Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-09 03:04:10 -04:00
Code Issues Packages Projects Releases Wiki Activity

Test tokenization with different UI langauges

Browse Source
This commit is contained in:
Kovid Goyal 2021-06-19 15:16:48 +05:30
parent 6f7766fbf4
commit 755b58d1f5
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 9 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
src/calibre/db/tests/fts.py
View File

@ -69,6 +69,8 @@ class FTSTest(BaseTest):
set_ui_language('en') set_ui_language('en')
def test_fts_tokenize(self): # {{{ def test_fts_tokenize(self): # {{{
from calibre_extensions.sqlite_extension import set_ui_language
def t(x, s, e, f=0): def t(x, s, e, f=0):
return {'text': x, 'start': s, 'end': e, 'flags': f} return {'text': x, 'start': s, 'end': e, 'flags': f}
@ -104,6 +106,13 @@ class FTSTest(BaseTest):
tt('你叫什么名字', '你', '叫', '什么', '名字') tt('你叫什么名字', '你', '叫', '什么', '名字')
tt('你叫abc', '你', '叫', 'abc') tt('你叫abc', '你', '叫', 'abc')
tt('a你b叫什么名字', 'a', '你', 'b', '叫', '什么', '名字') tt('a你b叫什么名字', 'a', '你', 'b', '叫', '什么', '名字')
for lang in 'de fr es sv it en'.split():
set_ui_language(lang)
tt("don't 'its' wörds", "don't", 'its', 'wörds', 'words')
tt("l'hospital", "l'hospital")
tt("x'bug'", "x'bug")
set_ui_language('en')
# }}} # }}}
def test_fts_basic(self): # {{{ def test_fts_basic(self): # {{{