Reputation: 3059
There is a brief explanation of validation based on XML here. I am trying to parse an XML file that refers to nested DTD's ie: XML file refers to a DTD, which refers to other DTD's.
The error I get is Namespace prefix SomeNameSpace on Config is not defined. All I am trying to do is parse the xml using etree.parse which is an API on lxml. My question is:
I see similar questions, but nothing that answers this question.
Upvotes: 0
Views: 1137
Reputation: 1455
A while back I tried to do something similar and wasn't able to find a solution. I finally wrote the script below which opens the XML file and looks for a DTD using a regex. It also has an override to take the DTD path on the command line, which was a requirement I had.
If lxml handles nested DTDs then the code below should work for you.
To be honest I thought it was a bit of a hack to read the file myself, but it was the only way I found.
import re
import sys
import os.path
import codecs
from lxml import etree
def main(args):
if len(args)<1:
print("Not enough arguments given. Expected:")
print("\tvalidatexml <xml file name> [<dtd file name>]\n")
exit(1)
dtdRe = re.compile('.*<!DOCTYPE .* ["\'](.*\.dtd)["\']>.*')
theDtd = None
inFile = args[0]
fdir = os.path.abspath(os.path.dirname(inFile))
if len(args)==2:
theDtd = os.path.abspath(args[1])
else:
with codecs.open(args[0], 'r', 'utf-8') as inf:
for ln in inf:
mtch = dtdRe.match(ln)
if mtch:
if os.path.isabs(mtch.group(1)):
theDtd = mtch.group(1)
else:
theDtd = os.path.abspath(fdir + '/' + mtch.group(1))
break
if theDtd is None:
print("No DTD specified!")
exit(2)
if not os.path.exists(theDtd):
print("The DTD ({}) does not exist!".format(theDtd))
exit(3)
print('Using DTD:', theDtd)
parser = etree.XMLParser(dtd_validation=True)
dtd = etree.DTD(open(theDtd))
tree = etree.parse(args[0])
valid = dtd.validate(tree)
if (valid):
print("XML was valid!")
else:
print("XML was not valid:")
print(dtd.error_log.filter_from_errors())
if __name__ == '__main__':
main(sys.argv[1:])
Upvotes: 1
Reputation: 1938
Can you try parse by Beautiful Soup? Errors still is exists?
Upvotes: 0