Experience with SOAP?

[bercobeute]

I'm thinking about using SOAP for my application, but therefore I need a SOAP processor. Anybody here that has successfully used SOAP messaging in their application?

[aaaaapo]

I'm thinking about using SOAP for my application - -

With full SOAP headers etc? What a waste of expensive bandwidth! :-)

but therefore I need a SOAP processor. Anybody here that has successfully used SOAP messaging in their application?

I'm using sgmllib.py to parse simple XML-like dataformat. Data looks like this:

<img time='1156230837' id='79/94/000005215' sender='Aapo' title='Stikkaa vodaa'></img>

And my parser returns a dictionary, where keys are tag names and values lists of dictionaries like this:

{'img': [[1156230837, {u'title': u'Stikkaa vodaa', u'sender': u'Aapo', u'id': u'79/94/000005215', u'time': u'1156230837'}], 
         [1156237022, {u'title': u'Hj l\xe4hettipy\xf6r\xe4', u'sender': u'Teemu', u'id': u'eb/cc/000005216', u'time': u'1156237022'}],
         ...
{'msg': [[...

Above is easy to use in Pys60 program later.

If somebody is interested in this, I can post those few lines of Python code to this forum.

[simo.salminen]

Please post away, maybe someone is interested.

[aaaaapo]

Ok, here it goes. If copypaste to your editor does not work, try this temporary link:
http://plok.in/tmp/PlokParser.py
Just running it ("/path/to/python2.2 PlokParser.py") should show some results. If this does not work and you need more instructions how to test PlokParser, please reply to this thread. NOTE: sgmllib.py and markupbase.py are needed, but usually they are found by python if python installation is successful.

Suggestions and improvements are appreciated.

# -*- coding: iso-8859-15 -*-
# $Id: PlokParser.py 213 2006-08-23 19:26:18Z arista $
# Copyright Aapo Rista and others 2006
# License GPL

import sgmllib
import urllib
import re

class PlokParser(sgmllib.SGMLParser):
    """A simple parser class. See exampledata.xml for data details."""

    # Characters which will be encoded to entities
    chartable = {
       u'&':u'&',
       u'"':u'"',
#       u"'":u"'", # This entity (single quote) went broken in nokia's forum-software
# Try to replace second single quote with & #039; (without space inside)
       u'<':u'&lt;',
       u'>':u'&gt;',
    }
    # Entities which will be decoded to characters (see init)
    entitytable = {}

    def __init__(self, verbose=0):
        """Initialise an object, passing 'verbose' to the superclass."""
        sgmllib.SGMLParser.__init__(self, verbose)
        # This will hold parsed tags until parser is clear()'ed
        self.lists = {}
        # Construct entitytable by swapping keys and values in chartable
        for (key, value) in self.chartable.items():
            self.entitytable[value] = key
        charlist = self.chartable.keys()
        # re which matches all chars which must be entities in XML
        self.char_re = re.compile("[%s]" % ("".join(charlist)))
        # re which matches all possible entities you'd like to decode
        self.ent_re = re.compile(r'&.*?;')

    def parse(self, s):
        """Parse the given string 's'."""
        self.feed(s)
        self.close()
        
    def clear(self):
        """Empty all lists."""
        self.lists = {}

    def decode_entity(self, match):
        """Callback function for re.sub in unknown_starttag().
        Replaces entities found with character equivalent in entitytable."""
        entity = match.group()
        if self.entitytable.has_key(entity):
            return self.entitytable[entity]
        else: 
            return entity

    def encode_entity(self, match):
        """Callback function for re.sub in create_tag().
        Replaces characters found in chartable with entities."""
        char = match.group()
        if self.chartable.has_key(char):
            return self.chartable[char]
        else: 
            return char

    def unknown_starttag(self, tag, attrs):
        """Handles all tags and puts their attributes into the appropriate list.
        Decodes also html entities."""
        attr_dict = {}
        for i in range(0, len(attrs)):
            attr_dict[attrs[i][0]] = self.ent_re.sub(self.decode_entity, attrs[i][1])
        if attr_dict.has_key("time"):
            # I hope sortfield will not confuse folks too much
            # It is used to keep lists in order (see python's sort())
            sortfield = int(attr_dict["time"])
        else:
            sortfield = 0
        if len(attr_dict) > 0:
            if not self.lists.has_key(tag):
                self.lists[tag] = []
            self.lists[tag].append([sortfield, attr_dict])

    def create_tag(self, tag, attr_dict):
        """Create XML-like tag including attributes found in attr_dict"""
        attr_list = []
        for k in attr_dict.keys():
            val = self.char_re.sub(self.encode_entity, attr_dict[k])
            attr_list.append(u'%s="%s"' % (k, val))
        return u"<%s %s></%s>" % (tag, " ".join(attr_list), tag)
        
if __name__ == "__main__":
    import time
    # Initialize parser
    pp = PlokParser()
    # Populate tags-list
    tags1 = []
    tags1.append(pp.create_tag(u"img", {u"id":u"1234", u"name":u"foo.jpg", u"title":u"Mom & Dad"}))
    tags1.append(pp.create_tag(u"img", {u"id":u"42", u"name":u"bar.jpg", u"title":u"Mom's shoe"}))
    tags1.append(pp.create_tag(u"msg", {u"id":u"3187", u"sender":u"Aapo", u"text":u"Ugh"}))
    tags1.append(pp.create_tag(u"msg", {u"id":u"3188", u"sender":u"Teemu", u"text":u"<:->"}))
    # Create single xml-string from tags-list
    xml1 = u"\r\n".join(tags1)
    # Parse xml, parsed data will be found in dictionary pp.lists
    pp.parse(xml1)
    list1 = pp.lists.copy()
    tags2 = []
    for (tag, data) in list1.items():
        for i in range(len(data)):
            tags2.append(pp.create_tag(tag, data[i][1]))
    xml2 = u"\r\n".join(tags2)
    pp.clear()
    pp.parse(xml2)
    list2 = pp.lists.copy()
    print xml1
    print
    print list1
    print
    print xml2
    print
    print list2