Benutzer Diskussion:Stefan Kühn/Check Wikipedia/Archiv/2008/Dez

Diese Seite ist ein Archiv abgeschlossener Diskussionen. Ihr Inhalt sollte daher nicht mehr verändert werden. Benutze bitte die aktuelle Diskussionsseite, auch um eine archivierte Diskussion weiterzuführen.

Um einen Abschnitt dieser Seite zu verlinken, klicke im Inhaltsverzeichnis auf den Abschnitt und kopiere dann Seitenname und Abschnittsüberschrift aus der Adresszeile deines Browsers, beispielsweise

[[Benutzer Diskussion:Stefan Kühn/Check Wikipedia/Archiv/2008/Dez#Thema 1]]

oder als Weblink zur Verlinkung außerhalb der Wikipedia

https://de.wikipedia.org/wiki/Benutzer_Diskussion:Stefan_K%C3%BChn/Check_Wikipedia/Archiv/2008/Dez#Thema_1

Difficult square bracket imbalance to find

We have no idea what's wrong here: nl:ASCII (tekenset). 82.171.148.8 20:51, 2. Dez. 2008 (CET)

We had the same problem in de. I think it is a problem with "[". But I am not sure. I will test it at the weekend. -- sk 21:36, 2. Dez. 2008 (CET)

False positive Image without caption

[[Image:Elbasan.jpg|Elbasan|right|thumb]] was detected as an image without a caption although it has it and Mediawiki displays it correctly. Best regards. Miraceti 21:06, 2. Dez. 2008 (CET)

I see this only tree times in all wikipedias. It is not often written like this. Please write the text at the end. Like this: [[Image:Elbasan.jpg|right|thumb|Elbasan]] At the moment my script don't find description, if the description not stand at the end. -- sk 21:30, 2. Dez. 2008 (CET)

This "error" occured on de.wp several times, approximateley 10 to 15 times, so I moved the caption to the end to fix it, since it makes the source easier to read. --STBR – !? 21:40, 2. Dez. 2008 (CET)

I agree. Leave the detection as it is, the syntax of images will be more regular. --66.131.214.76 00:29, 3. Dez. 2008 (CET) -- dixit LaddΩ parlons! ;)

Streamline detection of Titles with special letters and no DEFAULTSORT

Hi Stefan, let me first congratulate you again for the great outcome of Check Wikipedia. In French Wikipédia, 5 to 10 contributors participate in the daily removal of errors, and many sections have been entirely cleared, leaving only errors emerging from new articles -- or new types of detections. Lots of discussions are also taking place, which I try to report to you from time to time. Keep going!

This is about the new detection 037 for "Title with special letters and no DEFAULTSORT"; it is very useful but it could be refined if you can find some time: as you know, using DEFAULTSORT to override the default sorting of articles with titles having special letters is necessary in order to override a known bug in mediawiki that will eventually be corrected -- though it was reported multiple times in the past few years and has not yet been addressed -- :( -- In the mean time, we have to resort to DEFAULTSORT, though it was created for a totally different objective.

One problem with DEFAULTSORT is that it retains its original argument when the article changes title. To avoid accumulating discrepancies between titles and DEFAULTSORT in the long run, it is best to restrict use of DEFAULTSORT only to articles where the "special letters" appear close to the beginning of the title (in addition to using it for special sorting of course).

So an improvement we would suggest is that error 037 could only signal articles that have a title where the first, second or third character is one of those "special characters". That way we would not introduce DEFAULTSORT in articles where the special letter plays no role in the resulting sort.

No ruch of course, I see you have plenty of extra issues to address  ;) Bye --66.131.214.76 01:07, 3. Dez. 2008 (CET) -- dixit LaddΩ parlons! ;)

In the Czech Wikipedia, we have decided to live with this Mediawiki's error since we hope it will be fixed one day. The Czech language uses a lot of special characters + there are some sorting rules more. It would be really difficult to handle those manually. It is better to put some effort somewhere else. That's why testing special characters in DEFAULTSORT has no sense for us. I understand that the script is general. Miraceti 07:38, 3. Dez. 2008 (CET)

If this is a concensus for Czech Wikipedia, you can deactivate that detection by assigning value 0 to the czesh version of its priority variable: error_037_prio_cswiki=0. --199.22.61.2 20:07, 3. Dez. 2008 (CET) -- dixit LaddΩ parlons! ;)

It is now in our translated text. Thanks for a hint. Miraceti 11:58, 4. Dez. 2008 (CET)

Titles with no DEFAULTSORT but special character

This needs filtering with the same legal characters as in the DEFAULTSORT itself. Haros 07:15, 2. Dez. 2008 (CET)

Correct. I will do this. Maybee at the next weekend. -- sk 21:31, 2. Dez. 2008 (CET)

 Ok. -- sk 22:20, 4. Dez. 2008 (CET)

Reste aus Vorlagenprogrammierung

In der Erklärung steht, das auf #if exist geprüft wird, diese Parserfunktion gibt es nicht, richtig lautet sie ohne leerzeichen: #ifexist:. Zusätzlich sollte auch auf #ifeq: geprüft werden (Beispiel). In dem genannten Beispiel fällt auch auf, das der Parameter {{{40}}}nicht erkannt wurde. Zusätzlich könnte auch #tag: mit aufgenommen werden. Der Umherirrende 22:14, 2. Dez. 2008 (CET)

 Ok, eingefügt. -- sk 21:23, 4. Dez. 2008 (CET)

<ul>, <ol> and <li> in image captions

Todesstrafe in einzelnen Bundesstaaten:

Zur Zeit keine Anwendung

Für verfassungswidrig erklärt

Seit 1976 nicht mehr angewandt

Seit 1976 angewandt

I guess there is no other way how to create lists in image captions than to use html tags. So lists in image captions detected by the script are rather false positives. Miraceti 12:00, 4. Dez. 2008 (CET)

I have found a way: using special templates for these tags (see cs:template:ol, cs:template:ul, cs:template:li). And because using templates is usually easier for users, it could be good to use them. So, false positive is not a problem anymore, I guess. Miraceti 13:20, 4. Dez. 2008 (CET)

It works also with templates, as you can see in the picture to the right. --Manoridius 20:54, 4. Dez. 2008 (CET)

Check this

Hello! Can you check into Swedish Wikipedia version. It´s a problem with the Mellanprioritet (Middle priority). I do not know what it is for any error, but it would be good if you can fix it! I look at some other language versions and there is no problem with this. Have you more questions about this can you answer me at swedish wikipedia. Elfsborgarn 16:58, 4. Dez. 2008 (CET)

You need a nowiki in the translation. I have fix this. -- sk 21:20, 4. Dez. 2008 (CET)

Translation update

Hi! I updated our translation page and I expected it would be used last night. But it was not. Does it mean, the script does not take the translations automatically? Could you, please, use our newest translation? Miraceti 08:03, 5. Dez. 2008 (CET)

Yes, my script use automtically the translation page. I think tomorrow the update is in the output. -- sk 08:48, 5. Dez. 2008 (CET)

Renaming of project page?

In Japanese Wikipedia, we are discussing about renaming ja:Wikipedia:WikiProject Check Wikipedia. If agreed, we could rename it? If yes, what steps are needed? ja:User:fryed-peach --08:24, 5. Dez. 2008 (CET)

Send me only the name of the new page and also the name of the new translation page. That all. I will include this tonight. -- sk 08:49, 5. Dez. 2008 (CET)

References check in af.wikipedia

Hi Stefan.

1st: Thanks for this great tool.

2nd: Please add the following check for <ref> without {{references}}: In afrikaans both {{verwysings}} and {{Verwysings}} are being used. Your script of 20081202 only checked for {{verwysings}}, please add a check for {{Verwysings}} as well.

Thanks

Thanks.  Ok, I include this. --sk 21:12, 4. Dez. 2008 (CET)

Hinweis: Da es sich hier nur um einen Klein- und Großschreibung handelt, sollte dies für alle Sprachen möglich sein. Weitere Fälle sind auch das "Template" oder der Namensraumname der Sprache dabei steht in unterschiedlicher Groß- und Kleinschreibung. Dies kann man unter dem Stichpunkt "Normalisierung" zusammenfassen. Dazu gehört auch eventuelle Unterstriche als Leerzeichen zu interpretieren und eben die Groß- und Kleinschreibung des ersten Buchstabens, sowie die Namensraumangaben. Alles das sind mögliche Formen in MediaWiki Seiten einzubinden. Der Umherirrende 21:29, 4. Dez. 2008 (CET)

Ich versuche schon immer Klein- und Großschreibung mit abzuhandeln. Im vorliegenden Fall ist aber im Skript einfach bei einem Umbau aus Versehen das verwysings komplett rausgefallen. - Das mit den Unterstrichen in Links hab ich auch noch auf dem Plan. -- sk 22:23, 4. Dez. 2008 (CET)

Den Hinweis mit dem Unterstrich wollte ich nur geben, da es ja auch Vorlagen mit Leerzeichen für die references geben kann, damit diese auch erkannt werden. Alle Links ohne Unterstriche im Artikel ist auch wieder etwas sinnlos, ich glaube aufgrund der Maße kann man dies wohl so lassen. Bei den PD sah ich das anderes, da waren es ja nicht viele und dort ist eine Weiternutzung ohne MediaWiki wahrscheinlicher. Wenn du schon dabei ist, kannst du ja auch darüber nachdenken, urlencodierte links zu finden, diese müssten dann decodiert werden. Beides ist eher Botarbeit und auch nur Quelltextkosmetik. Falls die Anzahl aber gering ist, könnte man es machen (was ich aber bezweifle). Der Umherirrende 16:53, 5. Dez. 2008 (CET)

Missing « : » before Categories in report

We activated detection of error 011 in French (curly bracket not correct) and noticed that Categories that get reported miss the starting colon «:» character, causing lines that seem empty in the table of that section. Thanks --66.131.214.76 03:29, 5. Dez. 2008 (CET) -- dixit LaddΩ parlons! ;)

Yesterday I have deactivated all errors which not work corret. First I will make the script better, then you can use the error. -- sk 08:46, 5. Dez. 2008 (CET)

As you wish. Take note that contributors voted to activate this detection because it was perceived as maximum priority.  ;) --66.131.214.76 00:31, 6. Dez. 2008 (CET) -- dixit LaddΩ parlons! ;)

Übersetzungsvorlage

Hallo Stefan, heute stimmt die Übersetzungsvorlage auf dem Toolserver nicht, die neuen Fehler der HTML-Formatierungen erscheinen nicht. Andim 09:32, 5. Dez. 2008 (CET)

Ich schaue heute abend mal nach. -- sk 16:11, 5. Dez. 2008 (CET)

 Ok, beseitigt. Müsste ab morgen in de korrekt sein. -- sk 08:12, 6. Dez. 2008 (CET)

Headings without ending - many errors on pl.wiki

[1] - it seems that almost every error that your script reported is <ref> inserted between == and ==. Maybe some kind of removing refs and other disappearing things before checking this? Matma Rex 22:24, 5. Dez. 2008 (CET)

It is only the line break in the headline. If you delete the line break then my script see the end of the headline in the same line. It has nothing to do with the ref. -- sk 08:14, 6. Dez. 2008 (CET)

Conflicts between errors 032 & 033

Hi Stefan, errors on <u> are reported under section heading "double pipe in one link" -- that used to be error 032, whereas <u> used to be 033. Bye --66.131.214.76 00:22, 6. Dez. 2008 (CET) -- dixit LaddΩ parlons! ;)

 Ok, this was a bug. At this moment I have fixed this error. -- sk 08:15, 6. Dez. 2008 (CET)

False positive with ref problem

Hi Stefan,

The following page is on the list with missing references. The reference is actually there:

http://nl.wikipedia.org/wiki/Kleistpark_(metrostation)

82.171.148.8 13:47, 6. Dez. 2008 (CET)

When I scan an article I delete all things like math, pre, nowiki, ... and also tables. This reference is in a table! I thinks this is not normal. -- sk 15:07, 6. Dez. 2008 (CET)

The reference is set in a table by purpose because of a layoutproblems in Internet Explorer according to the writer of the page. 82.171.148.8 18:19, 6. Dez. 2008 (CET)

Please don't fix layout problems like this with tables in Wikipedia. -- sk 18:13, 8. Dez. 2008 (CET)

Statistics

Hi!

Could you, please, add a short summary to the script's output? A table with columns "error number" and "number of found problems / not tested" would be fine. Thank you very much! Miraceti 10:48, 8. Dez. 2008 (CET)

Not today, but at the next weekend or next week I have more time. I think it is a good idea. -- sk 18:12, 8. Dez. 2008 (CET)

Defaultsort false positive problem

• http://nl.wikipedia.org/wiki/S%C3%BCper_Lig_2006/07 (/)

82.171.148.8 18:32, 3. Dez. 2008 (CET)

I found 3 more on the Dutch wikipedia:

• http://nl.wikipedia.org/wiki/R.V._%26_A.V._Sparta (&)
• http://nl.wikipedia.org/wiki/Beatrice_d%27Este_(geb._1475) (')
• http://nl.wikipedia.org/wiki/Digg* (asteriks)

Maybe you can include those in the perlscript. Thanks in advance! 82.171.148.8 20:34, 5. Dez. 2008 (CET)

• Any updates on this one? 82.171.148.8 21:22, 9. Dez. 2008 (CET)

No scan on December 8

Hi, it seems that the scan did not complete on December 8 for all international versions following «en». Either abnormally slow or hung somewhere. --66.131.214.76 02:08, 9. Dez. 2008 (CET)

The reason is the big new enwiki-Dump http://dumps.wikimedia.org/enwiki/20081008/. It need time to download this dump, to extract this file and then to scan it. The scan of eswiki start at 2008-12-09 15:33. This is later than normal. -- sk 08:35, 10. Dez. 2008 (CET)

Missing </nowiki> in output

The output for <font></font> misses </nowiki> twice. Haros 19:23, 9. Dez. 2008 (CET)

I will change this. Thanks for the info. -- sk 08:37, 10. Dez. 2008 (CET)

 Ok -- sk 21:33, 11. Dez. 2008 (CET)

Restrict error 037 : articles missing DEFAULTSORT

Hi, in French Wikipedia, error 037 reports over 18000 articles missing a DEFAULTSORT statement. It is too many and not quite the right ones : it is inappropriate to report articles with accented letters in 10th position, since the tenth letter plays no role in the sorting of that article in Categories and putting DEFAULTSORT on those kind of polluting them.

Can you please only report articles with special letter in the first two or three letters ? Thanks --66.131.214.76 01:13, 11. Dez. 2008 (CET) -- dixit LaddΩ parlons! ;)

At the weekend I will change this. In de we have more then 20000 in this category. -- sk 08:55, 11. Dez. 2008 (CET)

 Ok -- sk 21:33, 11. Dez. 2008 (CET)

Much better and real quick ! Thanks -- One more thing relating to this, I noticed in the error description of the Afrikaans version that you report only articles that also comprise at least one category statement that does not have the syntax "[[Category:ABC|Text]]". I believe that this second condition is not appropriate: DEFAULTSORT is required anyhow because many templates put articles in categories, without an explicit [[Category]] statement in the text of the article, and they will use the accented article title unless a DEFAULTSORT is used -- thus DEFAULTSORT is necessary even if all Category statements have an explicit Text. Best regards --66.131.214.76 12:33, 12. Dez. 2008 (CET) -- dixit LaddΩ parlons! ;)

ruwiki

Can you please add {{ref-list}} and {{Ref-list}} as synonyms to "references"? I hope this is the last reflist-like template that is not in your script:) --77.79.175.67 04:46, 13. Dez. 2008 (CET)

• And add {{примечания}} as synonyms to <references />. --Pessimist2006 13:03, 13. Dez. 2008 (CET)

No defaultsort in redirect pages

In fi wiki there's redirect pages with an alternative spelling of the article name, that redirects to the correct article, which has a corrected defaultsort. For example this page: http://fi.wikipedia.org/w/index.php?title=De%60Angelo_Wilson&redirect=no . I think it's unnecessary to add defaultsorts to redirect pages, since redirect pages aren't categorized and thus there's no require for sorting it. Is it possible to leave out redirect pages from defaultsort error reporting. Thanks in advance --Kimmo Laine / 88.195.95.74 15:03, 14. Dez. 2008 (CET)

I do not quite agree, in fr wiki many redirections belong to categories, based on their purpose. Moreover, the existing detection only reports articles (and redirection pages) that comprise at least one [[Category]] statement. --66.131.214.76 23:53, 15. Dez. 2008 (CET) -- dixit LaddΩ parlons! ;)

Unterschied Fehler 016 und Fehler 027?

Hallo Stefan, habe mir die ganzen möglichen Fehler eben angesehen, dazu eine Frage aus reiner Neugierde heraus: Mir sind die Fehler 016 (Line very long) und 027 (Long line) aufgefallen. Gibt es einen bestimmten Grund warum zwischen 150 Zeichen und 80 Zeichen in einer Zeile mit Leerzeichen am Anfang unterschieden wird? Bei Fehler 027 (80 Zeichen) wird Quelltext vermutet, bei 016 (150 Zeichen) ein "Fehler" - aber Quelltext-Zeilen dürfen auch 150 Zeichen lang sein und Fehler gibt es auch bei 80 Zeichen ;-) Wäre es nicht einfacher das zusammenzufassen? --PasO 21:41, 14. Dez. 2008 (CET)

Das ist ein Überbleibsel aus zwei erfolglosen Versuchen diesen Fehler zu programmieren. Beide Fehler sind derzeit total deaktiviert. Vielleicht hab ich ja später mal Zeit dafür. Ich wollte damit alle Zeilen ausfindig machen, die an erster Stelle ein Leehrzeichen haben und dann mehr als eine Bildschirmbreite lang sind. Leider hab ich da zuviele Probleme gehabt und es erstmal gelassen.-- sk 09:16, 19. Dez. 2008 (CET)

error 036

Error 036 reports a form of broken redirects. Can you expand this feature to detect any illegal character between #REDIRECT and [[ ? For example, in Japanese Wikipedia many broken redirects have IDEOGRAPHIC SPACE (U+3000). --fryed-peach 15:11, 14. Dez. 2008 (CET)

Yes I can expand this. Can you give me an example. Thanks. -- sk 09:13, 19. Dez. 2008 (CET)

See old version of ja:商業放送. The source text is:

#REDIRECT　[[民間放送]]

, which contains IDEOGRAPHIC SPACE after #REDIRECT. --fryed-peach 13:43, 19. Dez. 2008 (CET)

<p> within <blockquote> (English Wikipedia)

Hi Stefan ... At en Wikipedia <blockquote> html is listed as acceptable "Wiki markup" language, and <p>s are necessary within <blockquote> for paragraphing (blank lines won't paragraph inside blockquote). QUESTION: Do you plan to adjust your Perl program to ignore <p>s within <blockquote> ... </blockquote>? (NOTE: I believe it is true that <blockquote> formatting can be replaced with {{quote| ... }} which does not need <p> to paragraph -- but as long as <blockquote> is listed as acceptable "Wiki markup," well, you get the point. :) Proofreader77 23:33, 15. Dez. 2008 (CET)

Can you give me an example. So I can better understand what you meen. I think it is possible to check for <blockquote> ... </blockquote>. -- sk 09:31, 19. Dez. 2008 (CET)

Sure ...

• Go to: en:Wikipedia:WikiProject_Check_Wikipedia#HTML_text_style_element_.3Cp.3E
• Then notice en:Alcopop table entry: "<blockquote><p>Flavored malt beverages are brewery products that differ from traditional..."
• NOTE: Within <blockquote>, Wikipedia does not treat blank lines as a new paragraph. If you want a paragraph inside <blockquote>, you must add explicit paragraph codes.
• As I mentioned, {{quote| ... could be used instead of <blockquote>with <p>s, BUT <blockquote> has apparently been on the list of accepted html in en.Wikipedia for a long time. :)

Cheers! Proofreader77 20:07, 19. Dez. 2008 (CET)

related: <p> within bullet (line beginning with *)

Far rarer, but a simpler one to program (IF line begins with "*" THEN ignore <p>s).

EXAMPLE: From en:Clinical psychology

*'''Existential.''' [[Existential therapy|Existential psychotherapy]] postulates that people are largely free to choose who we are and how we interpret and interact with the world. It intends to help the client find deeper meaning in life and to accept responsibility for living. As such, it addresses fundamental issues of life, such as death, aloneness, and freedom. The therapist emphasizes the client’s ability to be self-aware, freely make choices in the present, establish personal identity and social relationships, create meaning, and cope with the natural anxiety of living.<ref>Van Deurzen, Emmy. (2002). ''Existential Counseling & Psychotherapy in Practice.'' London; Thousand Oaks : Sage Publications. ISBN 0761962239</ref> Important writers in existential therapy include [[Rollo May]], [[Victor Frankl]], [[James Bugental]], and [[Irvin Yalom]].<P>One influential therapy that came out of Existential therapy is [[Gestalt Therapy]], primarily founded by [[Fritz Perls]] in the 1950s. It is well-known for techniques designed to increase various kinds of self-awareness—the best-known perhaps being the "empty chair technique"—which are generally intended to explore resistance to authentic contact, resolve internal conflicts, and help the client complete "unfinished business".<ref name="woldt">Woldt, Ansel and Toman, Sarah. (2005). ''Gestalt Therapy: History, Theory, and Practice''. Thousand Oaks, CA. : Sage Publications. ISBN 0761927913</ref>

In this case you should not use <p>. You should use <br />. This is better, because you need no </p>. If you write only <p> without </p>, than you have not valid XHTML. -- sk 09:30, 19. Dez. 2008 (CET)

Yes, <p> without matching </p> is not "correct" html (although many break this rule:), so let us use the code correctly (<p> ...</p>)... BUT THEN NOTE: </br> does not produce a blank line between paragraphs (as usual).

COMPARE: (<p>... </p>)

• Existential. Existential psychotherapy postulates that people are largely free to choose who we are and how we interpret and interact with the world. It intends to help the client find deeper meaning in life and to accept responsibility for living. As such, it addresses fundamental issues of life, such as death, aloneness, and freedom. The therapist emphasizes the client’s ability to be self-aware, freely make choices in the present, establish personal identity and social relationships, create meaning, and cope with the natural anxiety of living. Important writers in existential therapy include Rollo May, Victor Frankl, James Bugental, and Irvin Yalom.

  One influential therapy that came out of Existential therapy is Gestalt Therapy, primarily founded by Fritz Perls in the 1950s. It is well-known for techniques designed to increase various kinds of self-awareness—the best-known perhaps being the "empty chair technique"—which are generally intended to explore resistance to authentic contact, resolve internal conflicts, and help the client complete "unfinished business".

(</br>)

• Existential. Existential psychotherapy postulates that people are largely free to choose who we are and how we interpret and interact with the world. It intends to help the client find deeper meaning in life and to accept responsibility for living. As such, it addresses fundamental issues of life, such as death, aloneness, and freedom. The therapist emphasizes the client’s ability to be self-aware, freely make choices in the present, establish personal identity and social relationships, create meaning, and cope with the natural anxiety of living. Important writers in existential therapy include Rollo May, Victor Frankl, James Bugental, and Irvin Yalom.
  One influential therapy that came out of Existential therapy is Gestalt Therapy, primarily founded by Fritz Perls in the 1950s. It is well-known for techniques designed to increase various kinds of self-awareness—the best-known perhaps being the "empty chair technique"—which are generally intended to explore resistance to authentic contact, resolve internal conflicts, and help the client complete "unfinished business".

As I noted, <p> within a bullet line is is FAR less common than within <blockquote> ...</blockquote>, but is (apparently?)OK in English Wikipedia. Proofreader77 20:35, 19. Dez. 2008 (CET)

Not really a question, just highlighting an example of a page listed in your program's output. Proofreader77 05:13, 16. Dez. 2008 (CET)

Highlighting is a feature for the future. :-) --sk 09:32, 19. Dez. 2008 (CET)

Cheers! Proofreader77 20:35, 19. Dez. 2008 (CET)

Missing template errors

The section "Template programming element" doesn't discover all errors. It should look for any occurrence of "{{{" in the article namespace, not just "{{{1}}}". --LA2 09:00, 19. Dez. 2008 (CET)

I know that this are not all elements. Can you give me an example? Why not "{{{1}}}"? -- sk 09:08, 19. Dez. 2008 (CET)

Look at this syntax correction that I made. Some earlier user had by mistake used subst: on a template instead of just calling the template. As a result, the article included both <noinclude> and {{{width|200}}}. These elements should not appear in articles, but this error was not detected by Check Wikipedia. --LA2 09:26, 20. Dez. 2008 (CET)

error number 30, no image description

Found a false positive from fi.wiki. [[Image:Cis-2-butene.PNG|cis-2-buteeni|frame|right]] this image HAS a description: "cis-2-buteeni", yet it is listed as not having one. Why is that? Can this be fixed? -- Kimmo Laine / 88.195.95.74 06:55, 19. Dez. 2008 (CET)

At the moment my script find not the description in the middle of all the other parameters. Normally we use the description at the end of the image. For example: [[Image:xy.jpg|frame|right|cis-2-buteeni]]! It is better to reed. In the German Wikipedia we had only <20 errors like this. Please write the description at the end. -- sk 09:12, 19. Dez. 2008 (CET)

I took a peak the the script, and with my feeble understanding of perl and regexp, I think you are just testing if the last parameter of image call is 'thumb', 'left', 'right' or a pixel value. In that case, you are not checking if it's 'frame'. Wouldn't an image like [[Image:foo.jpg|frame]] pass as having a description, thought it doesn't? That'd be a false negative. I don't know if such cases exist, but it's a possibility. -- Kimmo Laine / 62.142.59.83 10:24, 19. Dez. 2008 (CET)

You are right. At the moment I check only for the last parameter. But we have so much errors, that I don't want explode the list. :-) "frame" is one and a normal image without all is the other one ([[image:xy.jpg]]). If the number of errors smaller I will insert this. -- sk 11:45, 19. Dez. 2008 (CET)

I insert "frame" in the script. -- sk 22:46, 20. Dez. 2008 (CET)

Translation

Hi! Thanks for the statistics! I have a question: do we really need lines matching error_\d\d\d_(prio|head|desc)_script in our translation? It seems, they are ignored and only lines matching error_\d\d\d_(prio|head|desc)_..wiki are used. We had had error_033_prio_script=0 END by accident in the translation but the test was run anyway. I do not know if it was run because we had had also error_033_prio_cswiki=3 END.[2] Miraceti 09:10, 23. Dez. 2008 (CET)

All parameter with end ..._script are only for information. This is the parameter which is set in the script by me. If you want change the procedure in cswiki then please change the parameter ..._cswiki. When you set error_033_prio_script=0 END then nothing will happen. If you set error_033_prio_cswiki=-1 END the error 33 will set "deactivated". -- sk 09:24, 23. Dez. 2008 (CET)

So it is as I supposed. We can take them out. Thanks.

Are you sure that error_033_prio_cswiki=-1 END causes error's deactivation? I thought it means "unknown" -> use the default _prio_script value. The deactivation should be done by error_033_prio_cswiki=0 END, shouldn't be? Miraceti 12:25, 23. Dez. 2008 (CET)

Yes, that is right. "0" means deactivated. -- sk 13:03, 23. Dez. 2008 (CET)

<references /> but no <ref>

Your script lists articles with <ref>s, but without <references />. Can it also perform inverse check: <references /> (or, of course, template like {{reflist}} on enwiki) but no <ref>? Matma Rex answer me on plwiki 14:13, 12. Dez. 2008 (CET)

So, what about it? Matma Rex answer me on plwiki 22:23, 19. Dez. 2008 (CET)

The problem is, that sometime the <ref>-tag is inside a template. I can not scan this. -- sk 16:44, 23. Dez. 2008 (CET)

I dont think it's so common. And if it is, detection can be turned off, right? Matma Rex answer me on plwiki 16:03, 25. Dez. 2008 (CET)

False positive in error 037 : articles missing DEFAULTSORT

An article like S/2003 J 10 is wrongly reported in section 037. Symbol «/» should not cause a report.

Comment #Defaultsort problem above also indicated an incorrect report in section 006 (DEFAULTSORT with special token) in NL Wikipedia.

--66.131.214.76 03:11, 24. Dez. 2008 (CET)

I will change this. -- sk 09:19, 27. Dez. 2008 (CET)

<center>

Can you add a detection of <center> tag? --fryed-peach 03:29, 25. Dez. 2008 (CET)

Yes, I can add this. -- sk 09:18, 27. Dez. 2008 (CET)

Defaultsort problem

How to fix this ones:

• http://nl.wikipedia.org/wiki/Wikipedia:Wikiproject/Check_Wikipedia#Code_006:_Defaultsort_met_speciale_tekens

82.171.148.8 18:29, 20. Dez. 2008 (CET)

 Ok I have fix this in the script. -- sk 11:03, 29. Dez. 2008 (CET)

error number 37 section on ca:wikipedia

Hi, there is a little error in error number 37 section on catalan wikipedia, the cawiki_output_for_wikipedia.txt contains the following sentence without the last </nowiki> tag.

El títol té algun caràcter especial i a l'article no hi ha ORDENA o DEFAULTSORT. O bé alguna categoria no té la sintaxi <nowiki>[[Categoria:ABC|Text]]</nowiki>. Regards. --Loupeter 08:44, 29. Dez. 2008 (CET)

This was an error in the cawiki translation page. See my change here. -- sk 10:51, 29. Dez. 2008 (CET)

Nº 10 false error

Hi, Stefan. First of all, thanks for this really really useful tool, and excuse my bad english. I'm user at es:wiki, and I would like to draw attention upon an error i've seen sometimes on item number 10, "not having the same amount of [[ and ]]". You can see it at es:Cantabria and es:Homosexualidad en los Estados Unidos. Can this be fixed? Ilfirin 20:49, 29. Dez. 2008 (CET)

Hello Ilfirin, I work on it, but it is very difficult. The problem is the line break in an image for ref or legend. If you delete the line break my script has no problem. -- sk 21:03, 29. Dez. 2008 (CET)

Thanks for the quick answer. What about counting how many [[ and ]] are there when the script finds an error? Or keeping "File"/"Image" (at es:wiki "Archivo"/"Imagen") out of this error list. Or maybe both at the same time: veryfing that the images have the correct amount of [[ and ]] indepently of the line breaks. Greetings, Ilfirin 22:59, 30. Dez. 2008 (CET)

Links in new «statistics» section

Hi, it would be useful if your new «statistics» section was providing links for errors having at least one detection, so that users can jump directly to sections of interest to them. Instead of:

| 6 || description_of_error_6 || middle || unkown || style="text-align:right;" | || style="text-align:right;" | 1 || ↗ || style="text-align:right;" | 1

use

| 6 || [[#description_of_error_6|description_of_error_6]] || middle || unkown || style="text-align:right;" | || style="text-align:right;" | 1 || ↗ || style="text-align:right;" | 1

Errors having 0 detections do not have sections on the page, so better not put a link for those errors, if possible.

Thanks --66.131.214.76 01:15, 24. Dez. 2008 (CET)

I will try this. -- sk 09:18, 27. Dez. 2008 (CET)

I have test this, and I have not good results. For me it is a problem with the different languages. -- sk 22:03, 3. Jan. 2009 (CET)

Titel mit Sonderzeichen und ohne DEFAULTSORT

In dieser Sektion tauchen auch Begriffe mit „/“, wie etwa TP/IX auf. Welche DEFAULTSORT-Angabe sollten die denn kriegen?--JFKCom 01:03, 28. Dez. 2008 (CET)

„/“ werden rausgenommen aus dem Skript. Siehe auch weiter oben. -- sk 09:16, 28. Dez. 2008 (CET)

Ah, jetzt seh' ich's auch weiter oben. Danke für die schnelle Auskunft!--JFKCom 09:44, 28. Dez. 2008 (CET)

Hallo Stefan, wie kann ich denn C++ aus der Fehlerliste rauskriegen? Weder das noch das hat geholfen. --UV 22:28, 28. Dez. 2008 (CET)

I think with {{DEFAULTSORT:C}} it should work. Please try it again. In the last day my script dont run at the toolserver. Maybe this was the reason. Normaly with DEFAULTSORT it should work. -- sk 11:03, 29. Dez. 2008 (CET)

Did not work. Why not? --UV 17:31, 2. Jan. 2009 (CET)

Is it correct that this only reports special characters within the first 3 positions of the title? I started to add DEFAULTSORT to many thousands of articles, but the reported number only fell with a few hundreds. --LA2 15:13, 2. Jan. 2009 (CET)

Yes, I have change this, after some request from other languages. -- sk 22:01, 3. Jan. 2009 (CET)

And it is the right way. Using DEFAULTSORT to resolve sorting problems caused by accents is a workaround for this very old bug of MediaWiki software. It will eventually get fixed and all such DEFAULTSORT will become useless and possibly conflict with titles of articles that will be renamed in the future. Thus limiting the detection of missing DEFAULTSORT to only those articles where the accented letter is at the beginning of the title (and thus impacts its sorting) is a good compromise to avoid damaging the encyclopedia. --66.131.214.76 03:51, 4. Jan. 2009 (CET)

The correct/not correct table

Hi Stefan, can you go here it:Discussioni Wikipedia:Elenchi generati offline/Check Wikipedia and complete the table? Or can you create a subpage /Errors and solutions with something like that table? Thanks a lot and happy new year! --Red Power 15:51, 31. Dez. 2008 (CET)

Update: I've split the table in 2 sections, since it's unfair to mix up the correct tags and the deprecated ones. I'm still doubtful about <p>; maybe we need 3 tables: A-correct and replaceable; B-discouraged; C-deprecated. And I think we should disable errors 41 and 42, because the replacements for 'big' and 'small' can produce only more mistakes. Bye. --Red Power 22:46, 4. Jan. 2009 (CET)