PyUNO Sax Module

So I am working on a script that parse ODF's XML. I have no
problem  
running it on Python 2.4 (my system python). However I cant
seem to  
work it on Pyuno (OOo's Python).

The script uses saxutils from the sax module.  I made a diff
on both  
modules and couldnt find much difference. I wonder if anyone
here have  
used python's sax module and if they are able to run the  
'DefaultHandler' class from the saxutils library.

I have googled and found similar people with the problem. I
found that  
ContentHandler might be an alternative. I am also reading
the  
doucmentation however I can't see to find anything related
with the  
DefaultHandler. Maybe I am looking somewhere else, anywhere
here is  
the link to those modules:
PyUNO: http://www.python.org/doc/2.3.4/lib/module-xml.sax.html
Python2.4: http://www.python.org/doc/2.4.3/lib/module-xml.sax.html

------------------------------------------------------------
---------
To unsubscribe, e-mail: dev-unsubscribexml.openoffice.org
For additional commands, e-mail: dev-helpxml.openoffice.org

Alexandro,

In case this can help you, Here an example of a SAX parsing
python script 
using 'zipfile' and 'xml.sax'
It works fine with OpenOffice and Word XML/Word docx
formats.

I do not know what would be the benefits of using the PyUNO
module, but I 
would be pleased to learn more on this.

Laurent


#!/usr/bin/python
# -*- coding: latin-1 -*-
#-----------------------------------------------------------
------------------
# (#) <TRAMway> converter for WordML (XML and
docx) an OpenOffice.org 
file
#-----------------------------------------------------------
------------------
"""
Three mode usage are available
1/ Command line; use options and arguments following the
getopt standard
2/ CGI, install the script on a Web server
3/ Python API, call API from any other code

API:
xml2tr.convert(option_tag,argument_tab)
an option element is a couple (option name, option value)
Type -h option for option list
"""

__version__ = '$Id: xml2tr.py 40 2007-01-08 16:26:04Z
saintax $'

from xml.sax import make_parser, handler, saxutils,
parseString
import string
import sys
import re
import os
import md5
import getopt # command line
import cgi    # web
import zipfile 

def usage():
    op = {'h':['help',      'This text', False],
          'v':['version',   'Display Tool Version', False],
          'o':['output=',   'Ouput in a file instead of
standard output', 
False],
          }

    sys.stderr.write("""USAGE: %s [options]
input_file...n
options:
-i string
--identification string
tDefines the regular expression matching the name of the
paragraph style 
to capture Requirement Identification.
tDefault string is 'reqidentification'n
-d string
--description string
tDefines the regular expression matching the name of the
paragraph style 
to capture Requirement Description.
tDefault string is 'reqdescription'n
-r string
--reference string
tDefines the regular expression matching the name of the
paragraph style 
to capture Requirement Reference.
tDefault string is 'reqreference'n
-a string
--attribute string
tDefines the regular expression matching the name of the
paragraph style 
to capture Requirement Reference.
tDefault string is 'reqreference'n
-p string
--prefix string
tAdd a prefix string to all detected requirement ids.
tDefault string is emptyn
-o string
--output string
tSpecifies an output file name.
tDefault is the standard output.n
Note:
nregular expression is not case sensitive and style name
can have prefix 
or suffix
input_file:
tWord 2003 XML or Opendocument file format
(WordML)n"""%(sys.argv[0]))
    sys.exit()

#__________#
class param:
    def __init__(self,val):
        self.value = val
 
#____________________#
class cmdFieldStorage:
    def __init__(self):
        self.dico  = {}

    def has_key(self, key):
        return self.dico.has_key(key)
 
    def add(self,key,val):
        self.dico[key] = param(val)

    def addlist(self,key,l):
        self.dico[key] = l

    def __getitem__(self, key):
        return self.dico[key]

    def getlist(self, key):
        if self.dico.has_key(key):
            return self.dico[key]
        else:
            return []

#__________________#
def header_appli(sum):
    return """<?xml
version="1.0" encoding="iso-8859-1"
standalone="yes"?>
<?topcased-application progid="TRAMway"?>
<?xml-stylesheet href="neutral.css"
type="text/css"?>
<tramway xmlns="http://www.w3.org/
1999/xhtml" md5="%s">
<title>TRAMway converter [%s] 
%s</title>n"""%(sum,os.path.basename(s
ys.argv[0]),__version__)

#__________________#
def footer_appli():
    return '</tramway>n'

#_______________________________________# 
class docHandler(handler.ContentHandler):
    def __init__(self,dico,fdo):
        self.intext = False
        self.buffer = ''
        self.style = ''
        self.id = ''
        self.sd = ''
        self.seen_id = {}
        self.reflist = ''
        self.desc = ''
        self.fo = dico
        self.keywords = '([: 
#\/^$,;?!&)({}=+|]|delete|derived|double|known|bad_id|co
vered|refined|resolved)'
        self.section = ''
        self.level = {}
        #self.secid = []
        self.attribute = {}
        self.o = fdo
        self.tl = {}
        self.auto = {}

    def startDocument(self):
        #print self.fo['i'].value
        pass

    def endDocument(self):
        self.record()
        self.newsection(1,True)
 
    def characters(self,str):
        if self.intext:
            str = saxutils.escape(str)
            self.buffer += str.encode('latin-1','replace')
 
    def startElement(self, name, attrs):
        if name == 'style:style':
            Pauto =
attrs.get('style:name').encode('latin-1')
            if re.match(r'Pd+', Pauto):
                self.auto[Pauto] = 
attrs.get('style:parent-style-name').encode('latin-1')
 
        if name == 'text:p':
            self.intext = True
            self.buffer = ''
            self.style =
attrs.get('text:style-name').encode('latin-1')
            if self.auto.has_key(self.style):
                self.style = self.auto[self.style]
        if name == 'w:t':
            self.intext = True
        elif name == 'w:p':
            self.style = 'none'
            self.buffer = ''
        elif name == 'w:pStyle':
            self.style =
attrs.get('w:val').encode('latin-1')
        elif name == 'w:style':
            self.section =
attrs.get('w:styleId').encode('latin-1')
        elif name == 'w:outlineLvl':
            self.level[self.section] = 
int(attrs.get('w:val').encode('latin-1'))+1

    def endElement(self, name):
        self.intext = False
        self.buffer = string.strip(self.buffer)
        if ((name == 'w:p') or (name == 'text:p')) and
self.buffer:
            if re.match(r'.*%s'%(self.fo['i'].value),
self.style, 
re.IGNORECASE):
                self.record() # record last requirement
                # new req
                self.desc = ''

                mat =
re.match(r'^s*(.+)s*:s*(.+)s*$',self.buffer)
                self.id = self.fo['p'].value
                if mat:
                    self.id += mat.group(1)
                    self.sd = ' 
shortdescription=%s'%(saxutils.quoteattr(mat.group(2))) 
                else:
                    self.id += self.buffer
                    self.sd = ''
 
                if self.seen_id.has_key(self.id):
                    sys.stderr.write('ERROROUBLE:%s
'%(self.id))
                    self.id += '_DEFINED_SEVERAL_TIMES'
                else:
                    self.seen_id[self.id] = True
            elif re.match(r'.*%s'%(self.fo['d'].value),
self.style, 
re.IGNORECASE):
                self.desc += 'n' + self.buffer
            elif re.match(r'.*%s'%(self.fo['r'].value),
self.style, 
re.IGNORECASE):
                if self.reflist:
                    self.reflist += '|' + self.buffer 
                else:
                    self.reflist = self.buffer
            elif re.match(r'.*%s'%(self.fo['a'].value),
self.style, 
re.IGNORECASE):
                mat =
re.match(r'^s*(.+)s*:s*(.+)s*$',self.buffer)
                if mat:
                    self.attribute[mat.group(1)] =
mat.group(2)
            elif self.level.has_key(self.style):
                self.newsection(self.level[self.style])

    def newsection(self,lvl,end=False):
        for i in self.tl.keys():
            if i >= lvl:
                self.o[0] += '</section>n'
                del self.tl[i]
        self.tl[lvl] = True
        if not end:
            #self.o[0] += '<section level="%d"

shortdescription=%s>n'%(lvl,saxutils.quoteattr(self.buff
er))
            self.o[0] += '<section 
shortdescription=%s>n'%(saxutils.quoteattr(self.buffer))


    def record(self):
        if re.match(r'.*#.*'+self.keywords+'.*', self.id,
re.IGNORECASE):
            sys.stderr.write('ERROR:BAD_ID:%s'%(self.id))
        elif self.id and re.match(r'^[swd_-]*$', self.id,

re.IGNORECASE):
            self.o[0] += 'n<requirement 
id=%s%s>n'%(saxutils.quoteattr(self.id),self.sd)
            self.o[0] += self.desc + 'n'
            if self.attribute:
                for i in self.attribute.keys():
                    self.o[0] += '<attribute name=%s 
value=%s/>n'%(saxutils.quoteattr(i),saxutils.quoteattr(s
elf.attribute[i]))
            if self.reflist:
                for i in self.reflist.split('|'):
                    self.o[0] += '<reference
type="cover" id="%s"/>n' 
%(i)
            self.o[0] += '</requirement>n'
        self.id = ''
        self.desc = ''
        self.reflist = ''
        self.attribute = {}
 
#__________________#
def get_cgi_param():
    return """<form
enctype="multipart/form-data"
method="post">
    <p> Input file <input type="file"
name="arg"/></p>
    <p> Option -i <input name="i"
value="reqidentification"/></p>
    <p> Option -d <input name="d"
value="reqdescription"/></p>
    <p> Option -r <input name="r"
value="reqreference"/></p>
    <p> Option -a <input name="a"
value="reqattribute"/></p>
    <input type="submit"
value="send"/>
    </form>"""

#____________________#
def convert(opts,str):
 
    op = {'h':['help',      'This text', False],
          'v':['version',   'Display Tool Version', False],
          'w':['web',       'HTML format output instead of
text', False],
          'o':['output=',   'Ouput in a file instead of
standard output', 
False],
          }

    sys.stderr = sys.stdout
    acgi = False
    if False and os.environ.has_key('HTTP_USER_AGENT') and
(opts==[]):
        acgi = True
        fo = cgi.FieldStorage()
        cgi_header = 'Content-type: application/xhtml+xml'
    else:
        fo = cmdFieldStorage()
        fo.add('i','(reqidentification|id)')
        fo.add('d','(reqdescription|requirement)')
        fo.add('r','reqreference')
        fo.add('a','reqattribute')
        fo.add('p','')

    fd = sys.stdout

    for o, a in opts:
        if o in ('-h', '--help'):
            usage()
        elif o in ('-v', '--version'): 
            print '%s v %s'%(sys.argv[0],__version__)
            sys.exit()
        elif o in ('-i', '--identification'):
            fo['i'].value = a
        elif o in ('-d', '--description'):
            fo['d'].value = a
        elif o in ('-r', '--reference'):
            fo['r'].value = a
        elif o in ('-a', '--attribute'):
            fo['a'].value = a
        elif o in ('-p', '--prefix'):
            fo['p'].value = a
        elif o in ('-o', '--output'):
            fd = open (a,'w')
            sys.stdout = fd
    sum = md5.new()
    sum.update(str)
    out = ''
    if acgi:
        out += cgi_header+'nn'
    out += header_appli(sum.hexdigest())
 
    if acgi and (not fo.has_key('i')):
        out += get_cgi_param()

    #if fo.has_key('arg'):
    #    args.append(fo['arg'].file)
 
    testout = ['']

    if str:
        dh = docHandler(fo,testout)
        parseString(str,dh)
 
    out += testout[0]
    out += footer_appli()
 
    if fd != sys.stdout:
        fd.close()
 
    return out

#_________________________#
def convert_zip(opts,f,content):
    zip = zipfile.ZipFile(f)
    print convert(opts,zip.read(content))
    zip.close()
 
#________________________#
if __name__ == '__main__':
    try:
        opts, args = getopt.getopt(
            sys.argv[1:], 'hi:d:r:a:p:o:', 
            ['help', 
'identification=','description=','reference=','attribute=','
prefix','output='])
    except getopt.GetoptError:
        usage()

    str,typ = '',''
    if (len(args) == 1) and os.path.isfile(args[0]):
        if args[0][-4:] == '.odt':
            convert_zip(opts,args[0],'content.xml')
        elif args[0][-5:] == '.docx':
            convert_zip(opts,args[0],'word/document.xml')
        else:
            str = open(args[0]).read()
            mat =
re.match(r'.*progid=(.*)?>.*$',str.split('n')[1])
            if mat:
                typ = mat.group(1)[1:-1]
            if typ == 'Word.Document':
                sys.stderr.write('word')
                print convert(opts,str)
            elif typ == 'TRAMway':
                sys.stderr.write('TRAMway')
                print str
            else:
                sys.stderr.write('ERROR Not supported
format')  

    # Test
    # example of API call without using getopt
    #print 
xml2tr.convert([('-i','requirementId'),('-d','requirementTex
t')],[sys.argv[1]])
 








Laurent



Alexandro Colorado <jzaopenoffice.org> 
29/03/2007 20:11
Veuillez répondre à
devxml.openoffice.org


A
devxml.openoffice.org
cc

Objet
[xml-dev] PyUNO Sax Module





So I am working on a script that parse ODF's XML. I have no
problem
running it on Python 2.4 (my system python). However I cant
seem to
work it on Pyuno (OOo's Python).

The script uses saxutils from the sax module.  I made a diff
on both
modules and couldnt find much difference. I wonder if anyone
here have
used python's sax module and if they are able to run the
'DefaultHandler' class from the saxutils library.

I have googled and found similar people with the problem. I
found that
ContentHandler might be an alternative. I am also reading
the
doucmentation however I can't see to find anything related
with the
DefaultHandler. Maybe I am looking somewhere else, anywhere
here is
the link to those modules:
PyUNO: http://www.python.org/doc/2.3.4/lib/module-xml.sax.html
Python2.4: http://www.python.org/doc/2.4.3/lib/module-xml.sax.html

------------------------------------------------------------
---------
To unsubscribe, e-mail: dev-unsubscribexml.openoffice.org
For additional commands, e-mail: dev-helpxml.openoffice.org