summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
...
* | | | | Merge pull request #9802 from sphinx-doc/bot/pull-translationsTakeshi KOMIYA2021-10-3125-52/+52
|\ \ \ \ \ | | | | | | | | | | | | Update message catalogs
| * | | | | Update message catalogstk0miya2021-10-3125-52/+52
|/ / / / /
* | | | | Merge pull request #9801 from tk0miya/docutils-0.18_traverse_2Takeshi KOMIYA2021-10-311-15/+15
|\ \ \ \ \ | | | | | | | | | | | | Support docutils-0.18: Consume generator of Element.traverse()
| * | | | | Support docutils-0.18: Consume generator of Element.traverse()Takeshi KOMIYA2021-10-311-15/+15
|/ / / / / | | | | | | | | | | | | | | | | | | | | The last fix in i18n module was incorrect because it compares the "already consumed" generators. It should compares the lists of nodes.
* | | | | Update CHANGES for PR #9772Takeshi KOMIYA2021-10-311-0/+2
| | | | |
* | | | | Merge pull request #9772 from gibsondan/loggingflushTakeshi KOMIYA2021-10-311-0/+5
|\ \ \ \ \ | |_|_|/ / |/| | | | Closes #9733: Fix for logging handler flushing warnings in the middle of the docs build
| * | | | Closes #9733: Fix for logging handler flushing warnings in the middle of the ↵gibsondan2021-10-251-0/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | build Summary: My project was mysteriously dropping warnings (see https://github.com/sphinx-doc/sphinx/issues/9733 for detailed repro) and I realized that it's becaues it imports libraries like airflow or mlflow that set up loggers automatically when they are imported. This causes this handler to flush even though shouldFlush is set to always return False. A simple workaround is to override flush to be a no-op. Test Plan: Repeat repro steps from #9733 - project now always includes warnings
* | | | | Merge pull request #9797 from tk0miya/9757_inherited_classmethodsTakeshi KOMIYA2021-10-313-1/+24
|\ \ \ \ \ | |_|/ / / |/| | | | Fix #9757: autodoc_inherit_docstrings does not effect to overriden classmethods
| * | | | Fix #9757: autodoc_inherit_docstrings does not effect to overriden classmethodsTakeshi KOMIYA2021-10-303-1/+24
|/ / / /
* | | | Merge pull request #9787 from dmares01/chocolatey-install-linkTakeshi KOMIYA2021-10-281-1/+1
|\ \ \ \ | | | | | | | | | | Update Chocolatey Install Link
| * | | | Update Chocolatey Install LinkDrew Mares2021-10-271-1/+1
|/ / / / | | | | | | | | The install URL returned a 404 but now it is updated to the new location
* | | | Merge pull request #9760 from tk0miya/9756_classmethod_not_having_funcTakeshi KOMIYA2021-10-272-1/+2
|\ \ \ \ | | | | | | | | | | Fix #9756: autodoc: Crashed if classmethod does not have __func__ attribute
| * \ \ \ Merge branch '4.x' into 9756_classmethod_not_having_funcTakeshi KOMIYA2021-10-27156-990/+1021
| |\ \ \ \ | |/ / / / |/| | | |
* | | | | Merge pull request #9769 from francoisfreitag/fix-linkcheck-auth-linkTakeshi KOMIYA2021-10-271-4/+2
|\ \ \ \ \ | | | | | | | | | | | | Fix linkcheck_auth link to Requests authentication
| * | | | | Fix linkcheck_auth link to Requests authenticationFrançois Freitag2021-10-251-4/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The link was directing to https://www.sphinx-doc.org/en/master/usage/requests-auth>. Prefer using the intersphinx module to generate the link, it’s more robust than directly linking to the page.
* | | | | | Merge pull request #9764 from tk0miya/9752_annotated_slots_attributeTakeshi KOMIYA2021-10-277-3/+22
|\ \ \ \ \ \ | | | | | | | | | | | | | | Fix #9752: autodoc: Failed to detect type annotation for slots attribute
| * | | | | | Fix #9752: autodoc: Failed to detect type annotation for slots attributeTakeshi KOMIYA2021-10-237-3/+22
| | | | | | |
* | | | | | | Merge pull request #9780 from tk0miya/docutils-0.18_traverseTakeshi KOMIYA2021-10-2715-27/+27
|\ \ \ \ \ \ \ | | | | | | | | | | | | | | | | Support docutils-0.18: Consume iterator of Element.traverse()
| * | | | | | | Support docutils-0.18: Consume iterator of Element.traverse()Takeshi KOMIYA2021-10-2715-27/+27
|/ / / / / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Since 0.18, Element.traverse() returns an iterator instead of intermediate object. As a result, the return value is always considered as truthy value. And it becomes fragile when the caller modifies the doctree on the loop.
* | | | | | | Merge pull request #9774 from sphinx-doc/bot/pull-translationsTakeshi KOMIYA2021-10-26125-945/+945
|\ \ \ \ \ \ \ | | | | | | | | | | | | | | | | Update message catalogs
| * | | | | | | Update message catalogstk0miya2021-10-25125-945/+945
|/ / / / / / /
* | | | | | | Merge pull request #9768 from tk0miya/fix_transifex_clientTakeshi KOMIYA2021-10-261-0/+2
|\ \ \ \ \ \ \ | |_|/ / / / / |/| | | | | | Fix CI: Downgrade python for transifex CI
| * | | | | | Fix CI: Downgrade python for transifex CITakeshi KOMIYA2021-10-251-0/+2
| | | | | | |
* | | | | | | Merge pull request #9767 from tk0miya/fix_transifex_clientTakeshi KOMIYA2021-10-251-0/+2
|\ \ \ \ \ \ \ | |/ / / / / / | | | | | | | Fix CI: Downgrade python for transifex CI
| * | | | | | Fix CI: Downgrade python for transifex CITakeshi KOMIYA2021-10-251-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The latest transifex-client could not be installed on python 3.10 environment. This downgrade python to 3.9 to be install the latest one. ref: https://github.com/transifex/transifex-client/pull/330
* | | | | | | Merge pull request #9765 from mitya57/typoTakeshi KOMIYA2021-10-242-4/+4
|\ \ \ \ \ \ \ | |/ / / / / / |/| | | | | | Fix a typo in variable name
| * | | | | | Fix a typo in variable nameDmitry Shachnev2021-10-232-4/+4
| | |_|/ / / | |/| | | |
* | | | | | Update CHANGES for PR #9737Takeshi KOMIYA2021-10-231-0/+1
| | | | | |
* | | | | | Merge pull request #9737 from ahippo/latex-aggedrightTakeshi KOMIYA2021-10-231-1/+1
|\ \ \ \ \ \ | | | | | | | | | | | | | | LaTeX: fix '\raggedright' escaping causing "aggedright" text
| * | | | | | LaTeX: fix '\raggedright' escaping causing "aggedright" textAndrey Mazo2021-10-151-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Sphinx version 4.0 introduced a bug in handling hlists in its LaTeX backend. Due to improper backslash escaping, LaTeX "\raggedright" command gets written as Carriage Return character (0x0D) followed by "aggedright". This results in stray "aggedright" text appearing in the resulting PDF and lack of effect \raggedright was supposed to achieve. Fix this by converting the remaining string to a raw string. This appears to be the only occurrence of such a missing escaping based on a quick grep. Fixes #9734. Fixes: 20884bb0c9f7: "refactor: LaTeX: Use raw strings for LaTeX macros"
* | | | | | | Update CHANGES for PR #9685Takeshi KOMIYA2021-10-231-0/+1
| | | | | | |
* | | | | | | Merge pull request #9685 from MatthijsBurgh/patch-1Takeshi KOMIYA2021-10-231-1/+2
|\ \ \ \ \ \ \ | |_|_|/ / / / |/| | | | | | Set refdoc to be used by missing-reference, intersphinx
| * | | | | | Set refdoc default to pending_xref nodes to be used by missing-referenceMatthijs van der Burgh2021-10-101-1/+2
| |/ / / / /
* | | | | | Update CHANGES for PR #9763Takeshi KOMIYA2021-10-231-0/+1
| | | | | |
* | | | | | Merge pull request #9763 from QuLogic/fix-parameter-breaksTakeshi KOMIYA2021-10-231-1/+2
|\ \ \ \ \ \ | | | | | | | | | | | | | | Fix selection of parameter names in HTML theme
| * | | | | | Fix selection of parameter names in HTML themeElliott Sales de Andrade2021-10-221-1/+2
|/ / / / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Parameters are written in HTML as (leaving out some internal classes): ``` <strong>name</strong><span class="classifier"><a><code><span>TypeName</span></code></a></span> ``` but in rendered form there's a colon between the name and type. This colon is inserted virtually using CSS, but since it doesn't exist, the browser thinks both sides are part of the same word. Styling the virtual text as inline block makes it be treated as a break, but also makes it apply vertical margins, so we need to set those to zero again. See https://github.com/matplotlib/matplotlib/issues/21432
* | | | | | Merge pull request #9724 from tk0miya/support_PreBibliographic_nodesTakeshi KOMIYA2021-10-231-3/+6
|\ \ \ \ \ \ | | | | | | | | | | | | | | Support docutils-0.18: allow PreBibliographic nodes before docinfo
| * | | | | | Support docutils-0.18: allow PreBibliographic nodes before docinfoTakeshi KOMIYA2021-10-231-3/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Since 0.18, `meta` directive inserts meta node into the top of the document. It confuses MetadataCollector. This allows doctree contains PreBibliographic nodes just before docinfo node.
* | | | | | | Merge pull request #9725 from tk0miya/set_auto_id_prefix_explicitlyTakeshi KOMIYA2021-10-231-0/+1
|\ \ \ \ \ \ \ | | | | | | | | | | | | | | | | Support docutils-0.18: Set auto_id_prefix explicitly
| * | | | | | | Support docutils-0.18: Set auto_id_prefix explicitlyTakeshi KOMIYA2021-10-111-0/+1
| |/ / / / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Since docutils-0.18, auto_id_prefix setting will be changed to `'%'` from `'id'`. To keep backward compatibility of node IDs, this sets `'id'` to settings explicitly.
* | | | | | | Update CHANGES for PR #9755Takeshi KOMIYA2021-10-231-0/+1
| | | | | | |
* | | | | | | Make util.typing.restify sanitise unreproducible output (eg. memory addresses)Chris Lamb2021-10-231-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Whilst working on the Reproducible Builds effort [0] I noticed that sphinx generates output that is not reproducible, causing a number of packages in Debian to unreproducible. Specifically, when Sphinx locates an alias of an instance when generating 'autodoc' documentation, it uses the raw Python repr(...) of the object and does not sanitise it for memory addresses (etc.) like elsewhere in Sphinx. This can result in documentation like this: -<dd><p>alias of &lt;webob.client.SendRequest object at 0x7fd769189df0&gt;</p> +<dd><p>alias of &lt;webob.client.SendRequest object at 0x7f0f02233df0&gt;</p> Patch attached that uses the object_description method, which was added to fix precisely this kind of issue. I originally filed this in Debian as bug #996948 [1]. [0] https://reproducible-builds.org/ [1] https://bugs.debian.org/996948
| | | | * | | Fix #9756: autodoc: Crashed if classmethod does not have __func__ attributeTakeshi KOMIYA2021-10-232-1/+2
| |_|_|/ / / |/| | | | |
* | | | | | Merge pull request #9745 from sphinx-doc/bot/pull-translationsTakeshi KOMIYA2021-10-2169-159/+159
|\ \ \ \ \ \ | | | | | | | | | | | | | | Update message catalogs
| * | | | | | Update message catalogstk0miya2021-10-1769-159/+159
| | |/ / / / | |/| | | |
* | | | | | Merge pull request #9753 from Marsfan/patch-1Takeshi KOMIYA2021-10-211-1/+1
|\ \ \ \ \ \ | |/ / / / / |/| | | | | Add missing space in intersphinx doc
| * | | | | Add missing space in intersphinx docGabe R2021-10-191-1/+1
|/ / / / / | | | | | | | | | | There was a space missing between `-m` and `sphinx.ext.intersphinx` in the example of showing all mappings in a intersphinx file. I added it in.
* | | | | Merge pull request #9719 from tk0miya/9718_intersphinxTakeshi KOMIYA2021-10-101-2/+2
|\ \ \ \ \ | | | | | | | | | | | | Fix #9718: intersphinx: Failed to invoke sphinx.ext.intersphinx module
| * | | | | Fix #9718: intersphinx: Failed to invoke sphinx.ext.intersphinx moduleTakeshi KOMIYA2021-10-091-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | A circular import was happened since #9701 merged. This goes lazy-import to avoid the error.
* | | | | | Merge pull request #9717 from tk0miya/9710_even_odd_rowsTakeshi KOMIYA2021-10-104-17/+83
|\ \ \ \ \ \ | | | | | | | | | | | | | | Fix #9710: html: Wrong styles for even/odd rows in nested tables