Hotfix id sunday (#1134)

* Add new rule for 'week' and remove existing 'week' translation from Indonesia (id) * Update id * Simplify extension of “week ago” support for Indonesian * Indonesian: interpret minggu as Sunday if not following a number --------- Co-authored-by: Adrián Chaves <[email protected]>
scrapinghub · Jan 31, 2023 · 350cafa · 350cafa
1 parent bedc881
commit 350cafa
Show file tree

Hide file tree

Showing 5 changed files with 20 additions and 6 deletions.
diff --git a/dateparser/data/date_translation_data/id.py b/dateparser/data/date_translation_data/id.py
@@ -202,7 +202,8 @@
         ],
         "\\1 week ago": [
             "(\\d+[.,]?\\d*) mgg lalu",
-            "(\\d+[.,]?\\d*) minggu yang lalu"
+            "(\\d+[.,]?\\d*) minggu yang lalu",
+            "(\\d+[.,]?\\d*) minggu lalu"
         ],
         "\\1 year ago": [
             "(\\d+[.,]?\\d*) tahun yang lalu",
@@ -235,6 +236,12 @@
         "in \\1 year": [
             "dalam (\\d+[.,]?\\d*) tahun",
             "dlm (\\d+[.,]?\\d*) thn"
+        ],
+        "\\1 week": [
+            "(\\d+[.,]?\\d*) minggu"
+        ],
+        "sunday": [
+            "minggu"
         ]
     },
     "locale_specific": {},

diff --git a/dateparser/languages/locale.py b/dateparser/languages/locale.py
@@ -134,6 +134,7 @@ def translate(self, date_string, keep_formatting=False, settings=None):
             for pattern, replacement in relative_translations.items():
                 if pattern.match(word):
                     date_string_tokens[i] = pattern.sub(replacement, word)
+                    break
             else:
                 if word in dictionary:
                     fallback = word if keep_formatting and not word.isalpha() else ''

diff --git a/dateparser_data/supplementary_language_data/date_translation_data/id.yaml b/dateparser_data/supplementary_language_data/date_translation_data/id.yaml
@@ -30,8 +30,13 @@ relative-type:
         - baru saja
 
 relative-type-regex:
+    \1 week ago:
+        - (\d+[.,]?\d*) minggu lalu
     \1 month ago:
         - (\d+[.,]?\d*) bulan lalu
-
     \1 year ago:
         - (\d+[.,]?\d*) tahun lalu
+    \1 week:
+        - (\d+[.,]?\d*) minggu
+    sunday:
+        - minggu
diff --git a/tests/test_data.py b/tests/test_data.py
@@ -62,10 +62,7 @@ def is_invalid_relative_regex_mapping(relative_regex_mapping):
     key, value = relative_regex_mapping
     if not (key and value and isinstance(key, str) and isinstance(value, list)):
         return True
-    if '\\1' not in key:
-        return True
-    return not (all([isinstance(x, str) for x in value])
-                and all(['(\\d+[.,]?\\d*)' in x for x in value]))
+    return not all([isinstance(x, str) for x in value])
 
 
 class TestLocaleInfo(BaseTestCase):

diff --git a/tests/test_languages.py b/tests/test_languages.py
@@ -131,9 +131,13 @@ def setUp(self):
         # Tagalog
         param('tl', "Biyernes Hulyo 3, 2015", "friday july 3 2015"),
         param('tl', "Pebrero 5, 2015 7:00 pm", "february 5 2015 7:00 pm"),
+
         # Indonesian
         param('id', "06 Sep 2015", "06 september 2015"),
         param('id', "07 Feb 2015 20:15", "07 february 2015 20:15"),
+        param('id', "Minggu, 18 Mar 2018 07:30", "sunday 18 march 2018 07:30"),
+        param('id', "3 minggu yang lalu", "3 week ago"),
+        param('id', "5 minggu", "5 week"),
 
         # Miscellaneous
         param('en', "2014-12-12T12:33:39-08:00", "2014-12-12 12:33:39-08:00"),