Improve field boundary detection

[software/subdivxget.git] / subdivxget

diff --git a/subdivxget b/subdivxget
index b5d01ff40b29e3e17fcb7826779fba85d011cf87..4ebcca5a71abf2bd94aa04b9c93b1003af36f240 100755 (executable)
--- a/subdivxget
+++ b/subdivxget
@@ -43,6 +43,8 @@ class SubDivXHTMLParser(HTMLParser.HTMLParser):
                self.parsing = False
                self.subs = []
                self.attr = None
                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):
                self.in_script_style = False
 
        def handle_starttag(self, tag, attrs):


@@ -61,12 +63,17 @@ class SubDivXHTMLParser(HTMLParser.HTMLParser):
                                self.parsing = True
                        elif attrs.get('id') == 'buscador_detalle_sub':
                                self.attr = 'desc'
                                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':
                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']
                        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):
                        self.depth += 1
 
        def handle_endtag(self, tag):


@@ -74,7 +81,12 @@ class SubDivXHTMLParser(HTMLParser.HTMLParser):
                        if tag == 'script' or tag == 'style':
                                self.in_script_style = False
                                return
                        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
 
                if self.depth == 0:
                        self.parsing = False
 


@@ -87,15 +99,23 @@ class SubDivXHTMLParser(HTMLParser.HTMLParser):
                if self.in_script_style:
                        return
                if self.attr is not None and data:
                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 = data[:-1].lower()

+                       self.attr_depth = 0
+                       self.cur[self.attr] = ''

                elif data == 'Subido por:':
                        self.attr = 'autor'
                elif data == 'Subido por:':
                        self.attr = 'autor'

+                       self.attr_depth = 0
+                       self.cur[self.attr] = ''

                elif data == 'el':
                        self.attr = 'fecha'
                elif data == 'el':
                        self.attr = 'fecha'

+                       self.attr_depth = 0
+                       self.cur[self.attr] = ''

 
 
 def subdivx_get_subs(query_str):
 
 
 def subdivx_get_subs(query_str):


@@ -126,7 +146,7 @@ def get_subs(query_str):
 
        for sub in subdivx_get_subs(query_str):
                print '''\
 
        for sub in subdivx_get_subs(query_str):
                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
          %(desc)s
                DOWNLOADING ...
        ''' % sub