test/test_write_annotations.py

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