Benutzer Diskussion:PerfektesChaos/js/lintHint/2020

aus Wikipedia, der freien Enzyklopädie

Pointers are pointing to the wrong place

For several weeks now, lintHint pointers have often been off. For example, editing en:Wikipedia:Biographies of living persons/Noticeboard/Archive109, lintHint detects 47 errors.

  • The first error is an obsolete <font> tag, and clicking on the down-arrow highlights text 10 characters after the actual <font> tag.
  • The last error is an obsolete <font> tag, and clicking on the down-arrow highlights text about 152 characters after the actual <font> tag.

What is going on? –Anomalocaris (Diskussion) 12:37, 3. Jan. 2020 (CET)

HNY.
Thanks – yes, it has been noticed already that the numbers are not precise any longer.
With beginning of December 2019 the executing parser system has been shifted from Parsoid/JS to Parsoid/PHP.
I guess this is the reason.
Currently the API answers are temporarily or for some wikis completely discarded due to server overload – see section above.
It is hoped that within January 2020 there will be a robust solution regarding servers.
For now partially Parsoid/JS and partially Parsoid/PHP are both producing answers and pointers.
Before the Parsoid migration has not been successfully completed and server load is stable I cannot start investigation, but I will do under stable conditions. LintHint does tell you just the answer received from server.
Greetings --PerfektesChaos 12:53, 3. Jan. 2020 (CET)
This is a Parsoid/PHP bug. We'll investigate this and fix it. SSastry (WMF) (Diskussion) 05:57, 4. Jan. 2020 (CET)
Archivierung dieses Abschnittes wurde gewünscht von: PerfektesChaos 13:48, 10. Okt. 2021 (CEST)

Why doesn't it detect this missing table close?

Hi. First, thanks for this great tool!

I noticed today that it finds no error in en:Special:PermaLink/980597500, yet the article is missing a table-closing at the end of the table in the Confederations section. Thanks! —[AlanM1(talk)]— 22:51, 27. Sep. 2020 (CEST)

I do think the missing of a table close |} is not in every case an lintError for the software. So the tool can’t find it. This is something I often aked me why the software can find missing tag like div, small, span, italic or bold, but not a missing end of a table. In your example the following content is interpreted as content of the table itself. The missing of the table ending should be found by the software, than lintHint will find it too. --Liebe Grüße, Lómelinde Diskussion 06:41, 28. Sep. 2020 (CEST)
I'm afraid what you're saying appears to be contradictory. Yes, lintHint routinely finds tables that are opened but not closed, as it does with div, italic, etc. markup. This is fairly straightforward to do, which is why I wonder why it fails in this case (and no others that I've seen). —[AlanM1(talk)]— 13:08, 28. Sep. 2020 (CEST)
Yes in many cases it will find the error, but this case is spacial, I know many pages with frames build up with tablesymtax on talkpages left open I’ll seach for an example.
see wikitext in Benutzer:Warhog/Vorlage:Diskussionsseite the table is open at the end an the pageinformation does not mark any LintError →info. I have often seen such pages. I hope you’ll understand I am not an English speaker. --Liebe Grüße, Lómelinde Diskussion 13:39, 28. Sep. 2020 (CEST)

Please note: This tool lintHint is just broadcasting reports of the Parsoid syntax processing.

  • This tool does not analyse any code.
  • It does nothing more than making analysis results visible.
  • It has no knowledge about meaning and circumstances of the messages forwarded to the reader.

Please address any content issues to mw:Help:Extension:Linter etc. THX --PerfektesChaos 15:15, 28. Sep. 2020 (CEST)

Archivierung dieses Abschnittes wurde gewünscht von: PerfektesChaos 13:49, 10. Okt. 2021 (CEST)
Abgerufen von „https://de.wikiup.org/index.php?title=Benutzer_Diskussion:PerfektesChaos/js/lintHint/2020&oldid=216401211“