Merge pull request #664 from deuteronomy-works/master

Merged
deuteronomy-works · Oct 6, 2023 · 6e3693f · 6e3693f
2 parents 7c2f65e + 2b5a4a7
commit 6e3693f
Show file tree

Hide file tree

Showing 3 changed files with 115 additions and 46 deletions.
diff --git a/local_probe_t_est.py b/local_probe_t_est.py
@@ -1,4 +1,5 @@
 import os
+import pprint
 from pyffmpeg import FFprobe
 
 
@@ -8,15 +9,16 @@
 #test_file = "G:\Entertainment\Movies\Interstellar (2014)\Interstellar.2014.720p.BluRay.x264.YIFY.mp4"
 flv_file = os.path.join(test_folder, 'sample_960x400_ocean_with_audio.flv')
 
-f = FFprobe('flv_file')
+f = FFprobe(flv_file)
 # ret = f.get_album_art('cover.png')
 
 
-print(f.duration)
+pprint.pprint(f.duration)
 for x in f.metadata:
     print(x, '\n')
 
-print(f.metadata)
-print(f.metadata[0])
+pprint.pprint(f.metadata)
+pprint.pprint(f.metadata[0])
 # print(f.metadata[0][0])
-print(f.metadata[0][0]['codec'])
+print("is codec flv1?")
+pprint.pprint(f.metadata[0]['codec'] == 'flv1')
diff --git a/pyffmpeg/pseudo_ffprobe.py b/pyffmpeg/pseudo_ffprobe.py
@@ -9,6 +9,7 @@
 import random
 import os
 import logging
+from collections import defaultdict
 # from base64 import b64decode
 
 from .misc import Paths, SHELL, ModifiedList
@@ -116,19 +117,18 @@ def _extract_all(self, stdout):
                 print("Multiple input files found.\
                      However only one will be probed")
             all_streams = all_streams[0]
-
         # individual streams
         streams = all_streams.split('Stream')
-        for x in range(len(streams)):
+        self.metadata[-1] = self._parse_input_meta(streams[0])
 
-            if x == 0:
-                if streams[x]:
-                    self.metadata[-1] = self._parse_input_meta(streams[x])
-            else:
-                if streams[x]:
-                    self.metadata[0].append(self._parse_meta(streams[x]))
+        tags = defaultdict(list)
+        for x in range(1, len(streams)):
+            if streams[x]:
+                tags.update(self._parse_meta(streams[x]))
+
+        if len(tags) > 0:
+            self.metadata[0] = tags
 
-        # parse other metadata
         self._parse_other_meta()
 
         # then handle stream 0:0 so
@@ -165,23 +165,8 @@ def get_album_art(self, out_file=None):
 
     def _parse_meta(self, stream):
         self.logger.info('Inside _parse_meta')
-        tags = {}
         metadata = self._strip_meta(stream)
-        # Previous key will be overriden
-        prev_key = ''
-
-        for x in range(len(metadata)):
-            line = metadata[x]
-            data = line.split(":", 1)
-            key = data[0].strip()
-            value = data[1].strip()
-            # this might be a continuation
-            if key == '':
-                tags[prev_key] += "\\r\\n" + data[1].strip()
-            else:
-                tags[key] = value
-                prev_key = key
-
+        tags = self._generate_tags(metadata)
         return tags
 
     def _parse_header(self, line):
@@ -202,23 +187,8 @@ def _parse_header(self, line):
 
     def _parse_input_meta(self, stream):
         self.logger.info('Inside _parse_input_meta')
-        tags = {}
         metadata = self._strip_input_meta(stream)
-
-        # previous key will be overriden
-        prev_key = ''
-        for x in range(len(metadata)):
-            line = metadata[x]
-            data = line.split(":", 1)
-            key = data[0].strip()
-            value = data[1].strip()
-            # this might be a continuation
-            if key == '':
-                tags[prev_key] += "\\r\\n" + data[1].strip()
-            else:
-                tags[key] = value
-                prev_key = key
-
+        tags = self._generate_tags(metadata)
         return tags
 
     def _parse_other_meta(self):
@@ -320,3 +290,23 @@ def _strip_input_meta(self, stdout):
                 meta.append(line)
 
         return meta
+
+    def _generate_tags(self, metadata):
+        prev_key = ''
+        tags = defaultdict(list)
+
+        for x in range(len(metadata)):
+            line = metadata[x]
+            data = line.split(":", 1)
+            key = data[0].strip() 
+            value = data[1].strip()
+            # this might be a continuation, if not we add it to orphaned key list
+            if key == '':
+                if prev_key == '':
+                    tags['unpaired_values'].append(value)
+                else:
+                    tags[prev_key].append("\\r\\n" + data[1].strip())
+            else:
+                tags[key] = value
+                prev_key = key
+        return tags
diff --git a/test_pseudo_ffprobe.py b/test_pseudo_ffprobe.py
@@ -1,6 +1,7 @@
 import pytest
 import os
 import requests
+from collections import defaultdict
 from pyffmpeg import FFprobe
 
 # test speed to make sure no convertion took place
@@ -31,3 +32,79 @@ def test_probe(file_name, duration):
 
 def test_album_art():
     pass
+
+@pytest.mark.parametrize(
+    'file_name',
+    [
+        ("countdown.mp4"),
+    ]
+)
+def test_generate_tags(file_name):
+    metadata_one = [
+        '    major_brand     : mp42',
+        '    minor_version   : 0',
+        '    compatible_brands: isommp42',
+        '    creation_time   : 2016-07-05T17:50:46.000000Z',
+        '  Duration: 00:00:04.37',
+        'start: 0.000000',
+        'bitrate: 322 kb/s'
+        ]
+    metadata_two = [
+        'codec: h264',
+        'data_rate: 223 kb/s',
+        'dimensions: 640x360',
+        'DAR: 16:9',
+        'SAR: 1:1',
+        'fps: 29.97',
+        'tbn: 30k',
+        'tbr: 29.97',
+        '      handler_name    : VideoHandler',
+        '      vendor_id       : [0][0][0][0]',
+        'codec: aac',
+        'bitrate: 96 kb/s',
+        'channels: stereo',
+        'sample_rate: 44100 Hz',
+        '      creation_time   : 2016-07-05T17:50:46.000000Z',
+        '      handler_name    : IsoMedia File Produced by Google, 5-11-2011',
+        '      vendor_id       : [0][0][0][0]',
+        'Parsing a group of options: output url /dev/null.',
+        'Opening an output file: /dev/null.',
+        '[h264 @ 0x7f916e046480] nal_unit_type: 7(SPS), nal_ref_idc: 3',
+        '[h264 @ 0x7f916e046480] nal_unit_type: 8(PPS), nal_ref_idc: 3'
+        ]   
+    tags_one = defaultdict(list,
+        {
+            'Duration': '00:00:04.37',
+            'bitrate': '322 kb/s',
+            'compatible_brands': 'isommp42',
+            'creation_time': '2016-07-05T17:50:46.000000Z',
+            'major_brand': 'mp42',
+            'minor_version': '0',
+            'start': '0.000000'
+        })
+    tags_two = defaultdict(list, 
+        {
+            'DAR': '16:9',
+            'Opening an output file': '/dev/null.',
+            'Parsing a group of options': 'output url /dev/null.',
+            'SAR': '1:1',
+            '[h264 @ 0x7f916e046480] nal_unit_type': '8(PPS), nal_ref_idc: '
+                                                        '3',
+            'bitrate': '96 kb/s',
+            'channels': 'stereo',
+            'codec': 'aac',
+            'creation_time': '2016-07-05T17:50:46.000000Z',
+            'data_rate': '223 kb/s',
+            'dimensions': '640x360',
+            'fps': '29.97',
+            'handler_name': 'IsoMedia File Produced by Google, 5-11-2011',
+            'sample_rate': '44100 Hz',
+            'tbn': '30k',
+            'tbr': '29.97',
+            'vendor_id': '[0][0][0][0]'
+        })    
+    test_file = os.path.join(TEST_FOLDER, file_name)
+    f = FFprobe(test_file)
+
+    assert tags_one == f._generate_tags(metadata_one)
+    assert tags_two == f._generate_tags(metadata_two)