Template talk:Unicode blocks

From three to two subtables

To simplify, I have folded the two Supplementary subtables with five planes into one subtable:

Part 1 Basic (BMP): 156 rows. Part 2 Supplementary (SMP...PUA-B): 64. The numbers allow this I think.[1].-DePiep (talk) 10:44, 9 February 2012 (UTC)

Table structure

The splitup, currently into 2 subtables, is not helpful. The table has a structure difficult to get across to readers (even experienced; I've read cmts over the times that this one is not easy to get). The splitup was done to reduce massiveness [2] ES: Table too long split bmp/smp/ssp. (this edit created three subtables, not two left)-DePiep (talk) 10:47, 9 February 2012 (UTC)

To make the table into a simple plain structure, I suggest to change it into one table, all 224 rows into one. I'd like to hear if that would be "too long" (the split up motivation), especially compared with the simplicity of overview. -DePiep (talk) 10:47, 9 February 2012 (UTC)

I can add that when the table is split, searching can be misleading for the eye (forgetting to open the second table) or automated on the page (no hits when a table is folded). While I am familiar with this template, this happened to me buggingly often. -DePiep (talk) 13:47, 9 February 2012 (UTC)

I'm OK with making it a single table, but there needs to be a way of indicating what plane each block belongs to ... new column on the left with abbrevaiated plane name ("BMP", "SMP", etc.) ? Also suggest removing the "Notes" column and put the C0/C1 notes in a footnote. BabelStone (talk) 18:38, 9 February 2012 (UTC)

re your edits:

- "No_Block" is defined by Unicode. It is used in "Block=No_Block" for each code point not in a block. @missing: 0000..10FFFF; No_Block in datafile Blocks.txt. So, a code point can have that value.

- es: "These are ISO/IEC 10646 terms not former Unicode names" re the first two blocks. "former" names indeed is wrong: The name is still used in Unicode [3] [4]. In general Aliases may very well be mentioned. On some level "Latin 1" is also used for "Latin-1 Supplement" (which is not insignificant because ISO/IEC 8859-1 is also known as "Latin 1", but that comprises the range 0000-00FF (256 points) i.e. not what Unicode aliases Latin_1). There are more aliases and names we might add to describe things, but those re C0, C1, Latin_1 are most confusingly. It does not help that Unicode keeps using all variants and more. -DePiep (talk) 21:25, 9 February 2012 (UTC)

 Back into one single table. Alternative names (including Unicode Aliases) could be re-added. -DePiep (talk) 20:06, 17 February 2012 (UTC)

Can we make collapsing an optional parameter?

Can we have a parameter that specifies whether the table should be collapsed or not, passed through when transcluding the template? This way the template could be shown uncollapsed in the Unicode blocks article where it makes no sense to have it collapsed, but shown collapsed in the Unicode character property page where readers may not want to see all the blocks. Same for Template:ISO 15924 script codes and Unicode. BabelStone (talk) 18:46, 9 February 2012 (UTC)

Yes. If it is a single table, we can do this: class="wikitable sortable collapsible {{{state|autocollapse}}}". This way it is uncollapsed when it is the first table on a page, collpsed when it is second or later table. And an editor can write in the article e.g. {{Unicode blocks| state=uncollapsed}} to overrule the autocollpase-mechanism and set any state (uncollapsed or collapsed) forcibly. This does not work with current stacked template situation (the header is always a first template). -DePiep (talk) 20:36, 9 February 2012 (UTC)

 as explained Default: state=autocollapse. Option: use {{Unicode blocks|state=uncollapsed}} to force a state on a page. -DePiep (talk) 20:08, 17 February 2012 (UTC)

Adding information about future versions of Unicode in advance as comments

I reverted commented out additions of blocks from the Unicode pipeline made by Vanisaac as I believe that adding in partial information about the version after next of Unicode is more likely to lead to confusion and mistakes than to be helpful. The added blocks included some that will certainly be in 7.0 but also some such as Tangut and CJK-E that certainly won't be as they are not far enough along the ISO process, and they may be mistakenly uncommented out when 7.0 is released. Also the commented out blocks included character counts which may well not be the same when the blocks are finally encoded, and if an editor simply uncomments the blocks when 7.0 is released then we will have incorrect counts. All in all I believe that providing inaccurate and incomplete information liable to change far in advance is unhelpful, and it is safest to wait until 7.0 is released before updating this and other Unicode-related pages. BabelStone (talk) 11:20, 3 February 2013 (UTC)

Grouping

@Drmccreedy: Why you has reverted my edit? Why doing away with the groupings of ten is bad? 217.117.125.72 (talk) 17:50, 29 September 2020 (UTC)

I reverted your edit because it changed the longstanding table format of groupings of ten without discussing it first. I didn't come up with the groupings but that format has the advantage of showing the plane while scrolling. There are so many blocks in a few planes, like BMP, that without the groupings you don't see the plane acronym until you scroll down several screens. With the groupings of ten you always see it. If there's a consensus to do away with the groupings, the acronyms should at least be aligned to the top of the cell so they're seen without initially scrolling. But I don't know if there is a consensus to change the format. DRMcCreedy (talk) 18:06, 29 September 2020 (UTC)

Update, use row-formatting template

I have edited the table to use new {{Unicode blocks/row}}. Basically, the row formats block-input entered in the table. refining of adding blocks can be done in the main table page (for regular content editing, no need to change the /row tempalte).

Basically: input is like

{{Unicode blocks/row |plane=SMP |planelink=yes
 |range=U+10000..U+1007F 
 |name=[[Linear B Syllabary]] 
 |cp=128 
 |char=88 
 |contains=[[Linear B]]
}}

The /row template produces the table row.

Details

1. |plane=SMP sets and formats the plane name, standardised. Defaults to |plane=BMP. |planelink=yes wikilinks to the plane article section (and adds anchor so one can use § SMP into the table). This is to be used at first appearance of the plane.
2. I have added the plane to each row. As discussion above decribed, the plane name sould be visible when scrolling 100+ rows. Otoh, when grouping by 10 or so, that suggests visually that they form some subgroup (they do not). Hope it is acceptable.
3. Numbers are right-aligned.
4. Sortorder in range-column is numerical (decimal) now.
5. Aim is that formatting is more easy to set, and content-editing (block data) has more overview. -DePiep (talk) 20:22, 17 February 2021 (UTC)

Adjusted plane bg colors to improve contrast with fontcolor (w3c). See /testcases. -DePiep (talk) 01:45, 18 February 2021 (UTC)