2429)
[dochtml] racquire()
[dochtml] racquire()
[dochtml] File "src/cysignals/signals.pyx", line 225, in cysignals.signals.python_check_interrupt (build/src/cysignals/signals.c:2429)
[dochtml] File "src/cysignals/signals.pyx", line 225, in cysignals.signals.python_check_interrupt (build/src/cysignals/signals.c:2429)
[dochtml] racquire()
[dochtml] racquire()
[dochtml] File "src/cysignals/signals.pyx", line 225, in cysignals.signals.python_check_interrupt (build/src/cysignals/signals.c:2429)
[dochtml] File "src/cysignals/signals.pyx", line 225, in cysignals.signals.python_check_interrupt (build/src/cysignals/signals.c:2429)
[dochtml] racquire()
[dochtml] File "src/cysignals/signals.pyx", line 225, in cysignals.signals.python_check_interrupt (build/src/cysignals/signals.c:2429)
[dochtml] File "src/cysignals/signals.pyx", line 96, in cysignals.signals.sig_raise_exception (build/src/cysignals/signals.c:1116)
[dochtml] File "src/cysignals/signals.pyx", line 96, in cysignals.signals.sig_raise_exception (build/src/cysignals/signals.c:1116)
[dochtml] File "src/cysignals/signals.pyx", line 96, in cysignals.signals.sig_raise_exception (build/src/cysignals/signals.c:1116)
[dochtml] File "src/cysignals/signals.pyx", line 96, in cysignals.signals.sig_raise_exception (build/src/cysignals/signals.c:1116)
[dochtml] File "src/cysignals/signals.pyx", line 96, in cysignals.signals.sig_raise_exception (build/src/cysignals/signals.c:1116)
[dochtml] KeyboardInterrupt
[dochtml] KeyboardInterrupt
[dochtml] File "src/cysignals/signals.pyx", line 96, in cysignals.signals.sig_raise_exception (build/src/cysignals/signals.c:1116)
[dochtml] KeyboardInterrupt
[dochtml] KeyboardInterrupt
[dochtml] KeyboardInterrupt
[dochtml] KeyboardInterrupt
[dochtml] Error building the documentation.
[dochtml] Traceback (most recent call last):
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/runpy.py", line 162, in _run_module_as_main
[dochtml] "__main__", fname, loader, pkg_name)
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/runpy.py", line 72, in _run_code
[dochtml] exec code in run_globals
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python2.7/site-packages/sage_setup/docbuild/__main__.py", line 2, in <module>
[dochtml] main()
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python2.7/site-packages/sage_setup/docbuild/__init__.py", line 1630, in main
[dochtml] builder()
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python2.7/site-packages/sage_setup/docbuild/__init__.py", line 298, in _wrapper
[dochtml] x.get(99999)
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/pool.py", line 561, in get
[dochtml] self.wait(timeout)
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/multiprocessing/pool.py", line 556, in wait
[dochtml] self._cond.wait(timeout)
[dochtml] File "/home/vincent/sage_patchbot/local/lib/python/threading.py", line 359, in wait
[dochtml] _sleep(delay)
[dochtml] File "src/cysignals/signals.pyx", line 225, in cysignals.signals.python_check_interrupt (build/src/cysignals/signals.c:2429)
[dochtml] File "src/cysignals/signals.pyx", line 96, in cysignals.signals.sig_raise_exception (build/src/cysignals/signals.c:1116)
[dochtml] KeyboardInterrupt
make[1]: *** [doc-html] Error 1
archive/issue_comments_300358.json:
{
"body": "<div id=\"comment:53\" align=\"right\">comment:53</div>\n\nMaybe due to swap (I need to check)...",
"created_at": "2016-07-22T23:20:09Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300358",
"user": "https://github.com/videlec"
}
Maybe due to swap (I need to check)...
archive/issue_comments_300359.json:
{
"body": "<div id=\"comment:54\" align=\"right\">comment:54</div>\n\nReplying to [@embray](#comment%3A38):\n> I don't know why you would run `./sage --patchbot` at all\n\nMainly just for convenience. It's easier to type `./sage --patchbot` than `python -m sage_patchbot.patchbot --sage-root=SAGE_ROOT`.\n\nSecond reason: on most systems, `python` is installed system-wide. But `pip` often is not.",
"created_at": "2016-07-25T08:52:00Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300359",
"user": "https://github.com/jdemeyer"
}
Replying to @embray:
I don't know why you would run
./sage --patchbot
at all
Mainly just for convenience. It's easier to type ./sage --patchbot
than python -m sage_patchbot.patchbot --sage-root=SAGE_ROOT
.
Second reason: on most systems, python
is installed system-wide. But pip
often is not.
archive/issue_events_289397.json:
{
"actor": "https://github.com/jdemeyer",
"created_at": "2016-07-25T09:06:40Z",
"event": "unlabeled",
"issue": "https://github.com/sagemath/sage/issues/20736",
"label": "https://github.com/sagemath/sage/labels/needs%20review",
"label_color": "7fff00",
"label_name": "needs review",
"label_text_color": "ffffff",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289397"
}
archive/issue_events_289398.json:
{
"actor": "https://github.com/jdemeyer",
"created_at": "2016-07-25T09:06:40Z",
"event": "labeled",
"issue": "https://github.com/sagemath/sage/issues/20736",
"label": "https://github.com/sagemath/sage/labels/needs%20info",
"label_color": "ffff00",
"label_name": "needs info",
"label_text_color": "ffffff",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289398"
}
archive/issue_comments_300360.json:
{
"body": "<div id=\"comment:55\" align=\"right\">comment:55</div>\n\nPlease update the ticket description with the recommended way to install and run the patchbot. Then I will test it.",
"created_at": "2016-07-25T09:06:40Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300360",
"user": "https://github.com/jdemeyer"
}
Please update the ticket description with the recommended way to install and run the patchbot. Then I will test it.
archive/issue_comments_300361.json:
{
"body": "<div id=\"comment:56\" align=\"right\">comment:56</div>\n\nYou should also decide whether you want to support `./sage --patchbot` or not. If you do, it should actually work (see [comment:37](#comment%3A37)). If you don't, remove the command (or better: make it print some message of which commands should actually be run).",
"created_at": "2016-07-25T09:10:20Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300361",
"user": "https://github.com/jdemeyer"
}
You should also decide whether you want to support ./sage --patchbot
or not. If you do, it should actually work (see comment:37). If you don't, remove the command (or better: make it print some message of which commands should actually be run).
archive/issue_comments_300362.json:
{
"body": "<div id=\"comment:57\" align=\"right\">comment:57</div>\n\n@jdemeyer\n\nI have tried to enhance the description of the ticket.\n\nI am not willing to avoid the dependency to pip.",
"created_at": "2016-07-28T08:29:18Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300362",
"user": "https://github.com/fchapoton"
}
@jdemeyer
I have tried to enhance the description of the ticket.
I am not willing to avoid the dependency to pip.
archive/issue_comments_300363.json:
{
"body": "Description changed:\n``````diff\n--- \n+++ \n@@ -1,20 +1,34 @@\n The patchbot is currently installing in a very not kosher way, as explained in\n \n https://github.com/robertwb/sage-patchbot/issues/79\n+\n+This ticket will change completely the installation mode. The patchbot will become a python package, to be installed using pip in\n+the system python as explained below.\n \n This ticket also updates the patchbot, in particular to cope with the new https trac.\n \n New tarball for pip: http://www-irma.u-strasbg.fr/~chapoton/sage-patchbot-2.6.1.tar.gz\n \n-installation:\n+installation (two equivalent methods at your disposal):\n \n ```\n-pip install --user sage-patchbot-2.6.0.tar.gz\n+upload the tarball\n+pip install --user sage-patchbot-2.6.1.tar.gz\n+```\n+or (recommended)\n+\n+```\n+pip install --user git+https://github.com/sagemath/[email protected]\n ```\n \n-usage (starting on the branch here):\n+There are now two possible ways to launch the patchbot:\n \n ```\n+python -m sage_patchbot.patchbot --sage-root=XXXX\n+```\n+or\n+\n+```\n+pull the branch here and run `make` (until this ticket is merged)\n ./sage -patchbot\n ```\n-\n``````\n",
"created_at": "2016-07-28T08:29:18Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300363",
"user": "https://github.com/fchapoton"
}
Description changed:
---
+++
@@ -1,20 +1,34 @@
The patchbot is currently installing in a very not kosher way, as explained in
https://github.com/robertwb/sage-patchbot/issues/79
+
+This ticket will change completely the installation mode. The patchbot will become a python package, to be installed using pip in
+the system python as explained below.
This ticket also updates the patchbot, in particular to cope with the new https trac.
New tarball for pip: http://www-irma.u-strasbg.fr/~chapoton/sage-patchbot-2.6.1.tar.gz
-installation:
+installation (two equivalent methods at your disposal):
```
-pip install --user sage-patchbot-2.6.0.tar.gz
+upload the tarball
+pip install --user sage-patchbot-2.6.1.tar.gz
+```
+or (recommended)
+
+```
+pip install --user git+https://github.com/sagemath/[email protected]
```
-usage (starting on the branch here):
+There are now two possible ways to launch the patchbot:
```
+python -m sage_patchbot.patchbot --sage-root=XXXX
+```
+or
+
+```
+pull the branch here and run `make` (until this ticket is merged)
./sage -patchbot
```
-
archive/issue_comments_300364.json:
{
"body": "<div id=\"comment:58\" align=\"right\">comment:58</div>\n\nLike I said before, `./sage --patchbot` does not actually work.",
"created_at": "2016-07-29T11:13:22Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300364",
"user": "https://github.com/jdemeyer"
}
Like I said before, ./sage --patchbot
does not actually work.
archive/issue_comments_300365.json:
{
"body": "Description changed:\n``````diff\n--- \n+++ \n@@ -7,28 +7,30 @@\n \n This ticket also updates the patchbot, in particular to cope with the new https trac.\n \n-New tarball for pip: http://www-irma.u-strasbg.fr/~chapoton/sage-patchbot-2.6.1.tar.gz\n+New tarball for patchbot: http://www-irma.u-strasbg.fr/~chapoton/sage-patchbot-2.6.1.tar.gz\n \n installation (two equivalent methods at your disposal):\n+1. Download the tarball and then\n \n ```\n-upload the tarball\n pip install --user sage-patchbot-2.6.1.tar.gz\n ```\n-or (recommended)\n+\n+2. (recommended)\n \n ```\n pip install --user git+https://github.com/sagemath/[email protected]\n ```\n \n There are now two possible ways to launch the patchbot:\n+1.\n \n ```\n python -m sage_patchbot.patchbot --sage-root=XXXX\n ```\n or\n+2. Checkout the branch here and run\n \n ```\n-pull the branch here and run `make` (until this ticket is merged)\n-./sage -patchbot\n+make && ./sage --patchbot\n ```\n``````\n",
"created_at": "2016-07-29T11:13:22Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300365",
"user": "https://github.com/jdemeyer"
}
Description changed:
---
+++
@@ -7,28 +7,30 @@
This ticket also updates the patchbot, in particular to cope with the new https trac.
-New tarball for pip: http://www-irma.u-strasbg.fr/~chapoton/sage-patchbot-2.6.1.tar.gz
+New tarball for patchbot: http://www-irma.u-strasbg.fr/~chapoton/sage-patchbot-2.6.1.tar.gz
installation (two equivalent methods at your disposal):
+1. Download the tarball and then
```
-upload the tarball
pip install --user sage-patchbot-2.6.1.tar.gz
```
-or (recommended)
+
+2. (recommended)
```
pip install --user git+https://github.com/sagemath/[email protected]
```
There are now two possible ways to launch the patchbot:
+1.
```
python -m sage_patchbot.patchbot --sage-root=XXXX
```
or
+2. Checkout the branch here and run
```
-pull the branch here and run `make` (until this ticket is merged)
-./sage -patchbot
+make && ./sage --patchbot
```
archive/issue_comments_300366.json:
{
"body": "<div id=\"comment:59\" align=\"right\">comment:59</div>\n\nSee the last line of [comment:37]",
"created_at": "2016-07-29T11:15:28Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300366",
"user": "https://github.com/jdemeyer"
}
See the last line of [comment:37]
archive/issue_comments_300367.json:
{
"body": "<div id=\"comment:60\" align=\"right\">comment:60</div>\n\nIndeed the command `./sage --patchbot` has no clue where the patchbot is actually installed. It might be on system python, in Sage python, in a virtualenv, or something else... I think it is much safer to disable it completely with an error message (as proposed in [comment:56](#comment%3A56)).",
"created_at": "2016-07-31T19:22:12Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300367",
"user": "https://github.com/videlec"
}
Indeed the command ./sage --patchbot
has no clue where the patchbot is actually installed. It might be on system python, in Sage python, in a virtualenv, or something else... I think it is much safer to disable it completely with an error message (as proposed in comment:56).
archive/issue_comments_300368.json:
{
"body": "<div id=\"comment:61\" align=\"right\">comment:61</div>\n\nI think it makes sense to assume that, if you run `./sage --patchbot`, you want to run the patchbot which is installed from a Sage shell. That would just require some part of `sage-env` to be executed before running the patchbot.",
"created_at": "2016-08-01T06:35:31Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300368",
"user": "https://github.com/jdemeyer"
}
I think it makes sense to assume that, if you run ./sage --patchbot
, you want to run the patchbot which is installed from a Sage shell. That would just require some part of sage-env
to be executed before running the patchbot.
archive/issue_comments_300369.json:
{
"body": "Changed commit from **[`897f5b2`](https://github.com/sagemath/sagetrac-mirror/commit/897f5b220c7eb63a994853092be376d62e40fe39)** to **[`14fb599`](https://github.com/sagemath/sagetrac-mirror/commit/14fb5998257fe1f7bc352d92c0de9571ae85a11d)**",
"created_at": "2016-08-02T13:05:25Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300369",
"user": "https://github.com/sagetrac-git"
}
Changed commit from 897f5b2
to 14fb599
archive/issue_comments_300370.json:
{
"body": "<div id=\"comment:62\"></div>\n\nBranch pushed to git repo; I updated commit sha1. New commits:\n<table><tr><td><a href=\"https://github.com/sagemath/sagetrac-mirror/commit/80a58ba864a96bfe64a9102ae2f935581ffab1bd\"><code>80a58ba</code></a></td><td><code>Merge branch 'public/20736' in 7.3.rc0</code></td></tr><tr><td><a href=\"https://github.com/sagemath/sagetrac-mirror/commit/14fb5998257fe1f7bc352d92c0de9571ae85a11d\"><code>14fb599</code></a></td><td><code>trac 20736 more information when calling sage --patchbot</code></td></tr></table>\n",
"created_at": "2016-08-02T13:05:25Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300370",
"user": "https://github.com/sagetrac-git"
}
Branch pushed to git repo; I updated commit sha1. New commits:
80a58ba | Merge branch 'public/20736' in 7.3.rc0 |
14fb599 | trac 20736 more information when calling sage --patchbot |
archive/issue_comments_300371.json:
{
"body": "<div id=\"comment:63\" align=\"right\">comment:63</div>\n\nIt seems that we are not making progress here. Jeroen, why do you insist so strongly on installing the patchbot package inside the sage shell ? It makes not much sense to me. I think one can reasonably expect anybody wanting to run a patchbot to be able to install pip first. There is no need to call \"sage -pip\".\n\nI propose the following alternative:\n\n(1) disable completely \"sage --patchbot\"\n\n(2) use \"sage --patchbot\" as a convenient way to launch the patchbot installed in the system python.\n\nI vote for (2), which is the current situation with the branch. I have just added some explanation text to appear when \"sage --patchbot\" is called.",
"created_at": "2016-08-02T13:13:46Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300371",
"user": "https://github.com/fchapoton"
}
It seems that we are not making progress here. Jeroen, why do you insist so strongly on installing the patchbot package inside the sage shell ? It makes not much sense to me. I think one can reasonably expect anybody wanting to run a patchbot to be able to install pip first. There is no need to call "sage -pip".
I propose the following alternative:
(1) disable completely "sage --patchbot"
(2) use "sage --patchbot" as a convenient way to launch the patchbot installed in the system python.
I vote for (2), which is the current situation with the branch. I have just added some explanation text to appear when "sage --patchbot" is called.
archive/issue_comments_300372.json:
{
"body": "<div id=\"comment:64\" align=\"right\">comment:64</div>\n\n> Jeroen, why do you insist so strongly on installing the patchbot package inside the sage shell ?\n\nI am not really insisting on that. I am insisting on not requiring a system install. But I can live with a user install outside of Sage.",
"created_at": "2016-08-02T13:28:43Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300372",
"user": "https://github.com/jdemeyer"
}
Jeroen, why do you insist so strongly on installing the patchbot package inside the sage shell ?
I am not really insisting on that. I am insisting on not requiring a system install. But I can live with a user install outside of Sage.
archive/issue_comments_300373.json:
{
"body": "<div id=\"comment:65\" align=\"right\">comment:65</div>\n\nTo turn the question around: why do you keep insisting on not supporting a patchbot installed within Sage?",
"created_at": "2016-08-02T13:34:33Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300373",
"user": "https://github.com/jdemeyer"
}
To turn the question around: why do you keep insisting on not supporting a patchbot installed within Sage?
archive/issue_comments_300374.json:
{
"body": "<div id=\"comment:66\" align=\"right\">comment:66</div>\n\nI do not insist on that either, I just do not know how to do it,\nand I am happy enough with the current state of things. I would prefer to stop spending\nmy time on this ticket, and to get more momentum towards python 3 instead.",
"created_at": "2016-08-02T13:39:35Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300374",
"user": "https://github.com/fchapoton"
}
I do not insist on that either, I just do not know how to do it, and I am happy enough with the current state of things. I would prefer to stop spending my time on this ticket, and to get more momentum towards python 3 instead.
archive/issue_comments_300375.json:
{
"body": "<div id=\"comment:67\" align=\"right\">comment:67</div>\n\nIt is supported with\n\n```\n$ sage -pip install my_patchbot_tarball.tar.gz\n$ sage -python -m sage_patchbot.patchbot --sage-root=my_sage_root\n```\nIt is just the command `sage -patchbot` which is half broken. So I would vote for (1) that is much cleaner and solve the issue.\n\nEDIT: in my initial post I wrote (2) instead of (1)... sorry",
"created_at": "2016-08-02T13:40:49Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300375",
"user": "https://github.com/videlec"
}
It is supported with
$ sage -pip install my_patchbot_tarball.tar.gz
$ sage -python -m sage_patchbot.patchbot --sage-root=my_sage_root
It is just the command sage -patchbot
which is half broken. So I would vote for (1) that is much cleaner and solve the issue.
EDIT: in my initial post I wrote (2) instead of (1)... sorry
archive/issue_comments_300376.json:
{
"body": "<div id=\"comment:68\" align=\"right\">comment:68</div>\n\nJeroen, do you also prefer (1) ?",
"created_at": "2016-08-02T13:47:57Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300376",
"user": "https://github.com/fchapoton"
}
Jeroen, do you also prefer (1) ?
archive/issue_comments_300377.json:
{
"body": "<div id=\"comment:69\" align=\"right\">comment:69</div>\n\nReplying to [@videlec](#comment%3A67):\n> It is supported with\n> \n> ```\n> $ sage -pip install my_patchbot_tarball.tar.gz\n> $ sage -python -m sage_patchbot.patchbot --sage-root=my_sage_root\n> ```\n\nNo, that does not really work since you cannot use the Sage Python to run the patchbot.",
"created_at": "2016-08-02T13:54:28Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300377",
"user": "https://github.com/jdemeyer"
}
Replying to @videlec:
It is supported with
$ sage -pip install my_patchbot_tarball.tar.gz $ sage -python -m sage_patchbot.patchbot --sage-root=my_sage_root
No, that does not really work since you cannot use the Sage Python to run the patchbot.
archive/issue_comments_300378.json:
{
"body": "<div id=\"comment:70\" align=\"right\">comment:70</div>\n\nReplying to [@jdemeyer](#comment%3A69):\n> Replying to [@videlec](#comment%3A67):\n> > It is supported with\n> > \n> > ```\n> > $ sage -pip install my_patchbot_tarball.tar.gz\n> > $ sage -python -m sage_patchbot.patchbot --sage-root=my_sage_root\n> > ```\n> \n> \n> No, that does not really work since you cannot use the Sage Python to run the patchbot.\n\nWhy is that? It (seems to) work for me.",
"created_at": "2016-08-02T13:59:26Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300378",
"user": "https://github.com/videlec"
}
Replying to @jdemeyer:
Replying to @videlec:
It is supported with
$ sage -pip install my_patchbot_tarball.tar.gz $ sage -python -m sage_patchbot.patchbot --sage-root=my_sage_root
No, that does not really work since you cannot use the Sage Python to run the patchbot.
Why is that? It (seems to) work for me.
archive/issue_comments_300379.json:
{
"body": "<div id=\"comment:71\" align=\"right\">comment:71</div>\n\nReplying to [@fchapoton](#comment%3A66):\n> I do not insist on that either, I just do not know how to do it\n\nGood. I will look into this then.",
"created_at": "2016-08-02T14:01:39Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300379",
"user": "https://github.com/jdemeyer"
}
Replying to @fchapoton:
I do not insist on that either, I just do not know how to do it
Good. I will look into this then.
archive/issue_comments_300380.json:
{
"body": "<div id=\"comment:72\"></div>\n\nBranch pushed to git repo; I updated commit sha1. This was a forced push. New commits:\n<table><tr><td><a href=\"https://github.com/sagemath/sagetrac-mirror/commit/654e5788ec2e32f06ef246f09ade24dc6ddf20bb\"><code>654e578</code></a></td><td><code>Improve --patchbot command</code></td></tr></table>\n",
"created_at": "2016-08-02T15:03:53Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300380",
"user": "https://github.com/sagetrac-git"
}
Branch pushed to git repo; I updated commit sha1. This was a forced push. New commits:
654e578 | Improve --patchbot command |
archive/issue_comments_300381.json:
{
"body": "Changed commit from **[`14fb599`](https://github.com/sagemath/sagetrac-mirror/commit/14fb5998257fe1f7bc352d92c0de9571ae85a11d)** to **[`654e578`](https://github.com/sagemath/sagetrac-mirror/commit/654e5788ec2e32f06ef246f09ade24dc6ddf20bb)**",
"created_at": "2016-08-02T15:03:53Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300381",
"user": "https://github.com/sagetrac-git"
}
Changed commit from 14fb599
to 654e578
archive/issue_comments_300382.json:
{
"body": "<div id=\"comment:73\" align=\"right\">comment:73</div>\n\nFirst attempt to support a patchbot installed within Sage, not fully tested yet.",
"created_at": "2016-08-02T15:04:52Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300382",
"user": "https://github.com/jdemeyer"
}
First attempt to support a patchbot installed within Sage, not fully tested yet.
archive/issue_comments_300383.json:
{
"body": "<div id=\"comment:74\" align=\"right\">comment:74</div>\n\nReplying to [@videlec](#comment%3A70):\n> Why is that? It (seems to) work for me.\n\n...until the patchbot tests a ticket which re-installs the Python within Sage.",
"created_at": "2016-08-02T15:26:23Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300383",
"user": "https://github.com/jdemeyer"
}
Replying to @videlec:
Why is that? It (seems to) work for me.
...until the patchbot tests a ticket which re-installs the Python within Sage.
archive/issue_comments_300384.json:
{
"body": "<div id=\"comment:75\" align=\"right\">comment:75</div>\n\nHang on, I have a better solution in mind to run the patchbot.",
"created_at": "2016-08-02T16:07:37Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300384",
"user": "https://github.com/jdemeyer"
}
Hang on, I have a better solution in mind to run the patchbot.
archive/issue_comments_300385.json:
{
"body": "<div id=\"comment:76\"></div>\n\nBranch pushed to git repo; I updated commit sha1. This was a forced push. New commits:\n<table><tr><td><a href=\"https://github.com/sagemath/sagetrac-mirror/commit/6d7990e88c9f194f28309bd0015c40ecc09080bf\"><code>6d7990e</code></a></td><td><code>Improve --patchbot command</code></td></tr></table>\n",
"created_at": "2016-08-02T21:18:37Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300385",
"user": "https://github.com/sagetrac-git"
}
Branch pushed to git repo; I updated commit sha1. This was a forced push. New commits:
6d7990e | Improve --patchbot command |
archive/issue_comments_300386.json:
{
"body": "Changed commit from **[`654e578`](https://github.com/sagemath/sagetrac-mirror/commit/654e5788ec2e32f06ef246f09ade24dc6ddf20bb)** to **[`6d7990e`](https://github.com/sagemath/sagetrac-mirror/commit/6d7990e88c9f194f28309bd0015c40ecc09080bf)**",
"created_at": "2016-08-02T21:18:37Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300386",
"user": "https://github.com/sagetrac-git"
}
Changed commit from 654e578
to 6d7990e
archive/issue_comments_300387.json:
{
"body": "<div id=\"comment:77\" align=\"right\">comment:77</div>\n\nCurrently testing...",
"created_at": "2016-08-02T21:39:59Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300387",
"user": "https://github.com/jdemeyer"
}
Currently testing...
archive/issue_events_289399.json:
{
"actor": "https://github.com/jdemeyer",
"created_at": "2016-08-02T21:39:59Z",
"event": "unlabeled",
"issue": "https://github.com/sagemath/sage/issues/20736",
"label": "https://github.com/sagemath/sage/labels/needs%20info",
"label_color": "ffff00",
"label_name": "needs info",
"label_text_color": "ffffff",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289399"
}
archive/issue_events_289400.json:
{
"actor": "https://github.com/jdemeyer",
"created_at": "2016-08-02T21:39:59Z",
"event": "labeled",
"issue": "https://github.com/sagemath/sage/issues/20736",
"label": "https://github.com/sagemath/sage/labels/needs%20review",
"label_color": "7fff00",
"label_name": "needs review",
"label_text_color": "ffffff",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289400"
}
archive/issue_comments_300388.json:
{
"body": "<div id=\"comment:78\" align=\"right\">comment:78</div>\n\nThis is good for me. Can anybody review my commit?",
"created_at": "2016-08-03T10:45:27Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300388",
"user": "https://github.com/jdemeyer"
}
This is good for me. Can anybody review my commit?
archive/issue_comments_300389.json:
{
"body": "<div id=\"comment:79\" align=\"right\">comment:79</div>\n\ndoes not work for me:\n\n```\n0) pull the branch here and make\n1) pip install --user git+https://github.com/sagemath/[email protected]\n2) ./sage --patchbot\nTraceback (most recent call last):\n File \"<string>\", line 1, in <module>\nImportError: No module named sage_patchbot\nError: cannot find installation path for sage_patchbot\nSee https://wiki.sagemath.org/buildbot/details for instructions\n```\nbut \"import sage_patchbot\" works in the system python",
"created_at": "2016-08-03T13:31:45Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300389",
"user": "https://github.com/fchapoton"
}
does not work for me:
0) pull the branch here and make
1) pip install --user git+https://github.com/sagemath/[email protected]
2) ./sage --patchbot
Traceback (most recent call last):
File "<string>", line 1, in <module>
ImportError: No module named sage_patchbot
Error: cannot find installation path for sage_patchbot
See https://wiki.sagemath.org/buildbot/details for instructions
but "import sage_patchbot" works in the system python
archive/issue_comments_300390.json:
{
"body": "<div id=\"comment:80\" align=\"right\">comment:80</div>\n\nInstead of a sloppy script, I would be very much happier with `sage -patchbot` returning an informative error message.",
"created_at": "2016-08-03T14:33:19Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300390",
"user": "https://github.com/videlec"
}
Instead of a sloppy script, I would be very much happier with sage -patchbot
returning an informative error message.
archive/issue_comments_300391.json:
{
"body": "<div id=\"comment:81\" align=\"right\">comment:81</div>\n\nReplying to [@videlec](#comment%3A80):\n> Instead of a sloppy script, I would be very much happier with `sage -patchbot` returning an informative error message.\n\n+1\n\nAs I understand it the purpose of `./sage -foo` is to run the `foo` program that's installed in the sage environment. I don't see why it should be doing anything else. At most if `./sage -patchbot` is meant to do *something* it can raise an error if `patchbot` does not happen to be installed in the Sage environment. Then if one wants they can `./sage -pip install sage-patchbot` and get it in their Sage environment, at which point `./sage -patchbot` is the equivalent of `. sage-env; patchbot`, etc...\n\nOtherwise all one needs to do to run the patchbot is to run `patchbot` however it happens to be installed on a (correctly configured) system.",
"created_at": "2016-08-04T09:45:14Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300391",
"user": "https://github.com/embray"
}
Replying to @videlec:
Instead of a sloppy script, I would be very much happier with
sage -patchbot
returning an informative error message.
+1
As I understand it the purpose of ./sage -foo
is to run the foo
program that's installed in the sage environment. I don't see why it should be doing anything else. At most if ./sage -patchbot
is meant to do something it can raise an error if patchbot
does not happen to be installed in the Sage environment. Then if one wants they can ./sage -pip install sage-patchbot
and get it in their Sage environment, at which point ./sage -patchbot
is the equivalent of . sage-env; patchbot
, etc...
Otherwise all one needs to do to run the patchbot is to run patchbot
however it happens to be installed on a (correctly configured) system.
archive/issue_comments_300392.json:
{
"body": "<div id=\"comment:82\" align=\"right\">comment:82</div>\n\nSo what exactly is necessary to make this branch work, i.e. to run `sage -patchbot` successfully?",
"created_at": "2016-08-05T06:54:25Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300392",
"user": "https://github.com/rwst"
}
So what exactly is necessary to make this branch work, i.e. to run sage -patchbot
successfully?
archive/issue_comments_300393.json:
{
"body": "<div id=\"comment:83\" align=\"right\">comment:83</div>\n\nAnswer: `./sage -pip install sage-patchbot` or the downloaded 2.6.1 tarball as argument. This means the wiki instructions need to change which I just did, please check.",
"created_at": "2016-08-05T07:03:39Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300393",
"user": "https://github.com/rwst"
}
Answer: ./sage -pip install sage-patchbot
or the downloaded 2.6.1 tarball as argument. This means the wiki instructions need to change which I just did, please check.
archive/issue_comments_300394.json:
{
"body": "<div id=\"comment:84\" align=\"right\">comment:84</div>\n\nPlease do not interfere with the wiki instructions until this ticket is closed. Revert your changes there, please.",
"created_at": "2016-08-05T07:05:52Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300394",
"user": "https://github.com/fchapoton"
}
Please do not interfere with the wiki instructions until this ticket is closed. Revert your changes there, please.
archive/issue_comments_300395.json:
{
"body": "<div id=\"comment:85\" align=\"right\">comment:85</div>\n\nThanks. The current branch should be considered as broken. The wiki instructions were\nperfectly working until Jeroen made his new branch.\n\n**EDIT**: a working branch is at **u/chapoton/20736**",
"created_at": "2016-08-05T07:17:58Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300395",
"user": "https://github.com/fchapoton"
}
Thanks. The current branch should be considered as broken. The wiki instructions were perfectly working until Jeroen made his new branch.
EDIT: a working branch is at u/chapoton/20736
archive/issue_comments_300396.json:
{
"body": "<div id=\"comment:86\" align=\"right\">comment:86</div>\n\nReplying to [@rwst](#comment%3A82):\n> So what exactly is necessary to make this branch work, i.e. to run `sage -patchbot` successfully?\n\nThat's easy: I just need to check both the Python-in-Sage and the system Python.\n\nHowever, there seems to be disagreement (see @videlec and @embray) about what `./sage --patchbot` **should** actually do. There is no point in fixing the branch if I have to throw it away anyway.",
"created_at": "2016-08-05T07:27:03Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300396",
"user": "https://github.com/jdemeyer"
}
Replying to @rwst:
So what exactly is necessary to make this branch work, i.e. to run
sage -patchbot
successfully?
That's easy: I just need to check both the Python-in-Sage and the system Python.
However, there seems to be disagreement (see @videlec and @embray) about what ./sage --patchbot
should actually do. There is no point in fixing the branch if I have to throw it away anyway.
archive/issue_comments_300397.json:
{
"body": "<div id=\"comment:87\" align=\"right\">comment:87</div>\n\nProposed solution:\n\n- do not allow `sage -patchbot` for running the bot\n- make so that `sage -patchbot` just prints:\n\n```\nThe patchbot can no longer be run in this way.\nPlease see https://wiki.sagemath.org/buildbot/details for up-to-date instructions.\n```\n\nI would accept this solution.",
"created_at": "2016-08-05T07:33:20Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300397",
"user": "https://github.com/fchapoton"
}
Proposed solution:
- do not allow
sage -patchbot
for running the bot - make so that
sage -patchbot
just prints:
The patchbot can no longer be run in this way.
Please see https://wiki.sagemath.org/buildbot/details for up-to-date instructions.
I would accept this solution.
archive/issue_comments_300398.json:
{
"body": "<div id=\"comment:88\" align=\"right\">comment:88</div>\n\nMy proposal would be to fix `./sage --patchbot` to work both with a patchbot installed in Sage as well as a patchbot installed system-wide. This is easy to do, but it would result in what Vincent calls \"a sloppy script\".",
"created_at": "2016-08-05T07:36:19Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300398",
"user": "https://github.com/jdemeyer"
}
My proposal would be to fix ./sage --patchbot
to work both with a patchbot installed in Sage as well as a patchbot installed system-wide. This is easy to do, but it would result in what Vincent calls "a sloppy script".
archive/issue_comments_300399.json:
{
"body": "<div id=\"comment:89\" align=\"right\">comment:89</div>\n\nI think the goal should be to make it as easy as possible to install and run the patchbot. We could even make it such that `./sage -i patchbot && ./sage --patchbot` still works as before.",
"created_at": "2016-08-05T07:38:29Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300399",
"user": "https://github.com/jdemeyer"
}
I think the goal should be to make it as easy as possible to install and run the patchbot. We could even make it such that ./sage -i patchbot && ./sage --patchbot
still works as before.
archive/issue_comments_300400.json:
{
"body": "<div id=\"comment:90\" align=\"right\">comment:90</div>\n\nI think it is more simple to propose a unique way to install and run the patchbot\n\n* install in the system python using `pip`\n* run using `python -m`\n\nSo that people have no choice to make. And this is still only two command lines.",
"created_at": "2016-08-05T07:51:46Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300400",
"user": "https://github.com/fchapoton"
}
I think it is more simple to propose a unique way to install and run the patchbot
- install in the system python using
pip
- run using
python -m
So that people have no choice to make. And this is still only two command lines.
archive/issue_comments_300401.json:
{
"body": "<div id=\"comment:91\" align=\"right\">comment:91</div>\n\nReplying to [@fchapoton](#comment%3A90):\n> And this is still only two command lines.\n\nActually, it's five commands:\n\n1. Download `pip`\n2. Install `pip`\n3. Download `sage_patchbot`\n4. Install `sage_patchbot`\n5. Run `sage_patchbot`",
"created_at": "2016-08-05T07:55:34Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300401",
"user": "https://github.com/jdemeyer"
}
Replying to @fchapoton:
And this is still only two command lines.
Actually, it's five commands:
- Download
pip
- Install
pip
- Download
sage_patchbot
- Install
sage_patchbot
- Run
sage_patchbot
archive/issue_comments_300402.json:
{
"body": "<div id=\"comment:92\" align=\"right\">comment:92</div>\n\nMeaning no offense, this comment is slightly small-minded, IMHO",
"created_at": "2016-08-05T07:57:44Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300402",
"user": "https://github.com/fchapoton"
}
Meaning no offense, this comment is slightly small-minded, IMHO
archive/issue_comments_300403.json:
{
"body": "<div id=\"comment:93\" align=\"right\">comment:93</div>\n\nLet's assume for a second that users are intelligent enough to run commands on their system that are not wrapped in Sage, and that they have working systems with working Python installations...\n\nWhat advantage is there to\n\n```\n$ ./sage -patchbot\n```\nover \n\n```\n$ patchbot\n```",
"created_at": "2016-08-05T09:22:33Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300403",
"user": "https://github.com/embray"
}
Let's assume for a second that users are intelligent enough to run commands on their system that are not wrapped in Sage, and that they have working systems with working Python installations...
What advantage is there to
$ ./sage -patchbot
over
$ patchbot
archive/issue_comments_300404.json:
{
"body": "<div id=\"comment:94\" align=\"right\">comment:94</div>\n\nReplying to [@embray](#comment%3A93):\n> that they have working systems with working Python installations...\n\nA working Python installation is not the same as a working `pip` installation.\n\n> What advantage is there to\n> \n> ```\n> $ ./sage -patchbot\n> ```\n> over \n> \n> ```\n> $ patchbot\n> ```\n\nThe fact that `./sage --patchbot` does not require the user to change his `$PATH`.",
"created_at": "2016-08-05T09:29:11Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300404",
"user": "https://github.com/jdemeyer"
}
Replying to @embray:
that they have working systems with working Python installations...
A working Python installation is not the same as a working pip
installation.
What advantage is there to
$ ./sage -patchbot
over
$ patchbot
The fact that ./sage --patchbot
does not require the user to change his $PATH
.
archive/issue_comments_300405.json:
{
"body": "<div id=\"comment:95\" align=\"right\">comment:95</div>\n\nReplying to [@jdemeyer](#comment%3A94):\n> Replying to [@embray](#comment%3A93):\n> > that they have working systems with working Python installations...\n> \n> \n> A working Python installation is not the same as a working `pip` installation.\n\nIt effectively is. `pip` is the official installer for Python packages and comes with modern distributions of Python (including most system distributions). While you can have a \"working\" Python installation *without* pip, a Python installation with a non-working pip installation is not a working Python installation.\n\n> > What advantage is there to\n> > \n> > ```\n> > $ ./sage -patchbot\n> > ```\n> > over \n> > \n> > ```\n> > $ patchbot\n> > ```\n> \n> \n> The fact that `./sage --patchbot` does not require the user to change his `$PATH`.\n\nWhy would a user need to change their path?",
"created_at": "2016-08-05T11:15:04Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300405",
"user": "https://github.com/embray"
}
Replying to @jdemeyer:
Replying to @embray:
that they have working systems with working Python installations...
A working Python installation is not the same as a working
pip
installation.
It effectively is. pip
is the official installer for Python packages and comes with modern distributions of Python (including most system distributions). While you can have a "working" Python installation without pip, a Python installation with a non-working pip installation is not a working Python installation.
What advantage is there to
$ ./sage -patchbot
over
$ patchbot
The fact that
./sage --patchbot
does not require the user to change his$PATH
.
Why would a user need to change their path?
archive/issue_events_289401.json:
{
"actor": "https://github.com/fchapoton",
"created_at": "2016-08-06T10:20:47Z",
"event": "unlabeled",
"issue": "https://github.com/sagemath/sage/issues/20736",
"label": "https://github.com/sagemath/sage/labels/needs%20review",
"label_color": "7fff00",
"label_name": "needs review",
"label_text_color": "ffffff",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289401"
}
archive/issue_events_289402.json:
{
"actor": "https://github.com/fchapoton",
"created_at": "2016-08-06T10:20:47Z",
"event": "labeled",
"issue": "https://github.com/sagemath/sage/issues/20736",
"label": "https://github.com/sagemath/sage/labels/needs%20work",
"label_color": "ffff00",
"label_name": "needs work",
"label_text_color": "ffffff",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289402"
}
archive/issue_events_289403.json:
{
"actor": "https://github.com/fchapoton",
"created_at": "2016-08-06T10:20:47Z",
"event": "demilestoned",
"issue": "https://github.com/sagemath/sage/issues/20736",
"milestone_number": null,
"milestone_title": "sage-7.3",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289403"
}
archive/issue_events_289404.json:
{
"actor": "https://github.com/fchapoton",
"created_at": "2016-08-06T10:20:47Z",
"event": "milestoned",
"issue": "https://github.com/sagemath/sage/issues/20736",
"milestone_number": null,
"milestone_title": "sage-7.4",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289404"
}
archive/issue_comments_300406.json:
{
"body": "<div id=\"comment:97\" align=\"right\">comment:97</div>\n\nThe `sage4` patchbot is broken:\n\n```\n[2016-09-08 09:56:55] Getting ticket list...\nTraceback (most recent call last):\n File \"/var/worker.sage/local/lib/python2.7/site-packages/sage_patchbot/patchbot.py\", line 1470, in main\n patchbot.test_a_ticket(ticket)\n File \"/var/worker.sage/local/lib/python2.7/site-packages/sage_patchbot/patchbot.py\", line 951, in test_a_ticket\n rating, ticket = self.get_one_ticket()\n File \"/var/worker.sage/local/lib/python2.7/site-packages/sage_patchbot/patchbot.py\", line 774, in get_one_ticket\n return all_tickets[-1]\nIndexError: list index out of range\n```",
"created_at": "2016-09-08T10:34:10Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300406",
"user": "https://github.com/jdemeyer"
}
The sage4
patchbot is broken:
[2016-09-08 09:56:55] Getting ticket list...
Traceback (most recent call last):
File "/var/worker.sage/local/lib/python2.7/site-packages/sage_patchbot/patchbot.py", line 1470, in main
patchbot.test_a_ticket(ticket)
File "/var/worker.sage/local/lib/python2.7/site-packages/sage_patchbot/patchbot.py", line 951, in test_a_ticket
rating, ticket = self.get_one_ticket()
File "/var/worker.sage/local/lib/python2.7/site-packages/sage_patchbot/patchbot.py", line 774, in get_one_ticket
return all_tickets[-1]
IndexError: list index out of range
archive/issue_comments_300407.json:
{
"body": "<div id=\"comment:98\" align=\"right\">comment:98</div>\n\nThere is something very wrong in how the patchbot gets its tickets from the server...",
"created_at": "2016-09-08T11:41:59Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300407",
"user": "https://github.com/jdemeyer"
}
There is something very wrong in how the patchbot gets its tickets from the server...
archive/issue_comments_300408.json:
{
"body": "<div id=\"comment:99\" align=\"right\">comment:99</div>\n\nReplying to [@jdemeyer](#comment%3A98):\n> There is something very wrong in how the patchbot gets its tickets from the server...\n\nThat's probably true. But your specific problem with sage4 is also that it has already recently given a report on every available ticket with \"BuildFailed\" status.",
"created_at": "2016-09-08T15:04:38Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300408",
"user": "https://github.com/fchapoton"
}
Replying to @jdemeyer:
There is something very wrong in how the patchbot gets its tickets from the server...
That's probably true. But your specific problem with sage4 is also that it has already recently given a report on every available ticket with "BuildFailed" status.
archive/issue_comments_300409.json:
{
"body": "<div id=\"comment:100\" align=\"right\">comment:100</div>\n\nReplying to [@fchapoton](#comment%3A99):\n> But your specific problem with sage4 is also that it has already recently given a report on every available ticket with \"BuildFailed\" status.\n\nRight, that is #21430. But still, that should be completely independent from how the patchbot itself works.",
"created_at": "2016-09-08T16:37:19Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300409",
"user": "https://github.com/jdemeyer"
}
Replying to @fchapoton:
But your specific problem with sage4 is also that it has already recently given a report on every available ticket with "BuildFailed" status.
Right, that is #21430. But still, that should be completely independent from how the patchbot itself works.
archive/issue_comments_300410.json:
{
"body": "<div id=\"comment:101\"></div>\n\nBranch pushed to git repo; I updated commit sha1. This was a forced push. New commits:\n<table><tr><td><a href=\"https://github.com/sagemath/sagetrac-mirror/commit/a3b040f026d3522d8f07e0339d86370e0824bcc4\"><code>a3b040f</code></a></td><td><code>Set JUPYTER_CONFIG_DIR to ~/.sage/jupyter and hardcode IPYTHONDIR</code></td></tr><tr><td><a href=\"https://github.com/sagemath/sagetrac-mirror/commit/78f964e494bd54b57b3dc456e536046ff4485e17\"><code>78f964e</code></a></td><td><code>Clarify comment (hopefully)</code></td></tr><tr><td><a href=\"https://github.com/sagemath/sagetrac-mirror/commit/fd3ab449128dc13ee74d2c3fab79da40ebebc0fd\"><code>fd3ab44</code></a></td><td><code>Update patchbot to 2.6.1 as system package</code></td></tr><tr><td><a href=\"https://github.com/sagemath/sagetrac-mirror/commit/fc8bcec8a2d753fcf1a9eaf451258081cf02726b\"><code>fc8bcec</code></a></td><td><code>Improve --patchbot command</code></td></tr></table>\n",
"created_at": "2016-09-09T09:48:49Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300410",
"user": "https://github.com/sagetrac-git"
}
Branch pushed to git repo; I updated commit sha1. This was a forced push. New commits:
a3b040f | Set JUPYTER_CONFIG_DIR to ~/.sage/jupyter and hardcode IPYTHONDIR |
78f964e | Clarify comment (hopefully) |
fd3ab44 | Update patchbot to 2.6.1 as system package |
fc8bcec | Improve --patchbot command |
archive/issue_comments_300411.json:
{
"body": "Changed commit from **[`6d7990e`](https://github.com/sagemath/sagetrac-mirror/commit/6d7990e88c9f194f28309bd0015c40ecc09080bf)** to **[`fc8bcec`](https://github.com/sagemath/sagetrac-mirror/commit/fc8bcec8a2d753fcf1a9eaf451258081cf02726b)**",
"created_at": "2016-09-09T09:48:49Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300411",
"user": "https://github.com/sagetrac-git"
}
Changed commit from 6d7990e
to fc8bcec
archive/issue_comments_300412.json:
{
"body": "<div id=\"comment:102\" align=\"right\">comment:102</div>\n\nRebased over #21430, partially squashed, fixed installations in system Python.",
"created_at": "2016-09-09T09:49:58Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300412",
"user": "https://github.com/jdemeyer"
}
Rebased over #21430, partially squashed, fixed installations in system Python.
archive/issue_comments_300413.json:
{
"body": "Dependencies: **#21430**",
"created_at": "2016-09-09T09:49:58Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300413",
"user": "https://github.com/jdemeyer"
}
Dependencies: #21430
archive/issue_events_289405.json:
{
"actor": "https://github.com/jdemeyer",
"created_at": "2016-09-09T09:49:58Z",
"event": "unlabeled",
"issue": "https://github.com/sagemath/sage/issues/20736",
"label": "https://github.com/sagemath/sage/labels/needs%20work",
"label_color": "ffff00",
"label_name": "needs work",
"label_text_color": "ffffff",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289405"
}
archive/issue_events_289406.json:
{
"actor": "https://github.com/jdemeyer",
"created_at": "2016-09-09T09:49:58Z",
"event": "labeled",
"issue": "https://github.com/sagemath/sage/issues/20736",
"label": "https://github.com/sagemath/sage/labels/needs%20review",
"label_color": "7fff00",
"label_name": "needs review",
"label_text_color": "ffffff",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289406"
}
archive/issue_comments_300414.json:
{
"body": "<div id=\"comment:103\" align=\"right\">comment:103</div>\n\nI am currently running the patchbot on `sage4` and `arando` with this branch.",
"created_at": "2016-09-09T13:44:35Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300414",
"user": "https://github.com/jdemeyer"
}
I am currently running the patchbot on sage4
and arando
with this branch.
archive/issue_comments_300415.json:
{
"body": "<div id=\"comment:104\" align=\"right\">comment:104</div>\n\nIf you guys are not on agreement, then I am +1 on removing or deprecating `sage --patchbot`. As a recent volunteer to run a patchbot, I thought it is illogical to invoke Sage to run a patchbot whose mission is to test Sage. Installing patchbot by \"pip ...\" and then running \"python -m ...\" is straightforward and simple. Why do we need an alternate and conceptually confusing way to launch a patchbot? \n\nIf this disagreement persists, then why don't we have a poll on sage-devel?\n\nBy the way, this\n`pip install --user git+https://github.com/sagemath/sage-patchbot.git`\ninstalls the latest patchbot, why does the statement in the wiki specify the version, which would be outdated sooner or later?",
"created_at": "2016-09-22T08:38:05Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300415",
"user": "https://github.com/kwankyu"
}
If you guys are not on agreement, then I am +1 on removing or deprecating sage --patchbot
. As a recent volunteer to run a patchbot, I thought it is illogical to invoke Sage to run a patchbot whose mission is to test Sage. Installing patchbot by "pip ..." and then running "python -m ..." is straightforward and simple. Why do we need an alternate and conceptually confusing way to launch a patchbot?
If this disagreement persists, then why don't we have a poll on sage-devel?
By the way, this
pip install --user git+https://github.com/sagemath/sage-patchbot.git
installs the latest patchbot, why does the statement in the wiki specify the version, which would be outdated sooner or later?
archive/issue_comments_300416.json:
{
"body": "<div id=\"comment:105\" align=\"right\">comment:105</div>\n\nReplying to [@kwankyu](#comment%3A104):\n> If you guys are not on agreement, then I am +1 on removing or deprecating `sage --patchbot`. As a recent volunteer to run a patchbot, I thought it is illogical to invoke Sage to run a patchbot whose mission is to test Sage.\n\nI don't see it as invoking Sage. I consider `sage --patchbot` just a simple convenience command. I changed it such that it should work in all cases: the `sage_patchbot` can be installed system-wide, by the user, or inside Sage. Why should we not support that?\n\n> Installing patchbot by \"pip ...\"\n\nThen pip becomes an additional requirement for running the Sage patchbot. Many systems come with Python installed system-wide, but without pip.",
"created_at": "2016-09-22T09:11:17Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300416",
"user": "https://github.com/jdemeyer"
}
Replying to @kwankyu:
If you guys are not on agreement, then I am +1 on removing or deprecating
sage --patchbot
. As a recent volunteer to run a patchbot, I thought it is illogical to invoke Sage to run a patchbot whose mission is to test Sage.
I don't see it as invoking Sage. I consider sage --patchbot
just a simple convenience command. I changed it such that it should work in all cases: the sage_patchbot
can be installed system-wide, by the user, or inside Sage. Why should we not support that?
Installing patchbot by "pip ..."
Then pip becomes an additional requirement for running the Sage patchbot. Many systems come with Python installed system-wide, but without pip.
archive/issue_comments_300417.json:
{
"body": "<div id=\"comment:106\" align=\"right\">comment:106</div>\n\n> I don't see it as invoking Sage. I consider `sage --patchbot` just a simple convenience command. \n\nYou can see it as such since you know the internals. But for those who do not know the details of how patchbot works, like me, the command gives the impression that I am about to run a patchbot installed inside Sage. \n\nOn the other hand, It is weird to control the patchbot installed in the system Python via sage interface.",
"created_at": "2016-09-22T09:54:54Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300417",
"user": "https://github.com/kwankyu"
}
I don't see it as invoking Sage. I consider
sage --patchbot
just a simple convenience command.
You can see it as such since you know the internals. But for those who do not know the details of how patchbot works, like me, the command gives the impression that I am about to run a patchbot installed inside Sage.
On the other hand, It is weird to control the patchbot installed in the system Python via sage interface.
archive/issue_comments_300418.json:
{
"body": "<div id=\"comment:107\" align=\"right\">comment:107</div>\n\nReplying to [@kwankyu](#comment%3A106):\n> the command gives the impression that I am about to run a patchbot installed inside Sage. \n\nWhich might be the case and which is not a problem (note however that it will always be run using the system Python).",
"created_at": "2016-09-22T12:27:18Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300418",
"user": "https://github.com/jdemeyer"
}
Replying to @kwankyu:
the command gives the impression that I am about to run a patchbot installed inside Sage.
Which might be the case and which is not a problem (note however that it will always be run using the system Python).
archive/issue_comments_300419.json:
{
"body": "<div id=\"comment:108\" align=\"right\">comment:108</div>\n\nReplying to [@kwankyu](#comment%3A106):\n> > I don't see it as invoking Sage. I consider `sage --patchbot` just a simple convenience command. \n> \n> \n> You can see it as such since you know the internals. But for those who do not know the details of how patchbot works, like me, the command gives the impression that I am about to run a patchbot installed inside Sage. \n> \n> On the other hand, It is weird to control the patchbot installed in the system Python via sage interface.\n\nI agree. I think \"invoking sage\" here should be read as \"activating the sage environment\" which is true. There's no more reason to run `sage --patchbot` than there is to run `sage --make` (which is not a thing, but offers the same \"convenience\").",
"created_at": "2016-09-22T16:40:10Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300419",
"user": "https://github.com/embray"
}
Replying to @kwankyu:
I don't see it as invoking Sage. I consider
sage --patchbot
just a simple convenience command.You can see it as such since you know the internals. But for those who do not know the details of how patchbot works, like me, the command gives the impression that I am about to run a patchbot installed inside Sage.
On the other hand, It is weird to control the patchbot installed in the system Python via sage interface.
I agree. I think "invoking sage" here should be read as "activating the sage environment" which is true. There's no more reason to run sage --patchbot
than there is to run sage --make
(which is not a thing, but offers the same "convenience").
archive/issue_comments_300420.json:
{
"body": "<div id=\"comment:109\" align=\"right\">comment:109</div>\n\nReplying to [@embray](#comment%3A108):\n> `sage --make`\n\nThat option is actually called `sage -i` :-)\n\nIn all seriousness: I really want to support a `sage_patchbot` installed within Sage itself.\n\nYou may argue that this is weird. But on the other hand: why not support it? You need good arguments to *explicitly not support* something. I never said that this should be the only way. I do not want to force everybody to use `sage --patchbot`. I am only saying: it costs little to support it, so why not?",
"created_at": "2016-09-23T09:20:46Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300420",
"user": "https://github.com/jdemeyer"
}
Replying to @embray:
sage --make
That option is actually called sage -i
:-)
In all seriousness: I really want to support a sage_patchbot
installed within Sage itself.
You may argue that this is weird. But on the other hand: why not support it? You need good arguments to explicitly not support something. I never said that this should be the only way. I do not want to force everybody to use sage --patchbot
. I am only saying: it costs little to support it, so why not?
archive/issue_comments_300421.json:
{
"body": "<div id=\"comment:110\" align=\"right\">comment:110</div>\n\nPersonally I see no problem with supporting `sage --patchbot`. Just that it should not try to run a patchbot module that's installed *outside* the sage environment. In other words, it should run the patchbot that's installed in sage, or do nothing at all (or at most give an error).",
"created_at": "2016-09-23T11:22:42Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300421",
"user": "https://github.com/embray"
}
Personally I see no problem with supporting sage --patchbot
. Just that it should not try to run a patchbot module that's installed outside the sage environment. In other words, it should run the patchbot that's installed in sage, or do nothing at all (or at most give an error).
archive/issue_comments_300422.json:
{
"body": "<div id=\"comment:111\" align=\"right\">comment:111</div>\n\nReplying to [@embray](#comment%3A110):\n> Personally I see no problem with supporting `sage --patchbot`. Just that it should not try to run a patchbot module that's installed *outside* the sage environment. In other words, it should run the patchbot that's installed in sage, or do nothing at all (or at most give an error).\n\nThat's fine for me... Fr\u00e9d\u00e9ric, Vincent: what you do think?",
"created_at": "2016-09-23T11:36:46Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300422",
"user": "https://github.com/jdemeyer"
}
Replying to @embray:
Personally I see no problem with supporting
sage --patchbot
. Just that it should not try to run a patchbot module that's installed outside the sage environment. In other words, it should run the patchbot that's installed in sage, or do nothing at all (or at most give an error).
That's fine for me... Frédéric, Vincent: what you do think?
archive/issue_comments_300423.json:
{
"body": "<div id=\"comment:112\" align=\"right\">comment:112</div>\n\nWhatever way this ticket is closed, I do not care. If you manage to agree on something, go on.",
"created_at": "2016-09-23T11:48:53Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300423",
"user": "https://github.com/fchapoton"
}
Whatever way this ticket is closed, I do not care. If you manage to agree on something, go on.
archive/issue_comments_300424.json:
{
"body": "<div id=\"comment:113\" align=\"right\">comment:113</div>\n\nSo Fr\u00e9d\u00e9ric, do you retract comment [comment:85]? Because then the easiest would be to go back to that branch (i.e. my branch from [comment:76]). Judging from Erik's comment, he should agree on that.",
"created_at": "2016-09-24T14:02:21Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300424",
"user": "https://github.com/jdemeyer"
}
So Frédéric, do you retract comment [comment:85]? Because then the easiest would be to go back to that branch (i.e. my branch from [comment:76]). Judging from Erik's comment, he should agree on that.
archive/issue_comments_300425.json:
{
"body": "<div id=\"comment:114\" align=\"right\">comment:114</div>\n\ngo on",
"created_at": "2016-09-24T16:10:01Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300425",
"user": "https://github.com/fchapoton"
}
go on
archive/issue_comments_300426.json:
{
"body": "<div id=\"comment:115\" align=\"right\">comment:115</div>\n\nUhh, the branch at https://github.com/sagemath/sagetrac-mirror/commit/6d7990e88c9f194f28309bd0015c40ecc09080bf is what I'm objecting to. It attempts to run `sage_patchbot` with the system python as well, which I'm saying it shouldn't do. `sage --whatever` shouldn't *touch* the system Python except when building sage itself.",
"created_at": "2016-09-26T11:58:14Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300426",
"user": "https://github.com/embray"
}
Uhh, the branch at https://github.com/sagemath/sagetrac-mirror/commit/6d7990e88c9f194f28309bd0015c40ecc09080bf is what I'm objecting to. It attempts to run sage_patchbot
with the system python as well, which I'm saying it shouldn't do. sage --whatever
shouldn't touch the system Python except when building sage itself.
archive/issue_comments_300427.json:
{
"body": "<div id=\"comment:116\" align=\"right\">comment:116</div>\n\nSo it seems that we cannot agree on anything then...",
"created_at": "2016-09-26T12:42:23Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300427",
"user": "https://github.com/jdemeyer"
}
So it seems that we cannot agree on anything then...
archive/issue_comments_300428.json:
{
"body": "<div id=\"comment:117\" align=\"right\">comment:117</div>\n\nI thought we did... please re-read [#20736 comment:110](https://github.com/sagemath/sage/issues/20736#comment:110)\n\nKeep in mind when I write \"run patchbot\" I'm looking at the `patchbot` program as a black box. I don't care that it happens to be written in Python--nor should anyone else running it. So it's rather bizarre for `sage --patchbot` to try to do anything with a patchbot module that is not installed in the `$SAGE_LOCAL` or with ~~any python that is not in `$SAGE_LOCAL`~~\n\nI think I see the hold-up in my own thinking. There may not *be* a Python in `$SAGE_LOCAL`.\n\nIn that case I'm back to not understanding why you even need or want `sage --patchbot`",
"created_at": "2016-09-26T15:38:14Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300428",
"user": "https://github.com/embray"
}
I thought we did... please re-read #20736 comment:110
Keep in mind when I write "run patchbot" I'm looking at the patchbot
program as a black box. I don't care that it happens to be written in Python--nor should anyone else running it. So it's rather bizarre for sage --patchbot
to try to do anything with a patchbot module that is not installed in the $SAGE_LOCAL
or with any python that is not in $SAGE_LOCAL
I think I see the hold-up in my own thinking. There may not be a Python in $SAGE_LOCAL
.
In that case I'm back to not understanding why you even need or want sage --patchbot
archive/issue_comments_300429.json:
{
"body": "<div id=\"comment:118\" align=\"right\">comment:118</div>\n\nReplying to [@jdemeyer](#comment%3A109):\n> You need good arguments to *explicitly not support* something.\n\nI would say this is rarely true, and it is almost always the opposite that is true.",
"created_at": "2016-09-26T15:42:47Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300429",
"user": "https://github.com/embray"
}
Replying to @jdemeyer:
You need good arguments to explicitly not support something.
I would say this is rarely true, and it is almost always the opposite that is true.
archive/issue_comments_300430.json:
{
"body": "<div id=\"comment:119\" align=\"right\">comment:119</div>\n\nReplying to [@embray](#comment%3A117):\n> I thought we did... please re-read [#20736 comment:110](https://github.com/sagemath/sage/issues/20736#comment:110)\n\nWell, I want to run the pacthbot installed **within Sage** using the **system** Python. You seem to agree with the first part but not the second.\n\nYou cannot use the Python-in-Sage to run the patchbot because that one might be reinstalled: the patchbot builds Sage so it might rebuild the Python-in-Sage too. And reinstalling a program that you are running is a recipe for trouble.",
"created_at": "2016-09-26T15:49:22Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300430",
"user": "https://github.com/jdemeyer"
}
Replying to @embray:
I thought we did... please re-read #20736 comment:110
Well, I want to run the pacthbot installed within Sage using the system Python. You seem to agree with the first part but not the second.
You cannot use the Python-in-Sage to run the patchbot because that one might be reinstalled: the patchbot builds Sage so it might rebuild the Python-in-Sage too. And reinstalling a program that you are running is a recipe for trouble.
archive/issue_comments_300431.json:
{
"body": "<div id=\"comment:120\" align=\"right\">comment:120</div>\n\nOkay, but when we say \"the patchbot installed within Sage\" you mean there's a copy of the `sage_patchbot` package in `local/lib/python2.7/site-packages`, right?\n\nWhat is the advantage to doing that? `site-packages` is a special path as far as Python is concerned. A typical Python executable knows where to find its `site-packages` (well, it gets it from `site.py` which it knows how to find), and that `site-packages` in a sense \"belongs\" to that Python. So it's very atypical to use one Python installation to run a package that happens to be in another Python's `site-packages`. One exception is virtualenvs that make use of the \"system site-packages\" feature.\n\nNot saying it won't work--of course ultimately it's just another directory full of Python modules one is adding to `sys.path` (assuming they are for the same Python version at least, which they might not be--your patch tries `python2.7`, followed by `python2`, followed by `python`, where the latter might be Python 2.6, or Python 3.4 or something).\n\nTL;DR I don't see what you gain from that.",
"created_at": "2016-09-26T16:01:38Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300431",
"user": "https://github.com/embray"
}
Okay, but when we say "the patchbot installed within Sage" you mean there's a copy of the sage_patchbot
package in local/lib/python2.7/site-packages
, right?
What is the advantage to doing that? site-packages
is a special path as far as Python is concerned. A typical Python executable knows where to find its site-packages
(well, it gets it from site.py
which it knows how to find), and that site-packages
in a sense "belongs" to that Python. So it's very atypical to use one Python installation to run a package that happens to be in another Python's site-packages
. One exception is virtualenvs that make use of the "system site-packages" feature.
Not saying it won't work--of course ultimately it's just another directory full of Python modules one is adding to sys.path
(assuming they are for the same Python version at least, which they might not be--your patch tries python2.7
, followed by python2
, followed by python
, where the latter might be Python 2.6, or Python 3.4 or something).
TL;DR I don't see what you gain from that.
archive/issue_comments_300432.json:
{
"body": "<div id=\"comment:121\" align=\"right\">comment:121</div>\n\nI haven't followed the full discussion, nor have I ever run a patchbot, but here's my 2 cents.\n\n+1 on using the system Python.\n\n-1 on including the patchbot as part of the sage distribution in any way.\n\n-1 on supporting `sage --patchbot`. \n\nI think it's inviting trouble to have the testing framework depend in any way on scripts of the software to be tested.",
"created_at": "2016-09-26T22:51:03Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300432",
"user": "https://github.com/mkoeppe"
}
I haven't followed the full discussion, nor have I ever run a patchbot, but here's my 2 cents.
+1 on using the system Python.
-1 on including the patchbot as part of the sage distribution in any way.
-1 on supporting sage --patchbot
.
I think it's inviting trouble to have the testing framework depend in any way on scripts of the software to be tested.
archive/issue_comments_300433.json:
{
"body": "<div id=\"comment:122\" align=\"right\">comment:122</div>\n\nReplying to [@jdemeyer](#comment%3A119):\n> You cannot use the Python-in-Sage to run the patchbot because that one might be reinstalled: the patchbot builds Sage so it might rebuild the Python-in-Sage too. And reinstalling a program that you are running is a recipe for trouble.\n\nReinstalled by what, a Sage upgrade or a patchbot run?",
"created_at": "2016-09-27T03:19:25Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300433",
"user": "https://github.com/rwst"
}
Replying to @jdemeyer:
You cannot use the Python-in-Sage to run the patchbot because that one might be reinstalled: the patchbot builds Sage so it might rebuild the Python-in-Sage too. And reinstalling a program that you are running is a recipe for trouble.
Reinstalled by what, a Sage upgrade or a patchbot run?
archive/issue_comments_300434.json:
{
"body": "<div id=\"comment:123\" align=\"right\">comment:123</div>\n\nReplying to [@rwst](#comment%3A122):\n> Replying to [@jdemeyer](#comment%3A119):\n> > You cannot use the Python-in-Sage to run the patchbot because that one might be reinstalled: the patchbot builds Sage so it might rebuild the Python-in-Sage too. And reinstalling a program that you are running is a recipe for trouble.\n> \n> \n> Reinstalled by what, a Sage upgrade or a patchbot run?\n\nA patchbot run (the patchbot obviously needs to build Sage, which might include a Python rebuild).",
"created_at": "2016-09-27T07:36:20Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300434",
"user": "https://github.com/jdemeyer"
}
Replying to @rwst:
Replying to @jdemeyer:
You cannot use the Python-in-Sage to run the patchbot because that one might be reinstalled: the patchbot builds Sage so it might rebuild the Python-in-Sage too. And reinstalling a program that you are running is a recipe for trouble.
Reinstalled by what, a Sage upgrade or a patchbot run?
A patchbot run (the patchbot obviously needs to build Sage, which might include a Python rebuild).
archive/issue_comments_300435.json:
{
"body": "<div id=\"comment:124\" align=\"right\">comment:124</div>\n\nReplying to [@embray](#comment%3A120):\n> TL;DR I don't see what you gain from that.\n\nI am sure I have said this already multiple times: what I gain is not requiring a system-wide `pip`.\n\nI just want the requirements to run the patchbot to be as minimal as possible. With my proposal, the only requirement is Sage.",
"created_at": "2016-09-27T07:40:55Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300435",
"user": "https://github.com/jdemeyer"
}
Replying to @embray:
TL;DR I don't see what you gain from that.
I am sure I have said this already multiple times: what I gain is not requiring a system-wide pip
.
I just want the requirements to run the patchbot to be as minimal as possible. With my proposal, the only requirement is Sage.
archive/issue_comments_300436.json:
{
"body": "<div id=\"comment:125\" align=\"right\">comment:125</div>\n\nReplying to [@jdemeyer](#comment%3A123):\n> Replying to [@rwst](#comment%3A122):\n> > Replying to [@jdemeyer](#comment%3A119):\n> > > You cannot use the Python-in-Sage to run the patchbot because that one might be reinstalled: the patchbot builds Sage so it might rebuild the Python-in-Sage too. And reinstalling a program that you are running is a recipe for trouble.\n> > \n> > \n> > Reinstalled by what, a Sage upgrade or a patchbot run?\n> \n> \n> A patchbot run (the patchbot obviously needs to build Sage, which might include a Python rebuild).\n\nJust FYI: that doesn't apply to unsafe tickets (EDITED). Those get built in a sandbox under /tmp. But they don't work anyway atm.",
"created_at": "2016-09-27T08:51:21Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300436",
"user": "https://github.com/rwst"
}
Replying to @jdemeyer:
Replying to @rwst:
Replying to @jdemeyer:
You cannot use the Python-in-Sage to run the patchbot because that one might be reinstalled: the patchbot builds Sage so it might rebuild the Python-in-Sage too. And reinstalling a program that you are running is a recipe for trouble.
Reinstalled by what, a Sage upgrade or a patchbot run?
A patchbot run (the patchbot obviously needs to build Sage, which might include a Python rebuild).
Just FYI: that doesn't apply to unsafe tickets (EDITED). Those get built in a sandbox under /tmp. But they don't work anyway atm.
archive/issue_comments_300437.json:
{
"body": "<div id=\"comment:126\" align=\"right\">comment:126</div>\n\nBut wait, a Python rebuild happens potentially only with unsafe tickets, as well.",
"created_at": "2016-09-27T08:53:30Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300437",
"user": "https://github.com/rwst"
}
But wait, a Python rebuild happens potentially only with unsafe tickets, as well.
archive/issue_comments_300438.json:
{
"body": "<div id=\"comment:127\" align=\"right\">comment:127</div>\n\nSo your precaution doesn't matter because Python rebuilds are done in /tmp, not in the Sage installation that runs the patchbot.",
"created_at": "2016-09-27T08:55:08Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300438",
"user": "https://github.com/rwst"
}
So your precaution doesn't matter because Python rebuilds are done in /tmp, not in the Sage installation that runs the patchbot.
archive/issue_comments_300439.json:
{
"body": "<div id=\"comment:128\" align=\"right\">comment:128</div>\n\nReplying to [@rwst](#comment%3A126):\n> But wait, a Python rebuild happens potentially only with unsafe tickets, as well.\n\nWrong. The patchbot also needs to build Sage in the usual way.",
"created_at": "2016-09-27T09:02:23Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300439",
"user": "https://github.com/jdemeyer"
}
Replying to @rwst:
But wait, a Python rebuild happens potentially only with unsafe tickets, as well.
Wrong. The patchbot also needs to build Sage in the usual way.
archive/issue_comments_300440.json:
{
"body": "<div id=\"comment:129\" align=\"right\">comment:129</div>\n\nBy the way, I was unaware that \"unsafe\" tickets were never tested. I never realized that.",
"created_at": "2016-09-27T09:20:30Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300440",
"user": "https://github.com/jdemeyer"
}
By the way, I was unaware that "unsafe" tickets were never tested. I never realized that.
archive/issue_comments_300441.json:
{
"body": "<div id=\"comment:130\" align=\"right\">comment:130</div>\n\nYou can test them but with some runs there were big problems. That's why it's discouraged to not use `--safe-only`.",
"created_at": "2016-09-27T11:09:37Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300441",
"user": "https://github.com/rwst"
}
You can test them but with some runs there were big problems. That's why it's discouraged to not use --safe-only
.
archive/issue_comments_300442.json:
{
"body": "<div id=\"comment:131\" align=\"right\">comment:131</div>\n\nReplying to [@mkoeppe](#comment%3A121):\n> I think it's inviting trouble to have the testing framework depend in any way on scripts of the software to be tested. \n\nMy idea is that the patchbot would not *depend* on anything from Sage. Only that Sage would provide convenient (but optional) scripts to run the patchbot.",
"created_at": "2016-10-18T13:04:16Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300442",
"user": "https://github.com/jdemeyer"
}
Replying to @mkoeppe:
I think it's inviting trouble to have the testing framework depend in any way on scripts of the software to be tested.
My idea is that the patchbot would not depend on anything from Sage. Only that Sage would provide convenient (but optional) scripts to run the patchbot.
archive/issue_comments_300443.json:
{
"body": "<div id=\"comment:132\" align=\"right\">comment:132</div>\n\nThis ticket seems stalled because jdemeyer insists to support running patchbot installed inside Sage by the system python using Sage interface, whose idea is objected by some others.\n\nThe main objective of the ticket is to make it official to install the patchbot only as a package to the system python. This objective is in direct conflict with the jdemeyer's idea.\n\nSo my suggestion is to separate the script implementing jdemeyer's idea from the present ticket, and he opens a ticket for his script. This will facilitate closing this ticket.",
"created_at": "2016-10-19T18:15:00Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300443",
"user": "https://github.com/kwankyu"
}
This ticket seems stalled because jdemeyer insists to support running patchbot installed inside Sage by the system python using Sage interface, whose idea is objected by some others.
The main objective of the ticket is to make it official to install the patchbot only as a package to the system python. This objective is in direct conflict with the jdemeyer's idea.
So my suggestion is to separate the script implementing jdemeyer's idea from the present ticket, and he opens a ticket for his script. This will facilitate closing this ticket.
archive/issue_comments_300444.json:
{
"body": "<div id=\"comment:133\" align=\"right\">comment:133</div>\n\nReplying to [@kwankyu](#comment%3A132):\n> This objective is in direct conflict with the jdemeyer's idea.\n\nAbsolutely not, there is no conflict! I just want one *additional* way to run the patchbot which does not harm other ways of running the patchbot.\n\nI don't see why people are so much against an additional feature. I'm not forcing anybody to run the patchbot \"my way\", I just want to keep the possibility.",
"created_at": "2016-10-19T20:55:10Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300444",
"user": "https://github.com/jdemeyer"
}
Replying to @kwankyu:
This objective is in direct conflict with the jdemeyer's idea.
Absolutely not, there is no conflict! I just want one additional way to run the patchbot which does not harm other ways of running the patchbot.
I don't see why people are so much against an additional feature. I'm not forcing anybody to run the patchbot "my way", I just want to keep the possibility.
archive/issue_comments_300445.json:
{
"body": "<div id=\"comment:134\" align=\"right\">comment:134</div>\n\n>`--sage-root=`\n\n1. the parameter should be called `--sage-testing-root=` (or similar) rather than `sage-root`: several sage installations may exists on a system. Especially sage-root might be menthally associated with the production-sage root directory.\n\n2. The option `sage-testing-root` should be added as a config file parameter.\n\nIf supporting `'sage -patchbot'` works as expected (starts the user-local or system installation of the patchbot and also uses the system python) and does not cause any serious harm (including maintenance issues) I would not insist on the complete removal of 'sage -patchbot'.\nA compromise between the two different opinions here could be to print a message that running `'sage -patchbot'` is no longer the preferred way to run the patchbot and point to the buildbot wiki.\n\n\n> sloppy script \n\nwhat is exactly the issue of the script by Jeroen?",
"created_at": "2016-10-20T21:32:07Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300445",
"user": "https://github.com/sagetrac-jakobkroeker"
}
--sage-root=
-
the parameter should be called
--sage-testing-root=
(or similar) rather thansage-root
: several sage installations may exists on a system. Especially sage-root might be menthally associated with the production-sage root directory. -
The option
sage-testing-root
should be added as a config file parameter.
If supporting 'sage -patchbot'
works as expected (starts the user-local or system installation of the patchbot and also uses the system python) and does not cause any serious harm (including maintenance issues) I would not insist on the complete removal of 'sage -patchbot'.
A compromise between the two different opinions here could be to print a message that running 'sage -patchbot'
is no longer the preferred way to run the patchbot and point to the buildbot wiki.
sloppy script
what is exactly the issue of the script by Jeroen?
archive/issue_events_289407.json:
{
"actor": "https://github.com/sagetrac-jakobkroeker",
"created_at": "2016-10-20T21:32:07Z",
"event": "unlabeled",
"issue": "https://github.com/sagemath/sage/issues/20736",
"label": "https://github.com/sagemath/sage/labels/needs%20review",
"label_color": "7fff00",
"label_name": "needs review",
"label_text_color": "ffffff",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289407"
}
archive/issue_events_289408.json:
{
"actor": "https://github.com/sagetrac-jakobkroeker",
"created_at": "2016-10-20T21:32:07Z",
"event": "labeled",
"issue": "https://github.com/sagemath/sage/issues/20736",
"label": "https://github.com/sagemath/sage/labels/needs%20work",
"label_color": "ffff00",
"label_name": "needs work",
"label_text_color": "ffffff",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289408"
}
archive/issue_comments_300446.json:
{
"body": "<div id=\"comment:135\" align=\"right\">comment:135</div>\n\n>The option sage-root should be added as a config file parameter. \n\n@fchapoton\n\nI just discovered that it is, but in the config file the parameter name has to `be sage_root` (underscore instead of a dash), so here is an inconsistency. Also please update the wiki instructions `https://wiki.sagemath.org/buildbot/details` accordingly.",
"created_at": "2016-10-21T05:53:36Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300446",
"user": "https://github.com/sagetrac-jakobkroeker"
}
The option sage-root should be added as a config file parameter.
@fchapoton
I just discovered that it is, but in the config file the parameter name has to be sage_root
(underscore instead of a dash), so here is an inconsistency. Also please update the wiki instructions https://wiki.sagemath.org/buildbot/details
accordingly.
archive/issue_comments_300447.json:
{
"body": "<div id=\"comment:136\" align=\"right\">comment:136</div>\n\nIMHO, it's obvious that the `--sage-root` when running the patchbot is the Sage installation that you want to test. I don't see the point of using a more obscure name like `--sage-testing-root` for the option.",
"created_at": "2016-10-21T06:58:30Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300447",
"user": "https://github.com/jdemeyer"
}
IMHO, it's obvious that the --sage-root
when running the patchbot is the Sage installation that you want to test. I don't see the point of using a more obscure name like --sage-testing-root
for the option.
archive/issue_comments_300448.json:
{
"body": "<div id=\"comment:137\" align=\"right\">comment:137</div>\n\nReplying to [@jdemeyer](#comment%3A136):\n\n> is the Sage installation that you want to test.\n\nIf I want to test a sage installation I just type './sage -t --all ' \n\nOf course you mean that 'sage-root' is here the folder where patchbot will test various tickets (and thus various sage versions)\n\n> IMHO, it's obvious that the `--sage-root` when running the patchbot is the Sage installation that you want to test. \n\nSince you formally failed to correcly explain what `--sage-root` means, I do not buy that 'it is obvious', at least not for a new patchbot user. So I find the 'sage-root' patchbot option should be at lieast documented (what it does) on the wiki and in the software and the inconsistency between 'sage_root' and 'sage-root' should be removed (underscore vs dash) \n\nBy the way; it is well known that naming is one important part of software development and it may create mental blockades or misunderstandings.\n\nIn summary: \nI expect as minimal action to gain positive review that 'sage_root' gets documented in the software and in the wiki\n(https://wiki.sagemath.org/buildbot/details) and the inconsistency between 'sage_root' and 'sage-root' is removed (underscore in config file vs dash in command line option)",
"created_at": "2016-10-22T18:30:08Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300448",
"user": "https://github.com/sagetrac-jakobkroeker"
}
Replying to @jdemeyer:
is the Sage installation that you want to test.
If I want to test a sage installation I just type './sage -t --all '
Of course you mean that 'sage-root' is here the folder where patchbot will test various tickets (and thus various sage versions)
IMHO, it's obvious that the
--sage-root
when running the patchbot is the Sage installation that you want to test.
Since you formally failed to correcly explain what --sage-root
means, I do not buy that 'it is obvious', at least not for a new patchbot user. So I find the 'sage-root' patchbot option should be at lieast documented (what it does) on the wiki and in the software and the inconsistency between 'sage_root' and 'sage-root' should be removed (underscore vs dash)
By the way; it is well known that naming is one important part of software development and it may create mental blockades or misunderstandings.
In summary: I expect as minimal action to gain positive review that 'sage_root' gets documented in the software and in the wiki (https://wiki.sagemath.org/buildbot/details) and the inconsistency between 'sage_root' and 'sage-root' is removed (underscore in config file vs dash in command line option)
archive/issue_comments_300449.json:
{
"body": "<div id=\"comment:138\" align=\"right\">comment:138</div>\n\nAnd the `'sage -patchbot'` command should print a warning that `'sage -patchbot'` is no longer the preferred way to use patchbot and print the preferred way to use the patchbot or link to the wiki (https://wiki.sagemath.org/buildbot/details)",
"created_at": "2016-10-22T18:35:21Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300449",
"user": "https://github.com/sagetrac-jakobkroeker"
}
And the 'sage -patchbot'
command should print a warning that 'sage -patchbot'
is no longer the preferred way to use patchbot and print the preferred way to use the patchbot or link to the wiki (https://wiki.sagemath.org/buildbot/details)
archive/issue_comments_300450.json:
{
"body": "<div id=\"comment:139\" align=\"right\">comment:139</div>\n\nReplying to [@jdemeyer](#comment%3A136):\n> IMHO, it's obvious that the `--sage-root` when running the patchbot is the Sage installation that you want to test. I don't see the point of using a more obscure name like `--sage-testing-root` for the option.\n\nHere we agree--the purpose of patchbot is testing sage installations. `--sage-root` is the path to the sage installation you want it to test. If there is some documentation that needs to be improved that's one thing, but putting \"testing\" in the option name is superfluous.\n\nAs for `sage -patchbot`, no there's no conflict. I'm -0.5 on it--I think it does no harm, ultimately, but it also comes across as code rot to me and doesn't serve any purpose I can glean except for \"backwards compatibility\", but with what I think is ultimately a misfeature anyways :/",
"created_at": "2016-11-07T12:58:14Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300450",
"user": "https://github.com/embray"
}
Replying to @jdemeyer:
IMHO, it's obvious that the
--sage-root
when running the patchbot is the Sage installation that you want to test. I don't see the point of using a more obscure name like--sage-testing-root
for the option.
Here we agree--the purpose of patchbot is testing sage installations. --sage-root
is the path to the sage installation you want it to test. If there is some documentation that needs to be improved that's one thing, but putting "testing" in the option name is superfluous.
As for sage -patchbot
, no there's no conflict. I'm -0.5 on it--I think it does no harm, ultimately, but it also comes across as code rot to me and doesn't serve any purpose I can glean except for "backwards compatibility", but with what I think is ultimately a misfeature anyways :/
archive/issue_comments_300451.json:
{
"body": "<div id=\"comment:140\" align=\"right\">comment:140</div>\n\nReplying to [@embray](#comment%3A139):\n> doesn't serve any purpose\n\nWell, if it didn't serve any purpose, I wouldn't insist so much. See [comment:54] and following.",
"created_at": "2016-11-07T13:17:28Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300451",
"user": "https://github.com/jdemeyer"
}
Replying to @embray:
doesn't serve any purpose
Well, if it didn't serve any purpose, I wouldn't insist so much. See [comment:54] and following.
archive/issue_comments_300452.json:
{
"body": "<div id=\"comment:141\" align=\"right\">comment:141</div>\n\nI'm not sure why you need to run `python -m sage_patchbot.patchbot`. Although that's not bad practice in cases where you want to make sure you're running the right `sage_patchbot` with the right `python`, in most cases (I think?) it suffices to just run `patchbot`.\n\nSo ISTM the main purpose of `./sage -patchbot` is to provide a sane default for the `--sage-root` option. First of all, this already gets set from the `SAGE_ROOT` environment variable if it is set. But what if the pwd is also a valid SAGE_ROOT? Then it could take the pwd to be the default (if `SAGE_ROOT` is not set). That would incorporate all the convenience of running `./sage -patchbot` from the same directory--more even. Compare:\n\n```\n$ cd my_sage\n$ ./sage -patchbot\n```\n\n```\n$ cd my_sage\n$ patchbot\n```",
"created_at": "2016-11-07T14:33:22Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300452",
"user": "https://github.com/embray"
}
I'm not sure why you need to run python -m sage_patchbot.patchbot
. Although that's not bad practice in cases where you want to make sure you're running the right sage_patchbot
with the right python
, in most cases (I think?) it suffices to just run patchbot
.
So ISTM the main purpose of ./sage -patchbot
is to provide a sane default for the --sage-root
option. First of all, this already gets set from the SAGE_ROOT
environment variable if it is set. But what if the pwd is also a valid SAGE_ROOT? Then it could take the pwd to be the default (if SAGE_ROOT
is not set). That would incorporate all the convenience of running ./sage -patchbot
from the same directory--more even. Compare:
$ cd my_sage
$ ./sage -patchbot
$ cd my_sage
$ patchbot
archive/issue_comments_300453.json:
{
"body": "Changed commit from **[`fc8bcec`](https://github.com/sagemath/sagetrac-mirror/commit/fc8bcec8a2d753fcf1a9eaf451258081cf02726b)** to **[`d7c83f1`](https://github.com/sagemath/sagetrac-mirror/commit/d7c83f1b06d1d851ef012acdf99a105cbe9db3ac)**",
"created_at": "2017-01-05T14:59:52Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300453",
"user": "https://github.com/sagetrac-git"
}
Changed commit from fc8bcec
to d7c83f1
archive/issue_comments_300454.json:
{
"body": "<div id=\"comment:142\"></div>\n\nBranch pushed to git repo; I updated commit sha1. This was a forced push. New commits:\n<table><tr><td><a href=\"https://github.com/sagemath/sagetrac-mirror/commit/e75e8def49a62f3c16d0233742ed76d69f37098e\"><code>e75e8de</code></a></td><td><code>Update patchbot to 2.6.1 as system package</code></td></tr><tr><td><a href=\"https://github.com/sagemath/sagetrac-mirror/commit/d7c83f1b06d1d851ef012acdf99a105cbe9db3ac\"><code>d7c83f1</code></a></td><td><code>Improve --patchbot command</code></td></tr></table>\n",
"created_at": "2017-01-05T14:59:52Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300454",
"user": "https://github.com/sagetrac-git"
}
Branch pushed to git repo; I updated commit sha1. This was a forced push. New commits:
e75e8de | Update patchbot to 2.6.1 as system package |
d7c83f1 | Improve --patchbot command |
archive/issue_comments_300455.json:
{
"body": "<div id=\"comment:143\" align=\"right\">comment:143</div>\n\nRebased to Sage-7.5.rc1",
"created_at": "2017-01-05T15:00:21Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300455",
"user": "https://github.com/jdemeyer"
}
Rebased to Sage-7.5.rc1
archive/issue_events_289409.json:
{
"actor": "https://github.com/jdemeyer",
"created_at": "2017-01-05T15:00:21Z",
"event": "demilestoned",
"issue": "https://github.com/sagemath/sage/issues/20736",
"milestone_number": null,
"milestone_title": "sage-7.4",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289409"
}
archive/issue_events_289410.json:
{
"actor": "https://github.com/jdemeyer",
"created_at": "2017-01-05T15:00:21Z",
"event": "milestoned",
"issue": "https://github.com/sagemath/sage/issues/20736",
"milestone_number": null,
"milestone_title": "sage-7.6",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289410"
}
archive/issue_comments_300456.json:
{
"body": "Changed dependencies from **#21430** to none",
"created_at": "2017-01-05T15:00:21Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300456",
"user": "https://github.com/jdemeyer"
}
Changed dependencies from #21430 to none
archive/issue_comments_300457.json:
{
"body": "<div id=\"comment:144\" align=\"right\">comment:144</div>\n\nSince this ticket seems not to reach a consensus, what about splitting it into two part:\n\n* remove patchbot optional spkg, and deprecate the related `sage -patchbot` command (this can be done for next beta)\n* discuss and write a convenient alternative command or documentation to run the patchbot (this seems the blocking point and might require more time)\n\nI can volonteer to implement the first point on top of the latest beta ;)",
"created_at": "2017-02-28T17:48:16Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300457",
"user": "https://github.com/sagetrac-tmonteil"
}
Since this ticket seems not to reach a consensus, what about splitting it into two part:
- remove patchbot optional spkg, and deprecate the related
sage -patchbot
command (this can be done for next beta) - discuss and write a convenient alternative command or documentation to run the patchbot (this seems the blocking point and might require more time)
I can volonteer to implement the first point on top of the latest beta ;)
archive/issue_comments_300458.json:
{
"body": "<div id=\"comment:145\" align=\"right\">comment:145</div>\n\nReplying to [@sagetrac-tmonteil](#comment%3A144):\n> * remove patchbot optional spkg, and deprecate the related `sage -patchbot` command (this can be done for next beta)\n> * discuss and write a convenient alternative command or documentation to run the patchbot (this seems the blocking point and might require more time)\n\nDoesn't \"deprecate the related `sage -patchbot` command\" relate more to the second point instead of the first?\n\nSo I would rather split up as\n\n1. remove patchbot optional spkg\n2. discuss how to run the patchbot and what to do with `sage --patchbot`",
"created_at": "2017-02-28T18:06:32Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300458",
"user": "https://github.com/jdemeyer"
}
Replying to @sagetrac-tmonteil:
- remove patchbot optional spkg, and deprecate the related
sage -patchbot
command (this can be done for next beta)- discuss and write a convenient alternative command or documentation to run the patchbot (this seems the blocking point and might require more time)
Doesn't "deprecate the related sage -patchbot
command" relate more to the second point instead of the first?
So I would rather split up as
- remove patchbot optional spkg
- discuss how to run the patchbot and what to do with
sage --patchbot
archive/issue_comments_300459.json:
{
"body": "<div id=\"comment:146\" align=\"right\">comment:146</div>\n\nOK, see #22473.\n\nReplying to [@jdemeyer](#comment%3A145):\n> Doesn't \"deprecate the related `sage -patchbot` command\" relate more to the second point instead of the first?\n> \n> So I would rather split up as\n> \n> 1. remove patchbot optional spkg\n> 2. discuss how to run the patchbot and what to do with `sage --patchbot`\n\nOK, I did not deprecate the command, but since the current `sage -patchbot` command does not work anyway, i replaced it with a short sentence pointing to https://wiki.sagemath.org/buildbot#Installing_the_patchbot (which are the most up-to-date instructions). It is just a suggestion until the current ticket get fixed, i can undo the last commit about the `sage -patchbot` command if required.",
"created_at": "2017-02-28T18:28:29Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300459",
"user": "https://github.com/sagetrac-tmonteil"
}
OK, see #22473.
Replying to @jdemeyer:
Doesn't "deprecate the related
sage -patchbot
command" relate more to the second point instead of the first?So I would rather split up as
- remove patchbot optional spkg
- discuss how to run the patchbot and what to do with
sage --patchbot
OK, I did not deprecate the command, but since the current sage -patchbot
command does not work anyway, i replaced it with a short sentence pointing to https://wiki.sagemath.org/buildbot#Installing_the_patchbot (which are the most up-to-date instructions). It is just a suggestion until the current ticket get fixed, i can undo the last commit about the sage -patchbot
command if required.
archive/issue_comments_300460.json:
{
"body": "<div id=\"comment:147\" align=\"right\">comment:147</div>\n\nReplying to [@sagetrac-tmonteil](#comment%3A146):\n> It is just a suggestion until the current ticket get fixed\n\nFor the record: in my opinion, nothing needs to be *fixed* since nothing is broken.",
"created_at": "2017-02-28T18:44:01Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300460",
"user": "https://github.com/jdemeyer"
}
Replying to @sagetrac-tmonteil:
It is just a suggestion until the current ticket get fixed
For the record: in my opinion, nothing needs to be fixed since nothing is broken.
archive/issue_comments_300461.json:
{
"body": "<div id=\"comment:148\" align=\"right\">comment:148</div>\n\nReplying to [@sagetrac-tmonteil](#comment%3A146):\n> https://wiki.sagemath.org/buildbot#Installing_the_patchbot (which are the most up-to-date instructions). It is just a suggestion until the current ticket get fixed\n\nIt's also an invitation to move the current discussion on this ticket to an edit war on that wiki page, so I am not so happy with that.",
"created_at": "2017-02-28T19:14:29Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300461",
"user": "https://github.com/jdemeyer"
}
Replying to @sagetrac-tmonteil:
https://wiki.sagemath.org/buildbot#Installing_the_patchbot (which are the most up-to-date instructions). It is just a suggestion until the current ticket get fixed
It's also an invitation to move the current discussion on this ticket to an edit war on that wiki page, so I am not so happy with that.
archive/issue_comments_300462.json:
{
"body": "<div id=\"comment:149\" align=\"right\">comment:149</div>\n\nReplying to [@jdemeyer](#comment%3A148):\n> Replying to [@sagetrac-tmonteil](#comment%3A146):\n> > https://wiki.sagemath.org/buildbot#Installing_the_patchbot (which are the most up-to-date instructions). It is just a suggestion until the current ticket get fixed\n> \n> \n> It's also an invitation to move the current discussion on this ticket to an edit war on that wiki page, so I am not so happy with that.\n\nOK, i removed that commit, the goal is to get rid of this broken optional package earlier, without entering the picky part of the current ticket.",
"created_at": "2017-02-28T21:44:24Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300462",
"user": "https://github.com/sagetrac-tmonteil"
}
Replying to @jdemeyer:
Replying to @sagetrac-tmonteil:
https://wiki.sagemath.org/buildbot#Installing_the_patchbot (which are the most up-to-date instructions). It is just a suggestion until the current ticket get fixed
It's also an invitation to move the current discussion on this ticket to an edit war on that wiki page, so I am not so happy with that.
OK, i removed that commit, the goal is to get rid of this broken optional package earlier, without entering the picky part of the current ticket.
archive/issue_comments_300463.json:
{
"body": "<div id=\"comment:150\" align=\"right\">comment:150</div>\n\nI think IIRC I was fine with keeping `sage -patchbot` as a shortcut for \"run `patchbot` with the `--sage-root` option set from this sage). I think that makes sense, and I think that's what Jeroen was trying to do originally. I just thought the solution was overcomplicated.\n\nI don't know why we're talking about `python -m` anything. When you install the `sage-patchbot` Python package (which Fr\u00e9d\u00e9ric did a great job putting together at my urging) it installs a `patchbot` script. If you run that it just works, you just have to supply the correct `--sage-root` argument.",
"created_at": "2017-03-01T08:21:29Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300463",
"user": "https://github.com/embray"
}
I think IIRC I was fine with keeping sage -patchbot
as a shortcut for "run patchbot
with the --sage-root
option set from this sage). I think that makes sense, and I think that's what Jeroen was trying to do originally. I just thought the solution was overcomplicated.
I don't know why we're talking about python -m
anything. When you install the sage-patchbot
Python package (which Frédéric did a great job putting together at my urging) it installs a patchbot
script. If you run that it just works, you just have to supply the correct --sage-root
argument.
archive/issue_comments_300464.json:
{
"body": "<div id=\"comment:151\" align=\"right\">comment:151</div>\n\nWhat I want in particular is to install the patchbot within the Sage environment and have `sage --patchbot` recognize that.\n\nOf course, I am totally open to supporting any other use cases too (unlike some other people who explicitly do not want to suppport my use case).",
"created_at": "2017-03-01T08:55:05Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300464",
"user": "https://github.com/jdemeyer"
}
What I want in particular is to install the patchbot within the Sage environment and have sage --patchbot
recognize that.
Of course, I am totally open to supporting any other use cases too (unlike some other people who explicitly do not want to suppport my use case).
archive/issue_comments_300465.json:
{
"body": "<div id=\"comment:152\" align=\"right\">comment:152</div>\n\nWas there some reason that wouldn't work? That is, if you ran `sage -pip install sage-patchbot`? Then, in principle, it would work if you ran `sage -sh -c patchbot` though I guess if `$SAGE_LOCAL/bin/patchbot` exists then `sage -patchbot` can be a shortcut for that. I'm just not sure why you'd need or want to do that but I don't oppose it.",
"created_at": "2017-03-01T10:22:21Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300465",
"user": "https://github.com/embray"
}
Was there some reason that wouldn't work? That is, if you ran sage -pip install sage-patchbot
? Then, in principle, it would work if you ran sage -sh -c patchbot
though I guess if $SAGE_LOCAL/bin/patchbot
exists then sage -patchbot
can be a shortcut for that. I'm just not sure why you'd need or want to do that but I don't oppose it.
archive/issue_comments_300466.json:
{
"body": "Dependencies: **#22473**",
"created_at": "2017-03-01T10:35:16Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300466",
"user": "https://github.com/jdemeyer"
}
Dependencies: #22473
archive/issue_comments_300467.json:
{
"body": "<div id=\"comment:154\" align=\"right\">comment:154</div>\n\nReplying to [@embray](#comment%3A150):\n> I don't know why we're talking about `python -m` anything. When you install the `sage-patchbot` Python package (which Fr\u00e9d\u00e9ric did a great job putting together at my urging) it installs a `patchbot` script.\n\n...in a directory which might not be the default `$PATH` (especially in the case of a `--user` install). So there are cases where `python -m sage_patchbot.patchbot` works but running `patchbot` does not.",
"created_at": "2017-03-01T10:52:19Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300467",
"user": "https://github.com/jdemeyer"
}
Replying to @embray:
I don't know why we're talking about
python -m
anything. When you install thesage-patchbot
Python package (which Frédéric did a great job putting together at my urging) it installs apatchbot
script.
...in a directory which might not be the default $PATH
(especially in the case of a --user
install). So there are cases where python -m sage_patchbot.patchbot
works but running patchbot
does not.
archive/issue_comments_300468.json:
{
"body": "<div id=\"comment:155\"></div>\n\nBranch pushed to git repo; I updated commit sha1. This was a forced push. New commits:\n<table><tr><td><a href=\"https://github.com/sagemath/sagetrac-mirror/commit/b2810d60b64becc44559d098358ad829964367c5\"><code>b2810d6</code></a></td><td><code>#22473 : remove the patchbot optional spkg.</code></td></tr><tr><td><a href=\"https://github.com/sagemath/sagetrac-mirror/commit/537849fc78ef4ce4029f50d12e4847d66b96d5b1\"><code>537849f</code></a></td><td><code>trac 20736 first try, failing</code></td></tr><tr><td><a href=\"https://github.com/sagemath/sagetrac-mirror/commit/b423b62329b9d028a8a6de11def5fbe1eae24519\"><code>b423b62</code></a></td><td><code>Improve --patchbot command</code></td></tr></table>\n",
"created_at": "2017-03-01T11:00:45Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300468",
"user": "https://github.com/sagetrac-git"
}
Branch pushed to git repo; I updated commit sha1. This was a forced push. New commits:
b2810d6 | #22473 : remove the patchbot optional spkg. |
537849f | trac 20736 first try, failing |
b423b62 | Improve --patchbot command |
archive/issue_comments_300469.json:
{
"body": "Changed commit from **[`d7c83f1`](https://github.com/sagemath/sagetrac-mirror/commit/d7c83f1b06d1d851ef012acdf99a105cbe9db3ac)** to **[`b423b62`](https://github.com/sagemath/sagetrac-mirror/commit/b423b62329b9d028a8a6de11def5fbe1eae24519)**",
"created_at": "2017-03-01T11:00:45Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300469",
"user": "https://github.com/sagetrac-git"
}
Changed commit from d7c83f1
to b423b62
archive/issue_comments_300470.json:
{
"body": "<div id=\"comment:156\" align=\"right\">comment:156</div>\n\nRebased to #22473.",
"created_at": "2017-03-01T11:02:48Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300470",
"user": "https://github.com/jdemeyer"
}
Rebased to #22473.
archive/issue_comments_300471.json:
{
"body": "<div id=\"comment:157\"></div>\n\nBranch pushed to git repo; I updated commit sha1. This was a forced push. New commits:\n<table><tr><td><a href=\"https://github.com/sagemath/sagetrac-mirror/commit/c9c4101b961c454b2acac23508a74ccfbcffb7c4\"><code>c9c4101</code></a></td><td><code>Improve --patchbot command</code></td></tr></table>\n",
"created_at": "2017-03-01T11:04:21Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300471",
"user": "https://github.com/sagetrac-git"
}
Branch pushed to git repo; I updated commit sha1. This was a forced push. New commits:
c9c4101 | Improve --patchbot command |
archive/issue_comments_300472.json:
{
"body": "Changed commit from **[`b423b62`](https://github.com/sagemath/sagetrac-mirror/commit/b423b62329b9d028a8a6de11def5fbe1eae24519)** to **[`c9c4101`](https://github.com/sagemath/sagetrac-mirror/commit/c9c4101b961c454b2acac23508a74ccfbcffb7c4)**",
"created_at": "2017-03-01T11:04:21Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300472",
"user": "https://github.com/sagetrac-git"
}
Changed commit from b423b62
to c9c4101
archive/issue_comments_300473.json:
{
"body": "<div id=\"comment:158\" align=\"right\">comment:158</div>\n\nLet's start over...",
"created_at": "2017-03-01T11:15:29Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300473",
"user": "https://github.com/jdemeyer"
}
Let's start over...
archive/issue_events_289411.json:
{
"actor": "https://github.com/jdemeyer",
"created_at": "2017-03-01T11:15:29Z",
"event": "unlabeled",
"issue": "https://github.com/sagemath/sage/issues/20736",
"label": "https://github.com/sagemath/sage/labels/needs%20work",
"label_color": "ffff00",
"label_name": "needs work",
"label_text_color": "ffffff",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289411"
}
archive/issue_events_289412.json:
{
"actor": "https://github.com/jdemeyer",
"created_at": "2017-03-01T11:15:29Z",
"event": "labeled",
"issue": "https://github.com/sagemath/sage/issues/20736",
"label": "https://github.com/sagemath/sage/labels/needs%20review",
"label_color": "7fff00",
"label_name": "needs review",
"label_text_color": "ffffff",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289412"
}
archive/issue_comments_300474.json:
{
"body": "<div id=\"comment:159\" align=\"right\">comment:159</div>\n\nReplying to [@jdemeyer](#comment%3A154):\n> Replying to [@embray](#comment%3A150):\n> > I don't know why we're talking about `python -m` anything. When you install the `sage-patchbot` Python package (which Fr\u00e9d\u00e9ric did a great job putting together at my urging) it installs a `patchbot` script.\n> \n> \n> ...in a directory which might not be the default `$PATH` (especially in the case of a `--user` install). So there are cases where `python -m sage_patchbot.patchbot` works but running `patchbot` does not.\n\nThat would imply someone has their `~/.local/lib/pythonX.Y/site-packages` on `sys.path`, but not `~/.local/bin` in their `$PATH`. That can certainly happen (I think it is the default in most cases) but I would consider it a slight misconfiguration. It is also not a problem at all if using virtualenvs instead.",
"created_at": "2017-03-03T13:03:15Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300474",
"user": "https://github.com/embray"
}
Replying to @jdemeyer:
Replying to @embray:
I don't know why we're talking about
python -m
anything. When you install thesage-patchbot
Python package (which Frédéric did a great job putting together at my urging) it installs apatchbot
script....in a directory which might not be the default
$PATH
(especially in the case of a--user
install). So there are cases wherepython -m sage_patchbot.patchbot
works but runningpatchbot
does not.
That would imply someone has their ~/.local/lib/pythonX.Y/site-packages
on sys.path
, but not ~/.local/bin
in their $PATH
. That can certainly happen (I think it is the default in most cases) but I would consider it a slight misconfiguration. It is also not a problem at all if using virtualenvs instead.
archive/issue_events_289413.json:
{
"actor": "https://github.com/embray",
"created_at": "2017-03-03T13:05:06Z",
"event": "unlabeled",
"issue": "https://github.com/sagemath/sage/issues/20736",
"label": "https://github.com/sagemath/sage/labels/needs%20review",
"label_color": "7fff00",
"label_name": "needs review",
"label_text_color": "ffffff",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289413"
}
archive/issue_events_289414.json:
{
"actor": "https://github.com/embray",
"created_at": "2017-03-03T13:05:06Z",
"event": "labeled",
"issue": "https://github.com/sagemath/sage/issues/20736",
"label": "https://github.com/sagemath/sage/labels/positive%20review",
"label_color": "dfffc0",
"label_name": "positive review",
"label_text_color": "ffffff",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289414"
}
archive/issue_comments_300475.json:
{
"body": "<div id=\"comment:161\" align=\"right\">comment:161</div>\n\nI wonder what made you change your mind...\n\nThis code (which has been there since [comment:101]) has been attacked by several people for 6 months and now it suddenly has positive_review?",
"created_at": "2017-03-03T13:52:41Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300475",
"user": "https://github.com/jdemeyer"
}
I wonder what made you change your mind...
This code (which has been there since [comment:101]) has been attacked by several people for 6 months and now it suddenly has positive_review?
archive/issue_comments_300476.json:
{
"body": "<div id=\"comment:162\" align=\"right\">comment:162</div>\n\nMy feeling at this point is along the lines of \"I'm not going to use it, but if it suits your needs then fine.\" :)",
"created_at": "2017-03-03T14:31:22Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300476",
"user": "https://github.com/embray"
}
My feeling at this point is along the lines of "I'm not going to use it, but if it suits your needs then fine." :)
archive/issue_comments_300477.json:
{
"body": "Changed branch from **[public/20736](https://github.com/sagemath/sagetrac-mirror/tree/public/20736)** to **[`c9c4101`](https://github.com/sagemath/sagetrac-mirror/commit/c9c4101b961c454b2acac23508a74ccfbcffb7c4)**",
"created_at": "2017-03-05T15:13:53Z",
"formatter": "markdown",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_comment",
"url": "https://github.com/sagemath/sage/issues/20736#issuecomment-300477",
"user": "https://github.com/vbraun"
}
Changed branch from public/20736 to c9c4101
archive/issue_events_289415.json:
{
"actor": "https://github.com/vbraun",
"created_at": "2017-03-05T15:13:53Z",
"event": "unlabeled",
"issue": "https://github.com/sagemath/sage/issues/20736",
"label": "https://github.com/sagemath/sage/labels/positive%20review",
"label_color": "dfffc0",
"label_name": "positive review",
"label_text_color": "ffffff",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289415"
}
archive/issue_events_289416.json:
{
"actor": "https://github.com/vbraun",
"commit_id": "7feb1fbe284c9ae034dc72167cf37cf994042cdd",
"commit_repository": "https://github.com/sagemath/sage",
"created_at": "2017-03-05T15:13:53Z",
"event": "closed",
"issue": "https://github.com/sagemath/sage/issues/20736",
"type": "issue_event",
"url": "https://github.com/sagemath/sage/issues/20736#event-289416"
}