fix paths for wmt18 extra test sets (#143)

* dataset: fix paths for wmt18 test-ts (closes #142)

* bump version to 1.5.1

* pin portalocker to 2.0.0 for 1.5.1 release

CHANGELOG.md

@@ -1,5 +1,8 @@
 # VERSION HISTORY
 
+- 1.5.1 (2021-03-04)
+  - Fix extraction error for WMT18 extra test sets (test-ts) (#142)
+
 - 1.5.0 (2021-01-15)
   - Fix an assertion error in chrF (#121)
   - Add missing `__repr__()` methods for BLEU and TER

sacrebleu/__init__.py

@@ -14,7 +14,7 @@
 # express or implied. See the License for the specific language governing
 # permissions and limitations under the License.
 
-__version__ = '1.5.0'
+__version__ = '1.5.1'
 __description__ = 'Hassle-free computation of shareable, comparable, and reproducible BLEU, chrF, and TER scores'
 
 

sacrebleu/dataset.py

@@ -260,20 +260,20 @@
         'data': ['http://data.statmt.org/wmt18/translation-task/test-ts.tgz'],
         'md5': ['5c621a34d512cc2dd74162ae7d00b320'],
         'description': 'Official evaluation sources with extra test sets interleaved.',
-        'cs-en': ['test/newstest2018-csen-src-ts.cs.sgm'],
-        'de-en': ['test/newstest2018-deen-src-ts.de.sgm'],
-        'en-cs': ['test/newstest2018-encs-src-ts.en.sgm'],
-        'en-de': ['test/newstest2018-ende-src-ts.en.sgm'],
-        'en-et': ['test/newstest2018-enet-src-ts.en.sgm'],
-        'en-fi': ['test/newstest2018-enfi-src-ts.en.sgm'],
-        'en-ru': ['test/newstest2018-enru-src-ts.en.sgm'],
-        'et-en': ['test/newstest2018-eten-src-ts.et.sgm'],
-        'fi-en': ['test/newstest2018-fien-src-ts.fi.sgm'],
-        'ru-en': ['test/newstest2018-ruen-src-ts.ru.sgm'],
-        'en-tr': ['test/newstest2018-entr-src-ts.en.sgm'],
-        'tr-en': ['test/newstest2018-tren-src-ts.tr.sgm'],
-        'en-zh': ['test/newstest2018-enzh-src-ts.en.sgm'],
-        'zh-en': ['test/newstest2018-zhen-src-ts.zh.sgm'],
+        'cs-en': ['test-ts/newstest2018-csen-src-ts.cs.sgm'],
+        'de-en': ['test-ts/newstest2018-deen-src-ts.de.sgm'],
+        'en-cs': ['test-ts/newstest2018-encs-src-ts.en.sgm'],
+        'en-de': ['test-ts/newstest2018-ende-src-ts.en.sgm'],
+        'en-et': ['test-ts/newstest2018-enet-src-ts.en.sgm'],
+        'en-fi': ['test-ts/newstest2018-enfi-src-ts.en.sgm'],
+        'en-ru': ['test-ts/newstest2018-enru-src-ts.en.sgm'],
+        'et-en': ['test-ts/newstest2018-eten-src-ts.et.sgm'],
+        'fi-en': ['test-ts/newstest2018-fien-src-ts.fi.sgm'],
+        'ru-en': ['test-ts/newstest2018-ruen-src-ts.ru.sgm'],
+        'en-tr': ['test-ts/newstest2018-entr-src-ts.en.sgm'],
+        'tr-en': ['test-ts/newstest2018-tren-src-ts.tr.sgm'],
+        'en-zh': ['test-ts/newstest2018-enzh-src-ts.en.sgm'],
+        'zh-en': ['test-ts/newstest2018-zhen-src-ts.zh.sgm'],
     },
     'wmt18/dev': {
         'data': ['http://data.statmt.org/wmt18/translation-task/dev.tgz'],

setup.py

@@ -132,7 +132,10 @@ def get_description():
     # your project is installed. For an analysis of "install_requires" vs pip's
     # requirements files see:
     # https://packaging.python.org/en/latest/requirements.html
-    install_requires = ['typing;python_version<"3.5"', 'portalocker'],
+    install_requires = [
+        'typing;python_version<"3.5"',
+        'portalocker==2.0.0',
+    ],
 
     # List additional groups of dependencies here (e.g. development
     # dependencies). You can install these using the following syntax,