From skip.montanaro at gmail.com Wed Jan 10 09:24:45 2018 From: skip.montanaro at gmail.com (Skip Montanaro) Date: Wed, 10 Jan 2018 08:24:45 -0600 Subject: [code-quality] A couple pylint feature requests Message-ID: As I work through code, especially while refactoring, I wish pylint was at times more stringent about lack of use of certain types of objects. At the moment, I'm thinking of two scenarios, main programs and private attributes. There's a clear difference between a program and a module, the main one being that a module "exports" its module-level symbols, while a main program doesn't (at least if it isn't intended to also serve as a module). It would be nice if there was a pylint checker which indicated for main programs module-level symbols which aren't used somewhere within that file. It might be off by default, but be enabled if you passed a --main flag on the command line, or if the the usual "if __name__" dance was found at the bottom of the file. In a similar fashion, I think it would be useful if private attributes (typically functions or methods, but also data attributes) were called out in any environment where they were defined but not used. It's considered bad practice to reach into an object and reference private attributes. Conversely, if a private attribute isn't "exported," it should almost certainly be used with that namespace. Skip From atodorov at mrsenko.com Tue Jan 16 10:54:51 2018 From: atodorov at mrsenko.com (Alexander Todorov) Date: Tue, 16 Jan 2018 17:54:51 +0200 Subject: [code-quality] Need help with debugging pylint tests for a plugin Message-ID: Hi folks, I've made several commits to django-pylint (https://github.com/MrSenko/django-pylint, branch next_release) in order to make it compatible with the latest available versions. However some of the existing tests fail and I'm not sure how to figure out what's wrong: https://travis-ci.org/MrSenko/django-pylint/jobs/329526377 For example the failure about file "func_noerror_ignore_meta_subclass.py" reports: Unexpected in testdata: 11: too-few-public-methods At the same time if I execute pylint + my plugin from the command line I don't get any errors reported: $ PYTHONPATH=. pylint --load-plugins pylint_django test/input/func_noerror_ignore_meta_subclass.py No config file found, using default configuration -------------------------------------------------------------------- Your code has been rated at 10.00/10 (previous run: 10.00/10, +0.00) Without going into much details about the code of pylint_django there's a function which is supposed to suppress this message if it is triggered by a class Meta: definition (standard thing for Django). When running manually from the command line this error message is suppressed. When running from tests (via pytest) the suppression doesn't seem to work. Any ideas on how to debug further ? Thanks, Alex From martmists at gmail.com Fri Jan 19 12:41:17 2018 From: martmists at gmail.com (Martmists) Date: Fri, 19 Jan 2018 18:41:17 +0100 Subject: [code-quality] SnekChek - Python Linter wrapper Message-ID: Hello members of PyCQA, I am martmists, current owner of the IzunaDevs organization, and I'm currently working on a library called SnekChek (which you can find here ) I was wondering if you were interested in taking over this package to your repository with me as main contributor, as I will keep updating this package as time goes by (unlike coala or ciocheck) while also supporting as many linters/checkers as possible. I'm also interested in what kind of input you guys have for me, as this is the first time I'm writing a wrapper for these things and I'm not sure if I'm doing it the correct way. I look forward to hearing your response. ~ Martmists -------------- next part -------------- An HTML attachment was scrubbed... URL: From graffatcolmingov at gmail.com Sat Jan 20 10:39:29 2018 From: graffatcolmingov at gmail.com (Ian Stapleton Cordasco) Date: Sat, 20 Jan 2018 09:39:29 -0600 Subject: [code-quality] SnekChek - Python Linter wrapper In-Reply-To: References: Message-ID: Hi there martmists, The PyCQA is generally open to accepting actively maintained and used projects. It seems that snekcheck is still in the early stages of development and there don't seem to be many contributors at the moment. I'd like to understand what value you would see in moving it to the PyCQA? As for input, it looks as if you're using an explicitly private API in Flake8 to wrap that utility. It's likely this will cause lots of pain for you and your users as that changes. You might want to contribute to Flake8's discussion of the design of a stable public API. Sent from my phone with my typo-happy thumbs. Please excuse my brevity On Jan 19, 2018 19:14, "Martmists" wrote: Hello members of PyCQA, I am martmists, current owner of the IzunaDevs organization, and I'm currently working on a library called SnekChek (which you can find here ) I was wondering if you were interested in taking over this package to your repository with me as main contributor, as I will keep updating this package as time goes by (unlike coala or ciocheck) while also supporting as many linters/checkers as possible. I'm also interested in what kind of input you guys have for me, as this is the first time I'm writing a wrapper for these things and I'm not sure if I'm doing it the correct way. I look forward to hearing your response. ~ Martmists _______________________________________________ code-quality mailing list code-quality at python.org https://mail.python.org/mailman/listinfo/code-quality -------------- next part -------------- An HTML attachment was scrubbed... URL: From carl.crowder at gmail.com Mon Jan 22 17:22:12 2018 From: carl.crowder at gmail.com (Carl Crowder) Date: Mon, 22 Jan 2018 14:22:12 -0800 Subject: [code-quality] pylint-django move to PyQCA? Message-ID: Hi all, I?ll be quick - I think pylint-django is better under PyCQA! If you agree I will help out wherever necessary I built it out of a hope I could help bring pylint to Django users so I hope that has helped, people seem to use it so I hope moving it to PyCQA helps users of it too! Carl -------------- next part -------------- An HTML attachment was scrubbed... URL: From pcmanticore at gmail.com Tue Jan 23 03:19:19 2018 From: pcmanticore at gmail.com (Claudiu Popa) Date: Tue, 23 Jan 2018 09:19:19 +0100 Subject: [code-quality] pylint-django move to PyQCA? In-Reply-To: References: Message-ID: Hi Carl, This sounds like a great idea! And it seems it was already moved last night :) From bashtanov at imap.cc Wed Jan 24 06:14:13 2018 From: bashtanov at imap.cc (Alexey Bashtanov) Date: Wed, 24 Jan 2018 11:14:13 +0000 Subject: [code-quality] pylint import succeeds or fails depending on a previous import statement Message-ID: <27c81560-26f1-53cd-8e8c-67840dbaf3a4@imap.cc> Hello, I came across the following problem with pylint: a import from a local directory works only if there's some `from ... import ...` statement above. My code is organized as follows: ~/pylinttest/ foo/ ??????? bar.py ??????? baz.py ??????? qux.py ??????? __init__.py ??? .pylintrc bar.py contains the following code: from urllib import request import qux baz.py contains the following code: import qux Here are the results of running pylint: alexey at ev9d9:~/pylinttest$ pylint foo.bar Using config file /home/alexey/pylinttest/.pylintrc ************* Module foo.bar W:? 1, 0: Unused request imported from urllib (unused-import) W:? 2, 0: Unused import qux (unused-import) alexey at ev9d9:~/pylinttest$ pylint foo.baz Using config file /home/alexey/pylinttest/.pylintrc ************* Module foo.baz E:? 1, 0: Unable to import 'qux' (import-error) W:? 1, 0: Unused import qux (unused-import) Why does it fail to load qux only if there was no from-import statement before? Changing urllib.request to e.g. collections.deque does not make a difference. My pylint version is 1.1.0, and it can be reproduced both with python 2.7.5 and 3.4.0 Regards, ? Alexey From bashtanov at imap.cc Wed Jan 24 12:20:54 2018 From: bashtanov at imap.cc (Alexey Bashtanov) Date: Wed, 24 Jan 2018 17:20:54 +0000 Subject: [code-quality] pylint import succeeds or fails depending on a previous import statement In-Reply-To: <27c81560-26f1-53cd-8e8c-67840dbaf3a4@imap.cc> References: <27c81560-26f1-53cd-8e8c-67840dbaf3a4@imap.cc> Message-ID: <563690f5-1e3a-21f6-0dd5-47f343760753@imap.cc> > My pylint version is 1.1.0, and it can be reproduced both with python > 2.7.5 and 3.4.0 pylint 1.8.1 and python 3.4.3 are affected as well From quantum.mechanic.1964 at gmail.com Fri Jan 26 07:05:23 2018 From: quantum.mechanic.1964 at gmail.com (Quantum Mechanic) Date: Fri, 26 Jan 2018 12:05:23 +0000 Subject: [code-quality] Pylint hosting on ReadTheDocs prevents successful searches Message-ID: Googling for "pylint pylintrc" gives me the first hit as https://docs.pylint.org/en/1.6.0/run.html. This lands me on https://docs.pylint.org/en/1.6.0/run.html, asking me to login. I have a login, and clicking "Login" sends me to https://readthedocs.org/dashboard/, spoiling the google search. In fact, searching on google only shows me that something might be there. I have to search *again* on ReadTheDocs, which isn't friendly at all. Poking around until I find a search box at https://pylint.readthedocs.io/en/latest/, I type in "pylintrc", which returns the *same bloody page* (OK, yes, it has this helpful bit: https://pylint.readthedocs.io/en/latest/?highlight=pylintrc, which does exactly...nothing.) So I'm stuck trolling around in the full docs there. Instead, I should just pull down the source code, and do my own grep -r. I feel like I've been violated. -- -QM Quantum Mechanics: The dreams stuff is made of -------------- next part -------------- An HTML attachment was scrubbed... URL: From graffatcolmingov at gmail.com Fri Jan 26 10:15:43 2018 From: graffatcolmingov at gmail.com (Ian Stapleton Cordasco) Date: Fri, 26 Jan 2018 09:15:43 -0600 Subject: [code-quality] Pylint hosting on ReadTheDocs prevents successful searches In-Reply-To: References: Message-ID: On Fri, Jan 26, 2018 at 6:05 AM, Quantum Mechanic wrote: > Googling for "pylint pylintrc" gives me the first hit as > https://docs.pylint.org/en/1.6.0/run.html. > > This lands me on https://docs.pylint.org/en/1.6.0/run.html, asking me to > login. I have a login, and clicking "Login" sends me to > https://readthedocs.org/dashboard/, spoiling the google search. I was not asked to log into anything on that site. You'll need to provide a whole lot more detail. > In fact, searching on google only shows me that something might be there. I > have to search *again* on ReadTheDocs, which isn't friendly at all. > > Poking around until I find a search box at > https://pylint.readthedocs.io/en/latest/, I type in "pylintrc", which > returns the *same bloody page* (OK, yes, it has this helpful bit: > https://pylint.readthedocs.io/en/latest/?highlight=pylintrc, which does > exactly...nothing.) > > So I'm stuck trolling around in the full docs there. Instead, I should just > pull down the source code, and do my own grep -r. > > I feel like I've been violated. This sounds like an attempt to be humorous at the expense of people who have actually been physically violated. That isn't acceptable on this list and is not tolerated by any mailing list hosted on python.org. Please refrain from this in the future. From marc at rintsch.de Fri Jan 26 23:49:24 2018 From: marc at rintsch.de (Marc 'BlackJack' Rintsch) Date: Sat, 27 Jan 2018 05:49:24 +0100 Subject: [code-quality] Pylint hosting on ReadTheDocs prevents successful searches In-Reply-To: References: Message-ID: <81c05e34-2b08-ee3c-41ab-5781da42ed6a@rintsch.de> On 26/01/18 13:05, Quantum Mechanic wrote: > Googling for "pylint pylintrc" gives me the first hit > as https://docs.pylint.org/en/1.6.0/run.html. > > This lands me on https://docs.pylint.org/en/1.6.0/run.html, asking me to > login. I have a login, and clicking "Login" sends me > to https://readthedocs.org/dashboard/, spoiling the google search. This page doesn't ask me for login. > In fact, searching on google only shows me that something might be > there. I have to search *again* on ReadTheDocs, which isn't friendly at all. This is something you should complain about to Google. That bloody thing, as good is it might be, is still not that good at mind reading, and presenting you the exact information you want, and only that information, even when you provide ample information about what exactly you are after, like in your case with even *two* words. ;-) > Poking around until I find a search box > at https://pylint.readthedocs.io/en/latest/, I type in "pylintrc", which > returns the *same bloody page* (OK, yes, it has this helpful > bit: https://pylint.readthedocs.io/en/latest/?highlight=pylintrc, which > does exactly...nothing.) This also doesn't happen for me. I get a page with search results (long link, might be broken into more than one line in Email): https://pylint.readthedocs.io/en/latest/search.html?q=pylintrc&check_keywords=yes&area=default Ciao, Marc 'BlackJack' Rintsch -- Lottery: A tax on people who are bad at math. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: OpenPGP digital signature URL: From steve at pearwood.info Fri Jan 26 11:14:34 2018 From: steve at pearwood.info (Steven D'Aprano) Date: Sat, 27 Jan 2018 03:14:34 +1100 Subject: [code-quality] Pylint hosting on ReadTheDocs prevents successful searches In-Reply-To: References: Message-ID: <20180126161434.GB22500@ando.pearwood.info> On Fri, Jan 26, 2018 at 09:15:43AM -0600, Ian Stapleton Cordasco wrote: > This sounds like an attempt to be humorous at the expense of people > who have actually been physically violated. That isn't acceptable on > this list and is not tolerated by any mailing list hosted on > python.org. Please refrain from this in the future. Ian, you are gate-keeping and telling the poster how he or she should feel. Please don't do that, it is a hostile and aggressive thing to do. You have no right to tell the OP that he or she *shouldn't* feel violated by an experience which he clearly experienced as frustrating and painful. Who are you to say that (s)he didn't, or shouldn't, have felt violated? And before you dismiss me (as I expect your initial reaction will be) both my wife and I are "people who have actually been physically violated". Your defence of our feelings, while well-intentioned, is neither needed nor wanted. I'm sure you thought you were doing the right thing, but we don't need you to act as gate-keeper telling us when we are being made fun of. We can decide for ourselves whether to interpret the OP's words as being at our expense or a genuine heartfelt sense of anguish at a hostile user-experience. Of course the OP's experience with Google and ReadTheDocs is not objectively as awful as what my wife has been through, or even for that matter me, but we shouldn't dismiss the mental pain of an ugly and frustrating user experience as any less real just because it didn't involve an actual knife being held to somebody's throat. Please don't be so hostile to a newcomer who has taken the time to report a problem on this list and did nothing to deserve the given reaction. At the *very least*, we as a community ought to give newcomers the benefit of the doubt and interpret the OP's comment as nothing more than hyperbole rather than assuming the worst ("...at the expense of..."). Thank you. Steve From peter.bittner at gmx.net Sat Jan 27 08:54:39 2018 From: peter.bittner at gmx.net (Peter Bittner) Date: Sat, 27 Jan 2018 14:54:39 +0100 Subject: [code-quality] Pylint hosting on ReadTheDocs prevents successful searches In-Reply-To: <20180126161434.GB22500@ando.pearwood.info> References: <20180126161434.GB22500@ando.pearwood.info> Message-ID: Alright everyone, so what are we going to do about that? The search functionality looks fine to me in general on [1]. And the search box is in a logical place, in the sidebar. Maybe using a different, more modern theme may improve the user experience slightly, though. I guess redirecting every access on "docs.pylint.org" to "pylint.readthedocs.io" (for both HTTP and HTTPS) will do away with the largest part of reader confusion. Is this something we can help fixing, e.g. by a PR? Regards, Peter [1] https://pylint.readthedocs.io/en/latest/search.html?q=pylintrc From peter.bittner at gmx.net Sat Jan 27 09:15:18 2018 From: peter.bittner at gmx.net (Peter Bittner) Date: Sat, 27 Jan 2018 15:15:18 +0100 Subject: [code-quality] Pylint hosting on ReadTheDocs prevents successful searches In-Reply-To: References: <20180126161434.GB22500@ando.pearwood.info> Message-ID: Oh! I was---almost---wrong! There _is_ a bug obviously in the docs (generator configuration?). Because the search results on "pylint.readthedocs.io" all include links to "docs.pylint.org", thus redirecting requests to the documentation index ("Pylint User Manual"). This issue has already been reported at https://github.com/PyCQA/pylint/issues/1600 "Almost" wrong, because: A permanent redirect (including the address path) would also cover that problem. Peter From peter.bittner at gmx.net Sat Jan 27 09:24:56 2018 From: peter.bittner at gmx.net (Peter Bittner) Date: Sat, 27 Jan 2018 15:24:56 +0100 Subject: [code-quality] Pylint hosting on ReadTheDocs prevents successful searches In-Reply-To: References: <20180126161434.GB22500@ando.pearwood.info> Message-ID: If you have a Bitbucket account please vote for this bug if you want the redirect (and the docs) to be working as described. - https://bitbucket.org/logilab/pylint.org/issues/16 Still yet, there is investigation needed why the old domain name ends up in the generated docs. Peter From willingc at willingconsulting.com Sat Jan 27 11:47:16 2018 From: willingc at willingconsulting.com (Carol Willing) Date: Sat, 27 Jan 2018 08:47:16 -0800 Subject: [code-quality] Pylint hosting on ReadTheDocs prevents successful searches In-Reply-To: <20180126161434.GB22500@ando.pearwood.info> References: <20180126161434.GB22500@ando.pearwood.info> Message-ID: > On Jan 26, 2018, at 8:14 AM, Steven D'Aprano wrote: > > On Fri, Jan 26, 2018 at 09:15:43AM -0600, Ian Stapleton Cordasco wrote: > >> This sounds like an attempt to be humorous at the expense of people >> who have actually been physically violated. That isn't acceptable on >> this list and is not tolerated by any mailing list hosted on >> python.org. Please refrain from this in the future. > > Ian, you are gate-keeping and telling the poster how he or she should > feel. Please don't do that, it is a hostile and aggressive thing to do. > You have no right to tell the OP that he or she *shouldn't* feel > violated by an experience which he clearly experienced as frustrating > and painful. Who are you to say that (s)he didn't, or shouldn't, have > felt violated? All, Jumping to conclusions about people's intents will continue along an unproductive path to improve the documentation experience. Let's try to keep this discussion on topic. Giving everyone the benefit of the doubt of being well intentioned, I would appreciate keeping any future discussion on the topic of the documentation search. Here's my thoughts for each of you as we move forward: - Thank you Quantum Mechanic for reporting the issue. - Thank you Ian for your empathy and respect for the community. - Thank you Steven for stressing that we "should" be a welcoming community to new and existing contributors. I just returned from PyCascades where Brett Cannon gave a great talk on open source community: https://www.youtube.com/watch?v=HiWfqMbJ3_8 Thank you, Carol Willing > > And before you dismiss me (as I expect your initial reaction will be) > both my wife and I are "people who have actually been physically > violated". Your defence of our feelings, while well-intentioned, is > neither needed nor wanted. I'm sure you thought you were doing the right > thing, but we don't need you to act as gate-keeper telling us when we > are being made fun of. We can decide for ourselves whether to interpret > the OP's words as being at our expense or a genuine heartfelt sense of > anguish at a hostile user-experience. > > Of course the OP's experience with Google and ReadTheDocs is not > objectively as awful as what my wife has been through, or even for that > matter me, but we shouldn't dismiss the mental pain of an ugly and > frustrating user experience as any less real just because it didn't > involve an actual knife being held to somebody's throat. > > Please don't be so hostile to a newcomer who has taken the time to > report a problem on this list and did nothing to deserve the given > reaction. At the *very least*, we as a community ought to give newcomers > the benefit of the doubt and interpret the OP's comment as nothing more > than hyperbole rather than assuming the worst ("...at the expense > of..."). > > > Thank you. > > > > Steve > _______________________________________________ > code-quality mailing list > code-quality at python.org > https://mail.python.org/mailman/listinfo/code-quality -------------- next part -------------- An HTML attachment was scrubbed... URL: From quantum.mechanic.1964 at gmail.com Tue Jan 30 11:46:06 2018 From: quantum.mechanic.1964 at gmail.com (Quantum Mechanic) Date: Tue, 30 Jan 2018 16:46:06 +0000 Subject: [code-quality] Pylint hosting on ReadTheDocs prevents successful searches In-Reply-To: References: Message-ID: Hi, I'm sorry I riled everyone up. My apologies. === Trying to reproduce === Getting back to this today, I'm not sent to readthedocs.org. I'll chalk that up to the interwebz demons fiddling around (could be anywhere between my $work, Google, and your end, including various intermediaries). Now, back to the next problem. There doesn't seem to be any documentation on the fields in .pylintrc. I'm specifically looking for init-hook help. I can't find anything of substance in the html docs, so I download the PDF and search for "init-hook". There are 6 hits, most only mentions. Finally, it occurs to me (from the nature of the change summaries in the PDF) that this is probably a command line argument. Low and behold, pylint --help has a blurb. (As an aside, I'm wondering how to format this well for command line, due to the significant whitespace feature, but that's a problem for another day.) === Problem 1 === Search not helpful =============== For me, the web docs are not that helpful for searching, as it doesn't get me any closer. Take this screenshot of a search for init-hook: The first link goes to http://docs.pylint.org/en/latest/user_guide/run.html?highlight=init-hook, which looks like this: ...which I had before. === Problem 2 === Command line help not online =============== The command line help does not exist online or in the PDF in except in brief, as far as I can see. Perhaps this is intentional? (It makes it harder to post Stack Exchange answers, but that's SEP.) === Problem 3 === How to include multiline code in .pylintrc file in the init-hook section. =============== This doesn't seem to be covered anywhere in the docs, but is an obvious use case. On Fri, Jan 26, 2018 at 12:05 PM, Quantum Mechanic < quantum.mechanic.1964 at gmail.com> wrote: > Googling for "pylint pylintrc" gives me the first hit as > https://docs.pylint.org/en/1.6.0/run.html. > > This lands me on https://docs.pylint.org/en/1.6.0/run.html, asking me to > login. I have a login, and clicking "Login" sends me to > https://readthedocs.org/dashboard/, spoiling the google search. > > In fact, searching on google only shows me that something might be there. > I have to search *again* on ReadTheDocs, which isn't friendly at all. > > Poking around until I find a search box at https://pylint.readthedocs. > io/en/latest/, I type in "pylintrc", which returns the *same bloody page* > (OK, yes, it has this helpful bit: https://pylint. > readthedocs.io/en/latest/?highlight=pylintrc, which does > exactly...nothing.) > > So I'm stuck trolling around in the full docs there. Instead, I should > just pull down the source code, and do my own grep -r. > > I feel like I've been violated. > > -- > -QM > Quantum Mechanics: The dreams stuff is made of > -- -QM Quantum Mechanics: The dreams stuff is made of -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: Screen Shot 2018-01-30 at 16.19.05.png Type: image/png Size: 61824 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: Screen Shot 2018-01-30 at 16.20.46.png Type: image/png Size: 134014 bytes Desc: not available URL: From willingc at willingconsulting.com Wed Jan 31 10:59:59 2018 From: willingc at willingconsulting.com (Carol Willing) Date: Wed, 31 Jan 2018 07:59:59 -0800 Subject: [code-quality] Pylint hosting on ReadTheDocs prevents successful searches In-Reply-To: References: Message-ID: Hi Quantum Mechanic, Web Search ------------- I went back and checked a Google search and a Duck Duck Go search on "pylint". - On Duck Duck Go, https://www.pylint.org and https://pylint.readthedocs.io/en/latest/ are the first two hits. - On Google, https://www.pylint.org and https://pypi.python.org/pypi/pylint are the first two hits. For a "pylint pylintrc" search: - On Duck Duck Go, the website and v2.0 ReadTheDocs page are in the top four hits. - On Google, the v1.6.5 ReadThe Docs page and the GitHub repo are in the top two hits. I'm not an SEO person but someone else may have a suggestion on how to improve the hit rate for v2.0 ReadTheDocs on Google. ReadTheDocs Search ----------------------- When I try searching on "pylintrc" from the search box, I do receive 6 hits. Unfortunately, each of those when clicked redirect to the docs home page. Searching for "init-hook", I see the same behavior. This is a bug/configuration error in Sphinx setup or the theme of some sort. There is an open issue on the GitHub repo: https://github.com/PyCQA/pylint/issues/1600 Next actions -------------- I've added this summary to the open issue: https://github.com/PyCQA/pylint/issues/1600#issuecomment-361977477 Perhaps labeling this doc issue as Help Wanted as well would be useful. Regards, Carol > On Jan 30, 2018, at 8:46 AM, Quantum Mechanic wrote: > > Hi, > > I'm sorry I riled everyone up. My apologies. > > === Trying to reproduce === > Getting back to this today, I'm not sent to readthedocs.org . I'll chalk that up to the interwebz demons fiddling around (could be anywhere between my $work, Google, and your end, including various intermediaries). > > Now, back to the next problem. There doesn't seem to be any documentation on the fields in .pylintrc. I'm specifically looking for init-hook help. > > I can't find anything of substance in the html docs, so I download the PDF and search for "init-hook". There are 6 hits, most only mentions. > > Finally, it occurs to me (from the nature of the change summaries in the PDF) that this is probably a command line argument. Low and behold, pylint --help has a blurb. (As an aside, I'm wondering how to format this well for command line, due to the significant whitespace feature, but that's a problem for another day.) > > === Problem 1 === > Search not helpful > =============== > > For me, the web docs are not that helpful for searching, as it doesn't get me any closer. Take this screenshot of a search for init-hook: > > The first link goes to http://docs.pylint.org/en/latest/user_guide/run.html?highlight=init-hook , which looks like this: > > ...which I had before. > > === Problem 2 === > Command line help not online > =============== > > The command line help does not exist online or in the PDF in except in brief, as far as I can see. Perhaps this is intentional? (It makes it harder to post Stack Exchange answers, but that's SEP.) > > === Problem 3 === > How to include multiline code in .pylintrc file in the init-hook section. > =============== > > This doesn't seem to be covered anywhere in the docs, but is an obvious use case. > > > > On Fri, Jan 26, 2018 at 12:05 PM, Quantum Mechanic >wrote: > Googling for "pylint pylintrc" gives me the first hit as https://docs.pylint.org/en/1.6.0/run.html . > > This lands me on https://docs.pylint.org/en/1.6.0/run.html , asking me to login. I have a login, and clicking "Login" sends me to https://readthedocs.org/dashboard/ , spoiling the google search. > > In fact, searching on google only shows me that something might be there. I have to search *again* on ReadTheDocs, which isn't friendly at all. > > Poking around until I find a search box at https://pylint.readthedocs.io/en/latest/ , I type in "pylintrc", which returns the *same bloody page* (OK, yes, it has this helpful bit: https://pylint.readthedocs.io/en/latest/?highlight=pylintrc , which does exactly...nothing.) > > So I'm stuck trolling around in the full docs there. Instead, I should just pull down the source code, and do my own grep -r. > > I feel like I've been violated. > > -- > -QM > Quantum Mechanics: The dreams stuff is made of > > > > -- > -QM > Quantum Mechanics: The dreams stuff is made of > _______________________________________________ > code-quality mailing list > code-quality at python.org > https://mail.python.org/mailman/listinfo/code-quality -------------- next part -------------- An HTML attachment was scrubbed... URL: