10BC0 Merge pull request #56 from rth/unicode-10 · unicode-rs/unicode-segmentation@0e09a6b · GitHub
[go: up one dir, main page]
Skip to content

Commit 0e09a6b

Browse files
ManishearthManishearth
authored
Merge pull request #56 from rth/unicode-10
Upgrade to Unicode 10
2 parents e2e111e + e765bea commit 0e09a6b

File tree

4 files changed

+1276
-1216
lines changed

4 files changed

+1276
-1216
lines changed

‎scripts/unicode.py‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -59,7 +59,7 @@ def is_surrogate(n):
5959

6060
def fetch(f):
6161
if not os.path.exists(os.path.basename(f)):
62-
os.system("curl -O http://www.unicode.org/Public/9.0.0/ucd/%s"
62+
os.system("curl -O http://www.unicode.org/Public/10.0.0/ucd/%s"
6363
% f)
6464

6565
if not os.path.exists(os.path.basename(f)):
@@ -357,4 +357,4 @@ def emit_break_module(f, break_table, break_cats, name):
357357
for cat in sentence_cats:
358358
sentence_table.extend([(x, y, cat) for (x, y) in sentence_cats[cat]])
359359
sentence_table.sort(key=lambda w: w[0])
360-
emit_break_module(rf, sentence_table, list(sentence_cats.keys()), "sentence")
360+
emit_break_module(rf, sentence_table, list(sentence_cats.keys()), "sentence")

‎scripts/unicode_gen_breaktests.py‎

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -172,7 +172,7 @@ def create_grapheme_data(f):
172172
stype = "&'static [(&'static str, &'static [&'static str])]"
173173
dtype = "&'static [(&'static str, &'static [&'static str], &'static [&'static str])]"
174174
f.write(" // official Unicode test data\n")
175-
f.write(" // http://www.unicode.org/Public/9.0.0/ucd/auxiliary/GraphemeBreakTest.txt\n")
175+
f.write(" // http://www.unicode.org/Public/10.0.0/ucd/auxiliary/GraphemeBreakTest.txt\n")
176176
unicode.emit_table(f, "TEST_SAME", test_same, stype, True, showfun, True)
177177
unicode.emit_table(f, "TEST_DIFF", test_diff, dtype, True, showfun, True)
178178

@@ -187,7 +187,7 @@ def create_words_data(f):
187187

188188
wtype = "&'static [(&'static str, &'static [&'static str])]"
189189
f.write(" // official Unicode test data\n")
190-
f.write(" // http://www.unicode.org/Public/9.0.0/ucd/auxiliary/WordBreakTest.txt\n")
190+
f.write(" // http://www.unicode.org/Public/10.0.0/ucd/auxiliary/WordBreakTest.txt\n")
191191
unicode.emit_table(f, "TEST_WORD", test, wtype, True, showfun, True)
192192

193193
def create_sentence_data(f):
@@ -201,7 +201,7 @@ def create_sentence_data(f):
201201

202202
wtype = "&'static [(&'static str, &'static [&'static str])]"
203203
f.write(" // official Unicode test data\n")
204-
f.write(" // http://www.unicode.org/Public/UNIDATA/auxiliary/SentenceBreakTest.txt\n")
204+
f.write(" // http://www.unicode.org/Public/10.0.0/ucd/auxiliary/SentenceBreakTest.txt\n")
205205
unicode.emit_table(f, "TEST_SENTENCE", test, wtype, True, showfun, True)
206206

207207
if __name__ == "__main__":

0 commit comments

Comments
 (0)
0