X-Git-Url: https://git.llucax.com/software/subdivxget.git/blobdiff_plain/c79b74764e72a29ce82d3ec489c2433e8ad6ba20..2dee376a09210989fb2211099fa023ef3bd0ba18:/subdivxget?ds=sidebyside

diff --git a/subdivxget b/subdivxget
index 5f67edb..597110c 100755
--- a/subdivxget
+++ b/subdivxget
@@ -1,10 +1,33 @@
 #!/usr/bin/env python
 
 import sys
-import urllib
+if sys.version_info[0] < 3:
+	from HTMLParser import HTMLParser
+	from urllib import urlopen, urlretrieve, urlencode
+	def get_encoding(info):
+		return info.getparam('charset')
+
+else:
+	from html.parser import HTMLParser
+	from urllib.request import urlopen, urlretrieve
+	from urllib.parse import urlencode
+	def get_encoding(info):
+		return info.get_content_charset('ascii')
 import zipfile
 import subprocess
-import HTMLParser
+
+
+def output(fo, fmt, *args, **kargs):
+	if not args:
+		args = kargs
+	fo.write((fmt % args) + '\n')
+
+def echo(fmt, *args, **kargs):
+	output(sys.stdout, fmt, *args, **kargs)
+
+def error(fmt, *args, **kargs):
+	output(sys.stderr, fmt, *args, **kargs)
+
 
 class SubDivXQuery:
 	def __init__(self, to_search, page_number):
@@ -22,27 +45,30 @@ class SubDivXQuery:
 	@property
 	def url(self):
 		return 'http://%s%s?%s' % (self.host, self.page,
-				urllib.urlencode(self.query))
+				urlencode(self.query))
 	@property
 	def page_uri(self):
-		return self.page + '?' + urllib.urlencode(self.query)
+		return self.page + '?' + urlencode(self.query)
 	@property
 	def down_uri(self):
 		return 'http://' + self.host + self.down_page
 
 
-class SubDivXHTMLParser(HTMLParser.HTMLParser):
+class SubDivXHTMLParser(HTMLParser):
 
 	IDLE = 1
 	HEADER = 2
 
 	def __init__(self, down_uri):
-		HTMLParser.HTMLParser.__init__(self)
+		HTMLParser.__init__(self)
 		self.down_uri = down_uri
 		self.depth = 0
 		self.parsing = False
 		self.subs = []
 		self.attr = None
+		self.attr_depth = 0
+		self.cur = None
+		self.in_script_style = False
 
 	def handle_starttag(self, tag, attrs):
 		attrs = dict(attrs)
@@ -52,38 +78,117 @@ class SubDivXHTMLParser(HTMLParser.HTMLParser):
 			self.parsing = True
 		if not self.parsing:
 			return
+		if tag == 'script' or tag == 'style':
+			self.in_script_style = True
+			return
 		if tag == 'div':
 			if attrs.get('id') == 'buscador_detalle':
 				self.parsing = True
 			elif attrs.get('id') == 'buscador_detalle_sub':
 				self.attr = 'desc'
+				self.attr_depth = self.depth + 1
+				self.cur[self.attr] = ''
 		elif tag == 'a':
 			if attrs.get('class') == 'titulo_menu_izq':
-				self.attr = 'title'
+				self.attr = 'titulo'
+				self.attr_depth = self.depth + 1
+				self.cur[self.attr] = ''
 			elif attrs.get('href', '').startswith(self.down_uri):
 				self.cur['url'] = attrs['href']
-		if self.parsing:
+		# br are usually not closed, so ignore them in depth calculation
+		if self.parsing and tag != 'br':
 			self.depth += 1
 
 	def handle_endtag(self, tag):
 		if self.parsing:
-			self.depth -= 1
+			if tag == 'script' or tag == 'style':
+				self.in_script_style = False
+				return
+			if self.depth == self.attr_depth:
+				self.attr = None
+				self.attr_depth = 0
+			# see comment in handle_starttag()
+			if tag != 'br':
+				self.depth -= 1
 		if self.depth == 0:
 			self.parsing = False
 
 	def handle_data(self, data):
-		if self.parsing:
-			data = data.strip()
-			if self.attr is not None and data:
-				self.cur[self.attr] = data
+		if not self.parsing:
+			return
+		data = data.strip()
+		# Hack to handle comments in <script> <style> which don't end
+		# up in handle_comment(), so we just ignore the whole tags
+		if self.in_script_style:
+			return
+		if self.attr is not None and data:
+			self.cur[self.attr] += ' ' + data
+			if self.attr_depth == 0:
+				self.cur[self.attr] = self.cur[self.attr].strip()
 				self.attr = None
-			elif data in ('Downloads:', 'Cds:', 'Comentarios:',
-					'Formato:'):
-				self.attr = data[:-1].lower()
-			elif data == 'Subido por:':
-				self.attr = 'autor'
-			elif data == 'el':
-				self.attr = 'fecha'
+				self.attr_depth = 0
+		elif data in ('Downloads:', 'Cds:', 'Comentarios:', 'Formato:'):
+			self.attr = data[:-1].lower()
+			self.attr_depth = 0
+			self.cur[self.attr] = ''
+		elif data == 'Subido por:':
+			self.attr = 'autor'
+			self.attr_depth = 0
+			self.cur[self.attr] = ''
+		elif data == 'el':
+			self.attr = 'fecha'
+			self.attr_depth = 0
+			self.cur[self.attr] = ''
+
+
+def filter_subtitles(subs, filters):
+	def is_good(sub, filter):
+		def is_any_good(sub, filter):
+			for value in sub.values():
+				if value.lower().find(filter) >= 0:
+					return True
+
+		field = None
+		if len(filter) > 2 and filter[1] == ':':
+			field = filter[0]
+			filter = filter[2:]
+		filter = filter.lower()
+
+		if field is None:
+			return is_any_good(sub, filter)
+		elif field == 't':
+			key = 'titulo'
+		elif field == 'd':
+			key = 'desc'
+		elif field == 'a':
+			key = 'autor'
+		elif field == 'f':
+			key = 'formato'
+		elif field == 'c':
+			key = 'comentarios'
+		elif field == 'C':
+			key = 'cds'
+		elif field == 'F':
+			key = 'fecha'
+		elif field == 'D':
+			key = 'downloads'
+		else:
+			# Not a recognizer field identifier, use the raw filter
+			return is_any_good(sub, field + ':' + filter)
+
+		return sub[key].lower().find(filter) >= 0
+
+	if not filters:
+		return subs
+
+	result = []
+	for sub in subs:
+		for filter in filters:
+			if not is_good(sub, filter):
+				break
+		else:
+			result.append(sub)
+	return result
 
 
 def subdivx_get_subs(query_str):
@@ -91,11 +196,16 @@ def subdivx_get_subs(query_str):
 	subs = []
 	while True:
 		query = SubDivXQuery(query_str, page_number)
-		url = urllib.urlopen(query.url)
+		url = urlopen(query.url)
 		parser = SubDivXHTMLParser(query.down_uri)
 
+		try:
+			encoding = get_encoding(url.info())
+		except:
+			encoding = 'ascii'
+
 		for line in url:
-			parser.feed(line)
+			parser.feed(line.decode(encoding))
 
 		url.close()
 
@@ -105,40 +215,78 @@ def subdivx_get_subs(query_str):
 		subs.extend(parser.subs)
 		page_number += 1
 
-	return sorted(subs, key=lambda s: int(s['downloads']), reverse=True)
+	return subs
+
 
+def unzip_subs(fname):
+	sub_exts = ('.srt', '.sub')
+	z = zipfile.ZipFile(fname, 'r')
+	z.printdir()
+	for fn in z.namelist():
+		if fn.endswith(sub_exts):
+			if '..' in fn or fn.startswith('/'):
+				error('Ignoring file with dangerous name: %s',
+						fn)
+				continue
+			echo('Extracting %s...', fn)
+			z.extract(fn)
 
-def get_subs(query_str):
+
+def get_subs(query_str, filters):
+	global opts
 	zip_exts = ('application/zip',)
 	rar_exts = ('application/rar', 'application/x-rar-compressed')
 
-	for sub in subdivx_get_subs(query_str):
-		print '''\
-	- %(title)s (%(autor)s - %(fecha)s - %(downloads)s - %(comentarios)s)
-	  %(desc)s
-		DOWNLOADING ...
-	''' % sub
-		fname, headers = urllib.urlretrieve(sub['url'])
+	subs = subdivx_get_subs(query_str)
+	subs = filter_subtitles(subs, filters)
+	subs.sort(key=lambda s: int(s['downloads']), reverse=True)
+
+	for sub in subs:
+		echo('''\
+- %(titulo)s (%(autor)s - %(fecha)s - %(downloads)s - %(comentarios)s)
+  %(desc)s
+	DOWNLOADING ...
+''', **sub)
+		if opts.list_only:
+			continue
+		fname, headers = urlretrieve(sub['url'])
 		if 'Content-Type' in headers:
 			if headers['Content-Type'] in zip_exts:
-				z = zipfile.ZipFile(fname, 'r')
-				z.printdir()
-				for fn in z.namelist():
-					if fn.endswith('.srt') or fn.endswith('.sub'):
-						if '..' in fn or fn.startswith('/'):
-							print 'Dangerous file name:', fn
-							continue
-						print 'Extracting', fn, '...'
-						z.extract(fn)
+				unzip_subs(fname)
 			elif headers['Content-Type'] in rar_exts:
 				if subprocess.call(['rar', 'x', fname]) != 0:
-					print 'Error unraring file %s' % fname
+					error('Error unraring file %s', fname)
 			else:
-				print 'Unrecognized file type:', headers['Content-Type']
+				error('Unrecognized file type:',
+						headers['Content-Type'])
 		else:
-			print 'No Content-Type!'
+			error('No Content-Type!')
+
+
+def parse_args(argv):
+	from optparse import OptionParser
+	parser = OptionParser(usage="%prog [OPTIONS] QUERY [FILTER ...]",
+			description="""
+Download subtitles from subdivx.com searching the string QUERY. If FILTERs are
+specified, only subtitles that matches all those filters are downloaded.
+Filters have the format "X:fitler", where X is a field specification: t=titulo,
+d=desc, a=autor, f=formato, c=comentarios, C=cds, F=fecha and D=downloads.
+filter is a string that should be found on that field (case insensitive). If
+the format specifier is not known (or there isn't one) the filter string is
+looked in all the fields.
+			""".strip())
+	parser.add_option("-l", "--list-only",
+			default=False, action='store_true',
+			help="Don't download the subtitles, just list them")
+
+	(opts, args) = parser.parse_args()
+	if not args:
+              parser.error("Missing query string")
+
+	return (args[0], args[1:], opts)
+
+(query_str, filters, opts) = parse_args(sys.argv)
 
+get_subs(query_str, filters)
 
-for q in sys.argv[1:]:
-	get_subs(q)