« back to all changes in this revision
Viewing changes to odftools/components/settings.py
- browse files at revision 33
- compare with another revision
- download diff
- download tarball
- view history from revision 33
- Committer: eric.talevich
- Date: 2007-07-17 01:32:15 UTC
- Revision ID: vcs-imports@canonical.com-20070717013215-zrfrecfiohvz7mg7
Start a class for each of the 5 XML-based document components, throw the files for components into a new subdirectory, chew up everything and make it ugly. HTML export is broken and I don't like the command-line options.
added
removed
odftools/components/settings.py
1
# -*- coding: iso-8859-15 -*-
2
3
"""Application-specific settings for the document."""
4
5
import os, sys
6
7
try:
8
import xml.etree.cElementTree as ET
9
except ImportError:
10
from elementtree.cElementTree import ElementTree as ET
11
12
13
# Prefix values with "application/vnd.oasis.opendocument." to get MIME types
14
15
odf_formats = {'odt':'text', 'ods':'spreadsheet', 'odp':'presentation',
16
'odg':'graphics', 'odc':'chart', 'odf':'formula', 'odi':'image',
17
'odm':'text-master', 'ott':'text-template',
18
'ots':'spreadsheet-template', 'otp':'presentation-template',
19
'otg':'graphics-template'}
20
21
odf_prefix = "application/vnd.oasis.opendocument."
22
23
24
# Exceptions for this module
25
26
class ReCompileError(Exception):
27
"""Thrown if regular expression cannot be compiled."""
28
pass
29
30
class PathNotFoundError(Exception):
31
"""Thrown if a file reference contains a nonexistant path."""
32
pass
33
34
# Main class
35
36
class Settings(object):
37
38
def __init__(self, text):
39
self.root = ET.fromstring(text)
40
41
42
43
# The Document tree and associated methods
44
45
class Document:
46
"""The ODF document object."""
47
# Map attribute names to file names
48
file_map = {'mimetype': 'mimetype',
49
'manifest': 'META-INF/manifest.xml',
50
'content': 'content.xml',
51
'styles': 'styles.xml',
52
'meta': 'meta.xml',
53
'settings': 'settings.xml'}
54
55
def __init__(self,
56
file='', # Document file name
57
mimetype='', # Mimetype string
58
manifest='', # Lists the contents of the ODF file
59
content='', # Content data (the text)
60
styles='', # Formatting data
61
meta='', # Metadata
62
settings='', # Application-specific data
63
additional={}, # Additional bundled files (e.g. images)
64
file_dates={} # File dates for all files and directories
65
):
66
67
# Get all method parameters
68
args = locals()
69
70
# Process all Document files
71
for key, filename in self.__class__.file_map.items():
72
if key not in args or 0 == len(args[key]):
73
setattr(self, key, '')
74
elif not filename or '.xml' != filename[-4:]:
75
setattr(self, key, args[key])
76
else:
77
try:
78
# Parse the XML string and set it as an ElementTree object
79
setattr(self, key, ET.XML(args[key]))
80
except Exception, e:
81
print >>sys.stderr, sysargs[key]
82
print >>sys.stderr, e
83
84
self.additional = additional
85
self.file_dates = file_dates
86
87
if not hasattr(self, 'file'):
88
self.file = None
89
90
def __del__(self): # XXX is this still necessary?
91
"""Clean up.
92
93
This was originally here to unlink each DOM component.
94
95
"""
96
for key in self.__class__.file_map:
97
attr = getattr(self, key)
98
if not isinstance(attr, basestring):
99
del attr
100
101
# ---------------------------
102
# Extract objects from the document
103
104
def getComponentAsString(self, component_name, #pretty_printing=False,
105
encoding=None):
106
"""Return document component as Unicode string."""
107
if component_name not in self.__class__.file_map:
108
return ""
109
filename = self.__class__.file_map[component_name]
110
attr = getattr(self, component_name)
111
if isinstance(attr, basestring):
112
return attr
113
#if pretty_printing:
114
# return attr.toprettyxml(encoding)
115
return ET.tostring(attr, encoding=encoding)
116
117
def getEmbeddedObjects(self, filter=None, ignore_case=False):
118
"""Return a dictionary of the objects embedded in the document.
119
120
A more general form of getImages. By default, this should return
121
all embedded objects; the list/dictionary can also be filtered
122
for a certain type, e.g. image files.
123
124
The filter currently supports UNIX glob patterns like "*a[bc]?.png"
125
and/or correct regular expressions like ".*a[bc].\.png$".
126
127
"""
128
# TODO: support other embedded objects
129
search = get_search_for_filter(filter, ignore_case)
130
return dict([(filename[9:], content)
131
for filename, content in self.additional.items()
132
if 'Pictures/' == filename[:9]
133
and search(filename[9:])])
134
135
def getElementsByType(self, elementtype):
136
"""Extract all elements of a given type from the document.
137
138
For example, formulas or code.
139
140
"""
141
pass
142
143
def getAuthor(self):
144
"""Return the author of this document if available."""
145
author = ''
146
if self.meta:
147
for node in self.meta.getElementsByTagName("dc:creator"):
148
if (node.firstChild.nodeType == node.TEXT_NODE) and node.firstChild.data:
149
author = node.firstChild.data
150
break
151
152
return author
153
154
def getExtension(self):
155
"""Return ODF extension for given mimetype."""
156
return get_extension(self.mimetype)
157
158
# ---------------------------
159
# Convert the document to other formats
160
161
def toXml(self, pretty_printing=False, encoding=None):
162
"""Return the content of the document as a XML Unicode string."""
163
if pretty_printing:
164
return self.content.toprettyxml(encoding)
165
return self.content.toxml(encoding)
166
167
def toText(self, skip_blank_lines=True):
168
"""Return the content of the document as a plain-text Unicode string."""
169
textlist = (node.text for node in self.content.getiterator()
170
if not skip_blank_lines or node.text)
171
return unicode(os.linesep).join(textlist)
172
173
def toHtml(self, title="", encoding="utf-8"):
174
"""Return an UTF-8 encoded HTML representation of the document."""
175
# TODO:
176
# First, convert to ET operations
177
# Then,
178
# - Scrape up meta tags and add to headnode
179
# '<meta http-equiv="content-type" content="text/html; charset=UTF-8">'
180
# '<meta type="Generator" content="python-odftools" />'
181
# - Title for the page, if applicable
182
# - Convert self.styles to CSS and add to headnode as a <style type="text/css"> element
183
# - see cssutils at the Python cheeseshop
184
# - Fix the unit test
185
#
186
# ENH:
187
# - Support encodings other than UTF-8, and maybe Unicode
188
# - Allow named elements
189
# - A more natural way of doing the doctype declaration, if possible
190
191
attrs_odf2html = {"style-name": "class"}
192
tags_odf2html = {
193
"a": "a",
194
"body": "body",
195
"p": "p",
196
"span": "span",
197
"table": "table",
198
"h": "h1",
199
"table-row": "tr",
200
"table-cell": "td",
201
"image": "img",
202
"list": "ol",
203
"list-item": "li" }
204
205
htmldoc = ET.Element("html")
206
headnode = ET.SubElement(htmldoc, "head")
207
titlenode = ET.SubElement(headnode, "title")
208
titlenode.text = title
209
# ENH: add meta etc. nodes to the head as needed
210
211
docbody = self.content.find("office:body")
212
if docbody:
213
bodynode = translate_nodes(docbody, tags_odf2html, attrs_odf2html)
214
else:
215
bodynode = ET.SubElement(htmldoc, "body")
216
217
doctypestr = '<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">\n'
218
htmlstr = ET.tostring(htmldoc, encoding=encoding) # .split("\n", 1)[1] # XXX kill 1st line?
219
return "\n".join((doctypestr, htmlstr))
220
221
def replace(self, search, replace):
222
"""Replace all occurences of search in content by replace.
223
224
Regular expressions are fully supported for search and replace.
225
226
Returns the number of replacements made.
227
228
"""
229
if not search:
230
return 0
231
232
import re, sre_constants
233
234
try:
235
_replace = re.compile(search).sub
236
search = lambda x, y: find(x, y)
237
except (sre_constants.error, TypeError), v:
238
print >>sys.stderr, 'Warning: could not compile regular expression:', v
239
return 0
240
241
count = 0
242
for node in self.content.getiterator():
243
if node.text:
244
try:
245
replaced = _replace(replace, node.text)
246
if replaced != node.text:
247
node.text = replaced
248
count += 1
249
except (sre_constants.error, TypeError), v:
250
print >>sys.stderr, 'Warning: could not compile regular expression:', v
251
return 0
252
return count
253
Loggerhead is a web-based interface for Breezy
Version: 2.0.1