summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
-rwxr-xr-xbenchsuite/benchsuite18
1 files changed, 10 insertions, 8 deletions
diff --git a/benchsuite/benchsuite b/benchsuite/benchsuite
index b849b454..9353cf49 100755
--- a/benchsuite/benchsuite
+++ b/benchsuite/benchsuite
@@ -23,13 +23,15 @@ import time
# strategies used to increase the relevance of results returned.
SUBTITLES_DIR = 'subtitles'
-SUBTITLES_EN_NAME = 'OpenSubtitles2016.raw.en'
-SUBTITLES_EN_NAME_SAMPLE = 'OpenSubtitles2016.raw.sample.en'
+SUBTITLES_EN_NAME = 'en.txt'
+SUBTITLES_EN_NAME_SAMPLE = 'en.sample.txt'
SUBTITLES_EN_NAME_GZ = '%s.gz' % SUBTITLES_EN_NAME
-SUBTITLES_EN_URL = 'http://opus.lingfil.uu.se/OpenSubtitles2016/mono/OpenSubtitles2016.raw.en.gz' # noqa
-SUBTITLES_RU_NAME = 'OpenSubtitles2016.raw.ru'
+# SUBTITLES_EN_URL = 'http://opus.lingfil.uu.se/OpenSubtitles2016/mono/OpenSubtitles2016.raw.en.gz' # noqa
+SUBTITLES_EN_URL = 'https://object.pouta.csc.fi/OPUS-OpenSubtitles/v2016/mono/en.txt.gz' # noqa
+SUBTITLES_RU_NAME = 'ru.txt'
SUBTITLES_RU_NAME_GZ = '%s.gz' % SUBTITLES_RU_NAME
-SUBTITLES_RU_URL = 'http://opus.lingfil.uu.se/OpenSubtitles2016/mono/OpenSubtitles2016.raw.ru.gz' # noqa
+# SUBTITLES_RU_URL = 'http://opus.lingfil.uu.se/OpenSubtitles2016/mono/OpenSubtitles2016.raw.ru.gz' # noqa
+SUBTITLES_RU_URL = 'https://object.pouta.csc.fi/OPUS-OpenSubtitles/v2016/mono/ru.txt.gz' # noqa
LINUX_DIR = 'linux'
LINUX_CLONE = 'git://github.com/BurntSushi/linux'
@@ -255,11 +257,11 @@ def bench_linux_unicode_greek_casei(suite_dir):
def bench_linux_unicode_word(suite_dir):
'''
- Benchmark Unicode aware \w character class.
+ Benchmark Unicode aware \\w character class.
Only ripgrep and git-grep (with LC_ALL=en_US.UTF-8) actually get
this right. Everything else uses the standard ASCII interpretation
- of \w.
+ of \\w.
'''
require(suite_dir, 'linux')
cwd = path.join(suite_dir, LINUX_DIR)
@@ -1088,7 +1090,7 @@ def download_subtitles_en(suite_dir):
# benchmarks finish in a reasonable time.
with open(path.join(subtitle_dir, en_path_sample), 'wb+') as f:
run_cmd(
- ['head', '-n', '32722372', en_path],
+ ['head', '-n', '55000000', en_path],
cwd=subtitle_dir, stdout=f)