mjohnson/espeak-ng
1
0
Fork 0
mirror of https://github.com/rhasspy/espeak-ng.git synced 2026-10-31 05:35:35 +00:00
Code Issues Packages Projects Releases Wiki Activity

tokenizer.c: Move general category classification that does not override property behaviour to the end, for generic classification.

Browse Source
This commit is contained in:
Reece H. Dunn
2017-05-02 18:58:24 +01:00
parent 683579f403
commit 0705844bf8
1 changed files with 10 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/libespeak-ng/tokenizer.c
+10 -5
View File
@@ -117,16 +117,13 @@ static espeakng_CTYPE codepoint_type(uint32_t c)
case 0x0085: return ESPEAKNG_CTYPE_NEWLINE; // NEW LINE (NEL)
}
// 2. Classify codepoints by their Unicode General Category.
// 2. Override property types for codepoints by their Unicode General Category.
ucd_category cat = ucd_lookup_category(c);
switch (cat)
{
case UCD_CATEGORY_Lu: return ESPEAKNG_CTYPE_UPPERCASE;
case UCD_CATEGORY_Ll: return ESPEAKNG_CTYPE_LOWERCASE;
case UCD_CATEGORY_Zl: return ESPEAKNG_CTYPE_NEWLINE;
case UCD_CATEGORY_Zp: return ESPEAKNG_CTYPE_PARAGRAPH;
case UCD_CATEGORY_Zs: return ESPEAKNG_CTYPE_WHITESPACE;
}
// 3. Classify codepoints by their Unicode properties.
@@ -156,7 +153,15 @@ static espeakng_CTYPE codepoint_type(uint32_t c)
return ESPEAKNG_CTYPE_ELLIPSIS;
}
// 4. Classify the remaining codepoints.
// 4. Classify codepoints by their Unicode General Category.
switch (cat)
{
case UCD_CATEGORY_Lu: return ESPEAKNG_CTYPE_UPPERCASE;
case UCD_CATEGORY_Ll: return ESPEAKNG_CTYPE_LOWERCASE;
}
// 5. Classify the remaining codepoints.
return ESPEAKNG_CTYPE_OTHER;
}