def get_text(self):
if self.conf_lyrics_done:
print 'Lyrics are already done'
return self.conf_lyrics_done
bing = BingSearchAPI()
tags = self.conf_tags
search = '%s lyrics %s' % (tags['title'], tags['performer'])
print 'Searching for lyrics. Search string: %s' % search
lyrics_search = bing.search('web', search.encode('utf-8'), {'$format': 'json'})
#print 'Lyrics search result: %s' % pformat(lyrics_search)
for result in lyrics_search.get('d', {}).get('results', [{}])[0].get('Web', []):
url = result['Url']
print 'lyrics in %s?' % url
for match, (good_attr, bad_part) in lyrics_matches.items():
if match in url:
# Good! We have a known site with lyrics - let's extract them.
print 'yes, lyrics are probably here'
browser = Browser()
browser.set_handle_robots(None)
browser.open(url)
text = browser.response().read()
soup = BeautifulSoup(text, convertEntities=BeautifulSoup.HTML_ENTITIES)
lyrics_el = soup.find(attrs=good_attr)
if not lyrics_el:
#print 'Not found lyrics in %s' % text
continue
#print 'full text: %s' % text
#print 'Found something like this: %s' % lyrics_el
parts = list(self.extract_text_parts(lyrics_el.contents, bad_part))
lyrics = '\n'.join(parts)
#print 'Found lyrics: \n%s' % lyrics
print 'Found lyrics: %s...' % lyrics[:150]
self.conf_lyrics = lyrics
self.conf_lyrics_done = True
return self.conf_lyrics_done
print 'Unsupported lyrics source: %s' % url
if not self.conf_lyrics_done:
print 'ERROR: lyrics not found! %s' % self.conf_tags['title']
return self.conf_lyrics_done
def get_pics(self):
if self.conf_pics_done:
print 'Pics are already done'
return self.conf_pics_done
imgdir = self.imgdir
if len(glob.glob1(imgdir, "*.png")) > REQUIRED_IMAGE_COUNT:
self.conf_pics_done = True
return self.conf_pics_done
bing = BingSearchAPI()
tags = self.conf_tags
search = '%s %s' % (tags['title'], tags['performer'])
print 'Searching for images. Search string: %s' % search
img_search = bing.search('image', search.encode('utf-8'), {'$format': 'json'})
print 'Images: %s' % pformat(img_search)
registry = processed_image_urls.setdefault(imgdir, set())
if not os.path.exists(imgdir):
os.makedirs(imgdir)
for result in img_search.get('d', {}).get('results', [{}])[0].get('Image', []):
if result['MediaUrl'] not in registry:
browser = Browser()
browser.set_handle_robots(None)
registry.add(result['MediaUrl'])
log.debug('%s images in %s' % (len(glob.glob1(imgdir, "*.png")), imgdir))
try:
#log.debug('Opening %s' % result['SourceUrl'])
browser.open(result['SourceUrl'])
#log.debug('Opening %s' % result['MediaUrl'])
img = Image.open(browser.open(result['MediaUrl']))
if img.size[0] >= DEFAULT_VIDEO_RESOLUTION and img.size[1] >= DEFAULT_VIDEO_RESOLUTION[1]:
print 'Found image: %s' % result['MediaUrl']
img.save(os.path.join(imgdir, ('image%03d.png'
% (len(glob.glob1(imgdir, "*.png"))) + 1)))
self.conf_pics_done = True
if len(glob.glob1(imgdir, "*.png")) > REQUIRED_IMAGE_COUNT:
self.conf_pics_done = True
break
except:
print_exc()
if len(glob.glob1(imgdir, "*.png")) < REQUIRED_IMAGE_COUNT:
search = tags['performer']
print 'Searching for images. Search string: %s' % search
img_search = bing.search('image', search.encode('utf-8'), {'$format': 'json'})
for result in img_search.get('d', {}).get('results', [{}])[0].get('Image', []):
if result['MediaUrl'] not in registry:
browser = Browser()
browser.set_handle_robots(None)
registry.add(result['MediaUrl'])
log.debug('%s images in %s' % (len(glob.glob1(imgdir, "*.png")), imgdir))
try:
#log.debug('Opening %s' % result['SourceUrl'])
browser.open(result['SourceUrl'])
#log.debug('Opening %s' % result['MediaUrl'])
img = Image.open(browser.open(result['MediaUrl']))
if img.size[0] >= DEFAULT_VIDEO_RESOLUTION[0] and img.size[1] >= DEFAULT_VIDEO_RESOLUTION[1]:
print 'Found image: %s' % result['MediaUrl']
img.save(os.path.join(imgdir, ('image%03d.png'
% (len(glob.glob1(imgdir, "*.png"))) + 1)))
if len(glob.glob1(imgdir, "*.png")) > REQUIRED_IMAGE_COUNT:
self.conf_pics_done = True
break
except:
print_exc()
return self.conf_pics_done