Trying to fix Segmentation fault caused by mediainfo in docker container. #2098v1.2.1-beta.9

14 files changed, 146 insertions, 92 deletions

diff --git a/libs/knowit/__init__.py b/libs/knowit/__init__.py
index eda706779..bf225e195 100644
--- a/libs/knowit/__init__.py
+++ b/libs/knowit/__init__.py
@@ -1,10 +1,9 @@
 """Know your media files better."""
 __title__ = 'knowit'
-__version__ = '0.4.0'
-__short_version__ = '.'.join(__version__.split('.')[:2])
+__version__ = '0.5.2'
+__short_version__ = '0.5'
 __author__ = 'Rato AQ2'
 __license__ = 'MIT'
-__copyright__ = 'Copyright 2016-2021, Rato AQ2'
 __url__ = 'https://github.com/ratoaq2/knowit'
 
 #: Video extensions
diff --git a/libs/knowit/__main__.py b/libs/knowit/__main__.py
index c30148421..d9255ffd4 100644
--- a/libs/knowit/__main__.py
+++ b/libs/knowit/__main__.py
@@ -169,7 +169,7 @@ def dumps(
     return convert(info, context)
 
 
-def main(args: typing.List[str] = None) -> None:
+def main(args: typing.Optional[typing.List[str]] = None) -> None:
     """Execute main function for entry point."""
     argument_parser = build_argument_parser()
     args = args or sys.argv[1:]
diff --git a/libs/knowit/api.py b/libs/knowit/api.py
index 4df780605..c6ebd3bd6 100644
--- a/libs/knowit/api.py
+++ b/libs/knowit/api.py
@@ -65,7 +65,7 @@ def know(
         raise KnowitException(debug_info(context=context, exc_info=True))
 
 
-def dependencies(context: typing.Mapping = None) -> typing.Mapping:
+def dependencies(context: typing.Optional[typing.Mapping] = None) -> typing.Mapping:
     """Return all dependencies detected by knowit."""
     deps = {}
     try:
diff --git a/libs/knowit/core.py b/libs/knowit/core.py
index 9736d7ba2..ede307dad 100644
--- a/libs/knowit/core.py
+++ b/libs/knowit/core.py
@@ -63,6 +63,17 @@ class Property(Reportable[T]):
         # Used to detect duplicated values. e.g.: en / en or [email protected] / [email protected] or Progressive / Progressive
         self.delimiter = delimiter
 
+    @classmethod
+    def _extract_value(cls,
+                       track: typing.Mapping,
+                       name: str,
+                       names: typing.List[str]):
+        if len(names) == 2:
+            parent_value = track.get(names[0], track.get(names[0].upper(), {}))
+            return parent_value.get(names[1], parent_value.get(names[1].upper()))
+
+        return track.get(name, track.get(name.upper()))
+
     def extract_value(
             self,
             track: typing.Mapping,
@@ -71,7 +82,7 @@ class Property(Reportable[T]):
         """Extract the property value from a given track."""
         for name in self.names:
             names = name.split('.')
-            value = track.get(names[0], {}).get(names[1]) if len(names) == 2 else track.get(name)
+            value = self._extract_value(track, name, names)
             if value is None:
                 if self.default is None:
                     continue
@@ -216,9 +227,10 @@ class MultiValue(Property):
 class Rule(Reportable[T]):
     """Rule abstract class."""
 
-    def __init__(self, name: str, override=False, **kwargs):
+    def __init__(self, name: str, private=False, override=False, **kwargs):
         """Initialize the object."""
         super().__init__(name, **kwargs)
+        self.private = private
         self.override = override
 
     def execute(self, props, pv_props, context: typing.Mapping):
diff --git a/libs/knowit/defaults.yml b/libs/knowit/defaults.yml
index 9dd7e46db..af6b79c32 100644
--- a/libs/knowit/defaults.yml
+++ b/libs/knowit/defaults.yml
@@ -455,46 +455,46 @@ profiles:
 
   VideoProfileLevel:
     L1:
-      default: "1"
+      default: '1'
       technical: Level 1
     L11:
-      default: "1.1"
+      default: '1.1'
       technical: Level 1.1
     L13:
-      default: "1.3"
+      default: '1.3'
       technical: Level 1.3
     L2:
-      default: "2"
+      default: '2'
       technical: Level 2
     L21:
-      default: "2.1"
+      default: '2.1'
       technical: Level 2.1
     L22:
-      default: "2.2"
+      default: '2.2'
       technical: Level 2.2
     L3:
-      default: "3"
+      default: '3'
       technical: Level 3
     L31:
-      default: "3.1"
+      default: '3.1'
       technical: Level 3.1
     L32:
-      default: "3.2"
+      default: '3.2'
       technical: Level 3.2
     L4:
-      default: "4"
+      default: '4'
       technical: Level 4
     L41:
-      default: "4.1"
+      default: '4.1'
       technical: Level 4.1
     L42:
-      default: "4.2"
+      default: '4.2'
       technical: Level 4.2
     L5:
-      default: "5"
+      default: '5'
       technical: Level 5
     L51:
-      default: "5.1"
+      default: '5.1'
       technical: Level 5.1
     LOW:
       default: Low
diff --git a/libs/knowit/properties/video.py b/libs/knowit/properties/video.py
index e1b293d01..60c5b8264 100644
--- a/libs/knowit/properties/video.py
+++ b/libs/knowit/properties/video.py
@@ -106,11 +106,12 @@ class Ratio(Property[Decimal]):
             if (width, height) == ('0', '1'):  # identity
                 return Decimal('1.0')
 
-            result = round_decimal(Decimal(width) / Decimal(height), min_digits=1, max_digits=3)
-            if self.unit:
-                result *= self.unit
+            if height:
+                result = round_decimal(Decimal(width) / Decimal(height), min_digits=1, max_digits=3)
+                if self.unit:
+                    result *= self.unit
 
-            return result
+                return result
 
         self.report(value, context)
         return None
diff --git a/libs/knowit/provider.py b/libs/knowit/provider.py
index f8c29f5f3..5306d8388 100644..100755
--- a/libs/knowit/provider.py
+++ b/libs/knowit/provider.py
@@ -103,10 +103,7 @@ class Provider:
 
             value = prop.extract_value(track, context)
             if value is not None:
-                if not prop.private:
-                    which = props
-                else:
-                    which = pv_props
+                which = props if not prop.private else pv_props
                 which[name] = value
 
         for name, rule in self.rules.get(track_type, {}).items():
@@ -116,8 +113,9 @@ class Provider:
 
             value = rule.execute(props, pv_props, context)
             if value is not None:
-                props[name] = value
-            elif name in props and not rule.override:
+                which = props if not rule.private else pv_props
+                which[name] = value
+            elif name in props and (not rule.override or props[name] is None):
                 del props[name]
 
         return props
diff --git a/libs/knowit/providers/enzyme.py b/libs/knowit/providers/enzyme.py
index 5dd3d8cef..6a06599d4 100644
--- a/libs/knowit/providers/enzyme.py
+++ b/libs/knowit/providers/enzyme.py
@@ -26,6 +26,7 @@ from knowit.rules import (
     LanguageRule,
     ResolutionRule,
 )
+from knowit.rules.general import GuessTitleRule
 from knowit.serializer import get_json_encoder
 from knowit.units import units
 from knowit.utils import to_dict
@@ -83,17 +84,20 @@ class EnzymeProvider(Provider):
             },
         }, {
             'video': {
-                'language': LanguageRule('video language'),
+                'guessed': GuessTitleRule('guessed properties', private=True),
+                'language': LanguageRule('video language', override=True),
                 'resolution': ResolutionRule('video resolution'),
             },
             'audio': {
-                'language': LanguageRule('audio language'),
+                'guessed': GuessTitleRule('guessed properties', private=True),
+                'language': LanguageRule('audio language', override=True),
                 'channels': AudioChannelsRule('audio channels'),
             },
             'subtitle': {
-                'language': LanguageRule('subtitle language'),
-                'hearing_impaired': HearingImpairedRule('subtitle hearing impaired'),
-                'closed_caption': ClosedCaptionRule('closed caption'),
+                'guessed': GuessTitleRule('guessed properties', private=True),
+                'language': LanguageRule('subtitle language', override=True),
+                'hearing_impaired': HearingImpairedRule('subtitle hearing impaired', override=True),
+                'closed_caption': ClosedCaptionRule('closed caption', override=True),
             }
         })
 
@@ -130,7 +134,8 @@ class EnzymeProvider(Provider):
 
         if logger.level == logging.DEBUG:
             logger.debug('Video {video_path} scanned using Enzyme {version} has raw data:\n{data}',
-                         video_path=video_path, version=enzyme.__version__, data=json.dumps(data))
+                         video_path=video_path, version=enzyme.__version__,
+                         data=json.dumps(data, cls=get_json_encoder(context), indent=4, ensure_ascii=False))
 
         result = self._describe_tracks(video_path, data.get('info', {}), data.get('video_tracks'),
                                        data.get('audio_tracks'), data.get('subtitle_tracks'), context)
diff --git a/libs/knowit/providers/ffmpeg.py b/libs/knowit/providers/ffmpeg.py
index 2474408cc..f19cea90b 100644
--- a/libs/knowit/providers/ffmpeg.py
+++ b/libs/knowit/providers/ffmpeg.py
@@ -34,6 +34,7 @@ from knowit.rules import (
     LanguageRule,
     ResolutionRule,
 )
+from knowit.rules.general import GuessTitleRule
 from knowit.serializer import get_json_encoder
 from knowit.units import units
 from knowit.utils import (
@@ -77,7 +78,7 @@ class FFmpegExecutor:
     def extract_info(self, filename):
         """Extract media info."""
         json_dump = self._execute(filename)
-        return json.loads(json_dump)
+        return json.loads(json_dump) if json_dump else {}
 
     def _execute(self, filename):
         raise NotImplementedError
@@ -144,7 +145,7 @@ class FFmpegProvider(Provider):
                 'id': Basic('index', data_type=int, allow_fallback=True, description='video track number'),
                 'name': Property('tags.title', description='video track name'),
                 'language': Language('tags.language', description='video language'),
-                'duration': Duration('duration', description='video duration'),
+                'duration': Duration('duration', 'tags.duration', description='video duration'),
                 'width': Quantity('width', unit=units.pixel),
                 'height': Quantity('height', unit=units.pixel),
                 'scan_type': ScanType(config, 'field_order', default='Progressive', description='video scan type'),
@@ -153,7 +154,7 @@ class FFmpegProvider(Provider):
                 'resolution': None,  # populated with ResolutionRule
                 'frame_rate': Ratio('r_frame_rate', unit=units.FPS, description='video frame rate'),
                 # frame_rate_mode
-                'bit_rate': Quantity('bit_rate', unit=units.bps, description='video bit rate'),
+                'bit_rate': Quantity('bit_rate', 'tags.bps', unit=units.bps, description='video bit rate'),
                 'bit_depth': Quantity('bits_per_raw_sample', unit=units.bit, description='video bit depth'),
                 'codec': VideoCodec(config, 'codec_name', description='video codec'),
                 'profile': VideoProfile(config, 'profile', description='video codec profile'),
@@ -166,13 +167,13 @@ class FFmpegProvider(Provider):
                 'id': Basic('index', data_type=int, allow_fallback=True, description='audio track number'),
                 'name': Property('tags.title', description='audio track name'),
                 'language': Language('tags.language', description='audio language'),
-                'duration': Duration('duration', description='audio duration'),
+                'duration': Duration('duration', 'tags.duration', description='audio duration'),
                 'codec': AudioCodec(config, 'profile', 'codec_name', description='audio codec'),
                 'profile': AudioProfile(config, 'profile', description='audio codec profile'),
                 'channels_count': AudioChannels('channels', description='audio channels count'),
                 'channels': None,  # populated with AudioChannelsRule
                 'bit_depth': Quantity('bits_per_raw_sample', unit=units.bit, description='audio bit depth'),
-                'bit_rate': Quantity('bit_rate', unit=units.bps, description='audio bit rate'),
+                'bit_rate': Quantity('bit_rate', 'tags.bps', unit=units.bps, description='audio bit rate'),
                 'sampling_rate': Quantity('sample_rate', unit=units.Hz, description='audio sampling rate'),
                 'forced': YesNo('disposition.forced', hide_value=False, description='audio track forced'),
                 'default': YesNo('disposition.default', hide_value=False, description='audio track default'),
@@ -190,17 +191,20 @@ class FFmpegProvider(Provider):
             },
         }, {
             'video': {
-                'language': LanguageRule('video language'),
+                'guessed': GuessTitleRule('guessed properties', private=True),
+                'language': LanguageRule('video language', override=True),
                 'resolution': ResolutionRule('video resolution'),
             },
             'audio': {
-                'language': LanguageRule('audio language'),
+                'guessed': GuessTitleRule('guessed properties', private=True),
+                'language': LanguageRule('audio language', override=True),
                 'channels': AudioChannelsRule('audio channels'),
             },
             'subtitle': {
-                'language': LanguageRule('subtitle language'),
-                'hearing_impaired': HearingImpairedRule('subtitle hearing impaired'),
-                'closed_caption': ClosedCaptionRule('closed caption'),
+                'guessed': GuessTitleRule('guessed properties', private=True),
+                'language': LanguageRule('subtitle language', override=True),
+                'hearing_impaired': HearingImpairedRule('subtitle hearing impaired', override=True),
+                'closed_caption': ClosedCaptionRule('closed caption', override=True),
             },
         })
         self.executor = FFmpegExecutor.get_executor_instance(suggested_path)
diff --git a/libs/knowit/providers/mediainfo.py b/libs/knowit/providers/mediainfo.py
index 39fd403ed..a19301bc6 100644
--- a/libs/knowit/providers/mediainfo.py
+++ b/libs/knowit/providers/mediainfo.py
@@ -1,5 +1,6 @@
-
+import ctypes
 import json
+import os
 import re
 from ctypes import c_void_p, c_wchar_p
 from decimal import Decimal
@@ -43,6 +44,7 @@ from knowit.rules import (
     LanguageRule,
     ResolutionRule,
 )
+from knowit.rules.general import GuessTitleRule
 from knowit.units import units
 from knowit.utils import (
     define_candidate,
@@ -77,7 +79,7 @@ class MediaInfoExecutor:
 
     locations = {
         'unix': ('/usr/local/mediainfo/lib', '/usr/local/mediainfo/bin', '__PATH__'),
-        'windows': ('__PATH__', ),
+        'windows': ('C:\\Program Files\\MediaInfo', 'C:\\Program Files (x86)\\MediaInfo', '__PATH__'),
         'macos': ('__PATH__', ),
     }
 
@@ -121,12 +123,28 @@ class MediaInfoCliExecutor(MediaInfoExecutor):
     }
 
     def _execute(self, filename):
-        return json.loads(check_output([self.location, '--Output=JSON', '--Full', filename]).decode())
+        data = check_output([self.location, '--Output=JSON', '--Full', filename]).decode()
+
+        return json.loads(data) if data else {}
+
+    @classmethod
+    def _is_gui_exe(cls, candidate: str):
+        if not candidate.endswith('MediaInfo.exe') or not os.path.isfile(candidate):
+            return False
+
+        try:
+            shell32 = ctypes.WinDLL('shell32', use_last_error=True)  # type: ignore
+            return bool(shell32.ExtractIconExW(candidate, 0, None, None, 1))
+        except Exception:
+            return False
 
     @classmethod
     def create(cls, os_family=None, suggested_path=None):
         """Create the executor instance."""
         for candidate in define_candidate(cls.locations, cls.names, os_family, suggested_path):
+            if cls._is_gui_exe(candidate):
+                continue
+
             try:
                 output = check_output([candidate, '--version']).decode()
                 version = cls._get_version(output)
@@ -154,7 +172,9 @@ class MediaInfoCTypesExecutor(MediaInfoExecutor):
 
     def _execute(self, filename):
         # Create a MediaInfo handle
-        return json.loads(MediaInfo.parse(filename, library_file=self.location, output='JSON'))
+        data = MediaInfo.parse(filename, library_file=self.location, output='JSON')
+
+        return json.loads(data) if data else {}
 
     @classmethod
     def create(cls, os_family=None, suggested_path=None):
@@ -254,19 +274,22 @@ class MediaInfoProvider(Provider):
             },
         }, {
             'video': {
-                'language': LanguageRule('video language'),
+                'guessed': GuessTitleRule('guessed properties', private=True),
+                'language': LanguageRule('video language', override=True),
                 'resolution': ResolutionRule('video resolution'),
             },
             'audio': {
-                'language': LanguageRule('audio language'),
+                'guessed': GuessTitleRule('guessed properties', private=True),
+                'language': LanguageRule('audio language', override=True),
                 'channels': AudioChannelsRule('audio channels'),
-                '_atmosrule': AtmosRule(config, 'atmos rule'),
-                '_dtshdrule': DtsHdRule(config, 'dts-hd rule'),
+                'atmos': AtmosRule(config, 'atmos rule', private=True),
+                'dtshd': DtsHdRule(config, 'dts-hd rule', private=True),
             },
             'subtitle': {
-                'language': LanguageRule('subtitle language'),
-                'hearing_impaired': HearingImpairedRule('subtitle hearing impaired'),
-                'closed_caption': ClosedCaptionRule('closed caption'),
+                'guessed': GuessTitleRule('guessed properties', private=True),
+                'language': LanguageRule('subtitle language', override=True),
+                'hearing_impaired': HearingImpairedRule('subtitle hearing impaired', override=True),
+                'closed_caption': ClosedCaptionRule('closed caption', override=True),
             }
         })
         self.executor = MediaInfoExecutor.get_executor_instance(suggested_path)
diff --git a/libs/knowit/providers/mkvmerge.py b/libs/knowit/providers/mkvmerge.py
index e5aca1550..ff422f8b4 100644
--- a/libs/knowit/providers/mkvmerge.py
+++ b/libs/knowit/providers/mkvmerge.py
@@ -28,6 +28,7 @@ from knowit.rules import (
     LanguageRule,
     ResolutionRule,
 )
+from knowit.rules.general import GuessTitleRule
 from knowit.serializer import get_json_encoder
 from knowit.units import units
 from knowit.utils import define_candidate, detect_os
@@ -67,7 +68,7 @@ class MkvMergeExecutor:
     def extract_info(self, filename):
         """Extract media info."""
         json_dump = self._execute(filename)
-        return json.loads(json_dump)
+        return json.loads(json_dump) if json_dump else {}
 
     def _execute(self, filename):
         raise NotImplementedError
@@ -166,17 +167,20 @@ class MkvMergeProvider(Provider):
             },
         }, {
             'video': {
+                'guessed': GuessTitleRule('guessed properties', private=True),
                 'language': LanguageRule('video language', override=True),
                 'resolution': ResolutionRule('video resolution'),
             },
             'audio': {
+                'guessed': GuessTitleRule('guessed properties', private=True),
                 'language': LanguageRule('audio language', override=True),
                 'channels': AudioChannelsRule('audio channels'),
             },
             'subtitle': {
+                'guessed': GuessTitleRule('guessed properties', private=True),
                 'language': LanguageRule('subtitle language', override=True),
-                'hearing_impaired': HearingImpairedRule('subtitle hearing impaired'),
-                'closed_caption': ClosedCaptionRule('closed caption'),
+                'hearing_impaired': HearingImpairedRule('subtitle hearing impaired', override=True),
+                'closed_caption': ClosedCaptionRule('closed caption', override=True),
             }
         })
         self.executor = MkvMergeExecutor.get_executor_instance(suggested_path)
diff --git a/libs/knowit/rules/general.py b/libs/knowit/rules/general.py
index b492c03a5..ad2c7734f 100644
--- a/libs/knowit/rules/general.py
+++ b/libs/knowit/rules/general.py
@@ -1,8 +1,6 @@
-
-import re
 from logging import NullHandler, getLogger
 
-import babelfish
+from trakit.api import trakit
 
 from knowit.core import Rule
 
@@ -10,22 +8,27 @@ logger = getLogger(__name__)
 logger.addHandler(NullHandler())
 
 
+class GuessTitleRule(Rule):
+    """Guess properties from track title."""
+
+    def execute(self, props, pv_props, context):
+        """Language detection using name."""
+        if 'name' in props:
+            language = props.get('language')
+            options = {'expected_language': language} if language else {}
+            guessed = trakit(props['name'], options)
+            if guessed:
+                return guessed
+
+
 class LanguageRule(Rule):
     """Language rules."""
 
-    name_re = re.compile(r'(?P<name>\w+)\b', re.IGNORECASE)
-
     def execute(self, props, pv_props, context):
         """Language detection using name."""
-        if 'language' in props:
+        if 'guessed' not in pv_props:
             return
 
-        if 'name' in props:
-            name = props.get('name', '')
-            match = self.name_re.match(name)
-            if match:
-                try:
-                    return babelfish.Language.fromname(match.group('name'))
-                except babelfish.Error:
-                    pass
-            logger.info('Invalid %s: %r', self.description, name)
+        guess = pv_props['guessed']
+        if 'language' in guess:
+            return guess['language']
diff --git a/libs/knowit/rules/subtitle.py b/libs/knowit/rules/subtitle.py
index fa16fdbc1..704109f99 100644
--- a/libs/knowit/rules/subtitle.py
+++ b/libs/knowit/rules/subtitle.py
@@ -10,18 +10,19 @@ class ClosedCaptionRule(Rule):
 
     def execute(self, props, pv_props, context):
         """Execute closed caption rule."""
-        for name in (pv_props.get('_closed_caption'), props.get('name')):
-            if name and self.cc_re.search(name):
-                return True
+        if '_closed_caption' in pv_props and self.cc_re.search(pv_props['_closed_caption']):
+            return True
+
+        if 'guessed' in pv_props:
+            guessed = pv_props['guessed']
+            return guessed.get('closed_caption')
 
 
 class HearingImpairedRule(Rule):
     """Hearing Impaired rule."""
 
-    hi_re = re.compile(r'(\bsdh\b)', re.IGNORECASE)
-
     def execute(self, props, pv_props, context):
         """Hearing Impaired."""
-        name = props.get('name')
-        if name and self.hi_re.search(name):
-            return True
+        if 'guessed' in pv_props:
+            guessed = pv_props['guessed']
+            return guessed.get('hearing_impaired')
diff --git a/libs/knowit/units.py b/libs/knowit/units.py
index 73ec16a5a..51e6cae73 100644
--- a/libs/knowit/units.py
+++ b/libs/knowit/units.py
@@ -1,10 +1,5 @@
 import typing
 
-try:
-    import pint
-except ImportError:
-    pint = False
-
 
 class NullRegistry:
     """A NullRegistry that masquerades as a pint.UnitRegistry."""
@@ -25,9 +20,18 @@ class NullRegistry:
 
 
 def _build_unit_registry():
-    registry = pint.UnitRegistry() if pint else NullRegistry()
-    registry.define('FPS = 1 * hertz')
-    return registry
+    try:
+        import pint
+
+        registry = pint.UnitRegistry()
+        registry.define('FPS = 1 * hertz')
+
+        pint.set_application_registry(registry)
+        return registry
+    except ModuleNotFoundError:
+        pass
+
+    return NullRegistry()
 
 
 units = _build_unit_registry()