User talk:Trappist the monk/Archive 20

Help with general citation fixing

I created a new discussion given that I almost emptied those categories and I intend to go on fixing other ones in which I may need your mentoring. I'm currently dealing with this one. I have 2 questions: First of all, continuing what I've asked above, I'm thinking of creating some regexes for Smallem for the most common problems with this category, examples mentioned in its description on EnWiki. Is that all right to do? Secondly, I noticed something in which the module "lacks": Although it does catch all (most?) occasions of extra text related to pages in English, it does nothing for extra text in other languages. In this article you'll see that it says there are no more errors related to that category although if you see the source code and search for the pages parameter you'll see many instances of "fq" being put in values, which is the Albanian equivalent of "p". You can also check the rendered results and see that many references have the "fq" part doubled. Can the module be internationalized at this particular point? I was thinking to make it automatically get the value from the translation table in the top of the config. page for each language (apart from the default English value). To be honest, I've always thought it did that. I was surprised to learn that that wasn't true. I'm thinking this problem will be present in some other aspects too but this is just the first I noticed. - Klein Muçi (talk) 23:47, 14 August 2021 (UTC)

Another suggestion: See the third citation here. Maybe CS1 can use "and" as delimiter beside comma to better understand grouped pages and be able to translate the "and" part? Naïve question: Why does the pages parameter accept non-numerical values? A lot of these problems could be solved if it only accepted numbers, commas and dashes. But I believe there must be cases I'm not thinking of here. - Klein Muçi (talk) 00:05, 15 August 2021 (UTC)

Right now, cs1|2 doesn't care what editors put in |pages= as long as they don't include anything that looks vaguely like the prefix that cs1|2 adds for the rendering. We could alarm on |page(s)= containing and where the only other text in the parameter value is digits and punctuation. I'll think about that. This (crude) search times out and returns results that we don't care about, but you can get a sense for how editors at en.wiki use 'and' in |page(s)=.

—Trappist the monk (talk) 00:53, 15 August 2021 (UTC)

i18n for volume, issue, page(s) is already there. No doubt it could use improvement but until someone tries to use it for their language and it doesn't work right, we'll never know.

—Trappist the monk (talk) 00:53, 15 August 2021 (UTC)

Yeah, I noticed that the page(s) parameter is not that strict. I'm a bit confused in regard to your last sentence though. Am I not that person that you mention there? I mean, the doubled "fq" part.

I was dealing with volume parameters now: Again, my first thought was that the parameter would only accept numbers as valid values but then I saw a lot of cases with Roman numbers and, most importantly, a lot of cases of long volume titles including almost whole sentences and I gave up on that idea. Actually I'm stuck right here: sq:Kategoria:Gabime CS1: Tekst shtesë: Vëllimi - I don't know how to deal with any of these remaining articles in which the volume value is more complex than just a number. Can you give me some examples of how to deal with those so I know how to go on? - Klein Muçi (talk) 01:20, 15 August 2021 (UTC)

There is on-and-off discussion about how cs1|2 should label |volume=, |issue=, and |number= in the rendering. There are those who argue that all of the templates should have the same style; there are those who argue that {{cite journal}} alone among the templates should retain it's unique volume-issue-page style while all of the others should adopt the {{cite magazine}} style. Thus far, those editors participating in the discussions have not achieved consensus. In the meantime, cs1|2 looks for specific text in the parameter values that looks like the label that cs1|2 uses (or might use in future) and alarms when that text is found. Often the correct fix is to remove that label-look-alike text. There are cases where I think that the whole |volume= label should be removed; for example, in this one from sq:Arkitektura Pallava:

{{cite book|first=Anna Aleksandra|last=Ślączka|year=2007|title=Temple consecration rituals in ancient India: text and archaeology|publisher=BRILL|volume=Volume 26 of Brill's indological library|isbn=9789004158436|url=http://books.google.com.sg/books?id=9LznwzyNE18C}}

|volume=Volume 26 of Brill's indological library is a series-volume number/title. cs1|2 does not support series volume per se, though I suppose that information, were it important, could be made part of |series=. For this example, I would argue that |volume=Volume 26 of Brill's indological library is not important so the whole parameter can go away.

—Trappist the monk (talk) 14:41, 15 August 2021 (UTC)

So, I updated Smallem's table in regex fixes that are not related to languages:

	(r"\|\s*authorfirst\s*=", r"|author-first="),
	(r"\|\s*authorgiven\s*=", r"|author-given="),
	(r"\|\s*authorlast\s*=", r"|author-last="),
	(r"\|\s*authormask\s*=", r"|author-mask="),
	(r"\|\s*authorsurname\s*=", r"|author-surname="),
	(r"\|\s*booktitle\s*=", r"|book-title="),
	(r"\|\s*chapterurl\s*=", r"|chapter-url="),
	(r"\|\s*conferenceurl\s*=", r"|conference-url="),
	(r"\|\s*contributionurl\s*=", r"|contribution-url="),
	(r"\|\s*dead-?url\s*=\s*(?:true|yes|y)\b", r"|url-status=dead"),
	(r"\|\s*dead-?url\s*=\s*no", r"|url-status=live"),
	(r"\|\s*displayauthors\s*=", r"|display-authors="),
	(r"\|\s*displayeditors\s*=", r"|display-editors="),
	(r"\|\s*doi-broken\s*=", r"|doi-broken-date="),
	(r"\|\s*doi-inactive-date\s*=", r"|doi-broken-date="),
	(r"\|\s*editorfirst\s*=", r"|editor-first="),
	(r"\|\s*editorgiven\s*=", r"|editor-given="),
	(r"\|\s*editorlast\s*=", r"|editor-last="),
	(r"\|\s*editorlink\s*=", r"|editor-link="),
	(r"\|\s*editorsurname\s*=", r"|editor-surname="),
	(r"\|\s*editormask\s*=", r"|editor-mask="),
	(r"\|\s*embargo\s*=", r"|pmc-embargo-date="),
	(r"\|\s*episodelink\s*=", r"|episode-link="),
	(r"\|\s*event-format\s*=", r""),
	(r"\|\s*eventurl\s*=", r""),
	(r"\|\s*event-url\s*=", r""),
	(r"\|\s*ref\s*=\s*Harv\b", r""),
	(r"\|\s*interviewerlink\s*=", r"|interviewer-link="),
	(r"\|\s*interviewermask\s*=", r"|interviewer-mask="),
	(r"\|\s*lastauthoramp=\s*=\s*(?:true|yes|y)\b", r"|name-list-style=amp"),
	(r"\|\s*last-author-amp=\s*=\s*(?:true|yes|y)\b", r"|name-list-style=amp"),
	(r"\|\s*laydate\s*=", r"|lay-date="),
	(r"\|\s*laysource\s*=", r"|lay-source="),
	(r"\|\s*layurl\s*=", r"|lay-url="),
	(r"\|\s*mailinglist\s*=", r"|mailing-list="),
	(r"\|\s*mapurl\s*=", r"|map-url="),
	(r"\|\s*name-list-format\s*=", r"|name-list-style="),
	(r"\|\s*nocat\s*=", r"|no-tracking="),
	(r"\|\s*no-cat\s*=", r"|no-tracking="),
	(r"\|\s*publicationdate\s*=", r"|publication-date="),
	(r"\|\s*publicationplace\s*=", r"|publication-place="),
	(r"\|\s*registration=\s*=\s*(?:true|yes|y)\b", r"|url-access=registration"),
	(r"\|\s*sectionurl\s*=", r"|section-url="),
	(r"\|\s*serieslink\s*=", r"|series-link="),
	(r"\|\s*seriesno\s*=", r"|series-number="),
	(r"\|\s*seriesnumber\s*=", r"|series-number="),
	(r"\|\s*series-separator\s*=", r""),
	(r"\|\s*subjectlink\s*=", r"|subject-link="),
	(r"\|\s*subscription=\s*=\s*(?:true|yes|y)\b", r"|url-access=subscription"),
	(r"\|\s*timecaption\s*=", r"|time-caption="),
	(r"\|\s*titlelink\s*=", r"|title-link="),
	(r"\|\s*transcripturl\s*=", r"|transcript-url="),

Can you please check it and tell me if you think it is correct and "all-inclusive"? Any kind of optimizations? - Klein Muçi (talk) 08:38, 15 August 2021 (UTC)

The problem with this list of find-and-replace regexes is that it will replace the found parameters wherever they exist in an article which may not be in cs1|2 templates. They should be rewritten to be like all of the language regexes so that they are cs1|2-template specific:

(r"(\{\{\s*cit[aeio][^\}]*\|\s*)authorfirst\s*=\s*", r"\1author-first")

I would rewrite the |dead-url=yes regex to simply delete that parameter because when |archive-url= has a value, |url-status=dead is the default state.

(r"(\{\{\s*cit[aeio][^\}]*)\|\s*dead-?url\s*=\s*(?:true|yes|y)\b", r"\1")

—Trappist the monk (talk) 14:12, 15 August 2021 (UTC)

Thanks for the explanation! What about more easier cases, for example: volume=1st volume? How should I act in this case? Should I leave just "1st"? How is that parameter typically supposed to work? Volume=2nd volume or Volume=2? Are both cases accepted? The first case would provide a problem in translation because you'd basically have to fully rewrite the value to make sense when rendered or make it so cs1|2 understands numbers like 1st, 2nd, etc. and automatically translates them in rendering. Moreover, touching a bit the subject that you mentioned yourself, what is the typical use of that parameter? What's typically considered a volume? The Harry Potter series has 7 books, if we are to cite something from the second book, are we supposed to write |volume=2? Maybe |volume=2nd? Or is that reserved only to didactic/encyclopedic kind of works?

And lastly, in regard to the regexes, thank you, I'll do that. Can we merge anyone of those though? For example:

	(r"\|\s*nocat\s*=", r"|no-tracking="),
	(r"\|\s*no-cat\s*=", r"|no-tracking="),

Also, I've left out of that list all the numbered parameters because I didn't know how to be correct with those. Should I add those? Can I add those? - Klein Muçi (talk) 20:37, 15 August 2021 (UTC)

PS:Is the regex about dead URLs correct? The first part is different from the other regexes on the list (I don't mean the group in the end). - Klein Muçi (talk) 20:56, 15 August 2021 (UTC)

Yes.

—Trappist the monk (talk) 22:22, 15 August 2021 (UTC)

I would not write |volume=1st; it's rare and I don't recall ever having seen it; I would write |volume=1. For periodicals, a volume is (typically) one year's worth of issues. For books, a volume is (typically) one part of a very lengthy work under a single title where binding the whole as a single 'book' would be cumbersome or prohibitively expensive; encyclopediae are a form of this kind of 'book'. Harry Potter is a series of individual books, not a multi-volume book.

Yes: (r"\|\s*no-?cat\s*=", r"|no-tracking=")

(r"\|\s*authorfirst(\d*)\s*=", r"|author-first\1="),
(r"\|\s*author(\d*)first\s*=", r"|author\1-first=")

—Trappist the monk (talk) 22:22, 15 August 2021 (UTC)

Thank you! Can you transform these regexes for me to only be used inside cs1|2 template?

(r"\|\s*dead-?url\s*=\s*no", r"|url-status=live"),
(r"\|\s*ref\s*=\s*Harv\b", r""),
(r"\|\s*last-?author-?amp=\s*=\s*(?:true|yes|y)\b", r"|name-list-style=amp"),
(r"\|\s*registration=\s*=\s*(?:true|yes|y)\b", r"|url-access=registration"),

I was too afraid to do it on my own because of their complexity. After that, I'll bring here the whole table for one final check. - Klein Muçi (talk) 23:14, 15 August 2021 (UTC)

(r"(\{\{\s*cit[aeio][^\}]*)\|\s*ref\s*=\s*harv\b", r"\1"),
(r"(\{\{\s*cit[aeio][^\}]*)\|\s*dead-?url\s*=\s*(?:true|yes|y)\b", r"\1"),
(r"(\{\{\s*cit[aeio][^\}]*\|\s*)dead-?url\s*=\s*no\b", r"\1url-status=live"),
(r"(\{\{\s*cit[aeio][^\}]*\|\s*)last-?author-?amp\s*=\s*=\s*(?:true|yes|y)\b", r"\1|name-list-style=amp"),
(r"(\{\{\s*cit[aeio][^\}]*\|\s*)registration\s*=\s*=\s*(?:true|yes|y)\b", r"\1|url-access=registration"),

—Trappist the monk (talk) 00:23, 16 August 2021 (UTC)

Thanks a lot! Time to activate MonkFixer3000™. :P

List of Regexes

(r"(\{\{\s*cit[aeio][^\}]*\|\s*)authorfirst\s*=\s*", r"\author-first"), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)authorfirst(\d*)\s*=\s*", r"\author-first\1="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)author(\d*)first\s*=\s*", r"\author\1-first="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)authorgiven\s*=\s*", r"|author-given="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)authorgiven(\d*)\s*=\s*", r"|author-given\1="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)author(\d*)given(\d*)\s*=\s*", r"|author\1-given="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)authorlast\s*=\s*", r"|author-last="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)authorlast(\d*)\s*=\s*", r"|author-last\1="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)author(\d*)last(\d*)\s*=\s*", r"|author\1-last="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)authormask\s*=\s*", r"|author-mask="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)authormask(\d*)\s*=\s*", r"|author-mask\1="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)author(\d*)mask(\d*)\s*=\s*", r"|author\1-mask="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)authorsurname\s*=\s*", r"|author-surname="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)authorsurname(\d*)\s*=\s*", r"|author-surname\1="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)author(\d*)surname(\d*)\s*=\s*", r"|author\1-surname="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)booktitle\s*=\s*", r"|book-title="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)chapterurl\s*=\s*", r"|chapter-url="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)conferenceurl\s*=\s*", r"|conference-url="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)contributionurl\s*=\s*", r"|contribution-url="), (r"(\{\{\s*cit[aeio][^\}]*)\|\s*dead-?url\s*=\s*(?:true|yes|y)\b", r"\1"), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)dead-?url\s*=\s*no\b", r"\1url-status=live"), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)displayauthors\s*=\s*", r"|display-authors="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)displayeditors\s*=\s*", r"|display-editors="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)doi-(?:broken|inactive)-date\s*=\s*", r"|doi-broken-date="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editorfirst\s*=\s*", r"|editor-first="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editorfirst(\d*)\s*=\s*", r"\editor-first\1="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editor(\d*)first\s*=\s*", r"\editor\1-first="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editorgiven\s*=\s*", r"|editor-given="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editorgiven(\d*)\s*=\s*", r"|editor-given\1="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editor(\d*)given(\d*)\s*=\s*", r"|editor\1-given="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editorlast\s*=\s*", r"|editor-last="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editorlast(\d*)\s*=\s*", r"|editor-last\1="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editor(\d*)last(\d*)\s*=\s*", r"|editor\1-last="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editorlink\s*=\s*", r"|editor-link="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editorlink(\d*)\s*=\s*", r"|editor-link\1="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editor(\d*)link(\d*)\s*=\s*", r"|editor\1-link="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editorsurname\s*=\s*", r"|editor-surname="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editorsurname(\d*)\s*=\s*", r"|editor-surname\1="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editor(\d*)surname(\d*)\s*=\s*", r"|editor\1-surname="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editormask\s*=\s*", r"|editor-mask="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editormask(\d*)\s*=\s*", r"|editor-mask\1="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editor(\d*)mask(\d*)\s*=\s*", r"|editor\1-mask="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)embargo\s*=\s*", r"|pmc-embargo-date="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)episodelink\s*=\s*", r"|episode-link="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)event-format\s*=\s*", r""), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)event-?url\s*=\s*", r""), (r"(\{\{\s*cit[aeio][^\}]*)\|\s*ref\s*=\s*harv\b", r"\1"), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)interviewerlink\s*=\s*", r"|interviewer-link="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)interviewerlink(\d*)\s*=\s*", r"|interviewer-link\1="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)interviewer(\d*)link(\d*)\s*=\s*", r"|interviewer\1-link="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)interviewermask\s*=\s*", r"|interviewer-mask="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)interviewermask(\d*)\s*=\s*", r"|interviewer-mask\1="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)interviewer(\d*)mask(\d*)\s*=\s*", r"|interviewer\1-mask="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)last-?author-?amp\s*=\s*=\s*(?:true|yes|y)\b", r"\1|name-list-style=amp"), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)laydate\s*=\s*", r"|lay-date="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)laysource\s*=\s*", r"|lay-source="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)laysurl\s*=\s*", r"|lay-url="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)mailinglist\s*=\s*", r"|mailing-list="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)mapurl\s*=\s*", r"|map-url="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)name-list-format\s*=\s*", r"|name-list-style="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)no-?cat\s*=\s*", r"|no-tracking="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)publicationdate\s*=\s*", r"|publication-date="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)publicationplace\s*=\s*", r"|publication-place="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)registration\s*=\s*=\s*(?:true|yes|y)\b", r"\1|url-access=registration"), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)sectionurl\s*=\s*", r"|section-url="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)serieslink\s*=\s*", r"|series-link="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)series(?:no|number)\s*=\s*", r"|series-number="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)series-separator\s*=\s*", r""), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)subjectlink\s*=\s*", r"|subject-link="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)subjectlink(\d*)\s*=\s*", r"|subject-link\1="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)subject(\d*)link(\d*)\s*=\s*", r"|subject\1-link="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)subscription\s*=\s*=\s*(?:true|yes|y)\b", r"\1|url-access=subscription"), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)timecaption\s*=\s*", r"|time-caption="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)titlelink\s*=\s*", r"|title-link="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)transcripturl\s*=\s*", r"|transcript-url="), (r"\[\[Kategoria:CS1\]\]", r""), (r"\[\[Kategoria:Gabime\ CS1[^\]]*\]\]", r""), (r"\[\[Kategoria:Mirëmbajtja\ CS1[^\]]*\]\]", r""), (r"\[\[Kategoria:Vetitë\ CS1[^\]]*\]\]", r""), (r"\[\[Kategoria:Gjuhë\ CS1\]\]", r""),

Every fix Smallem makes that is related to citations and not to languages per se. - Klein Muçi (talk) 00:32, 16 August 2021 (UTC)

This one (and others like it) is redundant; this one also has an error in the replacement that is repeated in other replacements: backslash (escape) should not precede the replacement parameter name (malformed \1):

(r"(\{\{\s*cit[aeio][^\}]*\|\s*)authorfirst\s*=\s*", r"\author-first"),

malformed \1 capture in replacement; the enumerator is the second capture so should use \2 in the replacement:

(r"(\{\{\s*cit[aeio][^\}]*\|\s*)authorfirst(\d*)\s*=\s*", r"\author-first\1="),

\1 capture in the replacement goes at the beginning of the replacement; when the pipe preceding the parameter name is included in the \1 capture, a pipe must not precede the parameter name in the replacement (there are many with this error); the enumerator is \2:

(r"(\{\{\s*cit[aeio][^\}]*\|\s*)authorgiven(\d*)\s*=\s*", r"|author-given\1="),

capture \1 should not include pipe and optional whitespace when deleting a parameter; replacement requires \1 capture

(r"(\{\{\s*cit[aeio][^\}]*\|\s*)event-format\s*=\s*", r""),

—Trappist the monk (talk) 01:09, 16 August 2021 (UTC)

List of Regexes Revised

(r"(\{\{\s*cit[aeio][^\}]*\|\s*)authorfirst(\d*)\s*=\s*", r"\2author-first="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)author(\d*)first\s*=\s*", r"\author\1-first="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)authorgiven(\d*)\s*=\s*", r"\2author-given="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)author(\d*)given(\d*)\s*=\s*", r"|author\1-given="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)authorlast(\d*)\s*=\s*", r"\2author-last="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)author(\d*)last(\d*)\s*=\s*", r"|author\1-last="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)authormask(\d*)\s*=\s*", r"\2author-mask="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)author(\d*)mask(\d*)\s*=\s*", r"|author\1-mask="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)authorsurname(\d*)\s*=\s*", r"\2author-surname="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)author(\d*)surname(\d*)\s*=\s*", r"|author\1-surname="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)booktitle\s*=\s*", r"|book-title="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)chapterurl\s*=\s*", r"|chapter-url="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)conferenceurl\s*=\s*", r"|conference-url="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)contributionurl\s*=\s*", r"|contribution-url="), (r"(\{\{\s*cit[aeio][^\}]*)\|\s*dead-?url\s*=\s*(?:true|yes|y)\b", r"\1"), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)dead-?url\s*=\s*no\b", r"\1url-status=live"), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)displayauthors\s*=\s*", r"|display-authors="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)displayeditors\s*=\s*", r"|display-editors="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)doi-(?:broken|inactive)-date\s*=\s*", r"|doi-broken-date="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editorfirst(\d*)\s*=\s*", r"\2editor-first="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editor(\d*)first\s*=\s*", r"\editor\1-first="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editorgiven(\d*)\s*=\s*", r"\2editor-given="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editor(\d*)given(\d*)\s*=\s*", r"|editor\1-given="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editorlast(\d*)\s*=\s*", r"\2editor-last="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editor(\d*)last(\d*)\s*=\s*", r"|editor\1-last="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editorlink(\d*)\s*=\s*", r"\2editor-link="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editor(\d*)link(\d*)\s*=\s*", r"|editor\1-link="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editorsurname(\d*)\s*=\s*", r"\2editor-surname="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editor(\d*)surname(\d*)\s*=\s*", r"|editor\1-surname="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editormask(\d*)\s*=\s*", r"\2editor-mask="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editor(\d*)mask(\d*)\s*=\s*", r"|editor\1-mask="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)embargo\s*=\s*", r"|pmc-embargo-date="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)episodelink\s*=\s*", r"|episode-link="), (r"(\{\{\s*cit[aeio][^\}]*)\|\s*event-format\s*=\s*", r"\1"), (r"(\{\{\s*cit[aeio][^\}]*)\|\s*event-?url\s*=\s*", r"\1"), (r"(\{\{\s*cit[aeio][^\}]*)\|\s*ref\s*=\s*harv\b", r"\1"), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)interviewerlink(\d*)\s*=\s*", r"\2interviewer-link="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)interviewer(\d*)link(\d*)\s*=\s*", r"|interviewer\1-link="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)interviewermask(\d*)\s*=\s*", r"\2interviewer-mask="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)interviewer(\d*)mask(\d*)\s*=\s*", r"|interviewer\1-mask="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)last-?author-?amp\s*=\s*=\s*(?:true|yes|y)\b", r"\1|name-list-style=amp"), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)laydate\s*=\s*", r"|lay-date="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)laysource\s*=\s*", r"|lay-source="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)laysurl\s*=\s*", r"|lay-url="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)mailinglist\s*=\s*", r"|mailing-list="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)mapurl\s*=\s*", r"|map-url="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)name-list-format\s*=\s*", r"|name-list-style="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)no-?cat\s*=\s*", r"|no-tracking="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)publicationdate\s*=\s*", r"|publication-date="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)publicationplace\s*=\s*", r"|publication-place="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)registration\s*=\s*=\s*(?:true|yes|y)\b", r"\1|url-access=registration"), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)sectionurl\s*=\s*", r"|section-url="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)serieslink\s*=\s*", r"|series-link="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)series(?:no|number)\s*=\s*", r"|series-number="), (r"(\{\{\s*cit[aeio][^\}]*)\|\s*series-separator\s*=\s*", r"\1"), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)subjectlink(\d*)\s*=\s*", r"\2subject-link="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)subject(\d*)link(\d*)\s*=\s*", r"|subject\1-link="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)subscription\s*=\s*=\s*(?:true|yes|y)\b", r"\1|url-access=subscription"), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)timecaption\s*=\s*", r"|time-caption="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)titlelink\s*=\s*", r"|title-link="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)transcripturl\s*=\s*", r"|transcript-url="), (r"\[\[Kategoria:CS1\]\]", r""), (r"\[\[Kategoria:Gabime\ CS1[^\]]*\]\]", r""), (r"\[\[Kategoria:Mirëmbajtja\ CS1[^\]]*\]\]", r""), (r"\[\[Kategoria:Vetitë\ CS1[^\]]*\]\]", r""), (r"\[\[Kategoria:Gjuhë\ CS1\]\]", r""),

I tried following your directions as best as I could. These are the results. I'm sorry if I've made any mistakes. - Klein Muçi (talk) 08:26, 16 August 2021 (UTC)

Nope. Position of the captures in the replacement is everything. Captures in a regex proceed from left to right so everything preceding the parameter in question is \1 which must be the first element of the replacement. For enumerated parameters, the enumeration is \2 and must be included in the new parameter and in the proper position. When both \1 and \2 are used, they must also proceed left to right in the replacement.

Do a text search for:

• r"| – \1 is not the first element in the replacement
• r"\1| – this will result in a citation with a doubled pipe
• r"\2 – this puts the enumerator ahead of the parameter and leaves out the leading part of the citation unless:
  • \1 – if \1 appears in the middle of the replacement, that is where the leading part of the citation will be
• r"\ – the \ must be followed by 1

\1 and \2 must never occur more than once in any replacement

—Trappist the monk (talk) 13:24, 16 August 2021 (UTC)

List of Regexes Revised (again)

(r"(\{\{\s*cit[aeio][^\}]*\|\s*)authorfirst?(\d*)\s*=\s*", r"\1author-first\2="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)author(\d*)first\s*=\s*", r"\1author\2-first="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)authorgiven?(\d*)\s*=\s*", r"\1author-given\2="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)author(\d*)given\s*=\s*", r"\1author\2-given="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)authorlast?(\d*)\s*=\s*", r"\1author-last\2="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)author(\d*)last\s*=\s*", r"\1author\2-last="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)authormask?(\d*)\s*=\s*", r"\1author-mask\2="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)author(\d*)mask\s*=\s*", r"\1author\2-mask="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)authorsurname?(\d*)\s*=\s*", r"\1author-surname\2="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)author(\d*)surname\s*=\s*", r"\1author\2-surname="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)booktitle\s*=\s*", r"\1book-title="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)chapterurl\s*=\s*", r"\1chapter-url="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)conferenceurl\s*=\s*", r"\1conference-url="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)contributionurl\s*=\s*", r"\1contribution-url="), (r"(\{\{\s*cit[aeio][^\}]*)\|\s*dead-?url\s*=\s*(?:true|yes|y)\b", r"\1"), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)dead-?url\s*=\s*no\b", r"\1url-status=live"), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)displayauthors\s*=\s*", r"\1display-authors="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)displayeditors\s*=\s*", r"\1display-editors="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)doi-(?:broken|inactive)-date\s*=\s*", r"\1doi-broken-date="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editorfirst?(\d*)\s*=\s*", r"\1editor-first\2="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editor(\d*)first\s*=\s*", r"\1editor\2-first="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editorgiven?(\d*)\s*=\s*", r"\1editor-given\2="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editor(\d*)given\s*=\s*", r"\1editor\2-given="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editorlast?(\d*)\s*=\s*", r"\1editor-last\2="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editor(\d*)last\s*=\s*", r"\1editor\2-last="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editorlink?(\d*)\s*=\s*", r"\1editor-link\2="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editor(\d*)link\s*=\s*", r"\1editor\2-link="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editorsurname?(\d*)\s*=\s*", r"\1editor-surname\2="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editor(\d*)surname\s*=\s*", r"\1editor\2-surname="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editormask?(\d*)\s*=\s*", r"\1editor-mask\2="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)editor(\d*)mask\s*=\s*", r"\1editor\2-mask="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)embargo\s*=\s*", r"\1pmc-embargo-date="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)episodelink\s*=\s*", r"\1episode-link="), (r"(\{\{\s*cit[aeio][^\}]*)\|\s*event-format\s*=\s*", r"\1"), (r"(\{\{\s*cit[aeio][^\}]*)\|\s*event-?url\s*=\s*", r"\1"), (r"(\{\{\s*cit[aeio][^\}]*)\|\s*ref\s*=\s*harv\b", r"\1"), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)interviewerlink?(\d*)\s*=\s*", r"\1interviewer-link\2="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)interviewer(\d*)link\s*=\s*", r"\1interviewer\2-link="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)interviewermask?(\d*)\s*=\s*", r"\1interviewer-mask\2="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)interviewer(\d*)mask\s*=\s*", r"\1interviewer\2-mask="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)last-?author-?amp\s*=\s*=\s*(?:true|yes|y)\b", r"\1name-list-style=amp"), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)laydate\s*=\s*", r"\1lay-date="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)laysource\s*=\s*", r"\1lay-source="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)laysurl\s*=\s*", r"\1lay-url="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)mailinglist\s*=\s*", r"\1mailing-list="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)mapurl\s*=\s*", r"\1map-url="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)name-list-format\s*=\s*", r"\1name-list-style="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)no-?cat\s*=\s*", r"\1no-tracking="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)publicationdate\s*=\s*", r"\1publication-date="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)publicationplace\s*=\s*", r"\1publication-place="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)registration\s*=\s*=\s*(?:true|yes|y)\b", r"\1url-access=registration"), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)sectionurl\s*=\s*", r"\1section-url="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)serieslink\s*=\s*", r"\1series-link="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)series(?:no|number)\s*=\s*", r"\1series-number="), (r"(\{\{\s*cit[aeio][^\}]*)\|\s*series-separator\s*=\s*", r"\1"), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)subjectlink?(\d*)\s*=\s*", r"\1subject-link\2="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)subject(\d*)link\s*=\s*", r"\1interviewer\2-link="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)subscription\s*=\s*=\s*(?:true|yes|y)\b", r"\1url-access=subscription"), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)timecaption\s*=\s*", r"\1time-caption="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)titlelink\s*=\s*", r"\1title-link="), (r"(\{\{\s*cit[aeio][^\}]*\|\s*)transcripturl\s*=\s*", r"\1transcript-url="), (r"\[\[Kategoria:CS1\]\]", r""), (r"\[\[Kategoria:Gabime\ CS1[^\]]*\]\]", r""), (r"\[\[Kategoria:Mirëmbajtja\ CS1[^\]]*\]\]", r""), (r"\[\[Kategoria:Vetitë\ CS1[^\]]*\]\]", r""), (r"\[\[Kategoria:Gjuhë\ CS1\]\]", r""),

I tried again. I'm sorry but I've only worked once with regex group captures in the past. I have yet to fully form the concept of those and how they work in my mind. Your explanations sure are helping. I really hope I have everything all right now. If not, would you be so kind as to give some (more) practical examples of what I should change in the format [how it is] - [how it should be]? I'm really trying to get it right but... - Klein Muçi (talk) 01:29, 17 August 2021 (UTC)

Nothing jumps out as obviously wrong.

—Trappist the monk (talk) 21:40, 17 August 2021 (UTC)

Finally! Thank you! I'll add it to the source code soon and try a test run, see if I can see anything strange/wrong.

Meanwhile, I'm exhausting the extra text in volume category and the only remaining articles there are articles which have extra specifications on the volume parameter, for example: |volume=Volume 1, part 3.4 or |volume=Volume 3, The Great Story. What am I supposed to do with these kind of values? Should I keep only the first part and erase the extra specifications? - Klein Muçi (talk) 22:03, 17 August 2021 (UTC)

Hello! :) Just returned from vacations. Apparently Smallem has finished its test run and everything works perfectly. Thanks for the help there! Could you help me in regard to the volumes request above so I can go on with other categories? - Klein Muçi (talk) 03:57, 24 August 2021 (UTC)

You could just ignore those errors (as we do) by setting err_extra_text_volume[hidden]=true. Perhaps, someday, the community here will take a decision that allows us to unhide those error messages.

—Trappist the monk (talk) 12:44, 24 August 2021 (UTC)

Ah, I see. It's something gray. What about the articles here? Any particular way I should fix them? Or nothing "set on stone" even here? - Klein Muçi (talk) 16:12, 24 August 2021 (UTC)

Remove the extraneous text because your version of the module suite adds bot. to the parameter value.

—Trappist the monk (talk) 16:25, 24 August 2021 (UTC)

Yes but do we usually go with |edition=1st or |edition=1? Klein Muçi (talk) 16:37, 24 August 2021 (UTC)

Whichever you choose. cs1|2 does not convert cardinal numerals to ordinal numerals or the other-'way-'round. In English we typically use ordinal numerals because, in the rendering, the edition abbreviation follows the numeral: '1st ed.' which we read as 'first edition'.

—Trappist the monk (talk) 16:56, 24 August 2021 (UTC)

I see... Any chance that functionality could be added in the near future? Everything you said about English applies to Albanian as well. It would be nice if we could have editors write a cardinal number and get it rendered in its ordinal form. But whatever the answer may be, thank you! I'll deal with those articles for a bit now. :P - Klein Muçi (talk) 17:22, 24 August 2021 (UTC)

We have discussed converting simple cardinal numerals to ordinal numerals a couple of times that I can find but never with much enthusiasm.

—Trappist the monk (talk) 19:46, 24 August 2021 (UTC)

Sad to hear that. I really think that part should be standardized and take into account ordinal and cardinal numbers, Latin numerals and words. If I want to start a discussion for this subject, where do I do so? Any particular reasons why that wouldn't be a wise thing to do?

Can you also help me fix the errors in these 2 articles here? I've read the help sections in EnWiki about the Vancouver style twice already but I'm still not sure how to work with those. - Klein Muçi (talk) 02:11, 25 August 2021 (UTC)

If I may chime in, FWIW these are the places I am aware of where this was discussed:

• Help_talk:Citation_Style_1/Archive_11#Suggestion_for_edition=_parameter_to_treat_raw_numbers
• Help_talk:Citation_Style_1/Archive_13#Suggestion_to_add_a_printing=_parameter_similar_to_edition=
• Help_talk:Citation_Style_1/Archive_19#"|edition="_output_difference_between_En_and_Vi_language_versions_of_{{Template:Cite_book}}
• Help_talk:Citation_Style_1/Archive_64#Proposal_to_enhance_edition=_parameter_to_support_special_numerical_symbols

As you can see, I am supporter of the on-the-fly "translation" idea of the |edition= contents, at least for the simple virtually non-conflictive cases (but they would cover the majority of cases already), so that, in the long run, the proposed parameter input could be documented as the cardinal form "1", "2", "3" etc. instead of ordinal numerals "1st, 2nd, 3rd, ..." - as with the other numerical input parameters. This would also make the translation of citations between languages easier as only citations with non-cardinal numbers would have to be treated specially.

In general, the proper talk space to make such suggestions would be here: Help_talk:Citation_Style_1

--Matthiaspaul (talk) 17:27, 25 August 2021 (UTC)

@Matthiaspaul, thank you! If you've read the whole discussion here, you'll see that I do agree with your idea. I also suggested adding Latin numerals (for example IV or X) or even transforming words (for example third or tenth). Maybe I'm wrong but I do think something similar should be done even for the volume parameter. In general I'm pro limiting value possibilities as much as possible because it makes way for more standardized environments. If people feel something should allow "free form completion", most likely that parameter could be divided into 2 or more other parameters with limited value possibilities. The first dialogue in this discussion was exactly related to this phenomenon in regard to the |pages= parameter. Very, very crude example to illustrate my point: Make it so pages can accept only numerals. Other editors will say: What about ranges of pages? Answer: Make a different parameter for that |page-range. What if users start confusing the said parameters? What if they write "Thirty-three" instead of "33"? Answer: Create error messages to clarify things and possibly bot jobs to help with the solutions. Voila! I believe we do have the technological power to be precise without needing to overburden human editors. This is, of course, a terrible idea because |pages= works fine in regard to ranges and I don't really believe we should split it into 2 because of that but I used it just as an example of my thoughts in regard to this.

As I've said above, it was strange for me to see how the module behaved in regard to |page=, |volume=, |edition= parameters. Compared to other parameters, where you can get 6 different errors for the same parameter, it felt like it lacked in verifiability. - Klein Muçi (talk) 21:16, 25 August 2021 (UTC)

The original:

{{cite journal|vauthors=J. S. Morrison, A. W. Gomme|year=1950|title=Pericles Monarchos|journal=Journal of Hellenic Studies|volume=70|pages=76–77|doi=10.2307/629294|url=http://jstor.org/stable/629294|publisher=The Journal of Hellenic Studies, Vol. 70}}

J SM, A WG (1950). "Pericles Monarchos". Journal of Hellenic Studies. 70. The Journal of Hellenic Studies, Vol. 70: 76–77. doi:10.2307/629294. {{cite journal}}: Vancouver style error: punctuation in name 1 (help)

retain |vauthors=:

{{cite journal |vauthors=Morrison JS, Gomme AW |date=1950 |title=Pericles Monarchos |journal=Journal of Hellenic Studies |volume=70 |pages=76–77 |doi=10.2307/629294 |jstor=629294}}

Morrison JS, Gomme AW (1950). "Pericles Monarchos". Journal of Hellenic Studies. 70: 76–77. doi:10.2307/629294. JSTOR 629294.

or switch:

{{cite journal |author=J. S. Morrison |author2=A. W. Gomme |date=1950 |title=Pericles Monarchos |journal=Journal of Hellenic Studies |volume=70 |pages=76–77 |doi=10.2307/629294 |jstor=629294}}

J. S. Morrison; A. W. Gomme (1950). "Pericles Monarchos". Journal of Hellenic Studies. 70: 76–77. doi:10.2307/629294. JSTOR 629294.

and the other original:

{{Cite book|vauthors = Loredana Chines, Giorgio Forni, Giuseppe Ledda, Elisabetta Menetti | title = Dalle Origini al Cinquecento |year = 2007 | publisher = Bruno Mondadori|location = Milano | ISBN = 978-88-424-9164-4 | series = La letteratura italiana|editor = Ezio Raimondi}}

Loredana Chines, Giorgio Forni, Giuseppe Ledda, Elisabetta Menetti (2007). Ezio Raimondi (ed.). Dalle Origini al Cinquecento. La letteratura italiana. Milano: Bruno Mondadori. ISBN 978-88-424-9164-4. {{cite book}}: Vancouver style error: name in name 1 (help)

retain |vauthors=

{{Cite book |vauthors=Chines L, Forni G, Ledda G, Menetti E |title=Dalle Origini al Cinquecento |date=2007 |publisher=Bruno Mondadori |location=Milano |isbn=978-88-424-9164-4 |veditors=Raimondi E}}

Chines L, Forni G, Ledda G, Menetti E (2007). Raimondi E (ed.). Dalle Origini al Cinquecento. Milano: Bruno Mondadori. ISBN 978-88-424-9164-4.

or switch:

{{Cite book |author=Loredana Chines |author2=Giorgio Forni |author3=Giuseppe Ledda |author4=Elisabetta Menetti |title=Dalle Origini al Cinquecento |year=2007 |publisher=Bruno Mondadori |location=Milano |ISBN=978-88-424-9164-4 |series=La letteratura italiana |editor=Ezio Raimondi}}

Loredana Chines; Giorgio Forni; Giuseppe Ledda; Elisabetta Menetti (2007). Ezio Raimondi (ed.). Dalle Origini al Cinquecento. La letteratura italiana. Milano: Bruno Mondadori. ISBN 978-88-424-9164-4.

—Trappist the monk (talk) 02:39, 25 August 2021 (UTC)

Thanks a lot! In regard to different identifying parameters like ASIN, ISBN, SBN, JSTOR, HDL, etc. ... Is there any way I can find those online so I can fix errors with those? - Klein Muçi (talk) 03:05, 25 August 2021 (UTC)

I'm understanding that my requests will continue for a while as I go through each category of CS1 and try to learn more about them so don't be afraid to leave me on hold whenever needed. All this time I've been learning about the technical aspect of cs1|2 and I know almost nothing in practical error fixing.

Having said that, I'm grouping my questions here and you can answer whenever you have some free time.

Can you help me create a regex that catches and fixes empty parameters like ||, |}} or other of the sort in cite templates? I was checking Smallem's code and saw that I had no regexes for that problem. I tried setting one up myself but I was too afraid to continue with it given that those symbols are used in common tables as well while also being technical symbols for the regex language on its own. - Klein Muçi (talk) 07:56, 25 August 2021 (UTC)

(talk page stalker) Mine looks like this: str = str.replace(/({{\s*[cC]it[ea](?:[^}{]*(?:\{\{[^}{]*}}[^}{]*)*))\|\s*([\}\|])/gi, '$1 $2'); //empty parameter Trappist probably has a better one. I do not run this check unsupervised, because (edited to add: I believe that) there are non-CS1 templates that start with "Cite" and make use of empty unnamed parameters. – Jonesey95 (talk) 14:16, 25 August 2021 (UTC)

@Jonesey95, oh... Are there... Can you give me one example? That would put the whole plan to the bin though as I was hoping to let it work unsupervised periodically. I guess this is why Trappist hasn't been able to help me on this in the past. - Klein Muçi (talk) 14:22, 25 August 2021 (UTC)

Not off the top of my head, but they are out there, I believe. I think there is one related to citing court cases; I would have to dig for it though. You could work around it by making the regex more specific: one for cite web, one for cite journal, one for cite news, etc. It becomes unmanageable, though, when you realize how many redirects there are. – Jonesey95 (talk) 15:00, 25 August 2021 (UTC)

@Jonesey95, yes. Only the templates themselves are a lot, let alone the redirects. Sad that this problems exists. That category looks like a perfect candidate to be kept all time empty by a bot. But apparently the best you can do is AWB. :P - Klein Muçi (talk) 15:16, 25 August 2021 (UTC)

While monkbot task 18 was alive, that task cleared Category:CS1 errors: empty unknown parameters. Task 18 hid all non-cs1|2 templates using a big non-capturing group to identify those templates that are cs1|2. Once that was accomplished, the only 'visible' templates in an article were cs1|2 so fixing the unknown empties used three relatively simple regexes to examine and fix each individual cs1|2 template:

find: \|[ \t]*([\r\n]+[ \t]*[\|\}]) replace $1 – the newline variant to retain newlines in vertical-formatted citations

find: \|\s*([\|\}]) replace $1 – fixes || and |}

find: \|\s*=\s*([\|\}]) replace $1 – fixes |=| and |=}

Since sq:Kategoria:Gabime CS1: Parametra të panjohur bosh has only one article, writing any tool to fix the occasional article that appears in that category seems to require more effort than it's worth.

With regard to the identifier errors, most times, I find, the error is simple like |doi=http:dx.doi.org/<doi> where the value should just be the <doi> itself, not a url – same applies to jstor. I generally don't fix isbn errors because I don't know what the editor actually consulted (various versions of a work may, and often do, have different isbns). Properly fixing identifiers requires some research. I don't think that there is, or should be, a shortcut to fixing truly broken identifiers.

—Trappist the monk (talk) 15:55, 25 August 2021 (UTC)

You're correct on your logic regarding the small number of articles but the problem is that I suspect that that error is far too common. The category is only empty now because I worked on it but in general users tend to think that every parameter must be between two straight lines so you're prone to get a lot of |}} cases. Anyway, given that there isn't a safe way to deal with it, I'll leave it for the moment. Maybe if it becomes "a problem" in the future, I'll be back.

As for the identifiers, that's what I suspected. Moving on with a different category, how do I work with the articles listed here? I checked the English homologue but it only says to fix them in ways described in the help page in regard to the |display-author/editor/etc. parameter. Tried checking there but I don't really understand what am I supposed to do because there's no explicit mention of the type of error that I'm trying to fix. Would you be so kind as to give me 2 examples of articles fixed from the Albanian category so I know what to do? - Klein Muçi (talk) 16:35, 25 August 2021 (UTC)

For sq:Ariu polar, use {{make cite iucn}} to update the {{cite iucn}} template to reflect the current IUCN citation:

{{make cite iucn |Wiig, Ø., Amstrup, S., Atwood, T., Laidre, K., Lunn, N., Obbard, M., Regehr, E. & Thiemann, G. 2015. Ursus maritimus. The IUCN Red List of Threatened Species 2015: e.T22823A14871490. https://dx.doi.org/10.2305/IUCN.UK.2015-4.RLTS.T22823A14871490.en. Downloaded on 25 August 2021.|x}}

{{cite iucn |author=Wiig, Ø. |author2=Amstrup, S. |author3=Atwood, T. |author4=Laidre, K. |author5=Lunn, N. |author6=Obbard, M. |author7=Regehr, E. |author8=Thiemann, G. |year=2015 |title=''Ursus maritimus'' |volume=2015 |page=e.T22823A14871490 |doi=10.2305/IUCN.UK.2015-4.RLTS.T22823A14871490.en |access-date=25 August 2021}}

For stuff like this:

{{cite journal|last=Giosan|first=L. et al|title=Fluvial landscapes of the Harappan Civilization|journal=Proceedings of the National Academy of Sciences, USA|year=2012|volume=109|issue=26|url=http://www.pnas.org/content/109/26/E1688.full}}

do this:

{{cite journal |last=Giosan |first=L. |display-authors=etal |title=Fluvial landscapes of the Harappan Civilization |journal=Proceedings of the National Academy of Sciences, USA |year=2012 |volume=109 |issue=26 |url=http://www.pnas.org/content/109/26/E1688.full}}

—Trappist the monk (talk) 16:52, 25 August 2021 (UTC)

Thank you! Was able to solve everything. Take a look at the 7th citation of this article. What have I messed up in the CS1/Config. page?

1 example of fixes in this category? (I believe this is the last error category I have to ask you about.) - Klein Muçi (talk) 18:10, 25 August 2021 (UTC)

Invalid <code class="cs1-code">&#124;$1=$2</code>

{{cite book|script-title=ИЗДЕЛИЕ 1ПН51 ТЕХНИЧЕСКОЕ ОПИСАНИЕ И ИНСТРУКЦИЯ ПО ЭКСПЛУАТАЦИИ |trans-title= PRODUCT 1PN51 TECHNICAL DESCRIPTION AND OPERATING INSTRUCTIONS|date=janar 1992 |page= 11|language=ru}}

{{cite book |script-title=ru:ИЗДЕЛИЕ 1ПН51 ТЕХНИЧЕСКОЕ ОПИСАНИЕ И ИНСТРУКЦИЯ ПО ЭКСПЛУАТАЦИИ |trans-title=PRODUCT 1PN51 TECHNICAL DESCRIPTION AND OPERATING INSTRUCTIONS |date=janar 1992 |page=11 |language=ru}}

ИЗДЕЛИЕ 1ПН51 ТЕХНИЧЕСКОЕ ОПИСАНИЕ И ИНСТРУКЦИЯ ПО ЭКСПЛУАТАЦИИ [PRODUCT 1PN51 TECHNICAL DESCRIPTION AND OPERATING INSTRUCTIONS] (in Russian). janar 1992. p. 11. {{cite book}}: Check date values in: |date= (help)

The help text for this error is missing from sq:Ndihmë:Gabimet CS1 but is available at Help:CS1 errors.

—Trappist the monk (talk) 18:45, 25 August 2021 (UTC)

Thank you! Yes, it is missing as are a lot of other things. That's why I discussed about having that help page at Meta some discussions above.

To end it with questions in regard to errors' categories: Is there an easy way to remove some articles from this category? A kind of change, be that regex or plain text I can do with AWB en-masse to remove some of those? Or am I forced to check each error one by one? My first thought was to ask you for a capture regex that transforms each |publisher=value or |publisher=value to |publisher=value but I think that wouldn't be a good solution, would it? Any other similar ideas? - Klein Muçi (talk) 20:52, 25 August 2021 (UTC)

I wrote an awb script for that where I laboriously listed the names of newspapers, websites, magazines, journals, etc and then had the script replace the template name, and the |publisher=, and |work= (and aliases) parameters with the appropriate parameter and at the same time removed the bold and italic markup. That script became User:Monkbot/task 14. You can see my lists in its code. Simply removing the markup is, in my view, the wrong solution.

—Trappist the monk (talk) 21:23, 25 August 2021 (UTC)

Hmm... Can I use it? What do I do, import it as a module on AWB and start the process? Any needed changes? - Klein Muçi (talk) 21:51, 25 August 2021 (UTC)

I can't stop you ... I won't guarantee that it will do what you want, and I won't do any work to maintain the script if it doesn't work for you.

—Trappist the monk (talk) 22:27, 25 August 2021 (UTC)

Tried it and it worked really good. The list went from 133 to 104 pages. Anything is better than 133. - Klein Muçi (talk) 22:47, 25 August 2021 (UTC)

Any smart ways I can work with this specific category other than looking at every citation one by one? Given that it is a maint category, we aren't even shown which citation is the problematic one, let alone the parameter. Adding on that, how do we work with maint categories in general considering the lack of information I mentioned above. Any tips?

Extra unrelated question: Does a full list of all parameters exist somewhere? I was thinking of creating a regex to remove parameters without values, for example: |location=|date=|title=|etc. with the change being |parameter=| → | . This would create articles with a lot of empty parameters, too many vertical bars in a row, (wouldn't it?) but we already have a category for that and can be fixed after that step is taken. Of course if you have a better idea, please do tell me. As you already do know, cosmetic changes aren't a problem at SqWiki. - Klein Muçi (talk) 05:53, 26 August 2021 (UTC)

To see maintenance messages, see Help:CS1 errors § Controlling error message display; why isn't that part of sq:Ndihmë:Gabimet CS1? I have never bothered to automate fixing of articles in Category:CS1 maint: extra punctuation mostly because it isn't simple; there are parameters that are allowed to have trailing punctuation – the meta parameters that are skipped during the text are listed at Line 389.

cs1|2 doesn't use named parameters without assigned values so having a list of all parameters (they are listed in sq:Moduli:Citation/CS1/Whitelist) isn't needed. You might try something like this:

(\{\{\s*[Cc]it[ae][^\}]*)\|[^=]+=\s*([\|\}])

which gets replaced with captures 1 and 2.

—Trappist the monk (talk) 12:16, 26 August 2021 (UTC)

Oh thanks! I didn't know about that. Any fast advice how to make the color red instead of that light green? Or should I ask at the TechPump?

Umm... So, this:

(r"(\{\{\s*cit[aeio][^\}]*)\|[^=]+=\s*([\|\}])", r"\1\2"),?

And to conclude it with the maintenance categories, can you give me 2 examples of error fixing in this category? - Klein Muçi (talk) 16:06, 26 August 2021 (UTC)

Yeah, I think that's right.

What is to show? For cs1|2 templates that are marked with the maintenance message, delete |ref=<whatever>.

I chose the green because these are not errors. If you want to change the color just for yourself, you can change the css:

.mw-parser-output span.cs1-maint {display: inline; color:red} – the color of the red error messages is #d33

If you want to change it for everyone at sq.wiki (I don't think that you should), then change this line.

—Trappist the monk (talk) 01:01, 27 August 2021 (UTC)

Oh, no. Wanted to change it just for myself. It's a bit hard for my eyes to see it. If not red, maybe I'll choose some other color now that I know how so I can distinguish it from the errors. Can we hope that in the future, the maint. messages includes a bit more info? For example, for this category, it can include the parameter where the extra punctuation is located and maybe what the specific symbol is? The kind of functionality that currently exists for the error messages.

Moving on to the property categories, 2 questions which very well may be my last:

1. This category is supposed to be temporary. I also read the discussion which lead to its creation. What's the future plan for it or still undecided? Just curious.
2. In regard to this category, what do you think the shortened date range means in this article? Second citation. What about the 2 shortened data ranges here? I have quite a few of these cases that got me really confused. Maybe you have more experience on the subject. - Klein Muçi (talk) 01:31, 27 August 2021 (UTC)

PS: Should the regex be changed

From this:

(r"(\{\{\s*cit[aeio][^\}]*)\|[^=]+=\s*([\|\}])", r"\1\2"),

To this:

(r"(\{\{\s*cit[aeio][^\}]*\|\s*)[^=]+=\s*([\|\}])", r"\1\2"),? Klein Muçi (talk) 02:30, 27 August 2021 (UTC)

1. Don't know. I would like to see two of these go away: |location=, |place=, |publication-place=; we don't need all three nor do we need the 'written at' stuff that the templates render when |location= or |place= is used with |publication-place= ...

2. Don't worry about the abbreviated year range properties category while there are date errors: |date=2013-03 and |date=2016-12 are malformed and there is an error message that says so:

• "Diabetes". World Health Organization. 2013-03. {{cite web}}: Check date values in: |date= (help)
• "What is Diabetes?". National Institute of Diabetes and Digestive and Kidney Diseases (NIDDKD). 2016-12. {{cite web}}: Check date values in: |date= (help)

Fix the errors first. I should probably refine that category handling so that the properties cat isn't added while there are date errors and at the same time make it a maintenance cat ...

Why do you think that the regex should be changed?

—Trappist the monk (talk) 13:26, 27 August 2021 (UTC)

Oh! So it is an error! I was thinking it wasn't one and I was misunderstanding it. That's relieving in a way. :P As for the regex, I just compared it with 99% of the other regex lines of Smallem's fixes which all have the same part in the beginning, be that in regard to languages or parameter fixes. I tried testing both versions at regex101 but apparently I'm bad at even testing because neither of the version looked like it was working so I asked here. Should I continue with your version?

And last question: The description at this category "bothers" me a bit. It just says what it is but it doesn't mention what should we be doing with the articles found there. How are we supposed to exactly handle unfit/usurped links? Also there I see that |url-status=dead and |url-status=live are not tracked. Any particular reason why is that so? I mean, we already track citations that use languages with an ISO 639-2 code for no special reason, if I've understood your correctly in the past. Wouldn't those 2 details above be interesting enough to also track in this point of view?

Thanks in advance because this has been a wild ride! :) Klein Muçi (talk) 18:41, 27 August 2021 (UTC)

If you compare the two regexes at regex101 or in the awb regex tester, you will see that your regex introduces an extra pipe and that will cause an unknown empty parameter error; not good. In retrospect, I suspect that neither of our regexes are ideal. I suspect, but don't know, that if you set smallem to doing this replacement, it will remove one parameter per template per edit so a template with five empty parameters (not at all unheard of) will take five edits to be stripped of empty parameters. Not sure that that is what you want.

For the articles in Category:CS1 maint: unfit URL, often there is nothing that can be done; if the url in |url= is unfit, it's unfit. That's why there is an archive url. Sometimes its possible to find a better source; sometimes not. It is not our place to tell editors what to do. Sometimes we recommend; sometimes we don't.

As far as I know, there has never been a need or desire to track |url-status=dead (a parameter/value pair that can be deleted whenever encountered) or to track |url-status=live. In the next version of the module suite it will be possible to track individual parameters with individual property cats. There isn't a mechanism to track specific parameter/value pairs (|url-status=live but not |url-status=dead); you will only be able to track all uses of |url-status= regardless of assigned value. The next version will also allow the use of css to highlight those citations that emit properties categories.

—Trappist the monk (talk) 00:45, 28 August 2021 (UTC)

That's good enough. We're not in a hurry. As for the unfit URLs, I don't know, I was thinking of cases where you can find pornographic material and I thought that maybe you should remove it altogether? Or other cases of the sort like just blatant auto-promoting marketing, totally unrelated to anything on the text.

I think you've wanted to write |url-status=dead above. Or if not, I'm not really sure what you're trying to say. Anyway, the next update looks really exciting. Thank you A LOT for your time in explaining things to me! :)) - Klein Muçi (talk) 05:49, 28 August 2021 (UTC)

Porn and other content unrelated to the 'citation' at a previously good url is why we created |url-status=unfit and |url-status=usurped and why the original url in |url= is not linked in the rendered citation – readers will never see the porn unless they actively seek it – of course, some bot may have chosen an archived snapshot of the porn, but that's the fault of the bot, not the fault of cs1|2.

Yeah, |url-status=dead. Fixed.

—Trappist the monk (talk) 11:45, 28 August 2021 (UTC)

I understand but... If we're not supposed to remove the unfit links, we're basically supposed to do nothing with that category? Why track it if it so? It feels strange compared with any other maintenance categories. - Klein Muçi (talk) 15:38, 28 August 2021 (UTC)

How good is this most-likely-monstrosity I've written to fix date format errors?

(r"(\{\{\s*cit[aeio][^\}]*\|\s*)((?:date|year)\s*=\s*(\w*))\-(\w*)", r"\1\–\2"), - Klein Muçi (talk) 21:14, 28 August 2021 (UTC)

If you are trying to fix something like:

{{cite book |title=Title |year=2020-2021}} – you want to change the hyphen to an endash

the regex fix: {{cite book |title=Title |–year=2020}}

Title. {{cite book}}: Unknown parameter |–year= ignored (help)

or trying to fix something like:

{{cite book |title=Title |year=2021-08}} – the proper fix for that is |date=August 2021

the regex fix: {{cite book |title=Title |–year=2021}}

Title. {{cite book}}: Unknown parameter |–year= ignored (help)

but this happens when there is nothing to be fixed:

{{cite book |title=Title |year=2021-08-28}} – I would use |date= because it is a whole date

the regex fix: {{cite book |title=Title |–year=2021-28}}

Title. {{cite book}}: Unknown parameter |–year= ignored (help)

—Trappist the monk (talk) 21:47, 28 August 2021 (UTC)

Ooh... How silly... I didn't think about the cases when the date was correct. :P I was just trying to automatize the fixes here. Now that you mentioned full correct dates I don't that there's a regex that could suffice for that, is there? :/ Considering Smallem works with every article on the Mainspace. - Klein Muçi (talk) 22:03, 28 August 2021 (UTC)

I once wrote an awb settings file to fix those issues. The is simple enough that I didn't bother to write a module for it just a bunch of individual regexes for all of the various date formats including CITEREF disambiguation. One regex to fix all date formats (if that is even possible) would be a nightmare. Multiple simple regexes is the way to go.

—Trappist the monk (talk) 00:19, 29 August 2021 (UTC)

Any chance you have a copy/can do that again? Haha! Considering my current knowledge on regex, that too would be a nightmare. - Klein Muçi (talk) 06:22, 29 August 2021 (UTC)

Silly detail but I thought asking: Should we consider for maintenance cases when |display-authors=X-number when X-number < number of authors? Klein Muçi (talk) 22:33, 29 August 2021 (UTC)

What needs to be maintained? The purpose of |display-authors=n where n < the number of authors is to truncate the rendered author list to a managable length. The purpose of a citation is to help readers locate a copy of a source; if a source has a lot of authors, the reader only needs the first few – and displaying all of them is a bit overwhelming:

Okbay, Aysu; Baselmans, Bart M L; De Neve, Jan-Emmanuel; Turley, Patrick; Nivard, Michel G; Fontana, Mark Alan; Meddens, S Fleur W; Linnér, Richard Karlsson; Rietveld, Cornelius A; Derringer, Jaime; Gratten, Jacob; Lee, James J; Liu, Jimmy Z; de Vlaming, Ronald; Ahluwalia, Tarunveer S; Buchwald, Jadwiga; Cavadino, Alana; Frazier-Wood, Alexis C; Furlotte, Nicholas A; Garfield, Victoria; Geisel, Marie Henrike; Gonzalez, Juan R; Haitjema, Saskia; Karlsson, Robert; van der Laan, Sander W; Ladwig, Karl-Heinz; Lahti, Jari; van der Lee, Sven J; Lind, Penelope A; Liu, Tian; Matteson, Lindsay; Mihailov, Evelin; Miller, Michael B; Minica, Camelia C; Nolte, Ilja M; Mook-Kanamori, Dennis; van der Most, Peter J; Oldmeadow, Christopher; Qian, Yong; Raitakari, Olli; Rawal, Rajesh; Realo, Anu; Rueedi, Rico; Schmidt, Börge; Smith, Albert V; Stergiakouli, Evie; Tanaka, Toshiko; Taylor, Kent; Thorleifsson, Gudmar; Wedenoja, Juho; Wellmann, Juergen; Westra, Harm-Jan; Willems, Sara M; Zhao, Wei; Amin, Najaf; Bakshi, Andrew; Bergmann, Sven; Bjornsdottir, Gyda; Boyle, Patricia A; Cherney, Samantha; Cox, Simon R; Davies, Gail; Davis, Oliver S P; Ding, Jun; Direk, Nese; Eibich, Peter; Emeny, Rebecca T; Fatemifar, Ghazaleh; Faul, Jessica D; Ferrucci, Luigi; Forstner, Andreas J; Gieger, Christian; Gupta, Richa; Harris, Tamara B; Harris, Juliette M; Holliday, Elizabeth G; Hottenga, Jouke-Jan; De Jager, Philip L; Kaakinen, Marika A; Kajantie, Eero; Karhunen, Ville; Kolcic, Ivana; Kumari, Meena; Launer, Lenore J; Franke, Lude; Li-Gao, Ruifang; Liewald, David C; Koini, Marisa; Loukola, Anu; Marques-Vidal, Pedro; Montgomery, Grant W; Mosing, Miriam A; Paternoster, Lavinia; Pattie, Alison; Petrovic, Katja E; Pulkki-Råback, Laura; Quaye, Lydia; Räikkönen, Katri; Rudan, Igor; Scott, Rodney J; Smith, Jennifer A; Sutin, Angelina R; Trzaskowski, Maciej; Vinkhuyzen, Anna E; Yu, Lei; Zabaneh, Delilah; Attia, John R; Bennett, David A; Berger, Klaus; Bertram, Lars; Boomsma, Dorret I; Snieder, Harold; Chang, Shun-Chiao; Cucca, Francesco; Deary, Ian J; van Duijn, Cornelia M; Eriksson, Johan G; Bültmann, Ute; de Geus, Eco J C; Groenen, Patrick J F; Gudnason, Vilmundur; Hansen, Torben; Hartman, Catharine A; Haworth, Claire M A; Hayward, Caroline; Heath, Andrew C; Hinds, David A; Hyppönen, Elina; Iacono, William G; Järvelin, Marjo-Riitta; Jöckel, Karl-Heinz; Kaprio, Jaakko; Kardia, Sharon L R; Keltikangas-Järvinen, Liisa; Kraft, Peter; Kubzansky, Laura D; Lehtimäki, Terho; Magnusson, Patrik K E; Martin, Nicholas G; McGue, Matt; Metspalu, Andres; Mills, Melinda; de Mutsert, Renée; Oldehinkel, Albertine J; Pasterkamp, Gerard; Pedersen, Nancy L; Plomin, Robert; Polasek, Ozren; Power, Christine; Rich, Stephen S; Rosendaal, Frits R; den Ruijter, Hester M; Schlessinger, David; Schmidt, Helena; Svento, Rauli; Schmidt, Reinhold; Alizadeh, Behrooz Z; Sørensen, Thorkild I A; Spector, Tim D; Starr, John M; Stefansson, Kari; Steptoe, Andrew; Terracciano, Antonio; Thorsteinsdottir, Unnur; Thurik, A Roy; Timpson, Nicholas J; Tiemeier, Henning; Uitterlinden, André G; Vollenweider, Peter; Wagner, Gert G; Weir, David R; Yang, Jian; Conley, Dalton C; Smith, George Davey; Hofman, Albert; Johannesson, Magnus; Laibson, David I; Medland, Sarah E; Meyer, Michelle N; Pickrell, Joseph K; Esko, Tõnu; Krueger, Robert F; Beauchamp, Jonathan P; Koellinger, Philipp D; Benjamin, Daniel J; Bartels, Meike; Cesarini, David; Cesarini, D (June 2016). "Genetic variants associated with subjective well-being, depressive symptoms, and neuroticism identified through genome-wide analyses". Nature Genetics. 48 (6): 624–633. doi:10.1038/ng.3552. PMC 4884152. PMID 27089181.

—Trappist the monk (talk) 23:37, 29 August 2021 (UTC)

LOL My bad. I meant the total opposite of that. X-number > number of authors. When, for example, we have only 2 authors but the display number is set to 29. - Klein Muçi (talk) 23:45, 29 August 2021 (UTC)

Still, what needs to be maintained? We have error messages for that condition:

{{cite journal |last1=Okbay |first1=Aysu |last2=Baselmans |first2=Bart M L |title=Genetic variants associated with subjective well-being, depressive symptoms, and neuroticism identified through genome-wide analyses |journal=Nature Genetics |date=June 2016 |volume=48 |issue=6 |pages=624–633 |doi=10.1038/ng.3552 |pmid=27089181 |pmc=4884152 |display-authors=29}}

Okbay, Aysu; Baselmans, Bart M L (June 2016). "Genetic variants associated with subjective well-being, depressive symptoms, and neuroticism identified through genome-wide analyses". Nature Genetics. 48 (6): 624–633. doi:10.1038/ng.3552. PMC 4884152. PMID 27089181. {{cite journal}}: Invalid |display-authors=29 (help)

—Trappist the monk (talk) 00:07, 30 August 2021 (UTC)

That's what I meant. Apparently that exists already. I was unlucky (and a bit hurried) to stumble upon this citation:

{{Cite journal|last1=Lindblad-Toh|first1=K.|last2=Wade|first2=C. M.|last3=Mikkelsen|first3=T. S.|last4=Karlsson|first4=E. K.|last5=Jaffe|first5=D. B.|last6=Kamal|first6=M.|last7=Clamp|first7=M.|last8=Chang|first8=J. L.|last9=Kulbokas|first9=E. J.|display-authors=29|year=2005|title=Genome sequence, comparative analysis and haplotype structure of the domestic dog|journal=Nature|volume=438|issue=7069|pages=803–819|bibcode=2005Natur.438..803L|doi=10.1038/nature04338|pmid=16341006|first10=M. C.|last27=Cuff|first24=J.|last25=Chin|first25=C.W.|last26=Cook|first26=A.|last30=Gnerre|first27=J.|last28=Daly|first28=M.J.|last29=Decaprio|first29=D.|first23=W.|first30=S.|last24=Butler|last23=Brockman|last11=Mauceli|first16=C. P.|first11=E.|last12=Xie|first12=X.|last13=Breen|first13=M.|last14=Wayne|first14=R. K.|last15=Ostrander|first15=E. A.|last16=Ponting|last17=Galibert|first22=T.|first17=F.|last18=Smith|last10=Zody|last19=Dejong|first19=P.J.|last20=Kirkness|first20=E.|last21=Alvarez|first21=P.|last22=Biagi|first18=D. R.}}

If you can see, the list of authors is cut off in the middle with other parameters and I thought that |display-authors=29 wasn't emitting an error message even though they weren't 29 authors because I didn't read until the end. Sorry, as I've said many times above, this is my first time I'm dealing with citation fixing manually. - Klein Muçi (talk) 00:25, 30 August 2021 (UTC)

question

Purplinko (talk) 00:33, 2 September 2021 (UTC) So you are from down under...?

Hyphenated / non-hyphenated parameters

Hi Trappist

I hope all is well. Just a friendly reminder that, per the results of the lengthy RFC earlier this year, the community determined that there is no preference between the hyphenated and non-hyphenated versions of parameters such as {{{access-date}}}, {{{archive-date}}} etc. and that the non-hyphenated versions are not deprecated, with bots forbidden to make such changes. As such, changes such as this one should not be switching from one valid form to another while doing other useful changes. I don't know if you're using an automated tool for such edits, but if so I'd suggest updating it so that it doesn't make such changes in future. Cheers  — Amakuru (talk) 13:21, 31 August 2021 (UTC)

I think that you are reading something into that close that is not there. The close says: Bot removal of non-hyphenated parameters from transclusions, i.e. Monkbot task 18, does not have community consensus and also Option C: Non-hyphenated parameters should not be deprecated; deprecation should not be continued and bot approval should be revoked. There is nothing in that close to prevent editors from switching to the hyphenated form of the lingering six parameters (switching is not deprecation it is merely switching). I am not a bot. In the diff that you provided, you will see that I made that edit, Monkbot did not. Monkbot/task 18 has not edited anything since 3 February 2021 (roughly 2.5 months before the closure).

If all I did in that edit was switch to the hyphenated forms, you would have grounds to chastise me. But, as you can see from the diff and by comparing §References before to §References after, I made fixes to three of the seven references that removed the red error messages and removed the article from Category:CS1 errors: extra text: volume, Category:CS1 errors: URL, and Category:CS1 errors: missing pipe.

—Trappist the monk (talk) 14:34, 31 August 2021 (UTC)

Please please don't re-open this can of worms, we've discussed this to death already this year. I get that you preferred a different outcome from the one that was reached, and you would prefer all params to be hyphenated, but that wasn't what was decided, that's the way it goes sometimes. But as much as the other fixes you made to that page were useful and needed, removing the error messages, it still isn't right for you to be switching the other params from hyphenated to non-hyphenated in the process. Unnecessary changes are still unnecessary changes even when accompanied by necessary changes. And see the top of WP:MOS for a spiel about switching between optional styles: "Where more than one style or format is acceptable under MoS, one should be used consistently within an article and should not be changed without good reason". Obviously this isn't a style question, but the same principle applies. Switching between two equally acceptable parameter formats adds noise to the diffs, and can cause irritation for people who prefer one version or the other. Note that it has been switched off in AWB too for exactly this reason. We can seek more opinions if you like, but it seems to me it would be far easier if you adjust your scripts to omit these hyphen switches, and then you can continue doing your excellent work fixing citations without such changes being made. Thanks  — Amakuru (talk) 15:21, 31 August 2021 (UTC)

Unless you are the Wikipedia Monarch, you cannot, with the one hand, make accusations of misbehavior and then, with the other hand, prevent self-defense.

You know, for a community that relies so heavily on RfCs, you would think that there would be a central suppository of all RfCs, indexed by topic and date and outcome and etc. There is not. I think that what you are trying to work your way around to is WP:CITEVAR. There was a discussion not too long ago that referenced a couple of RfCs that appear to say that WP:CITEVAR does not apply to the form of a wikitext citation when the differences undetectable by the reader. No one in that discussion suggested another interpretation. Perhaps you might want to start yet-another-citevar-rfc to nail down, to the last jot and tittle, that which editors may or may not do while fixing citation templates.

If you don't like something that I have done, revert me, I have no objections – so long as you retain the core fixes to the citations.

—Trappist the monk (talk) 18:20, 31 August 2021 (UTC)

@Amakuru, sorry for interfering but I do believe the discussion above can be rather simple to close. TTM made a personal change in good faith, you make a personal change in good faith back if you don't like that. The whole passive aggressive tone with phrases such as ...that's the way it goes sometimes. can be well reserved for other cases. - Klein Muçi (talk) 08:58, 1 September 2021 (UTC)

@Klein Muçi: my intention was to be friendly about this matter, on the grounds that (a) I don't like fights, and (b) it's a fairly minor issue in the grand scheme of things. And in saying "that's the way it goes sometimes", I was alluding to the fact that I've been on the "wrong" end of RfCs a few times myself over the years, and it's always annoying when it happens, but as Wikipedians we just have to move on. If it came across as passive aggressive, then I apologise to Trappist. That said, though, I thought it fair to ask that unnecessary changes to the parameters used in citations not be made in fashion, given all the hot air that occurred on this very topic earlier this year. I'm certainly not about to get into an edit war on the matter, I just thought it was a natural corollary of the decisions made earlier this year, that's all. Thanks  — Amakuru (talk) 22:04, 2 September 2021 (UTC)

@Amakuru, as a note to myself I must say that reader-writer interactions do require more attention than speaker-listener ones. Using jargon, in my eyes that was read as "tough luck, suck it up", which looked like an unneeded revanchist expression for this situation but given your explanations, I was wrong. Thank you for taking the time to clear that up. - Klein Muçi (talk) 22:31, 2 September 2021 (UTC)

Suggestions for unknown language parameters

Take a look at the cite errors here. Not many editors know of the ISO codes, they just tend to write whatever abbreviation they can think of and hope for the best. Sometimes it works, sometimes the abbreviation is not abbreviated enough (like in that case) and sometimes the abbreviation is not even close to the ISO code (a common error is |language=al (for Albanian) instead of |language=sq). Is there any chance that the module might start giving suggestions in regard to this phenomenon? Directing an user from |language=eng to |language=en should be fairly easy if they had a simple message saying that no? - Klein Muçi (talk) 09:06, 4 September 2021 (UTC)

Maintenance cats don't have messaging except at the category page because maint cats are hidden from readers and hidden from those editors who have not explicitly enabled the messaging. Even if we did include clues in the messaging, I don't know of a systematic way to make suggestions about should be in the language parameter. This is why we link to Template:Citation Style documentation/language/doc from Category:CS1 maint: unrecognized language.

—Trappist the monk (talk) 13:05, 4 September 2021 (UTC)

For me it looks straightforward to have clues about the most common errors for example, for "eng", "ita" or for the inclusion of "and". That list can be expanded as new common errors are found. But you have more experience and maybe understand that there are a lot more of different kind of errors which would be hard to "clue-solve" because of the risk of misdirections. - Klein Muçi (talk) 14:48, 4 September 2021 (UTC)

Wikilinks in citations

Today I stumbled upon this discussion and it made me think: Are there certain standards or common practices to follow when it comes to wikilinks in citations? I'm asking because before I used to think that wikiformatting was a good thing, italicizing certain names, until that became a specific error. So this discussion now made me curious about wikilinks. I'm sure your experience on the matter can be helpful. - Klein Muçi (talk) 21:56, 4 September 2021 (UTC)

regex for fq

Unrelated question: Can you give me a regex that removes all cases of cite templates without parameters while following the standard format I've sent many times above? For example {{cite web}}, {{cite journal}}, {{cite book}}, etc. This would be an extra step in helping against empty citations. I'm assuming it would be safe to procced with something like that without being afraid of ruining anything else, no? I'm usually wrong in these assumptions. - Klein Muçi (talk) 00:14, 2 September 2021 (UTC)

(r"(\{\{\s*cit[aeio][^\}]*\|\s*pages?)\s*=\s*fq\.?", r"\1="),

This catches cases like this:{{cite book|last=Sylejmani|first=Agim|title=Rexhep Mala, Zjarr lufte që nuk shuhet|date=janar 2014|publisher=Shoqata e të Burgosurve Politikë e Kosovës|location=Prishtinë|pages=Fq.39–40}} (check the last parameter, the "fq." part is extra unneeded text)

How can I make it so it also catches cases like these without creating specific regexes for each of them:

{{cite book|last=Sylejmani|first=Agim|title=Rexhep Mala, Zjarr lufte që nuk shuhet|date=janar 2014|publisher=Shoqata e të Burgosurve Politikë e Kosovës|location=Prishtinë|pages=39–40 fq.}}

{{cite book|last=Sylejmani|first=Agim|title=Rexhep Mala, Zjarr lufte që nuk shuhet|date=janar 2014|publisher=Shoqata e të Burgosurve Politikë e Kosovës|location=Prishtinë|pages=39–40, 40-72 fq.}}

{{cite book|last=Sylejmani|first=Agim|title=Rexhep Mala, Zjarr lufte që nuk shuhet|date=janar 2014|publisher=Shoqata e të Burgosurve Politikë e Kosovës|location=Prishtinë|pages=39 fq.}} And more...

Basically the "fq" part, with or without an extra dot, floating around the other possible values. - Klein Muçi (talk) 08:34, 2 September 2021 (UTC)

(r"(\{\{\s*cit[aeio][^\}]*\|\s*pages?\s*=[^\|\}]*)fq\.?", r"\1"),

—Trappist the monk (talk) 14:21, 2 September 2021 (UTC)

Thank you! Works very well. What about the cite templates without parameters? Can that be generalized with 1 regex? - Klein Muçi (talk) 17:07, 2 September 2021 (UTC)

Also, this regex works almost-fine for removing variations of (red.) (same as ed.) in editor related parameters:

(r"(\{\{\s*cit[aeio][^\}]*\|\s*(?:veditors|editor|editor(\d)|editor-surname|editor(\d)-surname|editor-first|editor(\d)-first|editor-last|editor(\d)-last)\s*=[^\|\}]*)\(?red\.?\)?", r"\1"),

It malfunctions at the left parentheses though, not sure why. :/ Also, can it be optimized somehow as to not get that long of an expression? I just got your regex above and hacked it for this occasion given that the logic is the same. - Klein Muçi (talk) 17:47, 2 September 2021 (UTC)

(r"(\{\{\s*cit[aeio][^\}]*\|\s*(?:p|pp|page|pages|quote-page|quote-pages)\s*=[^\|\}]*)(?:|p|pp|pgs|pages|fq\.?)", r"\1"),

Tried creating a format to cover all cases in English and Albanian. It works fine in matching (I believe) but malfunctions in replacing. What am I doing wrong? The idea is to do the same for editor, volume, edition, issue and volume. If you can fine-tune it, I can modify it for the other mentioned cases. - Klein Muçi (talk) 18:05, 2 September 2021 (UTC)

(\{\{\s*cit[aeio][^\}]*\|\s*v?editors?\d?(?:-surname|-first|-last)?\s*=[^\|\}]*)\(?red\.?\)?

Was able to further refine the above regex. It still malfunctions with the left parenthesis. I haven't taken into account all the extra text possibilities in it. Writing in real time while working. - Klein Muçi (talk) 18:23, 2 September 2021 (UTC)

Try this:

(\{\{\s*cit[aeio][^\}]*\|\s*(?:pp?|pages?|quote-pages?)\s*=[^\|\}]*?)(?:pages|pgs|pp?|fq)\.?

—Trappist the monk (talk) 21:52, 2 September 2021 (UTC)

Yes, it works. Can you also modify the other regexes down below? I don't wanna risk ruining them. Do you think they're good enough not to attract false positives? - Klein Muçi (talk) 22:03, 2 September 2021 (UTC)

This search suggests little need for an empty-citation finder. And, simply deleting an apparently empty template is often the wrong thing to do.

—Trappist the monk (talk) 20:06, 2 September 2021 (UTC)

Okay. I wasn't too sure myself for that.

As for removing ed/red in CS1 maint: extra text:<param>, I compiled this:

(r"(\{\{\s*cit[aeio][^\}]*\|\s*(?:v?authors?|contributor|v?editors?|interviewer|translator)\d?(?:-surname|-first|-last)?\d?\s*=[^\|\}]*)\(?r?ed\.?\)?", r"\1"),

Tried to include all known cases. What do you think? It malfunctions at (r. - Klein Muçi (talk) 21:14, 2 September 2021 (UTC)

(r"(\{\{\s*cit[aeio][^\}]*\|\s*edition\s*=[^\|\}]*)(?:edn?|edition)", r"\1"),

(r"(\{\{\s*cit[aeio][^\}]*\|\s*(?:issue|number)\s*=[^\|\}]*)(?:iss|no)", r"\1"),

(r"(\{\{\s*cit[aeio][^\}]*\|\s*(?:pp?|pages?|quote-pages?)\s*=[^\|\}]*)(?:pp?|pgs|pages|fq\.?)", r"\1"),

(r"(\{\{\s*cit[aeio][^\}]*\|\s*volume\s*=[^\|\}]*)(?:vol\.?|volume)", r"\1"),

Also these 4 for CS1 errors: extra text:<param> They all have the same malfunction in the end, trying to fix it. Klein Muçi (talk) 21:31, 2 September 2021 (UTC)

Try these:

(\{\{\s*cit[aeio][^\}]*\|\s*edition\s*=[^\|\}]*?)(?:edition|edn?\.?)\s*

(\{\{\s*cit[aeio][^\}]*\|\s*(?:issue|number)\s*=[^\|\}]*?)(?:issue|iss\.?|no\.?)\s*

(\{\{\s*cit[aeio][^\}]*\|\s*(?:pp?|pages?|quote-pages?)\s*=[^\|\}]*?)(?:pp?|pgs|pages|fq\.?)\s*

(\{\{\s*cit[aeio][^\}]*\|\s*volume\s*=[^\|\}]*?)(?:volume|vol\.?)\s*

—Trappist the monk (talk) 23:18, 2 September 2021 (UTC)

Thanks a lot! Was able to fix even this after your suggestion:

(r"(\{\{\s*cit[aeio][^\}]*\|\s*(?:v?authors?|first|last|contributor|v?editors?|interviewer|translator)\d?(?:-surname|-first|-last)?\d?\s*=[^\|\}]*?)\(r?ed\.?\)", r"\1"),

I made it a bit more specific (only catches extra text in parentheses now) so it doesn't give false positives in author/contributor/editor/interviewer/translator names. - Klein Muçi (talk) 23:41, 2 September 2021 (UTC)

To end it with regex requests, can you help me set up a spacing standardizing regex between cite template elements? We've talked about this in the past. Basically converting everything, horizontal and vertical cite styles, into this {{cite template |parameter1=value |parameter2=value |parameter3=value |parameter4=value...}} After a bit of consideration, especially after working with citation repairs manually these days I've noticed that no editor cares much about the style used. Truth be told, they rarely care about the content of the citations themselves for as long as it doesn't render anything red, which is a bit sad really. Imposing a certain standard and unifying the style between articles will at least improve readability and technical maintenance on them. - Klein Muçi (talk) 08:08, 3 September 2021 (UTC)

Not an easy task because cs1|2 templates can contain other templates where the piping rules may require no extra whitespace, wikilinks that have pipes, etc. Are you really sure that there is a need for such a tool?

—Trappist the monk (talk) 11:09, 3 September 2021 (UTC)

Oh, I had totally not thought about the templates inside templates scenario. Given that it is cosmetic in nature, there isn't a need per se but the fact is that currently it may be safe to say that no one maintains that part of articles in any article. You can put as much spaces as you want in any article you choose and if you don't create an error in rendering, that change will be stuck there forever. After much time in fixing cite errors manually, there have been A LOT of styling inconsistencies introduced in any article, many of those by me, so much that you rarely have articles that follow one single style from start to end. To be honest, in most of the cases, it changes many times even on the same template, with each parameter having a different kind of spacing so... At this situation, any kind of order would be better. But, as I said, I hadn't thought about that aforementioned scenario. I don't know if it is even possible considering that detail now. - Klein Muçi (talk) 23:21, 3 September 2021 (UTC)

Please, take a look at this now-reverted edit Smallem made. In the edit summary you can see the specific regex it used. Can you explain to me what's going wrong there? I tried using regex101 to understand it but with the same regex and citation it gives no match. I'm really confused. - Klein Muçi (talk) 22:52, 4 September 2021 (UTC)

I think I need to switch from:

(r"(\{\{\s*cit[aeio][^\}]*\|\s*(?:pp?|pages?|quote-pages?)\s*=[^\|\}]*?)(?:pp?|pgs|pages|fq\.?)\s*", r"\1"),

to:

(r"(\{\{\s*cit[aeio][^\}]*\|\s*(?:pp?|pages?|quote-pages?)\s*=[^\|\}]*?)(?:pp?|pgs|pages|fq\.?)\b\s*", r"\1"),, no? Added a word boundary. Even though I don't know why regex101 doesn't show any match in the above case. Klein Muçi (talk) 00:16, 5 September 2021 (UTC)

Regex101 will match using the original search string if you set it to be case insensitive. Does Smallem do case insensitive searches? Must do else we would have written:

(\{\{\s*[Cc]it[aeio][^\}]*\|\s*(?:pp?|pages?|quote-pages?)\s*=[^\|\}]*?)(?:pp?|pgs|pages|fq\.?)\s*", r"\1")

\b will work for the case where a dot follows p but it will leave the dot. Perhaps this:

(\{\{\s*[Cc]it[aeio][^\}]*\|\s*(?:pp?|pages?|quote-pages?)\s*=[^\|\}]*?)(?:pp?|pgs|pages?|fq)(?:\.\s+|\.|\s+)

where the extra text must be followed by a <dot><space> pair, a <dot>, or a <space>. This will skip anything that is like fq123 which could be a separate regex because p123 might be legit.

—Trappist the monk (talk) 00:55, 5 September 2021 (UTC)

Ooh, yes. Smallem performs case insensitive searches. I forgot that detail. And yes, I had to stop Smallem because it started doing just that.

Okay, thank you!

What about these other 2 regexes which are also problematic at the moment:

(r"(\{\{\s*cit[aeio][^\}]*\|\s*v?editors?\d?(?:-surname|-first|-last)?\d?\s*=[^\|\}]*?)\(?(?:red\.?|ed\.?|editors?)\)?", r"\1"),

(r"(\{\{\s*cit[aeio][^\}]*\|\s*edition\s*=[^\|\}]*?)(?:edition|edn?\.?)\s*", r"\1"),

Here are some problems they caused now reverted: [1], [2], [3]

I tried fixing them with word boundaries but I started having problems with the dots. Klein Muçi (talk) 08:42, 5 September 2021 (UTC)

I'm guessing I should do this, no?

(r"(\{\{\s*cit[aeio][^\}]*\|\s*v?editors?\d?(?:-surname|-first|-last)?\d?\s*=[^\|\}]*?)\(?(?:r?ed|editors?)(?:\.\s+|\.|\s+)\)?", r"\1"),

(r"(\{\{\s*cit[aeio][^\}]*\|\s*edition\s*=[^\|\}]*?)(?:edition|edn?)(?:\.\s+|\.|\s+)", r"\1"),

And I really thank you for your diligence in teaching regular expressions to me. I'm honestly surprised by how much I have improved in complexity ever since this conversation started. - Klein Muçi (talk) 08:56, 5 September 2021 (UTC)

I know for sure that at least the second regex isn't right because of this edit. :( - Klein Muçi (talk) 11:09, 5 September 2021 (UTC)

After many tests during this day I came to the conclusion that the edition, issue and volume parameter have far too unpredictable values given by the lack of constraints those parameters have and Smallem usually does more harm than good, even when the regexes are set up right. I decided to keep the regexes about pages and the one about editors (authors', interviewers' and contributors' mistakes, even though they have the same mistake, apparently can't be solved by Smallem for reasons you most likely already know). The pages ones look already good (I'm yet to run 1 full test with Smallem but I've run a lot of small tests and they show no errors) as for the one for editors, I'm currently trying this:

(r"(\{\{\s*cit[aeio][^\}]*\|\s*v?editors?\d?(?:-surname|-first|-last)?\d?\s*=[^\|\}]*?)\s+\(?r?ed(?:\.\s+|\.|\s+)\)?", r"\1"),

The new part is the \s+ before the supposed r/ed part. I'll report back if I stumble on any errors. - Klein Muçi (talk) 21:09, 6 September 2021 (UTC)

So far it seems to work good, I have to wait until the test is over though. Meanwhile I had to remove also the regex for the pages in English for the same reasons. Too unreliable. I'm keeping just the one for the Albanian abbreviation fq and the r/ed one, which seems to be working good, without any false positives. If my words are any worth though, I'd really suggest to start putting some constrains in those parameters (issue/volume/edition/page) or at least check for some more details than just plain parameter name replication. People really do use a lot of creativity when filling up those values, a lot of times putting values that belong somewhere else (chapter and page get merged quite often, just an example). Anyway, I'll wait until the test is completely finished. - Klein Muçi (talk) 16:48, 7 September 2021 (UTC)

Test is over. As a conclusion I'm only keeping those 2 regexes I mentioned above. Whenever you have some time, I'd appreciate it if you could clarify the details with the charts module so I know what to do with the old (sub)module. It is by no way an urgent matter so don't worry too much though. - Klein Muçi (talk) 15:11, 8 September 2021 (UTC)

graphs

I noticed something wrong with the graph in this category: Mirëmbajtja CS1: Emra të shumëfishtë 5 faqe (... 5 pages). Meanwhile, that said category has 5 subcategories with one of them having +200 pages). The graph is supposed to show recursive results but apparently does not. What have I done wrong on it? This is rather a big problem considering we have quite some grouped categories in maint. and error categories. - Klein Muçi (talk) 09:05, 1 September 2021 (UTC)

The {{PAGESINCATEGORY}} documentation doesn't mention recursion. Since the magic word is expensive, I guess I would be surprised if it were recursive.

—Trappist the monk (talk) 12:16, 1 September 2021 (UTC)

Uhm... Do I have any solution to get recursive results for subcategories apart from tracking each category alone? I've thought that "all pages" meant recursion. :/ - Klein Muçi (talk) 17:09, 1 September 2021 (UTC)

I can imagine a lua module that, if given a list of a category's subcategories would get the individual results from {{PAGESINCATEGORY}} and return a string that might read 'Mirëmbajtja CS1: Emra të shumëfishtë 220 pages, 0 files in 5 subcategories' or some such but the list of which category has which subcategories would have to be maintained manually because Scributo doesn't have access to category contents. There is some advantage to having a lua module do this because it has this library call: mw.site.stats.pagesInCategory ('Mirëmbajtja CS1: Emra të shumëfishtë', '*') which returns a table that looks something like this:

table#1 {
  ["all"] = 5,
  ["files"] = 0,
  ["pages"] = 0,
  ["subcats"] = 5,
  }

—Trappist the monk (talk) 19:03, 1 September 2021 (UTC)

That seems interesting but, first and foremost, that looks like too much work and secondly, would I be able to track the said results into a graph (automatically, like I'm currently doing)? - Klein Muçi (talk) 20:47, 1 September 2021 (UTC)

More automatically perhaps. Right now, sq:Stampa:Grafiku i kategorive të mirëmbajtjes CS1 has a lot of repeat stuff: each group n (there are 27) calls {{PAGESINCATEGORY}} with the category name; each tooltip n repeats the category name twice (once in another call to {{PAGESINCATEGORY}}); each group name in group names repeats the category name three times – twice to make the category link and yet another call to {{PAGESINCATEGORY}}. A module can import sq:Moduli:Citation/CS1/Configuration and fetch category names from error_conditions.error condition.category and error_conditions.maint condition.category so the new module would be guaranteed to use the correct category name (so long as the error condition and maint condition keys are correct).

Here is a hack that gets the essentials more-or-less right: sq:Moduli:Livadhi/trappist_the_monk/graph_experiment and you can see what the output looks like with this command in the debug console: =p.graph_params_create(). Alas, just dropping the invoke for this module into a copy of the invoke in Stampa:Grafiku i kategorive të mirëmbajtjes CS1 doesn't work but, you can copy the text from debug console and replace the matching bits in the template and it will work. I have to think about getting round that.

—Trappist the monk (talk) 23:42, 1 September 2021 (UTC)

Hahaha You're crazy man! XD Yes, excellent results! So, now we have a specific Lua module just for creating the technical side of the graphic? I see that the colors are the only one that I should add manually. Basically now I just copy-paste the results from the debug mode to the corresponding places in the template and I'm done? Can the same module be used for the other 2 graphs as well? This one and this one? - Klein Muçi (talk) 00:00, 2 September 2021 (UTC)

Also, given that the graph values are created automatically now... Maybe the whole subcategories being grouped together thing is not needed anymore. Those can be categories on its own and be tracked individually on the graph now that I don't risk messing up the whole graph while adding 5-10 new categories. If only it could take care of the colors as well... Maybe it can take values from here somehow while taking care not to put similar colors together? - Klein Muçi (talk) 08:51, 2 September 2021 (UTC)

Hacked on some more and now calls sq:Moduli:Chart. See at sq:Përdoruesi:Trappist the monk/Livadhi personal; still some work to do with the bar chart's x legend.

—Trappist the monk (talk) 13:53, 2 September 2021 (UTC)

And I think that the x legend is fixed; can't test it because I can't edit sq:Stampa:Grafiku i kategorive të mirëmbajtjes CS1. To test it, replace the {{#invoke:chart|...}} with {{#invoke:Livadhi/trappist the monk/graph experiment|graph_params_create}}.

I also changed the legends and tool tips for those categories that have subcats so that they read, for example: 'Mirëmbajtja CS1: Emra shifrorë 29 faqe from 5 subcategories'. No doubt, you'll want to translate that if it should be kept.

—Trappist the monk (talk) 14:13, 2 September 2021 (UTC)

Just did. Works perfectly. Even though I like the added functionality of subcategories I think it would be better if the results could be tracked individually. Will I need 1 specific module per graph? There are 3 graphs in total as I mentioned above. Or can that be merged in 1 module and I can invoke specific parts of it on specific templates? - Klein Muçi (talk) 17:16, 2 September 2021 (UTC)

Now uses the color scheme specified in sq:Stampa:Grafiku i kategorive të gabimeve CS1 (64 colors). Now does not chart empty categories. Supports both of the maint and error charts with slightly different invokes:

{{#invoke:chart|...}} with {{#invoke:Livadhi/trappist the monk/graph experiment|graph_params_create|maint}}

{{#invoke:chart|...}} with {{#invoke:Livadhi/trappist the monk/graph experiment|graph_params_create|error}}

demonstration at sq:Përdoruesi:Trappist the monk/Livadhi personal.

Is 'from' the same in Albanian as it is in English or does that also need translation?

—Trappist the monk (talk) 18:20, 2 September 2021 (UTC)

Took care of the translations. Perfect! I see you have removed the missing language error as well (that makes the chart impossible to be read currently because of its size). I think you should add that to the module though, as a possibility for better future times. Also, we're keeping the subcategories version eh? I'm not against per se. What happens with the colors? Does each category, even empty ones, get aligned to 1 specific color? Or do only not-empty ones get colors aligned to them? The reason I ask is because I'm worried for future cases when we might add new error categories. If there are only 64 colors and we add even just 1 more error category, it will start malfunctioning. In that case it would be good if we could have 70 colors "waiting there", just in case. But if only non-empty categories get aligned with a color, then we've solved that problem once and for all because there's no way all categories will be full at the same time. It has never happened until now. - Klein Muçi (talk) 18:38, 2 September 2021 (UTC)

It isn't that I removed the missing language cat, its not in sq:Stampa:Grafiku i kategorive të gabimeve CS1. Added, tested, and commented out. I fixed the bug in the x legend the error chart x legend was linking to the maint category. Added the total number of categories so: 'Mirëmbajtja CS1 (15 of 27 kategori bosh janë fshehur)' and 'Gabime CS1 (33 of 64 kategori bosh janë fshehur)'.

Colors are assigned from the top of bar_chart_colors_t to each non-empty category. If you compare the two charts at sq:Përdoruesi:Trappist the monk/Livadhi personal, you will see that the colors are the same left-to-right. I have added a limit so that the chart will not display more than 64 categories.

The code to support subcat tallying is still present but I have added the direct names for the subcats to the list so the subcat tallying is skipped.

—Trappist the monk (talk) 19:29, 2 September 2021 (UTC)

Everything appears good. Very naïve questions: Can the module be made to work with sq:Moduli:Citation/CS1/doc/Category list somehow so every module emitted category is there automatically, albeit in a red link until created? Of course, if this was possible, extra care would be needed for those specific categories that are always redlinked there.

Follow up question: Can/Should the said module be part of sq:Moduli:Cs1 documentation support when the work with it is finished? Assuming we'll be able to offer support even for the third mentioned chart. - Klein Muçi (talk) 21:54, 2 September 2021 (UTC)

I have added the code from sq:Moduli:Chart/cat properties to sq:Moduli:Livadhi/trappist the monk/graph experiment so that the new module can render sq:Stampa:Grafiku i gjuhëve të burimeve. Also added the ability to render a chart of script-language use; see sq:Përdoruesi:Trappist the monk/Livadhi personal.

I think that I am done with this experiment. If you want to keep it, move it to an appropriate place. I think that there is no need to keep Moduli:Chart/cat properties but if you do, it should be moved someplace else because it really isn't a submodule of sq:Moduli:Chart.

No, this experiment should not be part of sq:Moduli:cs1 documentation support because you are the only one who is using the charts (so far as I know).

—Trappist the monk (talk) 14:15, 4 September 2021 (UTC)

Moved it. Is there any short way to invoke and create all possible charts at the same time with a single invocation? - Klein Muçi (talk) 14:58, 4 September 2021 (UTC)

PS: Should the opening part of the said module be changed now? - Klein Muçi (talk) 15:22, 4 September 2021 (UTC)

Also, can you add a kind of not-too-obtrusive message explaining that the missing language category is missing from the chart so as to be able to actually use the chart? :P Maybe on the same place when we already notify of the hidden categories. And, can we make it so the categories in pie charts also work as wikilinks like they do in bar charts? - Klein Muçi (talk) 15:39, 4 September 2021 (UTC)

ALSO (apparently there was more to tell than I was expecting) I deleted and then had to restore Moduli:Chart/cat properties because it is used here and on its subcategories and associative templates. Take a look if you can. The idea was to start covering all kind of maint categories with specific charts, something that I proposed even on EnWiki, if you've seen the TechPump lately, but given that I'm the only editor on SqWiki looking after the tech side in general, I was only able to deal with categories in regard to citations and external links before I had to halt that project to deal with other stuff. Now I'm not really sure what to do with that module or its new name, whatever that might be. I could use any suggestions on this subject if you do have. I'm saying this more on a general manner, not in regard to the technical aspect per se. - Klein Muçi (talk) 15:53, 4 September 2021 (UTC)

I've done some cleanup and moved the 'data' into a separate data module sq:Moduli:CS1 charts/data.

Since the invokes are so short, it seems best to me to not have a 'dump-all-charts' mechanism. This is especially true if sq:Moduli:CS1 charts is going to get renamed to make it available for non-cs1|2 categories. And, this is an 'expensive' module. My sandbox, which displays the four cs1|2 charts, has an Expensive parser function count: 340/500.

I don't think that the module is the place for messages about that is and what is not included in the chart.

Bars in bar charts wikilinked.

—Trappist the monk (talk) 17:37, 4 September 2021 (UTC)

1. Thank you but, even though that was a needed detail, I was talking about the legend entries in the pie chart.
2. What road would I need to take if I was to be able to make it work for other categories as well? - Klein Muçi (talk) 19:49, 4 September 2021 (UTC)

Decide what chart (bar or pie) you want. Choose a 'keyword' (the existing keywords are 'error', 'lang', 'maint', and 'script'). Add the keyword to keyword_cat_map_t table at the bottom of sq:Moduli:CS1 charts/data. The rvalue in that table is the name of the sequence table that lists the category names to be charted. Add the new sequence table's name to the list of exported tables. While still at the bottom of ~/data, if you have chosen a bar chart, add the keyword and the text of the bar chart's x legend (parent category name without namespace and without wiki markup) to keyword_x_legends_map_t.

Somewhere above the exported-tables section of ~/data, create a new section and your chart's sequence table. The simplest form is a list of category names without namespace.

—Trappist the monk (talk) 23:34, 10 September 2021 (UTC)

Oh, so the information needs to be fed manually. I thought I could just put a category name in the invocation and we could get a graph in regard to its subcategories created, even though I saw that that wasn't what was happening with the CS1 categories. Apparently I need to duplicate the module's core code in a new specific module for each group of charts I want to create together with their categories because I can't have 1 big module for everything and put all categories, related or not, to it. Thanks for the information! :) - Klein Muçi (talk) 22:20, 11 September 2021 (UTC)

Uhm... The charts are behaving a bit strange though...

Check here, here, here and here. Check the message for the hidden categories. They all mention that X out 0(?!) categories are hidden? What is happening? Am I doing the invocations wrong or is there something bad with the module's code? - Klein Muçi (talk) 22:27, 11 September 2021 (UTC)

(edit conflict)

Yes, the category lists are manual because Scribunto does not have access to category content.

What? Why do you think that you need to duplicate sq:Moduli:CS1 charts? To do other chars, all that needs doing is to add the necessary stuff that I described to sq:Moduli:CS1 charts/data. If the module is going to support more than just cs1|2, you might want to move the two modules to a more neutral name, but there should be no reason to duplicate anything.

—Trappist the monk (talk) 22:32, 11 September 2021 (UTC)

The plan is to support with charts more than just cs1|2. My idea for that was to create specific modules with specific names and basically the same code apart from what you mentioned above. For example: Module: CS1 charts supports the charts for 'error', 'lang', 'maint', and 'script'. Module: X charts supports the charts for 'Y', and 'Z'. Module: Alfa charts supports the chart(s) for some other concept, etc. It looks strange to me to have just 1 module and have every category ever of every chart ever I want to create dumped in it. But if you think that's the right way to go... - Klein Muçi (talk) 23:04, 11 September 2021 (UTC)

It is poor programming practice to duplicate code because it increases the maintenance burden; a fix to one must be done to all and that is a pain. Really, there is nothing wrong with putting data for multiple charts into a single ~/data module. If that becomes burdensome, we can devise some sort of mechanism to support multiple data modules.

—Trappist the monk (talk) 23:25, 11 September 2021 (UTC)

Okay then. Any suggestions for a neutral name? My first guess was "Module:Chart" but then I remembered that...

The reason I ask is because users at SqWiki usually tend to import modules/templates from EnWiki preserving the original name so I don't want to choose something which is highly probable to be a module on its own here soon enough (given that currently this module only exists at SqWiki, a situation rather rare). If that happened, the new SqWiki user would just think that it is the same module with an outdated code and just overwrite everything with the "new" EnWiki code. (But on the same time, I'd like to have a common name for it, if possible...) Klein Muçi (talk) 00:00, 12 September 2021 (UTC)

Pick a name that is descriptive of what you use the module to do. The name does not have to be English and I doubt very much that anyone at en.wiki will create a module with an Albanian name so an accidental import that overwrites your existing module seems rather unlikely. Both sq:Moduli:CS1 charts and sq:Moduli:CS1 charts/data will need to be renamed and line 14 will need to be changed the new name.

—Trappist the monk (talk) 00:35, 12 September 2021 (UTC)

Yeah. I was thinking of using an English name given that 90%+ of our module names are like that but an Albanian one seems like the most perfect solution if I don't go with something descriptive in English with more than 1 word. Thanks a lot for the help! :)) - Klein Muçi (talk) 00:40, 12 September 2021 (UTC)

I was planning to fully protect Module:CS1 charts (like we have done with everything related to CS1) and thought to give it a last look before doing that, see if you had anything to do on it before proceeding. I saw 1 TODO comment. Given that apart from you it is highly unlikely that anyone else deals with it, what do you think I should do with that? Ignore it and leave it like that? Remove it? Maybe you know what TO DO now and want to do it before I protect it? Take a look at line 190. - Klein Muçi (talk) 09:11, 12 September 2021 (UTC)

TODO removed. If ever t[1] has anything that isn't a number, adding that value to a number will generate an error and we can see why that is then.

—Trappist the monk (talk) 11:34, 12 September 2021 (UTC)

Thank you! Saw and protected. :) - Klein Muçi (talk) 12:48, 12 September 2021 (UTC)

Happy Adminship Anniversary!

Wishing Trappist the monk a very the monk happy adminship anniversary on behalf of the Birthday Committee! Best wishes! CAPTAIN RAJU^(T) 19:10, 17 September 2021 (UTC)

Precious anniversary

Precious

Nine years!

--Gerda Arendt (talk) 06:53, 20 September 2021 (UTC)

Cite iucn template

Has the template been fixed for errata versions? Quetzal1964 (talk) 07:57, 24 September 2021 (UTC)

yes. here is the plain text errata citation from IUCN Panthera leo:

Bauer, H., Packer, C., Funston, P.F., Henschel, P. & Nowell, K. 2016. Panthera leo (errata version published in 2017). The IUCN Red List of Threatened Species 2016: e.T15951A115130419. https://dx.doi.org/10.2305/IUCN.UK.2016-3.RLTS.T15951A107265605.en. Downloaded on 24 September 2021.

give that to {{make cite iucn}}:

{{make cite iucn|Bauer, H., Packer, C., Funston, P.F., Henschel, P. & Nowell, K. 2016. Panthera leo (errata version published in 2017). The IUCN Red List of Threatened Species 2016: e.T15951A115130419. https://dx.doi.org/10.2305/IUCN.UK.2016-3.RLTS.T15951A107265605.en. Downloaded on 24 September 2021.|x}}

and get:

{{cite iucn |author=Bauer, H. |author2=Packer, C. |author3=Funston, P.F. |author4=Henschel, P. |author5=Nowell, K. |year=2016 |title=''Panthera leo'' (errata version published in 2017) |errata=2017 |volume=2016 |page=e.T15951A115130419 |doi=10.2305/IUCN.UK.2016-3.RLTS.T15951A107265605.en |access-date=24 September 2021}}

Bauer, H.; Packer, C.; Funston, P.F.; Henschel, P.; Nowell, K. (2017) [errata version of 2016 assessment]. "Panthera leo (errata version published in 2017)". IUCN Red List of Threatened Species. 2016: e.T15951A115130419. doi:10.2305/IUCN.UK.2016-3.RLTS.T15951A107265605.en. Retrieved 24 September 2021.

without |errata=:

{{cite iucn |author=Bauer, H. |author2=Packer, C. |author3=Funston, P.F. |author4=Henschel, P. |author5=Nowell, K. |year=2016 |title=''Panthera leo'' (errata version published in 2017) |volume=2016 |page=e.T15951A115130419 |doi=10.2305/IUCN.UK.2016-3.RLTS.T15951A107265605.en |access-date=24 September 2021}}

Bauer, H.; Packer, C.; Funston, P.F.; Henschel, P.; Nowell, K. (2016). "Panthera leo (errata version published in 2017)". IUCN Red List of Threatened Species. 2016: e.T15951A115130419. doi:10.2305/IUCN.UK.2016-3.RLTS.T15951A107265605.en. Retrieved 24 September 2021.

I should probably tweak {{make cite iucn}} so that it removes the (errata version published in year) text – thought I did that...

—Trappist the monk (talk) 11:27, 24 September 2021 (UTC)

Thanks, I didn't realise that there was a template that converted the plain text citation, I have been doing it manually.Quetzal1964 (talk) 13:45, 24 September 2021 (UTC)

Related question about the iucn template; I'm not sure if this is an issue with the template or the doi the iucn page is providing.

{{cite iucn |last1=Muller |first1=Z. (eliding rest of names) |title=''Giraffa camelopardalis'' |amends=2016 |volume=2018 |page=e.T9194A136266699 |date=2018 |doi=10.2305/IUCN.UK.2016-3.RLTS.T9194A136266699.en}}

Muller, Z. (2018) [amended version of 2016 assessment]. "Giraffa camelopardalis". IUCN Red List of Threatened Species. 2018: e.T9194A136266699. doi:10.2305/IUCN.UK.2016-3.RLTS.T9194A136266699.en.

For me, that's giving a green "|date= / |doi= mismatch" warning, which I'm presuming is because the doi says 2016 but the date says 2018. Is this because the template doesn't know to check the amends date instead, or because the IUCN has mucked up the doi somehow? --PresN 18:49, 26 September 2021 (UTC)

Thanks. Fixed.

—Trappist the monk (talk) 20:39, 26 September 2021 (UTC)

Help with a small error

Hey Trappist! Can you help me solve a problem I've caused in my version of the CS1 module? Check the third reference here. The word "original" (origjinali in Albanian) is doubled. Also, the link is on that word instead of the "archived" word (arkivuar in Albanian), unlike here. Where is that part fixed? - Klein Muçi (talk) 22:17, 30 September 2021 (UTC)

Probably this:

['archived-dead'] = 'Arkivuar nga origjinali $1 më $2',

should be this:

['archived-dead'] = 'Arkivuar nga $1 më $2',

The link is supposed to be on 'origjinali' just as it is on 'the original' here:

{{cite book |title=Title |url=//example.com |archive-url=//archive.org |archive-date=2021-09-30}}

Title. Archived from the original on 2021-09-30.

—Trappist the monk (talk) 23:20, 30 September 2021 (UTC)

Thank you!

Since you got me worried with this, can you also check line 30 on our module and see if I'm missing a space there? I mean, compared to the module here, I am but I'm not sure how that renders.

PS: The first link you've sent strangely doesn't work. - Klein Muçi (talk) 00:16, 1 October 2021 (UTC)

I see something like this:

Arkivuar nga origjinali |archive-url= ka nevojë për |url= (Ndihmë!) më 2 shtator 2018.

which I never really liked so whenever we get round to releasing the next version of the module suite that particular message won't be displayed:

{{cite book/new |title=Title |archive-url=//archive.org |archive-date=2021-09-30}}

Title. {{cite book}}: |archive-url= requires |url= (help)

—Trappist the monk (talk) 00:29, 1 October 2021 (UTC)

Oh, okay then. Lastly, can you take a look here at that citation? (There's only 1.) The message has a part of it in English. (missing prefix) Where can I translate it? Are there other possibilities the error detail can take in err_script_parameter? It's been literally months I look for it on and off. :P - Klein Muçi (talk) 00:56, 1 October 2021 (UTC)

In sq:Moduli:Citation/CS1 with three others where they shouldn't be. I'll update our sandboxen tomorrow.

—Trappist the monk (talk) 01:07, 1 October 2021 (UTC)

Ooh! Somehow it had never crossed my mind to look in the main page. When I search the page there for add error message I get 21 hits though. Are there only 4 translations to make? (I know most of them are variables and not translatable text strings.) I'll be on the lookout for your update there and then complete the translations. Thank you! :)) - Klein Muçi (talk) 01:21, 1 October 2021 (UTC)

Module:Cs1 documentation support

I did some localization on the above module, 133 bytes in total, not really sure if I've broken anything. Can you take a quick look, especially in regard to the sandbox → livadhi change? I wouldn't want to break the functionality of the category listing.

I noticed an odd thing on it though: Module:Cs1_documentation_support#L-1124

Is that intended or forgotten? - Klein Muçi (talk) 10:09, 1 October 2021 (UTC)

in line 922, error not gabim.

—Trappist the monk (talk) 11:24, 1 October 2021 (UTC)

Thank you! If you ever find time of moving the tables there in a separate ~/data submodule like you've commented, could you have the courtesy of mentioning that to me whenever we happen to be discussing the next time? I regularly check the cs1|2 module suite for updates but rarely do so for the said module. It would be helpful. - Klein Muçi (talk) 12:04, 1 October 2021 (UTC)

Module:CS1 charts

In regard to w:sq:Module:CS1 charts, is it possible to make the pie charts have their main category wikilinked below them like the bar charts do? You can see them all in action here for a comparison. (The module is currently fully protected. Tell me if I should remove it temporarily.) - Klein Muçi (talk) 00:21, 2 October 2021 (UTC)

Pie chart doesn't have chart label support. Bar chart doesn't either but we abuse |x legends= to make it look like it does.

We might incorporate something like the hack I made for the first pie chart at sq:Përdoruesi:Trappist the monk/Livadhi personal into sq:Module:CS1 charts.

—Trappist the monk (talk) 12:21, 2 October 2021 (UTC)

Ah, interesting... I hadn't put attention to the twisting that we were making to that parameter. Do we put the label below for standardization with the b-chart? While also creating some distance with the legends' links below, of course. If that's not possible, it's good how it is, just wikilink it when made functional. I'm removing the protection from the module. - Klein Muçi (talk) 12:34, 2 October 2021 (UTC)

We cannot put the label below the pie. |label= in the invoke adds the label. sq:Përdoruesi:Trappist the monk/Livadhi personal

—Trappist the monk (talk) 13:17, 2 October 2021 (UTC)

Hmm, can't it be made an incorporated (hardcoded) wikilinked part of those two p-charts we currently create with w:sq:Module:CS1 charts? (The languages and scripts ones.) And maybe |label=off/no/etc. can switch it off if needed (although I can't really think of a need for a function like that). - Klein Muçi (talk) 13:26, 2 October 2021 (UTC)

I don't know what you're talking about.

—Trappist the monk (talk) 13:34, 2 October 2021 (UTC)

You type: {{#invoke:sq:CS1 charts|bar_chart_create|maint}}

You get: a bar chart that below has a wikilink to w:sq:Kategoria:Mirëmbajtja CS1

You type: {{#invoke:CS1 charts|pie_chart_create|lang}}

You get: a pie chart that doesn't have a wikilink to w:sq:Kategoria:Gjuhë CS1 (neither below, nor above)

- After your last edit -

You type: {{#invoke:CS1 charts|pie_chart_create|lang|label=[[:Kategoria:Gjuhë CS1|Gjuhë CS1]]}}

You get: a pie chart that has a label above the chart, that, I'm assuming, can be wikilinked to w:sq:Kategoria:Gjuhë CS1

Can this function be integrated in w:sq:Module:CS1 charts (hardcoded) so when I type {{#invoke:CS1 charts|pie_chart_create|lang}}(without the |label= part), I get a pie chart that has a label above the chart which wikilinks you to w:sq:Kategoria:Gjuhë CS1? There's no need for the label functionality per se.

Can the same be done for {{#invoke:CS1 charts|pie_chart_create|script}}?

w:sq:Module:CS1 charts only serves for generating those 4 CS1 related charts and most likely that will be the case forever. So those 4 charts should be accompanied with a wikilink sending you to their accompanying categories. The 2 bar charts already do that by abusing |x legends= and wikilink to their corresponding categories, respectively to w:sq:Kategoria:Mirëmbajtja CS1 and w:sq:Kategoria:Gabime CS1. The 2 pie charts don't do that. They should do that and wikilink to their accompanying categories, respectively to w:sq:Kategoria:Gjuhë CS1 and w:sq:Kategoria:Vetitë CS1: Vlera në gjuhë me shkronja jo latine.

----

TL;DR Each of the 4 charts should have a wikilink integrated (preferably below it) to send you to its accompanying category. Bar charts have that functionality, pie charts don't. They should. The label functionality per se is not needed. - Klein Muçi (talk) 14:54, 2 October 2021 (UTC)

sq:Përdoruesi:Trappist the monk/Livadhi personal. no |label=.

—Trappist the monk (talk) 15:18, 2 October 2021 (UTC)

Yes, very good. Thanks a lot! :) - Klein Muçi (talk) 15:33, 2 October 2021 (UTC)

Lua banner

Minor pet peeve: Shouldn't Module:Citation/CS1 and its accompanying modules have the Lua banner in their documentation listing the other modules which they depend on like Module:Cs1 documentation support does? - Klein Muçi (talk) 23:21, 2 October 2021 (UTC)

Which conflicts with my pet peeve of unnecessary redundancy... There is a table on each module page listing all of the modules in the suite; surely that is sufficient.

—Trappist the monk (talk) 00:19, 3 October 2021 (UTC)

Hahaha, you never disappoint. That's the reason why I asked before acting. Even though, on the other hand, No globals is missing and so is "standardization". But I get your point. - Klein Muçi (talk) 00:26, 3 October 2021 (UTC)

Nomination for deletion of Template:Row numbers

Template:Row numbers has been nominated for deletion. You are invited to comment on the discussion at the entry on the Templates for discussion page. -- Tamzin^{[cetacean needed]} (she/they) 16:22, 9 October 2021 (UTC)

Quarters in local date_names

Hello! I've noticed that there is a part in CS1/config. that I've left for a long while now untranslated because I've been unsure on how to act on it. The 4 quarters part. First of all, I must disclose I'm not aware at all where that date format might be used or how. Secondly I don't know how much used is it throughout articles in general. Given that I have nothing better to do these days, can you give me some background information on it? - Klein Muçi (talk) 08:53, 10 October 2021 (UTC)

Used in cs1|2 when that is the date given by the source; just like any other date. Not used much as can be seen from these en.wiki searches:

• Q1 18 articles
• Q2 53 articles
• Q3 22 articles
• Q4 22 articles

Used not at all at sq.wiki:

• Q1 0 articles
• Q2 0 articles
• Q3 0 articles
• Q4 0 articles

—Trappist the monk (talk) 11:33, 10 October 2021 (UTC)

Hmm, so... "Quarter" is "çerek" in Albanian (çereku in the definite form). And the first ordinals are "i parë", "i dytë", "i tretë", "i katërt". Would I be fine going with "çereku i parë/i dytë/i tretë/i katërt"? I'm just asking because I'm not familiar with the term, as I said. Do you have any other information how other languages might have solved this? - Klein Muçi (talk) 11:47, 10 October 2021 (UTC)

I also had an idea of equaling each quarter to its corresponding seasonal term but that doesn't feel right given that the seasons have their own place elsewhere. - Klein Muçi (talk) 11:49, 10 October 2021 (UTC)

Equating seasons to some other form of date is problematic because seasons are geographic. And, are we talking about astronomical seasons or meteorological seasons?

—Trappist the monk (talk) 12:05, 10 October 2021 (UTC)

Since I don't speak or read Albanian, I cannot tell you if 'çereku i parë' and the others are acceptable. In English, 'First Quarter' is similar to 'First Semester' (for school sessions) or 'First Trimester' (for human gestation). I would imagine that those same quarter / semester / trimester time measures are similar in Albanian society. Right?

—Trappist the monk (talk) 12:05, 10 October 2021 (UTC)

Thanks for the examples! We use "semestri i parë" and "trisemestri i parë" respectively for school and human gestation. This inspired me to go with "semester" even on this case (maybe accompanying it with the word "year") but I was reading about the word etymology now and it seems strange choosing that path. Semester = sex mensis in Latin, literally 6 months, half year (not a quarter). This makes words like the examples I wrote above seem even more weird as "trisemester" would be 3 six months, 18 months. :P But I believe that's a rabbit hole I need to discuss with another Albanian. (It would be much easier if I had other Albanians interested in matters of the sort in SqWiki.)

Thank you for the explanations! :) Klein Muçi (talk) 12:21, 10 October 2021 (UTC)

Empty categories

Hello, Trappist the monk,

Three categories you created, Category:CS1 Swati-language sources (ss), Category:CS1 Tsonga-language sources (ts) and Category:CS1 Venda-language sources (ve), keep showing up on the Empty Category list despite having a tag saying to keep them, even if they are empty. This tag usually prevents categories with the tag from showing up on the list. I could add a second empty category tag, but that would look pretty silly.

Do you know what the problem might be or should these categories be tagged for CSD C1 deletion despite the tag? Thank you. Liz ^{Read! Talk!} 15:52, 10 October 2021 (UTC)

What do you mean keep showing up on the Empty Category list? Do these three categories appear, disappear, and then reappear or are they always there? Do other of the various language categories exhibit the same problem?

All three categories use {{CS1 language sources}} just like all of the other empty categories that I sampled. Because these three and all other categories use the same template to provide the category documentation, I suspect that the problem is not in the categories themselves but somewhere else. Have you tried purging these categories?

—Trappist the monk (talk) 16:27, 10 October 2021 (UTC)