Module talk:character info

|caption= parameter

The display of |caption= needs to be made independent of |image=. Legitimate uses of |caption= without |image= include {{character info/new|codepoint=0x3008|caption={{normalization|2=2329|3=LEFT-POINTING ANGLE BRACKET}}}} in 〈; the content of |caption= is currently not displayed, but that content is displayed (albeit with some undesirable results) by calling a blank |image= parameter. Pinging Kephir. — I.S.M.E.T.A. 14:06, 30 January 2015 (UTC)

@Kephir Thanks for that. — I.S.M.E.T.A. 14:14, 30 January 2015 (UTC)

Autogeneratation of hexadecimal codepoints’ decimal equivalents

Please see Wiktionary:Tea room/2015/April#=. Would it be possible to have this module automatically generate the decimal equivalent of whatever hexadecimal codepoint it displays? That would be a desirable feature. — I.S.M.E.T.A. 18:01, 27 April 2015 (UTC)

@Dixtosa That’s perfect; thank you. :-)  — I.S.M.E.T.A. 07:05, 28 April 2015 (UTC)

Blanking reserved codepoints

@Keφr: I'm really impressed by whatever you did to this module that causes it to autogenerate Greek characters' compositions; that is helpful information. Thank you. There's a small, unrelated problem, however: See, for example, this use of yours of {{character info/new}}; is it possible to get rid of the links to reserved codepoints, like the one to U+038D in the character-info box for U+038E GREEK CAPITAL LETTER UPSILON WITH TONOS in Ύ? If not, how about a note stating that the codepoint is reserved? It looks like the arrow points to a spacing character otherwise. — I.S.M.E.T.A. 18:34, 6 July 2015 (UTC)

Bold codename

I believe codenames are too geeky to be bold and that large.--Giorgi Eufshi (talk) 11:20, 18 November 2016 (UTC)

@Giorgi Eufshi: Allright. The text-size of the codenames was 110%, now it's just 100%. I also removed the bold formatting. --Daniel Carrero (talk) 11:55, 20 November 2016 (UTC)

@Giorgi Eufshi: Let mw know if you would change anything else. --Daniel Carrero (talk) 19:01, 21 November 2016 (UTC)

I would hide unassigned ranges under three dots (put between the left/right arrow and the name of the current range) in a {{comment}}-like fashion. --Giorgi Eufshi (talk) 06:16, 22 November 2016 (UTC)

I did it myself. Does it not look better?--Giorgi Eufshi (talk) 07:00, 22 November 2016 (UTC)

@I'm so meta even this acronym, what do you think about this change? See entries like ԯ and Σ, the unassigned ranges are hidden under three dots. What do you think? (naturally, I'm asking you because you were the person who suggested that the template should recognize and inform people whenever there are unassigned ranges in the first place) --Daniel Carrero (talk) 09:09, 22 November 2016 (UTC)

Let's see. I'll use the entry ଌ as an example. Personally, I'd prefer displaying the full "[U+0B0D–U+0B0E]" as originally proposed by  I.S.M.E.T.A. I like the geeky stuff, but I wonder if you and/or others would prefer hiding it specifically because it is too geeky. Meanwhile, I just formatted the "..." as bold text to make it easier to see. --Daniel Carrero (talk) 20:47, 22 November 2016 (UTC)

@Giorgi Eufshi, Daniel Carrero: IMO, that is far too easy to miss (users have complained that the button to toggle quotations is too easy to miss). The whole character-box thing is pretty geeky — it already gives the codepoint (in decimal and hexadecimal), the character's Unicode name, canonical composition, character block, and previous/next codepoints by default — the only non-geeky thing is the large image. The codepoint stuff is useful technical information, and I see no value in hiding it. — I.S.M.E.T.A. 20:56, 22 November 2016 (UTC)

If it's OK with everyone, I removed the ellipsis and readded the full text like "[U+0B0D–U+0B0E]". Feel free to discuss further. We can restore the ellipsis if people want, though I find Meta's argument convincing: I agree, the whole box is geeky. (geeky is also a compliment; many of the good things is life are geeky, but I digress) --Daniel Carrero (talk) 14:30, 23 November 2016 (UTC)

@Giorgi Eufshi: I see you reverted my last edit to the module, with this edit summary: "You also restored the part that I made efforts to converted to a better piece of code."

I love your use of "mw.html.create". I even think the whole module could use it, it has multiple tables that use either HTML syntax or Mediawiki syntax; your way is better. But the only piece of code currently using "mw.html.create" is the one that hides unassigned codepoints and causes ellipses to appear. Assuming that we don't want this to happen (maybe we should ask at the Beer parlour for the opinions of more people?), then I think we can safely revert to the other version without the ellipses and without the better table syntax. Currently, the template is hiding unassigned codepoints and displaying ellipses again, per your last edit. --Daniel Carrero (talk) 14:55, 23 November 2016 (UTC)

@Giorgi Eufshi: I see that we are back to showing the "[U+0B0D–U+0B0E]", this time with "mw.html.create". Looks good to me, thanks. --Daniel Carrero (talk) 12:48, 24 November 2016 (UTC)

@Daniel Carrero, Dixtosa, Giorgi Eufshi: It's looking great, guys; thanks. It even looks good when there are unassigned codepoints on both sides; see, for example, Ό, Ⴭ, and ⴭ. — I.S.M.E.T.A. 17:40, 29 November 2016 (UTC)

Diacritics

It would be nice if combining characters could be displayed over the dotted circle (◌) in the composition display. Currently, they combine with the preceding space, which looks clunky. See, for instance, the character box at Á, where the composition thingy shows the following: A [U+0041] + ́ [U+0301]. It would be neater if it displayed thus: A [U+0041] + ◌́ [U+0301].

Not sure if there's any way to determine which codepoints are combining characters. Probably we would have to create data modules listing all the combining characters in Unicode. — Eru·tuon 03:13, 3 May 2017 (UTC)

@Erutuon:  Done! I mean, the code that generates the dotted circle is done, at least. Check Á again, it should have the dotted circle as you asked. We'll need to fill Module:Unicode data/combining characters with a complete list of combining characters. Let me know if there's any problem with the code.

When possible, as discussed before (link) I'd certainly like to move all the character info to Wikidata so other Wiktionaries and projects can use it, too. But for now it seems we'll still have to use data modules. --Daniel Carrero (talk) 03:46, 3 May 2017 (UTC)

Why don't you use of Module:Unicode data/combining? It already covers combining characters from Unicode data; just check if a char is in there. --Octahedron80 (talk) 03:52, 3 May 2017 (UTC)

@Octahedron80: That module does seem to have a complete list, but how can it be used? I see the function is_combining in Module:Unicode data uses it, but I don't understand the function or what it does. It's rather cryptic. — Eru·tuon 04:01, 3 May 2017 (UTC)

It is not hard to do with Lua. :-) I'll do if you can't. --Octahedron80 (talk) 04:04, 3 May 2017 (UTC)

@Octahedron80: If you want to do it, that would be great! Feel free to adapt our modules and use the existing data module somehow. I don't understand how Module:Unicode data/combining works. I don't know why "[0x000315] = 232," has a "232" in it, and there's no documentation. --Daniel Carrero (talk) 04:04, 3 May 2017 (UTC)

The number 230 etc. is a property for combining characters that is like ordering value used for Unicode normalization. We can just ignore it.--Octahedron80 (talk) 04:10, 3 May 2017 (UTC)

Hey! I got the name Canonical_Combining_Class --Octahedron80 (talk) 05:40, 3 May 2017 (UTC)

Oh, that's nice. Thanks for finding that and for making the module work with the old data module. --Daniel Carrero (talk) 05:44, 3 May 2017 (UTC)

Heh! It was as simple as changing the name of the function. I was trying the same thing when you made your edit, @Octahedron80. Yay! — Eru·tuon 04:47, 3 May 2017 (UTC)

However, let the system clear my error msg for some time...--Octahedron80 (talk) 04:50, 3 May 2017 (UTC)

@Daniel Carrero: Thanks! That gives the desired result! Obviously, the list isn't complete. — Eru·tuon 04:00, 3 May 2017 (UTC)

You're welcome. :) --Daniel Carrero (talk) 04:05, 3 May 2017 (UTC)

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘ @Erutuon, Daniel Carrero, Octahedron80: The presentation of characters' compositions is greatly improved now. Thank you all for your work on this. — I.S.M.E.T.A. 22:11, 11 May 2017 (UTC)

worrisome markup

I suspect that a lot of the CSS styles and nested tables could be eliminated or made more elegant, but the code is spaghetti-ish and I can't tell what's happening. —suzukaze (t・c) 23:50, 3 December 2017 (UTC)

@Suzukaze-c: require("Module:fun").logAll(export) is a way to show the input and output of the exported functions. I can see some empty HTML. — Eru·tuon 00:55, 4 December 2017 (UTC)

worrisomest markup

Why are the table widths hardcoded? (Twice, to 25em / 70px). To force multiple character boxes to have the same width? It makes smaller boxes wider than necessary.– Jberkel 18:29, 3 March 2021 (UTC)

Unseemly uptack image

The character info box for ⊥ has an unseemly svg image; it is much too tall ( instead of ⊥). Why does it use an image in the first place? The character info box for the antonymous character ⊤ does not use an image and is just fine.  --Lambiam 09:50, 19 February 2024 (UTC)

It's for the astrological decile aspect. Might be the two should be split, or kept unified but with the astrological symbol looking like the mathematical symbol.

But yeah, the dict entry doesn't require a display. kwami (talk) 06:41, 23 February 2024 (UTC)