Ignore script and style tags

[software/subdivxget.git] / subdivxget

diff --git a/subdivxget b/subdivxget
index 1aecc66b904811165ec5b5f1bfd44798cd551d42..b5d01ff40b29e3e17fcb7826779fba85d011cf87 100755 (executable)
--- a/subdivxget
+++ b/subdivxget
@@ -43,6 +43,7 @@ class SubDivXHTMLParser(HTMLParser.HTMLParser):
                self.parsing = False
                self.subs = []
                self.attr = None
+               self.in_script_style = False
 
        def handle_starttag(self, tag, attrs):
                attrs = dict(attrs)
@@ -52,6 +53,9 @@ 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
@@ -67,6 +71,9 @@ class SubDivXHTMLParser(HTMLParser.HTMLParser):
 
        def handle_endtag(self, tag):
                if self.parsing:
+                       if tag == 'script' or tag == 'style':
+                               self.in_script_style = False
+                               return
                        self.depth -= 1
                if self.depth == 0:
                        self.parsing = False
@@ -75,6 +82,10 @@ class SubDivXHTMLParser(HTMLParser.HTMLParser):
                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
                        self.attr = None