~bzr-pqm/bzr/bzr.dev

« back to all changes in this revision

Viewing changes to bzrlib/xml.py

  • Committer: Martin Pool
  • Date: 2005-05-03 08:00:27 UTC
  • Revision ID: mbp@sourcefrog.net-20050503080027-908edb5b39982198
doc

Show diffs side-by-side

added added

removed removed

Lines of Context:
 
1
#! /usr/bin/env python
1
2
# -*- coding: UTF-8 -*-
2
3
 
3
4
# This program is free software; you can redistribute it and/or modify
19
20
# "XML is like violence: if it doesn't solve your problem, you aren't
20
21
# using enough of it." -- various
21
22
 
22
 
# importing this module is fairly slow because it has to load several
23
 
# ElementTree bits
24
23
 
25
 
from bzrlib.trace import mutter, warning
 
24
__copyright__ = "Copyright (C) 2005 Canonical Ltd."
 
25
__author__ = "Martin Pool <mbp@canonical.com>"
26
26
 
27
27
try:
28
 
    from cElementTree import (ElementTree, SubElement, Element,
29
 
                              XMLTreeBuilder, fromstring, tostring)
 
28
    from cElementTree import Element, ElementTree, SubElement
30
29
except ImportError:
31
 
    mutter('WARNING: using slower ElementTree; consider installing cElementTree'
32
 
           " and make sure it's on your PYTHONPATH")
33
 
    from util.elementtree.ElementTree import (ElementTree, SubElement,
34
 
                                              Element, XMLTreeBuilder,
35
 
                                              fromstring, tostring)
36
 
 
37
 
from bzrlib.errors import BzrError
38
 
 
39
 
 
40
 
class Serializer(object):
41
 
    """Abstract object serialize/deserialize"""
42
 
    def write_inventory(self, inv, f):
43
 
        """Write inventory to a file"""
44
 
        elt = self._pack_inventory(inv)
45
 
        self._write_element(elt, f)
46
 
 
47
 
    def write_inventory_to_string(self, inv):
48
 
        return tostring(self._pack_inventory(inv)) + '\n'
49
 
 
50
 
    def read_inventory_from_string(self, xml_string):
51
 
        return self._unpack_inventory(fromstring(xml_string))
52
 
 
53
 
    def read_inventory(self, f):
54
 
        return self._unpack_inventory(self._read_element(f))
55
 
 
56
 
    def write_revision(self, rev, f):
57
 
        self._write_element(self._pack_revision(rev), f)
58
 
 
59
 
    def write_revision_to_string(self, rev):
60
 
        return tostring(self._pack_revision(rev)) + '\n'
61
 
 
62
 
    def read_revision(self, f):
63
 
        return self._unpack_revision(self._read_element(f))
64
 
 
65
 
    def read_revision_from_string(self, xml_string):
66
 
        return self._unpack_revision(fromstring(xml_string))
67
 
 
68
 
    def _write_element(self, elt, f):
69
 
        ElementTree(elt).write(f, 'utf-8')
 
30
    from elementtree.ElementTree import Element, ElementTree, SubElement
 
31
 
 
32
import os, time
 
33
from trace import mutter
 
34
 
 
35
class XMLMixin:
 
36
    def to_element(self):
 
37
        raise Exception("XMLMixin.to_element must be overridden in concrete classes")
 
38
    
 
39
    def write_xml(self, f):
 
40
        ElementTree(self.to_element()).write(f, 'utf-8')
70
41
        f.write('\n')
71
42
 
72
 
    def _read_element(self, f):
73
 
        return ElementTree().parse(f)
74
 
 
75
 
 
76
 
# performance tuning for elementree's serialiser. THis should be
77
 
# sent upstream - RBC 20060523.
78
 
# the functions here are patched into elementree at runtime.
79
 
import elementtree.ElementTree
80
 
import re
81
 
escape_re = re.compile("[&'\"<>]")
82
 
escape_map = {
83
 
    "&":'&amp;',
84
 
    "'":"&apos;", # FIXME: overkill
85
 
    "\"":"&quot;",
86
 
    "<":"&lt;",
87
 
    ">":"&gt;",
88
 
    }
89
 
def _escape_replace(match, map=escape_map):
90
 
    return map[match.group()]
91
 
 
92
 
def _escape_attrib(text, encoding=None, replace=None):
93
 
    # escape attribute value
94
 
    try:
95
 
        if encoding:
96
 
            try:
97
 
                text = elementtree.ElementTree._encode(text, encoding)
98
 
            except UnicodeError:
99
 
                return elementtree.ElementTree._encode_entity(text)
100
 
        if replace is None:
101
 
            return escape_re.sub(_escape_replace, text)
102
 
        else:
103
 
            text = replace(text, "&", "&amp;")
104
 
            text = replace(text, "'", "&apos;") # FIXME: overkill
105
 
            text = replace(text, "\"", "&quot;")
106
 
            text = replace(text, "<", "&lt;")
107
 
            text = replace(text, ">", "&gt;")
108
 
            return text
109
 
    except (TypeError, AttributeError):
110
 
        elementtree.ElementTree._raise_serialization_error(text)
111
 
 
112
 
elementtree.ElementTree._escape_attrib = _escape_attrib
113
 
 
114
 
escape_cdata_re = re.compile("[&<>]")
115
 
escape_cdata_map = {
116
 
    "&":'&amp;',
117
 
    "<":"&lt;",
118
 
    ">":"&gt;",
119
 
    }
120
 
def _escape_cdata_replace(match, map=escape_cdata_map):
121
 
    return map[match.group()]
122
 
 
123
 
def _escape_cdata(text, encoding=None, replace=None):
124
 
    # escape character data
125
 
    try:
126
 
        if encoding:
127
 
            try:
128
 
                text = elementtree.ElementTree._encode(text, encoding)
129
 
            except UnicodeError:
130
 
                return elementtree.ElementTree._encode_entity(text)
131
 
        if replace is None:
132
 
            return escape_cdata_re.sub(_escape_cdata_replace, text)
133
 
        else:
134
 
            text = replace(text, "&", "&amp;")
135
 
            text = replace(text, "<", "&lt;")
136
 
            text = replace(text, ">", "&gt;")
137
 
            return text
138
 
    except (TypeError, AttributeError):
139
 
        elementtree.ElementTree._raise_serialization_error(text)
140
 
 
141
 
elementtree.ElementTree._escape_cdata = _escape_cdata
 
43
    def read_xml(cls, f):
 
44
        return cls.from_element(ElementTree().parse(f))
 
45
 
 
46
    read_xml = classmethod(read_xml)
 
47
 
 
48
 
 
49