| <<<Back 1 day (to 2020/03/15) | Fwd 1 day (to 2020/03/17) >>> | 20200316 |
zdohnal | chrisl: Hi Chris, regarding the issue (gs stops enumerating fonts from fontconfig if one fails) you fixed last week - it seems the error from fontconfig call is expected - according to our font person, Akira Tagoh, Cantarell-VF font is a variable font, which needs different treatment from fontconfig | 05:27.17 |
| chrisl: FcPatternGetInteger() call for FC_WEIGHT fails correctly, because FC_WEIGHT is not integer, but a range of values | 05:28.46 |
| the same it is for FC_SIZE, FC_WIDTH | 05:29.09 |
| and these variable fonts are recognizable by FC_VARIABLE being set to true | 05:30.09 |
| chrisl: so my question is - does gs suppose to support variable fonts? Or is it aware of its existence? | 05:33.25 |
| chrisl: regardless of previous questions, does gs want to support variable fonts? If it does not, FC_VARIABLE can be set to false in query pattern, so variable fonts will be filtered out | 05:40.11 |
| chrisl: if it does, it will probably need more things to change, but I'm not sure - I was recommended to see pango source code (it seems there is no official howto...) | 05:43.20 |
chrisl | zdohnal: At least for now, we won't be looking to support variable fonts (beyond the already support, but deprecated mulitple-master fonts) | 07:14.15 |
zdohnal | chrisl: ok, so the fix seems simple, but I rather test it | 07:18.53 |
| chrisl: thank you! | 07:19.04 |
chrisl | zdohnal: Given that we're now tolerant of those errors, there's no urgency in fixing the enumeration specification | 07:21.14 |
zdohnal | chrisl: you're right, the fix would be more about performance then urgent issue | 07:23.38 |
su91 | Hello, a quick question, on page https://www.ghostscript.com/doc/9.51/Use.htm, at the end of section 2 there is a table with "Invocation Names". For "MS Windows 95 and later" there is "gswin32c" - shouldn't it be "gswin32c or gswin64c"? | 20:03.27 |
| <<<Back 1 day (to 2020/03/15) | Forward 1 day (to 2020/03/17)>>> | |