def test_xml_cv():
s = open('./samples/cv.xml')
t = len(xml(tok(s)).children)
e = None
assert_not_equal(t, e)
def test_xml():
s = open('./samples/dbus.xml')
t = xml(tok(s))
e = Root(children=[
Instruction(instruction='<?xml version="1.0"?>'),
Text(text=' '),
Comment(comment='<!--*-nxml-*-->'),
Text(text='\n'),
Doctype(
doctype='<!DOCTYPE busconfig PUBLIC "-//freedesktop//DTD D-BUS '
'Bus Configuration 1.0//EN"\n "http://www.freedesktop.org/'
'standards/dbus/1.0/busconfig.dtd">'),
Text(text='\n\n'),
Comment(comment='<!--\n This file is part of systemd.\n\n systemd '
'is free software; you can redistribute it and/or modify '
'it\n under the terms of the GNU Lesser General Public '
'License as published by\n the Free Software Foundation; '
'either version 2.1 of the License, or\n (at your option) '
'any later version.\n-->'),
Text(text='\n\n'),
Tag(name='<busconfig>',
attrs=[],
children=[
Text(text='\n\n '),
Tag(name='<policy user="******">',
attrs=[],
children=[
Text(text='\n '),
Tag(name='<allow own="org.freedesktop.'
'timedate1"/>',
attrs=[],
children=[]),
Text(text='\n '),
Tag(name='<allow send_destination="org.'
'freedesktop.timedate1"/>',
attrs=[],
children=[]),
Text(text='\n '),
Tag(name='<allow receive_sender="org.'
'freedesktop.timedate1"/>',
attrs=[],
children=[]),
Text(text='\n ')
]),
Text(text='\n\n '),
Tag(name='<policy context="default">',
attrs=[],
children=[
Text(text='\n '),
Tag(name='<allow send_destination="'
'org.freedesktop.timedate1"/>',
attrs=[],
children=[]),
Text(text='\n '),
Tag(name='<allow receive_sender="org.'
'freedesktop.timedate1"/>',
attrs=[],
children=[]),
Text(text='\n ')
]),
Text(text='\n\n')
]),
Text(text='\n')
])
assert_equal(t, e)
def test_1():
'''bug'''
s = io.StringIO('<foo><bar>duh</bar></foo>')
t = xml(tok(s))
e = Root([Tag('<foo>', [], [Tag('<bar>', [], [Text('duh')])])])
assert_equal(t, e)
def test_malformed():
s = io.StringIO('<foo><bar>xxx</foo></bar>')
try:
t = xml(tok(s))
except MalformedXML:
pass
def test_0():
s = io.StringIO('<foo>x</foo>')
t = xml(tok(s))
e = Root([Tag('<foo>', [], [Text('x')])])
assert_equal(t, e)
import os
import click
import sax.tokenizer.gen as gt
import sax.tokenizer.loop as lt
src = './samples/dbus-systemd1.xml'
tsg = list(gt.tok(open(src)))
tsl = list(lt.tok(open(src)))
def k(n=32):
for k, t in tsl[:n]:
print(k, t)
diffs = [(g, l) for g, l in zip(tsg, tsl) if g != l]
'''
Intersting, parsers have different streams. Mostly out of
sync because of the selfclosing `tokenistic sugar` trick,
causing more tokens emitted by gen.tok:
<sctag/> -> (opening, sctag), (closing, sctag)
instead of
<sctag/> -> (selfclosing, sctag)
but may resolve the same trees when parsed.