[00:54:17] https://en.wikisource.org/wiki/Page:Chronological_Table_and_Index_of_the_Statutes.djvu/29 [00:54:26] Where's the missing DIV? [00:54:33] I've tried expanding templates [00:54:41] I do not see a missing DIV tag [00:54:53] So do I flag this is a mis-detection? [00:55:17] File a linter bug? [00:55:18] I'm getting very frustrated with having to play hunt the pedant [00:55:25] Reedy [00:55:37] And wait for it to not be fixed? [00:55:39] XD [00:55:41] Well [00:55:48] You're currently complaining to people who aren't listening [00:55:50] So... [00:55:53] Two sides of the same coin [00:56:08] Though, you might be wrong... And there might be a missing div [00:56:30] Can someone else PLEASE examine the code of the page and tell me what's wrong? [00:56:46] because I've tried very way I can think of to find it [00:56:52] (being the missing tag) [00:57:30] https://en.wikisource.org/wiki/Page:Chronological_Table_and_Index_of_the_Statutes.djvu/29 [00:57:39] https://en.wikisource.org/w/index.php?title=Page:Chronological_Table_and_Index_of_the_Statutes.djvu/29&action=edit&lintid=746549 [01:01:10] subbu: ^ [01:09:05] There seem to be out of channel right now :( [01:24:05] ShakespeareFan00: or people that work on the linter extension can see the task instead of just hoping their irc client is connected to the channel [01:24:12] and are paying attention [03:29:52] Reedy, thanks .. will take a look later tonight / tomorrow. [03:36:12] Maybe ShakespeareFan00 is seeing a stale lint issue .. most likely because of delayed refresh in the list of reported issues .. there are no lint errors in the current version of that page. [08:01:40] будьте добры прокоментировать следующий текст: https://en.wikipedia.org/wiki/User:NikitaSadkov [09:17:09] будьте добры прокоментировать следующий текст: https://en.wikipedia.org/wiki/User:NikitaSadkov [09:21:03] Nikita_Sadkov: why? [09:21:40] собираю расстрельный список [09:22:21] Why should I comment on en.wp about ru.wp? That does not make sense. [09:22:53] andre__: меня забанили на русской, посему я вынужден задать вопрос на английской [09:23:42] Nikita_Sadkov, except that you do NOT ask in English. [09:23:52] so that does not make any sense. [09:24:42] again: it is off-topic. en.wp is not a venue for discussing ru.wp issues. [09:25:15] Sorry. Thought that is Russian language channel, as it was pointed at https://ru.wikipedia.org/wiki/%D0%92%D0%B8%D0%BA%D0%B8%D0%BF%D0%B5%D0%B4%D0%B8%D1%8F:IRC#%D0%A1%D0%B5%D1%82%D0%B8_%D0%B8_%D0%BA%D0%B0%D0%BD%D0%B0%D0%BB%D1%8B [09:25:38] andre__, Russian pages link here for some reason. [09:26:29] andre__, and they say that is russian language channel. [11:47:45] https://phabricator.wikimedia.org/T185203 [11:47:49] Feedbakc sought [11:52:17] ShakespeareFan00: If you don't expect hundreds of people to each click your link: context welcome and which kind of 'feedback' you are looking for [11:54:38] andre__: On wikisource I cam across some table lines starting with || which is a subtle [11:54:47] but obvious coding mistake [11:55:03] Table data lines should start with a single | [11:55:41] The PHAb ticket/issue is asking if something Linter could be extended to look for these [11:56:08] Or if there could be an extension (like the external Checkwiki tool) implemented as Special:Page to hunt for these [11:56:14] and other coding errors [11:56:46] and the feedback you are looking for is....? [11:56:58] "Yes, it's a valid idea."? :) [11:57:20] The feebback I am wanting is a consensus that this is an issue (albiet low priority) [11:57:38] and ideally some kind of time-scale on when it might be implemented XD [11:58:22] * ShakespeareFan00 waits for the men in white coasts to arrive due to my insane enthusiasm for it actually happening) XD [12:12:05] ShakespeareFan00: I recommend keeping tasks to single goals, eg: the checkwiki stype stuff should be seperate to adding new linter checks [12:19:20] You want me to split the bug? [12:19:25] Compliant... [12:23:10] And why is it not currently possible to use [[]] syntax in Phab tickets? [12:23:27] ShakespeareFan00, it is. [12:23:36] How? [12:23:40] see https://www.mediawiki.org/wiki/Phabricator/Help#Formatting [12:23:47] [[ https://foo.bar | Text ]] [12:23:54] use Markdown syntax. [12:24:47] https://phabricator.wikimedia.org/T185207 [12:24:59] Okay. I can fix that [12:25:18] I thought it was possible to do [[en:Wikipedia:Page]] directly :( [12:26:11] no, Phab developers were sane enough to not support some non-standardized confusing MediaWiki syntax. [12:26:22] XD :) [12:27:02] I split out the "Integrate Checkwiki as Special:Page" from the more spefcifc table stuff [12:27:21] The thinking is that many of the coding errors sought by Checkwiki are generic [12:27:55] and thus should perhaps be part of some kind of more general reports on a Special:page generated by an appropriate extension [12:28:38] At present it's a high level feature request [12:28:53] I can appreciate other details might get quite complex [12:29:40] A reg-exp / rule or Linter like detection method could all play a part [12:30:15] Mediawiki to some extent already has support for rule based matching as that's how edit filters work [12:30:35] Not sure if it regexp based though [14:41:15] Very strange - {|width=420 {{ts|sm|bc|bt2|bl2|br2}} [14:41:21] in a header [14:41:27] and then content in a body [14:41:30] causes problems [14:41:46] https://en.wikisource.org/w/index.php?title=Page:Chronological_Table_and_Index_of_the_Statutes.djvu/844&action=edit&lintid=746941 [14:42:02] For some reason a spare DIV gets generated [14:45:52] When doing a manual expansion of what should be genereated shows no tag mismatch [14:54:47] Hello, I have a question. [14:55:06] Usually editcount gives me total number of edit. is it possible to see via API what was my 100th edit? when i edit that & what page? [15:41:11] https://phabricator.wikimedia.org/T185221 [15:42:41] Posting links without any context don't make anybody understand your intention, I am afraid. [15:42:59] its definatly related to linter [15:45:38] ShakespeareFan00, hi .. [15:46:20] subbu [15:46:22] Hi [15:46:29] Your name came up in disscusions [15:46:37] so, there are no lint errors in the current version of that page ( Page:Chronological_Table_and_Index_of_the_Statutes.djvu/844 ) [15:46:47] maybe you are seeing a stale entry in the linter issues list. [15:46:54] Okay [15:47:10] sometimes, jobs get stuck or there are database errors that introduce staleness. [15:48:29] Which isn't helpful to a contributor like me that fixes something [15:48:50] Goes to check what else broke, finds the Special:Pgae hasn't updated and then gets [15:48:57] yo-uo edit/reverts going on [15:49:06] In attempt to find a stable repair [15:49:23] subbu: Would you mind if I file a concern.. [15:49:48] "Special:LintErrors" gives no indication as to freshness of the dataset used." [15:49:51] ? [15:49:52] ShakespeareFan00, we know about it .. [15:50:09] Which ticket no? [15:50:23] https://en.wikisource.org/w/index.php?title=Special:LintErrors/missing-end-tag&dir=prev [15:50:27] when pages are edited, it takes a while for them to be reparsed [15:50:43] doesn't indicate when a page was last parsed. [15:50:45] and only when they are reparsed, linter issues are updated. [15:51:19] Certain other special pages have a warning indication about lag... [15:51:25] ok .. we could maybe add some indication about when it was last updated .. but, right now, we aren't maintaining that timestamp in the database, so we cannot do it right away. [15:52:04] Understood, That's why I was suggesting opening a ticket to track a possible patch [15:52:18] given that whats needed is reasonably obvious :) [15:52:54] subbu [15:52:57] On something else... [15:53:52] we have https://phabricator.wikimedia.org/T171791, https://phabricator.wikimedia.org/T179854, https://phabricator.wikimedia.org/T170313 [15:54:52] subbu: Page:Cowie's Printer's pocket-book and manual.djvu/70 renders one way in Page namespace [15:55:01] https://en.wikisource.org/wiki/Page%3ACowie's_Printer's_pocket-book_and_manual.djvu/70 [15:55:16] but renders differently when transcluded here- [15:55:29] https://en.wikisource.org/wiki/Cowie%27s_Printer%27s_Pocket-Book_and_Manual/A_Table,_shewing_the_Number_of_Thousands_in_a_Sheet_of_Eighteens,_Twelves,_Octavo,_and_Quarto,_of_various_Sizes [15:55:47] This isn't a structural error but a logical one. [15:56:07] I would appreciate a second set of eyes examining the relevant markup [16:02:26] ShadowReine, so, on that Page:Cowie's page ... the problem is the {{nop}} in the first line ... it generates a

.. but it is embedded inside a table outside a table-cell .. i.e. there is no preceding | [16:02:51] so, it gets "fostered" out of the table and also causes the missing error .. but, the core problem is that the {{nop}} there is out of place. it doesn't belong there. [16:03:05] Subbu, you’re highlighting the wrong person. [16:03:13] ShadowReine, oops sorry :) [16:03:16] ShakespeareFan00, ^^ [16:03:17] It’s ShakespeareFan00 you need ;) [16:03:39] NBd, just didn’t want you to talk for ten minutes without realizing you’d highlighted the wrong S name *goes back to work* [16:03:59] subbu: Yet that {{nop}} is the advised approach for split tables [16:04:14] According to the disscussion at Wikisource [16:04:30] https://en.wikisource.org/wiki/Wikisource:Scriptorium#LintErrors [16:04:58] If you'd like to tell the long-standing and respected contributor there that THEY are wrong [16:05:05] I wish you the best of luck [16:05:10] ShakespeareFan00, maybe so .. but it has to be proper markup. if it is broken markup, nothing the parser / browser can do with it. i cannot wade into that discussion now, but maybe the instructions are missing steps / are being misinterpreted. [16:05:51] subbu: Without the {{nop}} there's no line-feed so that the starting |- [16:06:05] gets rendered into the last cell of the preceeding table row [16:06:15] due to the way Proofread Page works [16:06:27] This mangles all kinds of tables [16:06:54] The correct way of doing this would be to have "proper" directives to use changing the parser handling [16:07:08] but when I suggested the same a while back there absolutely no interest [16:07:11] (sigh) [16:07:33] okay .. i don't understand the details now ... but i'll copy this irc discussion onto the bug report and it can be followed up later. have to get to other things now. [16:07:46] I know the {{nop}} there is wrong [16:08:14] It shouldn't however be causing the Missing DIV end though as the code inside the {{nop}} is balanced [16:08:55] If on the other hand the markup prior to adding the {{nop}} was... and LintErrors isn't updating due to lags [16:08:57] .. [16:09:00] Hmm [16:09:30] I find it odd that sometimes Editing a page causes a Lint Error to pop up almost immediately [16:09:41] but that adding a fix takes ages for a page to reparse [16:09:52] This is demotivating to contributors :( [16:11:04] Maybe a race condition of some sort [16:11:16] bawolff: Entirly possible [16:11:41] Ideally, to mark special situations like this I'd like to see something like a [16:11:56] |~ |~ tags to mark continuation points [16:12:26] |~|~ is a sequence unlikely to occur naturally I feel and so the parser could look for it [16:12:39] amend it's handling in respect of transcluded content [16:12:50] Well the parser or proofread page... [16:13:23] Dual tags which get parsed out is how hypenated words between pages are handled on Wikisource IIRC [16:13:45] subbu: BTW I treat conversations here as quotable on Phabricator or wiki [16:13:57] You don't need to ask me specfically about log quoting [16:14:21] ShakespeareFan00, ok. [16:14:44] Thats Issue 1 [16:14:47] Issue 2 [16:15:19] It would be nice in the UI for Editing (extensions) aside, to have a button that does a check for coding errors [16:15:43] (Which could include running a single page through something like Linter) [16:15:51] Would this be possible or feasible? [16:21:28] bard, what you are asking for is a wikitext markup validater to be built into mediawiki as an extension, am I right? [16:21:35] Yes [16:21:53] You have validaters for HTML/CSS etc.. [16:22:21] this extension would be equipped with a Special:markupcheck page ? [16:22:25] Essentialy as I've said in an early ticket it would be nice if Checkwiki were made a Special:Page generated by an extension [16:22:33] Dysklyver: Yes [16:22:45] sounds cool [16:23:03] As well as a group or reports that would be automatically updated on a specfic basis [16:23:19] Let me find the earlier ticket so yu can add your last comment [16:23:25] You simply cannot validate something that has no definition, I am afraid. [16:24:17] https://phabricator.wikimedia.org/T185207 [16:24:29] ^^ Dysklyver [16:24:57] yes [16:25:00] andre__: HTML wasn't as such formally defined initally [16:25:07] But OKAY I see what you mean [16:26:22] my thoughts are to request this as a feature on mediawiki at somewhere like https://www.mediawiki.org/wiki/Project:WikiProject_Extensions [16:30:34] Also I see no reason why this is not possible, all it has to do is check for linter errors and tell you where the damn things are [16:30:47] that can't be too hard [17:02:36] Dysklyver, ShakespeareFan00, andre__: one of the design axioms of wikitext is that any input is valid wikitext. [17:02:42] So, there's nothing to validate [17:02:52] (Sigh) :( [17:03:35] if any input is valid, then why are there missing

tags or whatever it is that bard is fixing ? [17:11:27] Dysklyver: because there is no actual parser/generator pair. just regular expressions that amngle the input until it looks something like html. [17:11:54] and yes, this is very hard to fix, because we need to be bug-by-bug compatible. otherwise we could no longer render old revisions [17:12:11] this is a long standing issue. it has been debated for mroe than a decade. and peopel have tried to fix it for more than a decade [17:12:16] that sounds very awkard all round [17:12:17] parsoid is a partial fix. [17:12:26] yes, it is indeed. [17:12:38] look for the "wikitext 2.0" project [17:12:45] well thanks for explaining that [17:13:35] welcome to the wonderful world of staying compatible with all the mistakes we have made since 2001, forever. [17:19:03] At least you don't have to figure out the mistakes a hardware engineer made 40 years afgo [17:27:59] ShakespeareFan00: ''How to '''interpret'' this line's markup, for''' example... [17:28:21] andre__: That's mismatched [17:28:44] Which is one of the things that should be flagged as coding erorr [17:28:48] ShakespeareFan00: no, not necessarily. [17:28:58] that's just your interpretation that it's mismatched. [17:29:44] I read this as How to interpret this line [17:29:54] Which is mal formed [17:30:24] yeah, one interpretation [17:31:07] If you actually meant How to interpret this line's markup [17:31:17] Then it's harder to spot that intent [17:31:50] Do you need a precedence order for tags? [17:32:16] I.E is more important than so intterupts it. [17:32:55] Well, it''''s complicated'''. Now is that apostrophe bold or not? [17:32:57] Any parser/rules inserting an additional sequence to reblance.. [17:33:00] maybe that is a better example ^ [17:33:12] there are enough examples like this. [17:33:16] andre__: Quite [17:33:26] you *cannot* define one behavior. that's how it is. [17:33:40] Wikisource says in that instance that the apostrophe has to be {{'}} [17:33:56] which is less ambiguous [17:34:24] Arguably I'd like to see escaped versions of some standard puntuation [17:34:39] *punctuation so it can't be confused with markup [17:34:59] {{!}} is already used to avoid | confusion [17:35:21] {{'}} is used for apostrophe [17:35:28] and so on [17:49:48] addshore: Hey! Around? [18:28:14] d3r1ck: partly [18:30:03] addshore: Don't know if you can quickly have a look at this please: https://gerrit.wikimedia.org/r/#/c/404880/ :) [18:30:26] addshore: And maybe this too: https://gerrit.wikimedia.org/r/#/c/404748/. [20:34:29] Hello! Please, I wish to know if there is a function like wfArrayToObject() in Mediawiki core, that converts Arrays to objects. [20:41:21] Why? [20:41:33] $object = (object)$myArray;