You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
<?xml version='1.0' encoding='UTF-8'?>
<rssxmlns:atom="http://www.w3.org/2005/Atom"xmlns:content="http://purl.org/rss/1.0/modules/content/"version="2.0">
<channel>
<title>Newest Python PEPs</title>
<link>https://peps.python.org/peps.rss</link>
<description>Newest Python Enhancement Proposals (PEPs) - Information on new language features, and some meta-information like release procedure and schedules.</description>
<atom:linkhref="https://peps.python.org/peps.rss"rel="self"/>
<docs>https://cyber.harvard.edu/rss/rss.html</docs>
<language>en</language>
<lastBuildDate>Thu, 10 Mar 2022 08:28:04 GMT</lastBuildDate>
</channel>
</rss>
I would prioritise fixing this in the new PEP build system.
A quick look: the main problem is in generate_rss.py:
defmain():
# get the directory with the PEP sourcesout_dir=Path(__file__).parent/"build"# get list of peps with creation time (from "Created:" string in pep source)peps_with_dt=sorted((pep_creation(path), path) forpathinout_dir.glob("pep-????.*"))
peps_with_dt is empty because there pep-????.*" matches no files in build/; it now contains directories of files like build/pep-0008/index.html.
It's also expecting to run on source rst/txt files containing Created: lines, not target html files like:
I've also been noticing a bunch of error messages (at least on the CI) when running it related to stuff it can't parse, but I'm not sure if those have anything to do with the problem reported here.
I've also been noticing a bunch of error messages (at least on the CI) when running it related to stuff it can't parse, but I'm not sure if those have anything to do with the problem reported here.
Like this?
$ python3 generate_rss.py<rst-doc>:556: (ERROR/3) PEP number must be a number from 0 to 9999; "PEP 586 mentions<586#backwards-compatibility>" is invalid.<rst-doc>:4: (ERROR/3) Unexpected indentation.<rst-doc>:5: (WARNING/2) Block quote ends without a blank line; unexpected unindent.<rst-doc>:195: (ERROR/3) PEP number must be a number from 0 to 9999; "PEP 557 <557#id7>" is invalid.<rst-doc>:527: (ERROR/3) PEP number must be a number from 0 to 9999; "557 <557#inheritance>" is invalid.<rst-doc>:532: (ERROR/3) PEP number must be a number from 0 to 9999; "PEP 557 indicates <557#post-init-parameters>" is invalid.<rst-doc>:4: (ERROR/3) Unexpected indentation.<rst-doc>:5: (WARNING/2) Block quote ends without a blank line; unexpected unindent.<rst-doc>:1202: (ERROR/3) PEP number must be a number from 0 to 9999; "PEP 484 specifies<484#suggested-syntax-for-python-2-7-and-straddling-code>" is invalid.
Same stuff with the old one:
$ python3 pep2rss.py .<rst-doc>:4: (ERROR/3) Unexpected indentation.<rst-doc>:5: (WARNING/2) Block quote ends without a blank line; unexpected unindent.<rst-doc>:1202: (ERROR/3) PEP number must be a number from 0 to 9999; "PEP 484 specifies<484#suggested-syntax-for-python-2-7-and-straddling-code>" is invalid.<rst-doc>:4: (ERROR/3) Unexpected indentation.<rst-doc>:5: (WARNING/2) Block quote ends without a blank line; unexpected unindent.<rst-doc>:195: (ERROR/3) PEP number must be a number from 0 to 9999; "PEP 557 <557#id7>" is invalid.<rst-doc>:527: (ERROR/3) PEP number must be a number from 0 to 9999; "557 <557#inheritance>" is invalid.<rst-doc>:532: (ERROR/3) PEP number must be a number from 0 to 9999; "PEP 557 indicates <557#post-init-parameters>" is invalid.<rst-doc>:556: (ERROR/3) PEP number must be a number from 0 to 9999; "PEP 586 mentions<586#backwards-compatibility>" is invalid.
Sure! It's coming from the docutils RST parser; some warnings/errors may be invalid, but some might be things to fix in our code or the PEPs themselves. (I'm not sure if Sphinx uses the same parser.)
Also, we're parsing the 10 most recent PEPs here, which will change over time.
Activity
hugovk commentedon Mar 10, 2022
For reference https://peps.python.org/peps.rss shows only:
Jackenmen commentedon Mar 10, 2022
Oh ye, this is what I've shown on the screenshot, I should have copied that instead, thanks!
hugovk commentedon Mar 10, 2022
I would prioritise fixing this in the new PEP build system.
A quick look: the main problem is in
generate_rss.py:peps_with_dtis empty because therepep-????.*"matches no files inbuild/; it now contains directories of files likebuild/pep-0008/index.html.It's also expecting to run on source rst/txt files containing
Created:lines, not target html files like:(Also,
make pagesrunsmake rssbefore generating the target html files, so build is actually empty, but it's expecting source rst/txt files.)CAM-Gerlach commentedon Mar 10, 2022
I've also been noticing a bunch of error messages (at least on the CI) when running it related to stuff it can't parse, but I'm not sure if those have anything to do with the problem reported here.
hugovk commentedon Mar 10, 2022
Please see PR #2407 for a fix.
Like this?
Same stuff with the old one:
CAM-Gerlach commentedon Mar 11, 2022
I see, thanks, so its not a regression. Should we open another issue for that?
hugovk commentedon Mar 11, 2022
Sure! It's coming from the docutils RST parser; some warnings/errors may be invalid, but some might be things to fix in our code or the PEPs themselves. (I'm not sure if Sphinx uses the same parser.)
Also, we're parsing the 10 most recent PEPs here, which will change over time.