]> Raphaƫl G. Git Repositories - youtubedl/blob - test/test_write_annotations.py
Merge tag 'upstream/2014.12.01'
[youtubedl] / test / test_write_annotations.py
1 #!/usr/bin/env python
2 # coding: utf-8
3 from __future__ import unicode_literals
4
5 # Allow direct execution
6 import os
7 import sys
8 import unittest
9 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
10
11 from test.helper import get_params, try_rm
12
13
14 import io
15
16 import xml.etree.ElementTree
17
18 import youtube_dl.YoutubeDL
19 import youtube_dl.extractor
20
21
22 class YoutubeDL(youtube_dl.YoutubeDL):
23 def __init__(self, *args, **kwargs):
24 super(YoutubeDL, self).__init__(*args, **kwargs)
25 self.to_stderr = self.to_screen
26
27 params = get_params({
28 'writeannotations': True,
29 'skip_download': True,
30 'writeinfojson': False,
31 'format': 'flv',
32 })
33
34
35 TEST_ID = 'gr51aVj-mLg'
36 ANNOTATIONS_FILE = TEST_ID + '.flv.annotations.xml'
37 EXPECTED_ANNOTATIONS = ['Speech bubble', 'Note', 'Title', 'Spotlight', 'Label']
38
39
40 class TestAnnotations(unittest.TestCase):
41 def setUp(self):
42 # Clear old files
43 self.tearDown()
44
45 def test_info_json(self):
46 expected = list(EXPECTED_ANNOTATIONS) # Two annotations could have the same text.
47 ie = youtube_dl.extractor.YoutubeIE()
48 ydl = YoutubeDL(params)
49 ydl.add_info_extractor(ie)
50 ydl.download([TEST_ID])
51 self.assertTrue(os.path.exists(ANNOTATIONS_FILE))
52 annoxml = None
53 with io.open(ANNOTATIONS_FILE, 'r', encoding='utf-8') as annof:
54 annoxml = xml.etree.ElementTree.parse(annof)
55 self.assertTrue(annoxml is not None, 'Failed to parse annotations XML')
56 root = annoxml.getroot()
57 self.assertEqual(root.tag, 'document')
58 annotationsTag = root.find('annotations')
59 self.assertEqual(annotationsTag.tag, 'annotations')
60 annotations = annotationsTag.findall('annotation')
61
62 # Not all the annotations have TEXT children and the annotations are returned unsorted.
63 for a in annotations:
64 self.assertEqual(a.tag, 'annotation')
65 if a.get('type') == 'text':
66 textTag = a.find('TEXT')
67 text = textTag.text
68 self.assertTrue(text in expected) # assertIn only added in python 2.7
69 # remove the first occurance, there could be more than one annotation with the same text
70 expected.remove(text)
71 # We should have seen (and removed) all the expected annotation texts.
72 self.assertEqual(len(expected), 0, 'Not all expected annotations were found.')
73
74 def tearDown(self):
75 try_rm(ANNOTATIONS_FILE)
76
77 if __name__ == '__main__':
78 unittest.main()