Newbie, list has no attribute iteritems

[rabad]

Hi,
I've created a custom filter based on HTMLParser, with the following
source:

class Filter(HTMLPars er):

def __init__(self, keyfile):
HTMLParser.__in it__(self)
mykwfile = open(keyfile, 'r')
self._keywords = []
for kw in mykwfile.read() .split('\n'):
self._keywords. append(kw)
print kw
mykwfile.close( )
self._toProcess = False
self.stack = []

def handle_starttag (self, tag, attrs):
if 'a' != tag:
self.stack.appe nd(self.__html_ start_tag(tag, attrs))
return
attrs = dict(attrs)
self._toProcess = True
for key in self._keywords:
if 'a' == tag:
p = re.compile(key, re.IGNORECASE)
if 'href' in attrs:
attrs['href'] = p.sub(r'XXXXX', attrs['href'])
self.stack.appe nd(self.__html_ start_tag(tag, attrs))

def handle_startend tag(self, tag, attrs):
if 'img' != tag and 'meta' != tag:
self.stack.appe nd(self.__html_ startend_tag(ta g, attrs))
return
attrs = dict(attrs)
self._toProcess = True
for key in self._keywords:
p = re.compile(key, re.IGNORECASE)
if 'img' == tag:
if 'src' in attrs:
attrs['src'] = p.sub(r'XXXXX', attrs['src'])
if 'alt' in attrs:
attrs['alt'] = p.sub(r'XXXXX', attrs['alt'])
if 'meta' == tag:
if 'description' in attrs:
attrs['description'] =
p.sub(r'XXXXX', attrs['description'])
if 'content' in attrs:
attrs['content'] =
p.sub(r'XXXXX', attrs['content'])
if 'meta' == tag or 'img' == tag:
self._toProcess = False
self.stack.appe nd(self.__html_ startend_tag(ta g, attrs))

def handle_endtag(s elf, tag):
self.stack.appe nd(self.__html_ end_tag(tag))
if self._toProcess :
self._toProcess = False

def handle_data(sel f, data):
if self._toProcess :
for key in self._keywords:
p = re.compile(key, re.IGNORECASE)
data = p.sub(r'XXXXX', data)
self.stack.appe nd(data)

def __html_start_ta g(self, tag, attrs):
return '<%s%s>' % (tag, self.__html_att rs(attrs))

def __html_startend _tag(self, tag, attrs):
return '<%s%s/>' % (tag, self.__html_att rs(attrs))

def __html_end_tag( self, tag):
return '</%s>' % (tag)

def __html_attrs(se lf, attrs):
_attrs = ''
if attrs:
_attrs = ' %s' % (' '.join([('%s="%s"' % (k,v)) for k,v in
attrs.iteritems ()]))
return _attrs

But when I use it, it gives me the following error message:
ERROR Processor exception: AttributeError: 'list' object has no
attribute 'it
eritems'
Traceback (most recent call last):
File "d:\esp\lib\pyt hon2.3\processo rs\DocDumpF.py" , line 87, in
Process
p.feed(document .GetValue("data "))
File "HTMLParser.py" , line 108, in feed
File "HTMLParser.py" , line 148, in goahead
File "HTMLParser.py" , line 281, in parse_starttag
File "d:\esp\lib\pyt hon2.3\processo rs\DocDumpF.py" , line 121, in
handle_startt
ag
self.stack.appe nd(self.__html_ start_tag(tag, attrs))
File "d:\esp\lib\pyt hon2.3\processo rs\DocDumpF.py" , line 167, in
__html_start_
tag
return '<%s%s>' % (tag, self.__html_att rs(attrs))
File "d:\esp\lib\pyt hon2.3\processo rs\DocDumpF.py" , line 178, in
__html_attrs
_attrs = ' %s' % (' '.join([('%s="%s"' % (k,v)) for k,v in
attrs.iteritems ()
]))

Anybody knows why it says attrs is not a list element?
Thanks,
Rubén

[Justin Ezequiel]

def handle_starttag (self, tag, attrs): # <-- attrs here is a
list
if 'a' != tag:
self.stack.appe nd(self.__html_ start_tag(tag, attrs)) #
<-- attrs here is still a list
return
attrs = dict(attrs) # <-- now attrs is a dictionary

[Bruno Desthuilliers]

rabad a écrit :

Hi,
I've created a custom filter based on HTMLParser, with the following
source:

(snip)

But when I use it, it gives me the following error message:
ERROR Processor exception: AttributeError: 'list' object has no
attribute 'iteritems'

(snip)

File "d:\esp\lib\pyt hon2.3\processo rs\DocDumpF.py" , line 178, in
__html_attrs
_attrs = ' %s' % (' '.join([('%s="%s"' % (k,v)) for k,v in
attrs.iteritems ()
]))

Anybody knows why it says attrs is not a list element?

Actually, what the traceback says is that
1/ attrs is a list object
2/ list objects have no attribute named iteritems

If you assumed it was a dict, then it's probably time to re-read
HTMLParser's doc. Else if you assumed list had an iteritems method, then
it's probably time to re-read the Python's tutorial !-)

IIRC, HTMLParser represents attributes as a list of (attrname, value)
pairs. If so (please check it out), your method should be rewritten as

return ' %s' % (' '.join(('%s="%s "') % attr for attr in attrs)
As a side note: __double_leadin g_undescores is probably a bit extrem.
The convention for implementation attributes is _single_leading _underscore.