[WIP] Route Tibetan through USE #933
Conversation
src/gen-use-table.py
Outdated
| @@ -338,6 +338,9 @@ def map_to_use(data): | |||
| if 0x1CF2 <= U <= 0x1CF3: UIPC = Right | |||
| if 0x1CF8 <= U <= 0x1CF9: UIPC = Top | |||
|
|
|||
| if not (UIPC in [Not_Applicable, Visual_Order_Left] or USE in use_positions): | |||
| print ("%s %s %s %s %s" % (hex(U), UIPC, USE, UISC, UGC), file=sys.stderr) | |||
| continue | |||
There was a problem hiding this comment.
I should figure out what is going on to know why this is failing, which of course has some reason.
There was a problem hiding this comment.
0xf18 Bottom O Other Mn
0xf19 Bottom O Other Mn
0xf3e Right O Other Mc
0xf3f Left O Other Mc
0xf86 Top O Other Mn
0xf87 Top O Other Mn
It’s failing because Unicode assigns each of those characters an Indic_Positional_Category but not an Indic_Syllabic_Category.
There was a problem hiding this comment.
And any hint what should be done here also? :)
There was a problem hiding this comment.
Set UISC for these characters after line 297.
|
This breaks clusters with multiple vowels like ⟨ཀིུ⟩: it needs the USE’s Tibetan overrides. |
src/gen-use-table.py
Outdated
| @@ -293,6 +293,9 @@ def map_to_use(data): | |||
|
|
|||
| # Resolve Indic_Syllabic_Category | |||
|
|
|||
| # TODO: There are Tibetan signs that don't have UISC | |||
| if U in [0xF18, 0xF19, 0xF3E, 0xF3F, 0xF86, 0xf87]: UISC = Cantillation_Mark | |||
There was a problem hiding this comment.
Always minimum four digits for Unicode hex values. Also, last one has lowercase 'f' while others are uppercase.
There was a problem hiding this comment.
Sure, can you help on finding correct syllable category for those also? They were just signs
There was a problem hiding this comment.
At least for the U+0F3F to be moved to left it has to be marked as a matra or medial consonant. If it's neither, it won't be moved.
There was a problem hiding this comment.
@PeterCon Peter, can you double check the overrides for Tibetan against the list posted on the spec page? Better yet, can you refresh the override list for all scripts?
src/gen-use-table.py
Outdated
| @@ -338,9 +341,6 @@ def map_to_use(data): | |||
| if 0x1CF2 <= U <= 0x1CF3: UIPC = Right | |||
| if 0x1CF8 <= U <= 0x1CF9: UIPC = Top | |||
|
|
|||
| if not (UIPC in [Not_Applicable, Visual_Order_Left] or USE in use_positions): | |||
There was a problem hiding this comment.
Oh that was just added, I probably should squash the changes.
There was a problem hiding this comment.
Yeah, looked at the squashed diff and looks good.
src/hb-ot-shape-complex-private.hh
Outdated
| @@ -233,7 +232,7 @@ hb_ot_shape_complex_categorize (const hb_ot_shape_planner_t *planner) | |||
| /* Unicode-2.0 additions */ | |||
| case HB_SCRIPT_TIBETAN: | |||
There was a problem hiding this comment.
This case should be removed and other commented-out TIBETAN uncommented.
|
Also, did you test the reported bug at all? |
Yep, and this by its own wasn't fixing it IIRC. I was considering this as just a starting point... |
|
No response to what to do with the signs and I guess I can't reach to this soon so probably better to go lights off for this. |
|
It's still a valid issue and we want to fix it. Please don't close valid issue and PRs just because of inactivity. |
|
Well, I though I can't reach to them soon and better to focus on some limited ones, but if you want to, sure :) |
Hopefully to fix #861