Update test_strings.json #135
Conversation
Lib/glyphsets/test_strings.json
Outdated
| "H.Y-Y.T,F.P,V:Y„V-T-A*A’A“A’A-AL–L—L-", | ||
| "n.r.r,y.y,(o)(i)(d)(j)(f)[j]{f}", | ||
| "080 076 474 973 94 7078", | ||
| "¿o,O? w@y «n•n» n*", | ||
| "080 076 474 973 94 7078 54 24 272 679", | ||
| "P4T47A .47,9.7-»4", | ||
| "42°21′29″N 71°03′49″W", |
There was a problem hiding this comment.
Minute and second (2032 and 2033) ar not actually included in the GF Latin Core. We should either add them there or remove them from here. I would advocate for the before (adding them into the glyphset) wdyt?
There was a problem hiding this comment.
I don't know what is best. I would see them in Core, but they can be replaced by ' and " so I don't know how essential they are. Maybe we can add an entry GF_Latin_Plus in there and add test for the different symbols (like the bullets are the same width etc) so it may motivates designer to add all these symbols when they see notdef?
There was a problem hiding this comment.
I've been motivating designers to add minute and second following the reported notdef. But I've been saying "despite they are not required in Latin Core, this would be a good addition etc" Precisely cause they differ from quotation marks and could be really used.
So at least for these two we could add them to Core, or change this string to use the quotes instead, so notdef reported.
And eventually, yes, also add a .json file for Latin Plus which is more related to symbols and others, like Vietnamese
Lib/glyphsets/test_strings.json
Outdated
| "010 080 030 020 050 060 070" | ||
| "010 080 030 020 050 060 070", | ||
| "3+4=7 5>2 2x²−(8×7)÷9≈1", | ||
| "ĿL·L ŀl·l" |
There was a problem hiding this comment.
Similar to minute and second, we don't have Ldot 013F or ldot 0140 included in the Latin Core.
So we should add them.
There was a problem hiding this comment.
yeah I don't know how to do with this one. Monospace font would do with Ldot, while the other should do with localised middle dot
There was a problem hiding this comment.
Well, in that case, we should probably
- Add the localized middle dot to this sample (I saw your attempt to do that in a previous commit, did you undo that because of the monospace fonts?)
- Alternatively, make the
diffbrowsers_proofer.htmlto use the localised glyph if present in the font so we don't receive tofus in QA results. Similar to what it seemsdiffbrowsers_text.htmldoes. We have arebel·lióword in the latter, and it works to shape the correct glyph.
- Alternatively, make the
- Add a FB check to ensure monospace letters include Ldot ones.
There was a problem hiding this comment.
Yep! my attempt failed cause the json didn't seem to like <span lang=cat> </span>. @m4rc1e how can we localise this string to avoid tofu in the proofer when the the Ldot codepoint is missing?
There was a problem hiding this comment.
Unfortunately the current implementation only works as strings. Perhaps I can allow us to add language tags as well?
There was a problem hiding this comment.
may be useful of we extend the proofer to more languages
There was a problem hiding this comment.
I'm updating the test_strings.json file so you have to provide a lang as well for each string. This guide may be useful to understand how it works, https://developer.mozilla.org/en-US/docs/Web/HTML/Global_attributes/lang
m4rc1e
force-pushed
the
Latin_Core_Strings
branch
from
0bf408c to
25e6221
Compare
|
Closing since work is now happening in #146 |
Modified test string template for diff2