From d7de47c06770e55969fc6cfde6e3529b67021e02 Mon Sep 17 00:00:00 2001
From: jbj <devnull@localhost>
Date: Sun, 13 Jun 2004 17:10:18 +0000
Subject: [PATCH] Add libxml2 handler.

CVS patchset: 7313
CVS date: 2004/06/13 17:10:18
---
 tools/expatparser.py  | 25 +++++++++++----
 tools/libxmlparser.py | 88 +++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 106 insertions(+), 7 deletions(-)
 create mode 100644 tools/libxmlparser.py

diff --git a/tools/expatparser.py b/tools/expatparser.py
index aa75e5d..23e192c 100644
--- a/tools/expatparser.py
+++ b/tools/expatparser.py
@@ -2,13 +2,18 @@ import sys, xml.parsers.expat
 
 class RpmExpatParser:
     def __init__(self, fn):
-	self.f = open(fn)
+	try:
+	    self.f = open(fn)
+	except:
+	    print "unable to open %s" % (fn)
+	    return
 	self.p = xml.parsers.expat.ParserCreate()
 	self.p.StartElementHandler = self.start_element
 	self.p.EndElementHandler = self.end_element
 	self.p.CharacterDataHandler = self.char_data
 	self.n = 2
 	self.lvl = 0
+	self.cdata = 0
 
     def spew(self, l):
 	sys.stdout.write(l)
@@ -20,30 +25,36 @@ class RpmExpatParser:
     def start_element(self, name, attrs):
 	l = self.pad() + '<' + name
 	if attrs.has_key(u'name'):
-	    l = l + ' name=' + attrs[u'name']
+	    l = l + ' name="' + attrs[u'name'] + '"'
 	l = l + '>'
 	if self.lvl < 2:
-	    l = l + '\r\n'
+	    l = l + '\n'
 	self.spew(l)
 	self.lvl = self.lvl + 1
+	self.cdata = 1
     
     def end_element(self, name):
 	self.lvl = self.lvl - 1
 	l = '</' + name + '>'
 	if self.lvl < 2:
 	    l = self.pad() + l
-	l = l + '\r\n'
+	l = l + '\n'
 	self.spew(l)
+	self.cdata = 0
 
     def char_data(self, data):
-	if not data.isspace():
+	if self.cdata == 1:
+	    if not data.isspace():
+		self.cdata = 2
+	if self.cdata > 1:
 	    self.spew(data)
 
     def read(self, *args):
 	return self.f.read(*args)
 
     def ParseFile(self):
-	self.p.ParseFile(self)
+	return self.p.ParseFile(self)
 
 p = RpmExpatParser('time.xml')
-p.ParseFile()
+ret = p.ParseFile()
+print ret
diff --git a/tools/libxmlparser.py b/tools/libxmlparser.py
new file mode 100644
index 0000000..c816cb9
--- /dev/null
+++ b/tools/libxmlparser.py
@@ -0,0 +1,88 @@
+import sys, libxml2
+
+class RpmLibxml2Parser:
+    def __init__(self, fn):
+        try:
+	    self.p = libxml2.newTextReaderFilename(fn)
+        except:
+	    print "unable to open %s" % (fn)
+	    return
+	self.p.SetParserProp(libxml2.PARSER_VALIDATE, 1)
+	self.p.SetParserProp(libxml2.PARSER_SUBST_ENTITIES, 1)
+	self.n = 2
+	self.lvl = 0
+	self.cdata = 0
+
+    def spew(self, l):
+	sys.stdout.write(l)
+	sys.stdout.flush()
+
+    def pad(self):
+	return (' ' * (self.n * self.lvl))
+
+    def start_element(self, name, attrs):
+	l = self.pad() + '<' + name
+	if attrs.has_key(u'name'):
+	    l = l + ' name="' + attrs[u'name'] + '"'
+	l = l + '>'
+	if self.lvl < 2:
+	    l = l + '\n'
+	self.spew(l)
+	self.lvl = self.lvl + 1
+	self.cdata = 1
+    
+    def end_element(self, name):
+	self.lvl = self.lvl - 1
+	l = '</' + name + '>'
+	if self.lvl < 2:
+	    l = self.pad() + l
+	l = l + '\n'
+	self.spew(l)
+	self.cdata = 0
+
+    def char_data(self, data):
+	if self.cdata == 1:
+	    if not data.isspace():
+		self.cdata = 2
+	if self.cdata > 1:
+	    self.spew(data)
+
+    def processNode(self):
+#	self.n = self.p.Depth() + 1
+	if self.p.NodeType() == 1:	# Element
+	    name = self.p.Name()
+	    attrs = {}
+	    while self.p.MoveToNextAttribute():
+		attrs[self.p.Name()] = self.p.Value()
+	    self.start_element(name, attrs)
+	elif self.p.NodeType() == 3:	# Text within element
+	    self.char_data(self.p.Value())
+	elif self.p.NodeType() == 10:	# Start element
+	    self.char_data(self.p.Value())
+	elif self.p.NodeType() == 14:	# Text
+	    self.char_data(self.p.Value())
+	elif self.p.NodeType() == 15:	# End element
+	    self.end_element(self.p.Name())
+	else:
+	    print "%d %d %s %d %s" % (self.p.Depth(), self.p.NodeType(),
+                              self.p.Name(), self.p.IsEmptyElement(),
+                              self.p.Value())
+	    while self.p.MoveToNextAttribute():
+		print "-- %d %d (%s) [%s]" % (self.p.Depth(), self.p.NodeType(),
+                                          self.p.Name(), self.p.Value())
+
+    def read(self, *args):
+	return self.p.Read(*args)
+
+    def ParseFile(self):
+	ret = self.read()
+	while ret == 1:
+	    self.processNode()
+	    ret = self.read()
+	return ret
+
+fn = 'time.xml'
+p = RpmLibxml2Parser(fn)
+ret = p.ParseFile()
+if ret != 0:
+    print "Error parsing and validating %s" % (fn)
-- 
2.7.4