Add fairly fine grained filtering support

[software/subdivxget.git] / subdivxget

diff --git a/subdivxget b/subdivxget
index bd6fc6ec6b6a427685b85c41b9c8ee92c309a792..da731c48c610de7ccc8bf1460c9245fa0ecbac7d 100755 (executable)
--- a/subdivxget
+++ b/subdivxget
@@ -43,6 +43,7 @@ class SubDivXHTMLParser(HTMLParser.HTMLParser):
                self.parsing = False
                self.subs = []
                self.attr = None
+               self.attr_depth = 0
                self.cur = None
                self.in_script_style = False
 
@@ -62,12 +63,17 @@ class SubDivXHTMLParser(HTMLParser.HTMLParser):
                                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):
@@ -75,7 +81,12 @@ class SubDivXHTMLParser(HTMLParser.HTMLParser):
                        if tag == 'script' or tag == 'style':
                                self.in_script_style = False
                                return
-                       self.depth -= 1
+                       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
 
@@ -88,15 +99,73 @@ class SubDivXHTMLParser(HTMLParser.HTMLParser):
                if self.in_script_style:
                        return
                if self.attr is not None and data:
-                       self.cur[self.attr] = data
-                       self.attr = None
-               elif data in ('Downloads:', 'Cds:', 'Comentarios:',
-                               'Formato:'):
+                       self.cur[self.attr] += ' ' + data
+                       if self.attr_depth == 0:
+                               self.cur[self.attr] = self.cur[self.attr].strip()
+                               self.attr = None
+                               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):
@@ -121,13 +190,16 @@ def subdivx_get_subs(query_str):
        return sorted(subs, key=lambda s: int(s['downloads']), reverse=True)
 
 
-def get_subs(query_str):
+def get_subs(query_str, filters):
        zip_exts = ('application/zip',)
        rar_exts = ('application/rar', 'application/x-rar-compressed')
 
-       for sub in subdivx_get_subs(query_str):
+       subs = subdivx_get_subs(query_str)
+       subs = filter_subtitles(subs, filters)
+
+       for sub in subs:
                print '''\
-       - %(title)s (%(autor)s - %(fecha)s - %(downloads)s - %(comentarios)s)
+       - %(titulo)s (%(autor)s - %(fecha)s - %(downloads)s - %(comentarios)s)
          %(desc)s
                DOWNLOADING ...
        ''' % sub
@@ -152,6 +224,5 @@ def get_subs(query_str):
                        print 'No Content-Type!'
 
 
-for q in sys.argv[1:]:
-       get_subs(q)
+get_subs(sys.argv[1], sys.argv[2:])