Template talk:Script/Archive 1

Comments
Latnf is a problem, Fraktur letters are separately encoded at 1D56C-1D59F (bold), 1D504-1D537, pluis C: 0212D, H: 0210C, I: 02111, R: 0211C, Z: 02128, long s: 017F. Supported e.g. by Code2001.

Choosing "Latf" should however somehow map an ascii string on these letters.

another problem is Nasta'liq script, considered a font variant by both Unicode and ISO 15924. This will need a solution like cuneiform where the font variant can be selected by an additional parameter.
 * The "Fraktur" block is intended for mathematical symbols only. Latnf should not map to those, but enforce various Fraktur typefaces. A difficult choice, since there exists a large number of these but none seems to be somehow the 'standard Fraktur' font recommended for Latnf. dab (𒁳) 07:27, 18 July 2007 (UTC)

Language codes
Hi, I've notice that in all scripts supported the language code is added to the HTML code (either a fixed language code for script/Runic, script/Gaelic or script/Coptic, or 'und' for undermited if not specified in Script/Nastaliq) but Hebrew and Cuneiform. Any reason why for these two scripts the language code is not included, not even 'und'? Thanks. &mdash;surue&ntilde;a 21:15, 22 July 2007 (UTC)
 * I am unsure if language codes should at all be included, or if we should give 'und' whenever this template is used (see above...). I guess I missed Hebrew. Cuneiform should also give 'und'. Maybe we could allow an optiona paramenter for the langauge code, and give 'und' whenever it is omitted. Feel free to fiddle with this. dab (𒁳) 15:12, 23 July 2007 (UTC)

Template:lang
Hello, I'd like to ask in which cases this template should be used instead of lang/rtl-lang. There are a lot of multilingual templates, and this confuses the user if they are for the same purposes. Couldn't this template be merged with these other more general one (via CSS styles, for example)? Best regards &mdash;surue&ntilde;a 20:34, 18 July 2007 (UTC) dab (𒁳) 19:16, 22 July 2007 (UTC)
 * this template is intended to address rendering issues, e.g. switching between various modes of cuneiform. There is currently no ISO code that would allow you, say, to distinguish between Old Assyrian and Neo-Assyrian cuneiform, it's all just "Xsux". A similar case is Nastaliq. Also, there are ISO scripts that are not necessarily used with a particular language, such as "Latf".
 * it is intended for cases where the script itself is under discussion. E.g., when discussing Cyrillic letters, there is no pertinent language code except 'und' "undetermined", since the string  is not in any particular language. Granted, there is nothing wrong with just using 'und' А Б В Г Д Є Ж since it is clear anyway that these characters are Cyrillic, no need to annotate that.


 * I see, these aren't easy use cases... :-) First of all, it is worth noting that these are very "rare" cases, I mean, only a few people will really write about these topics. My main objection with this template is that it is not needed to enforce the language code (because in some cases this is the purpose of the template, of course, adding the 'und' language code). So I'm afraid of wikipedians widely using this script template to tag normal words, i.e. where the language should be specified. So as you said, for the second case, when the script itself is being discussed it is easy to document that the editors should use lang with the undetermined language code. For the case about specifying the cuneiform epoch, I think that this is a difficult topic: few people has installed a font for rendering cuneiform (there are even no fonts available for all the different epochs), so I would rather use images rather than forcing the wikipedians to install multiple fonts. However, if a template is really needed for this usage, I must say that I suppose this is one of those few cases where a specific cuneiform template should be created (again). Even if there are more ancient languages where the epoch should be specified, I not sure whether a common template should be created to handle them because it would be misused very easily. Finally, for the case about Nasta'liq, I would like to ask if a possible solution is to employ this callygraphy when the language is Persian, Pashto or Urdu (in the CSS, e.g. via lang). I really think your template is awesome, and in fact I think a lot of its functionality should be added to template lang, but in my opinion only the minimum number of multi-language templates should be available otherwise editors would be confused, and this script template can be easily (and it is being) misused. We can create private language/script codes for the Wikipedia (in fact, we will have to create and document private language code for transliterations, until the W3C creates a standard solution). What do you think? Best regards, &mdash;surue&ntilde;a 21:04, 22 July 2007 (UTC)

surueña, this template is intended to address the "difficult cases". It should only be used by people who know what they are doing, and I do not suggest that it should be used widely. However, there are a few cases where it is (still) necessary to use it: the enforcing of script variants that are considered font variants by Unicode (Latf, Latg, variants of Xsux and Ital). These are the rare cases where we are forced to give a list of known fonts for those users who have them installed. Obviously, if a system doesn't have a font with the script, there is nothing we can do to render it. How this template should and should not be used should be stated in its documentation, as with every other template. I would be happy with a good css solution for Nastaliq though. For Latg, since we don't want Irish rendered in Latg by default, this could be done in the css for "ga-Latg", "mga-Latg" and "sga-Latg", exclusively. Similarly, "de-Latf", "en-Latf" etc. Similarly, for Xsux, we could distinguish "sux-Xsux" and "akk-Xsux". Unfortunately, "akk-Xsux" could be either Old or Neo-Assyrian (but the difference between the Old and the Neo-Assyrian script is considerable -- I know there are no widespread fonts yet, I'm looking into the future here). As I say in the doc, I never even got round to addressing Hani, this should be done cleanly in the css. Since Unicode is now recognizing Coptic as separate, this is not a problem any more, but we still do need a disambiguation of the various Old Italic scripts (Ital). No, I do not think we should introduce idiosyncratic language or script codes. If there are no ISO codes, we should use specialized templates, not the css. dab (𒁳) 15:16, 23 July 2007 (UTC)


 * there are other cases, such as Coptic and Hebrew that are here only for historical reasons. These should indeed be addressed in the css, and after that, use of the script template should be deprecated. But they have to be addressed properly first. Thus, for Hebrew, the css should anticipate all of "he", "yi", "yi-Hebr", "lad-Hebr", "ar-Hebr", "und-Hebr", and if possible any "...-Hebr" code. dab (𒁳) 11:44, 24 July 2007 (UTC)

Protection?
Maybe the template needs Template:pp-template potection. But: about 100 links, no more. Anyway, I separated the documentation. -DePiep (talk) 22:11, 28 July 2009 (UTC)

FreeSans vs Free Sans
See MediaWiki_talk:Common.css -- John Vandenberg (chat) 16:14, 11 April 2011 (UTC)

Automatic right-to-left handling?
Is there any way text in an individual language can be automatically enclosed within a right-to-left span tag using this template? ᛭ LokiClock (talk) 23:49, 19 June 2010 (UTC)


 * I finally found the subpages that hold script-specific code, such as Template:Script/Runic, so this is solved. ᛭ LokiClock (talk) 13:07, 18 April 2011 (UTC)

Nastaliq musn't be used for unrelated languages
That phrase  specifies that the computer must show Arabic letters with a Nastaliq font. It's a fault, because Nastaliq isn't the regular way of rendering Arabic letters. Do something, to make the code, instead, choose an appropriate font, such as those specified there script/Arabic. The problem also affects rendering Arabic letters with that template Phoenician glyph. If it is intended that Nastaliq fonts be used for Urdu or languages which may use that style regularly, then its OK, but please, make sure that not all Arabic script languages be shown that way. --Mahmudmasri (talk) 04:40, 23 June 2011 (UTC)

Aegean font-face for Old Persian
Could someone add the Aegean font-face for the Xpeo language code? Thanks! 96.241.150.64 (talk) 18:52, 4 December 2012 (UTC)
 * done.--El caballero de los Leones (talk) 03:25, 28 April 2015 (UTC)

Andalus Script Style
HI, I SEE YOU USE LOT OF STYLE, IN TEMPLATE…
 * &lrm; &rarr; &lrm; (uses Script/Arabic)
 * &lrm; &rarr; (uses Script/Nastaliq)
 * &lrm; &rarr; (uses Script/Nastaliq)
 * &lrm; &rarr; (uses Script/Nastaliq)

BUT I NEED, THE ANDA‑LUSIA ONE (WINDOWS MONOTYPE ANDALUS), WHO IS THE PERFECT SYNTHESIS BETWEEN ARABIC & HEBREW, UNIFYING FONT SHARING COMMON VALUE, THAT WOULD BE MY FONTS IN TAÔQODÊKRÊSË PROJECT… I NEED THAT TEMPLATE, TO AVOID TO WRITE LONG COMMAND : سفراء I ALREADY VRITTEN TO MONOTYPE TO ASK TO ADD PERSIAN, URDU KURDISH & SUPPLEMENT IN IT…

--ͰΑΘϷΕΝΑΝΑΤΕΣ‑ΥΣ (dôr.) ΗΤϷΑΝΑΝΘΕΣ‑ΟΣ (att.) ͿΗϷΑΟΛΙΕΥϷΟΛϜƏ Α̃ΜΑΝΓΖΕΔͿƏ - حطشأنانثس يشأعليهوسعلۋ أمنجزهدي  (talk) 11:33, 11 August 2016 (UTC)

Runes going the wrong way
Could someone help us out over at Talk:Thor? The runes are going the wrong way now, apparently due to a recent template edit. Haukur (talk) 18:49, 12 July 2019 (UTC)


 * After some trial and error in the sandboxes I have managed to fix the template for script/runr to take an "rtl" parameter in order to display runic text with right-to-left directionality (of course in rtl mode the individual runic letters should be mirrored, but that is done at the font level using OpenType, which may not work even if the runic font supports rtl mirroring ...). BabelStone (talk) 00:51, 13 July 2019 (UTC)


 * Thank you! Haukur (talk) 11:45, 13 July 2019 (UTC)

Inscriptional Parthian
I'm confused by the change from "Inscriptional Parthian" to "Pahlavi" on Template:Script and its doc page for ISO 15924 code  (Inscriptional Parthian). Can you shed some light on it? What am I missing? DRMcCreedy (talk) 23:07, 9 February 2021 (UTC)
 * This edit I made. The change is, for Prti: instead of calling
 * Script/Inscriptional Parthian ❌,
 * Script/Pahlavi ✅
 * is called (transcluded).


 * Obviously, Template:Script/Inscriptional Parthian is a redirect. The edit makes skipping that redirect, calling the target template directly. Since all is in the background (not visible for editores using  in an article), for regular editors nothing changes, they can use  as before. This is also why I think this is not controversial. HTH -DePiep (talk) 23:28, 9 February 2021 (UTC)
 * I understand now. Was confusing script Prti with ISO 15924 Prti.  I now see that the fonts for Inscriptional Parthian are covered.  Thanks. DRMcCreedy (talk) 23:49, 9 February 2021 (UTC)
 * Yes, this is causing headaches over here too ;-)
 * We have four ID's for a single script: enwiki-article, Wikidata-item, ISO-alpha4, Unicode alias. (ISO 15924/overview-4id)
 * Then, with these four, there is still the issue: show them correct! Is where comes in, and css and preferred fonts.
 * So we need more edits sure ;-) -DePiep (talk) 01:47, 10 February 2021 (UTC)

Mtei, Meithei worsening in chrome
In Chrome, the Mtei/Meithei font worsens by using the script template. Demo from the /doc:


 * &rarr; (uses Script/Meithei)

In Chrome, the lefthand Mtei text shows its characters, the righthand does not (placeholders only). -DePiep (talk) 13:17, 19 February 2021 (UTC)

az-Arab, fa-Arab
In article Miandoab (Azerbaijan city), Script is called with az-Arab. Which Arab specification should  use?


 * مياندوآب (fa, so pre-1929?)
 * Qoşaçay (post-1991 Latin?)


 * I have set (added):  . -DePiep (talk) 19:21, 20 February 2021 (UTC)

Smrt, Samaritan

 * I have removed faulty 'ISO code' Smrt ❌. Code Samr ✅ is still available for Samaritan. "Smrt" was not used as input anywhere in 1322 enwiki pages. -DePiep (talk) 20:14, 9 February 2021 (UTC)

Hano, Hanunoo

 * Hano for Hanunuo script appears twice in the #switch-list:
 * Hano &rarr; Script/Hanunoo)
 * default= ... &rarr; class="Script-Hano" (btw uppercase S?)
 * The second one will never fire.


 * Template explansion demo (Xpeo=via #default):


 * &rarr;

-DePiep (talk) 20:28, 9 February 2021 (UTC)
 *  FooBar
 *  FooBar


 * Compare option added . Now Hanunoo is recognised and will be solved throug the #default=-route (class=...).


 * &rarr;


 *  FooHanoBar
 *  Foo-hnoo-Bar
 *  FooXpeoBar


 * I cannot judge which option serves the best font rendering (Firefox, Chrome: no difference). -DePiep (talk) 20:48, 9 February 2021 (UTC)

Using Nanunoo script



 *  ᜥ ᜥᜲ ᜥᜳ ᜥ᜴
 *  ᜥ ᜥᜲ ᜥᜳ ᜥ᜴
 *  ᜥ ᜥᜲ ᜥᜳ ᜥ᜴

result:
 * (late sign: -02:37, 10 February 2021 (UTC))
 * (late sign: -02:37, 10 February 2021 (UTC))
 * (late sign: -02:37, 10 February 2021 (UTC))
 * (late sign: -02:37, 10 February 2021 (UTC))

Geok and Geor, Georgian
In Unicode, Geok (Khutsuri, ?!) and Geor  are not distinguished: both are named "Georgian" (Unicode Alias name). Therefor it is plausible that the 'covering' scripts will cover both scripts together (I cannot prove though). However, in Geor now is handled as non-specified.
 * &rarr;


 *  ⴌⴓⴑⴞⴓⴐⴈ
 *  ⴌⴓⴑⴞⴓⴐⴈ
 * -- Geok
 * -- Geor


 * Questions
 * Shouldn't they be merged (=treated as one by )? -DePiep (talk) 21:19, 9 February 2021 (UTC)

Cakm and Chak, Chakma

 * Chak ❌ is not an existing ISO code. To be removed. Cakm ✅ is OK, and should handle Chakma script without change.
 * Chak is not used as input in the ~1300 transcluding pages. -DePiep (talk) 22:19, 9 February 2021 (UTC)

Syre, Surj, Syrn, Syrc: Syriac
In Unicode, Syrc envelopes all four scripts. Interestingly, Syrc (-c) is not recognised in and so is handled by "class=Unicode" (uppercase?).


 * --e
 * --j
 * --n
 * --c
 * &rarr;


 *  ܐܣܛܪܢܓܠܐ &lrm; --e
 *  ܣܪܛܐ &lrm; --j
 *  ܡܕܢܚܝܐ &lrm; --n
 *  e-ܣܛܪܢܓܠܐj-ܣܪܛܐn-ܡܕܢܚܝܐ --c
 * --e
 * --j
 * --n
 * --c


 * Todo: consider & research merging all four. -DePiep (talk) 23:01, 9 February 2021 (UTC)


 * Renamed, technical move only
 * Script/Serto &rarr; Script/Syrj
 * Script/Mdnh &rarr; Script/Syrn
 * Script/Strng &rarr; Script/Syre
 * -DePiep (talk) 01:59, 10 February 2021 (UTC)

Latn, Latin
It confuses me that script ID "Latn" (=Latin script) makes to use Script/Classical and Medieval Latin. Does indicate it uses specific scripts.

As I understand it, the Latin script is the universal script at enwiki, enveloping all Latn characters, including those derived (like Polish diacritics and IPA). Yes it could include Medieval forms, as it also includes --per Unicode spources-- the IPA characters.

So it would be problematric when an editor enters "1" in the reasonable expectation that it will render all characters, not just a-b-c, as expected. Not changing the font to medieval, and maybe even start missing out IPA-characters. -DePiep (talk) 00:00, 10 February 2021 (UTC)

Bugi, Lontara
Unicode chart Buginese has local font family set. To disentangle, check Bugi <=?=> Lontara. -DePiep (talk) 21:00, 18 February 2021 (UTC)

Mong, Mongolian

 * About Mongolian script, MongolUnicode. Is in Unicode, alias:"Mongolian"; WD d:Q1055705.

MongolUnicode is a a well-working Script template at enwiki. It nicely handles vert/hor options, and style settings like font-size. Used in 900+ articles (and 5000 times in there). I am looking to get it channeled through this, without disruption. -DePiep (talk) 22:33, 23 February 2021 (UTC)

Shrd, Shahara
About Script/Sharada, used here. Script/styles sharada.css exists, but is not called by (the css is not transcluded/used at all). Is this intended or by mistake? If intended, we should delete the css file; otherwise we should update, I assume. any idea? -DePiep (talk) 15:07, 1 March 2021 (UTC)
 * If I recall correctly, I created the css and only after that I realized that Script/Sharada already existed without any css transclusion. Malurian123 (talk) 16:33, 1 March 2021 (UTC)
 * OK, thanks. Now, should use /styles sharada.css? If that was your plan, please make the edit. I am not very familiar with TemplateStyles. -DePiep (talk) 21:54, 1 March 2021 (UTC)
 * Yeah, I guess so. Malurian123 (talk) 03:41, 2 March 2021 (UTC)

Hmng, Pahawh Hmong
About Script/Pahawh Hmong and Script/styles pahawh hmong.css.

The /styles subtemplate is not used in the Script/Pahawh Hmong parent. By intention, or to be corrected?

(BTW, as of today main template Script/Pahawh Hmong is not used wrong ). -DePiep (talk) 23:13, 1 March 2021 (UTC)
 * Sorry, I was not involved in creating either template. Malurian123 (talk) 03:41, 2 March 2021 (UTC)

more protection?
This template currently has 1600+ transclusions, and it often appears in lead sections. Does anyone mind if we elevate its protection, as the risk of vandalism seems high? --Joy &#91;shallot&#93; (talk) 07:59, 13 September 2021 (UTC)
 * 11k tc's in 862 articles. However, I did not see any vandalisme. -DePiep (talk) 11:55, 13 September 2021 (UTC)

Cuneiform template errors
Template:Script/Cuneiform, which renders Akkadian cuneiform characters, is supposed to be able to display both Classical Sumerian and Neo-Assyrian signs, by adding a number, 4 for Sumerian and 7 for Neo-Assyrian. Currently, the template only displays Classical Sumerian signs regardless of the input: and  look the exact same when they are supposed to look quite different (see numbers 4 and 7 in the image to the right). This is an issue since articles using the cuneiform template for Neo-Assyrian stuff, such as the Neo-Assyrian kings, produce correct Akkadian spellings but with antiquated signs that would not have been used by these figures themselves. The article on Sinsharishkun renders his name as 𒀭𒈗𒃻𒌦 while he himself rendered it as (the same four signs having undergone ~2000 years of development). Hopefully this is a fixable issue given that the template is supposed to support both renditions. Ichthyovenator (talk) 09:48, 11 September 2021 (UTC)
 * Apparently this issue is not experienced by everyone (see here) - but for me the template does not display Sumerian and Neo-Assyrian differently (showing Sumerian regardless of input), which is an issue since some readers will thus not see them differently either. Even if it does work, it also appears that some signs (particularly 𒌶 apparently) do not display in Neo-Assyrian even for those who do see the template correctly. Ichthyovenator (talk) 10:30, 11 September 2021 (UTC)
 * All this depends on what fonts you have installed. For me, according to my browser's Inspector, the  text displays with Segoe UI Historic (a font not specified by the CSS added by the template, but chosen by my browser) and the   text with Noto Sans Cuneiform. (They are not perceptibly different, which is odd; perhaps one has borrowed glyphs from the other or they both derived glyphs from a common source.) If you haven't got any of the fonts specified in either the   or the   CSS, you'll probably have the same font applied to both. For how to identify the font used by the browser, I wrote wikt:Wiktionary:Unicode because this keeps coming up in Wiktionary discussions.
 * From my experience on Wiktionary, I've concluded that the only way to really solve this issue is with webfonts. Without webfonts users have to install one of the fonts specified in the CSS. Neither Wiktionary nor Wikipedia loads webfonts anywhere that I've noticed, but Wikisource has mw:Extension:UniversalLanguageSelector installed, which loads webfonts, as described in s:Template:Lang/doc. — Eru·tuon 17:31, 11 September 2021 (UTC)

Unicode

 * Does Unicode have this distinction? DePiep (talk) 20:18, 11 September 2021 (UTC)
 * As in different code points for different historical forms of cuneiform characters? I'm not very familiar with cuneiform, but no in the case of the character mentioned above, 𒊕 (U+12295 CUNEIFORM SIGN SAG), and perhaps no in general. — Eru·tuon 23:23, 11 September 2021 (UTC)
 * Yes, that distinction to look for. Cuneiform in Unicode is described in Chapter 11.
 * 11.1 Cuneiform, Sumero-Akkadian; 11.2 Ugaritic; 11.3 Old Persian.
 * Identified as script (ISO 15924), but currently not in Unicode: Proto-cuneiform numerals, . -DePiep (talk) 08:04, 12 September 2021 (UTC)


 * I see. I'll look into trying to install the right fonts and see if I can see the change. Given that the fonts in this case can distinguish between different stages of the cuneiform script and I'd wager many readers miss the needed fonts in the same way I do, I still think that perhaps there should be some solution which doesn't require readers to install new fonts to see the text correctly. Without the correct fonts, the names of the Neo-Assyrian kings display in Classic Sumerian cuneiform, which is anachronistic to say the least. As I also mentioned, some signs (𒌶) apparently do not display correctly in Neo-Assyrian even if the fonts are installed. Ichthyovenator (talk) 16:26, 12 September 2021 (UTC)
 * The solution that doesn't require users to install fonts is an extension that loads webfonts, which I mentioned above. — Eru·tuon 02:11, 13 September 2021 (UTC)
 * Right. I'm not able to implement something like that but perhaps someone more knowledgeable would be able to take a look when they have the time. Ichthyovenator (talk) 08:05, 13 September 2021 (UTC)
 * I think webfonts require the help of the Wikimedia developers. The extension that provides webfonts on Wikisource, UniversalLanguageSelector extension is already installed on Wikipedia, but I haven't seen evidence that it provides webfonts and the page on Wikisource seems to indicate that it doesn't provide webfonts for different styles of cuneiform (I only see Akkadian mentioned, not Sumerian or Hittite or Old Persian). So perhaps you would need to find free fonts for different styles of cuneiform, obtain consensus for webfonts through a Wikipedia discussion, and then write a Phabricator request to ask the Wikimedia developers to turn on the webfont feature of the extension and to add free webfonts for different styles of cuneiform to the extension. I'm guessing here though because I haven't seen how they got it set up on Wikisource. — Eru·tuon 18:02, 13 September 2021 (UTC)


 * I think when Unicode does not differentiate, the font-setting will be complicated (unsolved) forever. These are the current options in :


 * from documentation:


 * sux|4 = Akkadian" title="Classical Sumerian cuneiform"


 * akk|elx|xeb|xhu|5 = Akkadian" title="Neo-Sumerian cuneiform"


 * hit = UllikummiA, Assurbanipal, CuneiformNA,'Noto Sans Cuneiform'" title="Hittite cuneiform"


 * 7 = CuneiformNA,Assurbanipal,Akkadian,'Noto Sans Cuneiform','Noto Sans Sumero-Akkadian Cuneiform','Segoe UI Historic';" title="Neo-Assyrian cuneiform"


 * peo = Behistun, Aegean, 'Old Persian';" title="Old Persian cuneiform"


 * default = 'Segoe UI Historic','Akkadian','Noto Sans Cuneiform','Noto Sans Sumero-Akkadian Cuneiform';" title="cuneiform text"

That is: six ID's + default. How dow we ever think to solve this? -DePiep (talk) 19:21, 12 September 2021 (UTC)

Cuneiform template problems
Template:Script/Cuneiform renders the Neo-Assyrian forms of most cuneiform signs, however it refuses to render some signs such as 𒌶 and 𒉢, which appear no different from the Sumerian forms when I use the code for Neo-Assyrian in the template. Antiquistik (talk) 15:44, 14 September 2021 (UTC)