|
| 1 | +# -*- coding: utf-8 -*- |
| 2 | +# |
| 3 | +# oxml/base.py |
| 4 | +# |
| 5 | +# Copyright (C) 2012, 2013 Steve Canny scanny@cisco.com |
| 6 | +# |
| 7 | +# This module is part of python-docx and is released under the MIT License: |
| 8 | +# http://www.opensource.org/licenses/mit-license.php |
| 9 | + |
| 10 | +""" |
| 11 | +Common and utility code used by other modules in the docx.oxml subpackage. |
| 12 | +""" |
| 13 | + |
| 14 | +from lxml import etree, objectify |
| 15 | + |
| 16 | + |
| 17 | +nsmap = { |
| 18 | + 'w': 'http://schemas.openxmlformats.org/wordprocessingml/2006/main', |
| 19 | +} |
| 20 | + |
| 21 | +# configure objectified XML parser |
| 22 | +_fallback_lookup = objectify.ObjectifyElementClassLookup() |
| 23 | +element_class_lookup = etree.ElementNamespaceClassLookup(_fallback_lookup) |
| 24 | +oxml_parser = etree.XMLParser(remove_blank_text=True) |
| 25 | +oxml_parser.set_element_class_lookup(element_class_lookup) |
| 26 | + |
| 27 | + |
| 28 | +# =========================================================================== |
| 29 | +# utility functions |
| 30 | +# =========================================================================== |
| 31 | + |
| 32 | +def _Element(tag, nsmap=None): |
| 33 | + return oxml_parser.makeelement(qn(tag), nsmap=nsmap) |
| 34 | + |
| 35 | + |
| 36 | +def nsdecls(*prefixes): |
| 37 | + return ' '.join(['xmlns:%s="%s"' % (pfx, nsmap[pfx]) for pfx in prefixes]) |
| 38 | + |
| 39 | + |
| 40 | +def oxml_fromstring(text): |
| 41 | + """``etree.fromstring()`` replacement that uses oxml parser""" |
| 42 | + return objectify.fromstring(text, oxml_parser) |
| 43 | + |
| 44 | + |
| 45 | +def oxml_tostring(elm, encoding=None, pretty_print=False, standalone=None): |
| 46 | + # if xsi parameter is not set to False, PowerPoint won't load without a |
| 47 | + # repair step; deannotate removes some original xsi:type tags in core.xml |
| 48 | + # if this parameter is left out (or set to True) |
| 49 | + objectify.deannotate(elm, xsi=False, cleanup_namespaces=False) |
| 50 | + # objectify.deannotate(elm, xsi=False, cleanup_namespaces=True) |
| 51 | + return etree.tostring(elm, encoding=encoding, pretty_print=pretty_print, |
| 52 | + standalone=standalone) |
| 53 | + |
| 54 | + |
| 55 | +def qn(tag): |
| 56 | + """ |
| 57 | + Stands for "qualified name", a utility function to turn a namespace |
| 58 | + prefixed tag name into a Clark-notation qualified tag name for lxml. For |
| 59 | + example, ``qn('p:cSld')`` returns ``'{http://schemas.../main}cSld'``. |
| 60 | + """ |
| 61 | + prefix, tagroot = tag.split(':') |
| 62 | + uri = nsmap[prefix] |
| 63 | + return '{%s}%s' % (uri, tagroot) |
| 64 | + |
| 65 | + |
| 66 | +def register_custom_element_class(tag, cls): |
| 67 | + """ |
| 68 | + Register *cls* to be constructed when the oxml parser encounters an |
| 69 | + element with matching *tag*. *tag* is a string of the form |
| 70 | + ``nspfx:tagroot``, e.g. ``'w:document'``. |
| 71 | + """ |
| 72 | + nspfx, tagroot = tag.split(':') |
| 73 | + namespace = element_class_lookup.get_namespace(nsmap[nspfx]) |
| 74 | + namespace[tagroot] = cls |
| 75 | + |
| 76 | + |
| 77 | +def _SubElement(parent, tag): |
| 78 | + return objectify.SubElement(parent, qn(tag), nsmap=nsmap) |
| 79 | + |
| 80 | + |
| 81 | +class OxmlBaseElement(objectify.ObjectifiedElement): |
| 82 | + """ |
| 83 | + Base class for all custom element classes, to add standardized behavior |
| 84 | + to all classes in one place. |
| 85 | + """ |
| 86 | + @property |
| 87 | + def xml(self): |
| 88 | + """ |
| 89 | + Return XML string for this element, suitable for testing purposes. |
| 90 | + Pretty printed for readability and without an XML declaration at the |
| 91 | + top. |
| 92 | + """ |
| 93 | + return oxml_tostring(self, encoding='unicode', pretty_print=True) |
0 commit comments