Fix #4532: language sensitivity of autocap #10119
Conversation
This includes the text suggestions found in whatwg#4532, which are editorial and explain that autocapitalization is language or script sensitive.
| @@ -81064,7 +81064,8 @@ body { display:none } | |||
|
|
|||
| <h4>Autocapitalization</h4> | |||
|
|
|||
| <p>Some methods of entering text, for example virtual keyboards on mobile devices, and also voice | |||
| <p>A few languages and writing systems distinguish between upper, lower, and titlecase letters. | |||
There was a problem hiding this comment.
"titlecase letter" needs a reference. Assuming this is about the Unicode code point category.
There was a problem hiding this comment.
It isn't necessarily about the code point category. Titlecasing is more complicated than just the explicit mappings, since some languages use digraphs and titlecase them in different ways (compare Serbian dz and Dutch ij). Anyway, happy to add a reference. Where should it point? Unicode glossary?
There was a problem hiding this comment.
This seems like a reasonable reference, but it needs to be introduced above as part of the Unicode and Encoding section. It seems we currently don't reference any terms from Unicode, but you can use introductory wording similar to what we do for Encoding. Let me know if you need help.
There was a problem hiding this comment.
Thanks. Part of me is twitching because we generally use the i18n-glossary as a proxy for Unicode terminology. We don't have titlecase defined there either, at the moment, but could do, especially as we have extensive illustrations of the problems being discussed here in charmod-norm.
In the interim 52d6dd8 includes a link to the Unicode glossary. I will look at the Unicode and Encoding section in HTML and make an additional push after that.
There was a problem hiding this comment.
i18n-glossary has reached production (TR) with 'titlecase' exported.
I looked at the "Unicode and Encoding" section, but that seems to be where terms exported from Encoding are brought in. Encoding has no reason to define upper/lower/titlecase, so I feel like this doesn't belong there.
I see that there is a section about Infra (and a list of imported terms) there (and this list naturally includes the ASCII casing terms (ASCII lowercase, ASCII uppercase, ASCII case-insensitive) that are core to HTML. Unicode case handling isn't there either, but could be incorporated.
Given that this is a "fleeting reference", can we keep it local? Failing that, we could link the I18N glossary or we could link Character Model: String Matching which explains case folding.
There was a problem hiding this comment.
It's also where in theory we could bring in references from Unicode (it's both Encoding and Unicode after all). Would need an equivalent paragraph and list.
I'm not entirely sure how I feel about referencing the i18n glossary or the Character Model as those documents feel like they're summarizing more authoritative sources and I'd rather we reference the latter.
source
Outdated
| validation.)</p> | ||
| validation.) Virtual keyboards or other input systems normally interpret this attribute in a | ||
| script or language-specific manner. In some languages the hints are ignored or can receive | ||
| special treatment.</p> |
There was a problem hiding this comment.
We already talk about how it works when implemented for virtual keyboards above: "A virtual keyboard that implements". If anything we should modify that.
There was a problem hiding this comment.
Will do. Watch for revision.
|
PS> I signed the Participation agreement as an individual, but it has been pending for a while. Is there a manual process behind that? |
|
You're manually verified the moment your first contribution is accepted for merging. |
This includes the text suggestions found in #4532, which are editorial and explain that autocapitalization is language or script sensitive.
(See WHATWG Working Mode: Changes for more details.)
/interaction.html ( diff )