Fix an issue detected by OSSFuzz (#1203)

Author: ennamarie19

.github/workflows/cifuzz.yml

@@ -0,0 +1,35 @@
+name: CIFuzz
+on: [pull_request]
+permissions: {}
+jobs:
+  Fuzzing:
+    runs-on: ubuntu-latest
+    permissions:
+      security-events: write
+    steps:
+    - name: Build Fuzzers
+      id: build
+      uses: google/oss-fuzz/infra/cifuzz/actions/build_fuzzers@master
+      with:
+        oss-fuzz-project-name: 'dateparser'
+        language: python
+    - name: Run Fuzzers
+      uses: google/oss-fuzz/infra/cifuzz/actions/run_fuzzers@master
+      with:
+        oss-fuzz-project-name: 'dateparser'
+        language: python
+        fuzz-seconds: 600
+        output-sarif: true
+    - name: Upload Crash
+      uses: actions/upload-artifact@v3
+      if: failure() && steps.build.outcome == 'success'
+      with:
+        name: artifacts
+        path: ./out/artifacts
+    - name: Upload Sarif
+      if: always() && steps.build.outcome == 'success'
+      uses: github/codeql-action/upload-sarif@v2
+      with:
+        # Path to SARIF file relative to the root of the repository
+        sarif_file: cifuzz-sarif/results.sarif
+        checkout_path: cifuzz-sarif

dateparser/parser.py

@@ -564,7 +564,7 @@ def _correct_for_time_frame(self, dateobj, tz):
             try:
                 tz = tz or get_timezone_from_tz_string(self.settings.TIMEZONE)
                 tz_offset = tz.utcoffset(dateobj)
-            except pytz.UnknownTimeZoneError:
+            except (pytz.UnknownTimeZoneError, pytz.NonExistentTimeError):
                 tz_offset = timedelta(hours=0)
 
             if "past" in self.settings.PREFER_DATES_FROM:

tests/test_clean_api.py

@@ -111,6 +111,69 @@ def test_dates_which_match_locales_are_parsed(
         self.when_date_is_parsed(date_string, locales=locales)
         self.then_parsed_date_is(expected_date)
 
+    @parameterized.expand(
+        [
+            param(
+                date_string="0:4",
+                locales=["fr-PF"],
+                languages=["en"],
+                region="",
+                date_formats=["%a", "%a", "%a", "%a"],
+                expected_date=datetime(1969, 12, 31, 14, 4),
+            )
+        ]
+    )
+    def test_dates_parse_utc_offset_does_not_throw(
+        self, date_string, locales, languages, region, date_formats, expected_date
+    ):
+        """
+        Bug discovered by OSSFuzz that caused an exception in pytz to halt parsing
+        Regression test to ensure that this is not reintroduced
+        """
+        self.when_date_is_parsed_with_args_and_settings(
+            date_string,
+            languages=languages,
+            locales=locales,
+            region=region,
+            date_formats=date_formats,
+            settings={
+                "CACHE_SIZE_LIMIT": 1000,
+                "DATE_ORDER": "YDM",
+                "DEFAULT_LANGUAGES": [
+                    "mzn",
+                    "as",
+                    "af",
+                    "fur",
+                    "sr-Cyrl",
+                    "kw",
+                    "ne",
+                    "en",
+                    "vi",
+                    "teo",
+                    "sr",
+                    "cgg",
+                ],
+                "LANGUAGE_DETECTION_CONFIDENCE_THRESHOLD": 0.18823535008398845,
+                "NORMALIZE": True,
+                "PARSERS": ["custom-formats", "absolute-time"],
+                "PREFER_DATES_FROM": "past",
+                "PREFER_DAY_OF_MONTH": "first",
+                "PREFER_LOCALE_DATE_ORDER": True,
+                "PREFER_MONTH_OF_YEAR": "current",
+                "RELATIVE_BASE": datetime(
+                    year=1970, month=1, day=1, hour=0, minute=0, second=0
+                ),
+                "REQUIRE_PARTS": [],
+                "RETURN_AS_TIMEZONE_AWARE": False,
+                "RETURN_TIME_AS_PERIOD": False,
+                "SKIP_TOKENS": [],
+                "STRICT_PARSING": False,
+                "TIMEZONE": "America/Hermosillo",
+                "TO_TIMEZONE": "Asia/Almaty",
+            },
+        )
+        self.then_parsed_date_and_time_is(expected_date)
+
     @parameterized.expand(
         [
             param(date_string="January 24, 2014", locales=["pt-AO"]),
@@ -133,6 +196,24 @@ def when_date_is_parsed(self, date_string, languages=None, locales=None):
     def when_date_is_parsed_with_settings(self, date_string, settings=None):
         self.result = dateparser.parse(date_string, settings=settings)
 
+    def when_date_is_parsed_with_args_and_settings(
+        self,
+        date_string,
+        languages=None,
+        locales=None,
+        region=None,
+        date_formats=None,
+        settings=None,
+    ):
+        self.result = dateparser.parse(
+            date_string,
+            languages=languages,
+            locales=locales,
+            region=region,
+            date_formats=date_formats,
+            settings=settings,
+        )
+
     def then_parsed_date_is(self, expected_date):
         self.assertEqual(
             self.result, datetime.combine(expected_date, datetime.min.time())