| 1 | ############################################################################### |
|---|
| 2 | # |
|---|
| 3 | # Copyright (c) 2007, 2008 OpenHex SPRL. (http://openhex.com) All Rights |
|---|
| 4 | # Reserved. |
|---|
| 5 | # |
|---|
| 6 | # This program is free software; you can redistribute it and/or modify it under |
|---|
| 7 | # the terms of the GNU General Public License as published by the Free Software |
|---|
| 8 | # Foundation; either version 2 of the License, or (at your option) any later |
|---|
| 9 | # version. |
|---|
| 10 | # |
|---|
| 11 | # This program is distributed in the hope that it will be useful, but WITHOUT |
|---|
| 12 | # ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS |
|---|
| 13 | # FOR A PARTICULAR PURPOSE. See the GNU General Public License for more |
|---|
| 14 | # details. |
|---|
| 15 | # |
|---|
| 16 | # You should have received a copy of the GNU General Public License along with |
|---|
| 17 | # this program. If not, see <http://www.gnu.org/licenses/>. |
|---|
| 18 | # |
|---|
| 19 | ############################################################################### |
|---|
| 20 | |
|---|
| 21 | __metaclass__ = type |
|---|
| 22 | |
|---|
| 23 | import os |
|---|
| 24 | import re |
|---|
| 25 | import md5 |
|---|
| 26 | import urllib |
|---|
| 27 | import zipfile |
|---|
| 28 | from cStringIO import StringIO |
|---|
| 29 | |
|---|
| 30 | import lxml.etree |
|---|
| 31 | import genshi |
|---|
| 32 | import genshi.output |
|---|
| 33 | from genshi.template import MarkupTemplate |
|---|
| 34 | |
|---|
| 35 | GENSHI_TAGS = re.compile(r'''relatorio://((/)?(for|choose|otherwise|when|if|with)( (\w+)=["'](.*)["']|)|.*)''') |
|---|
| 36 | EXTENSIONS = {'image/png': 'png', |
|---|
| 37 | 'image/jpeg': 'jpg', |
|---|
| 38 | 'image/bmp': 'bmp', |
|---|
| 39 | 'image/gif': 'gif', |
|---|
| 40 | 'image/tiff': 'tif', |
|---|
| 41 | 'image/xbm': 'xbm', |
|---|
| 42 | } |
|---|
| 43 | |
|---|
| 44 | _encode = genshi.output.encode |
|---|
| 45 | ETElement = lxml.etree.Element |
|---|
| 46 | |
|---|
| 47 | |
|---|
| 48 | class ImageHref: |
|---|
| 49 | |
|---|
| 50 | def __init__(self, zipfile): |
|---|
| 51 | self.zip = zipfile |
|---|
| 52 | |
|---|
| 53 | def __call__(self, expr, name): |
|---|
| 54 | bitstream, mimetype = expr |
|---|
| 55 | bitstream.seek(0) |
|---|
| 56 | file_content = bitstream.read() |
|---|
| 57 | name = md5.new(file_content).hexdigest() |
|---|
| 58 | path = 'Pictures/%s.%s' % (name, EXTENSIONS[mimetype]) |
|---|
| 59 | if path not in self.zip.namelist(): |
|---|
| 60 | self.zip.writestr(path, file_content) |
|---|
| 61 | return {'{http://www.w3.org/1999/xlink}href': path} |
|---|
| 62 | |
|---|
| 63 | |
|---|
| 64 | class Template(MarkupTemplate): |
|---|
| 65 | |
|---|
| 66 | def __init__(self, source, filepath=None, filename=None, loader=None, |
|---|
| 67 | encoding=None, lookup='strict', allow_exec=True): |
|---|
| 68 | self.namespaces = {} |
|---|
| 69 | super(Template, self).__init__(source, filepath, filename, loader, |
|---|
| 70 | encoding, lookup, allow_exec) |
|---|
| 71 | |
|---|
| 72 | def _parse(self, source, encoding): |
|---|
| 73 | inzip = zipfile.ZipFile(self.filepath) |
|---|
| 74 | content = inzip.read('content.xml') |
|---|
| 75 | styles = inzip.read('styles.xml') |
|---|
| 76 | inzip.close() |
|---|
| 77 | |
|---|
| 78 | content = super(Template, self)._parse(self.add_directives(content), |
|---|
| 79 | encoding) |
|---|
| 80 | styles = super(Template, self)._parse(self.add_directives(styles), |
|---|
| 81 | encoding) |
|---|
| 82 | return [(genshi.core.PI, ('relatorio', 'styles.xml'), None)] +\ |
|---|
| 83 | styles +\ |
|---|
| 84 | [(genshi.core.PI, ('relatorio', 'content.xml'), None)] +\ |
|---|
| 85 | content |
|---|
| 86 | |
|---|
| 87 | def add_directives(self, content): |
|---|
| 88 | tree = lxml.etree.parse(StringIO(content)) |
|---|
| 89 | root = tree.getroot() |
|---|
| 90 | self.namespaces = root.nsmap.copy() |
|---|
| 91 | self.namespaces['py'] = 'http://genshi.edgewall.org/' |
|---|
| 92 | |
|---|
| 93 | self._handle_text_a(tree) |
|---|
| 94 | self._handle_images(tree) |
|---|
| 95 | return StringIO(lxml.etree.tostring(tree)) |
|---|
| 96 | |
|---|
| 97 | def _handle_text_a(self, tree): |
|---|
| 98 | """ |
|---|
| 99 | Will treat all text:a tag (py:if/for/choose/when/otherwise) |
|---|
| 100 | tags |
|---|
| 101 | """ |
|---|
| 102 | # Some tag name constants |
|---|
| 103 | table_cell_tag = '{%s}table-cell' % self.namespaces['table'] |
|---|
| 104 | attrib_name = '{%s}attrs' % self.namespaces['py'] |
|---|
| 105 | office_name = '{%s}value' % self.namespaces['office'] |
|---|
| 106 | office_valuetype = '{%s}value-type' % self.namespaces['office'] |
|---|
| 107 | genshi_name = '{%s}replace' % self.namespaces['py'] |
|---|
| 108 | xlink_href_attrib = '{%s}href' % self.namespaces['xlink'] |
|---|
| 109 | |
|---|
| 110 | # First we create the list of all the text:a nodes. |
|---|
| 111 | # If this node href matches the relatorio URL it is kept. |
|---|
| 112 | # If this node href matches a genshi directive it is kept for further |
|---|
| 113 | # processing. |
|---|
| 114 | genshi_directives, text_a = [], [] |
|---|
| 115 | for statement in tree.xpath('//text:a', namespaces=self.namespaces): |
|---|
| 116 | href = urllib.unquote(statement.attrib[xlink_href_attrib]) |
|---|
| 117 | match_obj = GENSHI_TAGS.match(href) |
|---|
| 118 | if match_obj is None: |
|---|
| 119 | continue |
|---|
| 120 | expr, closing, directive, _, attr, attr_val = match_obj.groups() |
|---|
| 121 | if directive is not None: |
|---|
| 122 | genshi_directives.append((statement, href)) |
|---|
| 123 | text_a.append((statement, |
|---|
| 124 | (expr, closing, directive, attr, attr_val))) |
|---|
| 125 | |
|---|
| 126 | # Then we match the opening and closing directives together |
|---|
| 127 | idx = 0 |
|---|
| 128 | genshi_pairs, inserted = [], [] |
|---|
| 129 | for statement, href in genshi_directives: |
|---|
| 130 | if not href.startswith('relatorio:///'): |
|---|
| 131 | genshi_pairs.append([statement, None]) |
|---|
| 132 | inserted.append(idx) |
|---|
| 133 | idx += 1 |
|---|
| 134 | else: |
|---|
| 135 | genshi_pairs[inserted.pop()][1] = statement |
|---|
| 136 | |
|---|
| 137 | for a_node, parsed in text_a: |
|---|
| 138 | expr, c_dir, directive, attr, a_val = parsed |
|---|
| 139 | |
|---|
| 140 | if directive is not None: |
|---|
| 141 | # If the text:a is a genshi directive statement: |
|---|
| 142 | # - we operate only on opening statement |
|---|
| 143 | # - we find the nearest ancestor of the closing and opening |
|---|
| 144 | # statement |
|---|
| 145 | # - we create a <py:xxx> node |
|---|
| 146 | # - we add all the node between the opening and closing |
|---|
| 147 | # statements to this new node |
|---|
| 148 | # - we replace the opening statement by the <py:for> node |
|---|
| 149 | # - we delete the closing statement |
|---|
| 150 | |
|---|
| 151 | if c_dir is not None: |
|---|
| 152 | # pass the closing statements |
|---|
| 153 | continue |
|---|
| 154 | for pair in genshi_pairs: |
|---|
| 155 | if pair[0] == a_node: |
|---|
| 156 | break |
|---|
| 157 | opening, closing = pair |
|---|
| 158 | |
|---|
| 159 | o_ancestors = list(opening.iterancestors()) |
|---|
| 160 | c_ancestors = list(closing.iterancestors()) |
|---|
| 161 | for n in o_ancestors: |
|---|
| 162 | if n in c_ancestors: |
|---|
| 163 | ancestor = n |
|---|
| 164 | break |
|---|
| 165 | |
|---|
| 166 | genshi_node = ETElement('{%s}%s' % (self.namespaces['py'], |
|---|
| 167 | directive), |
|---|
| 168 | attrib={attr: a_val}, |
|---|
| 169 | nsmap=self.namespaces) |
|---|
| 170 | can_append = False |
|---|
| 171 | for node in ancestor.iterchildren(): |
|---|
| 172 | if node in o_ancestors: |
|---|
| 173 | outermost_o_ancestor = node |
|---|
| 174 | can_append = True |
|---|
| 175 | continue |
|---|
| 176 | if node in c_ancestors: |
|---|
| 177 | outermost_c_ancestor = node |
|---|
| 178 | break |
|---|
| 179 | if can_append: |
|---|
| 180 | genshi_node.append(node) |
|---|
| 181 | ancestor.replace(outermost_o_ancestor, genshi_node) |
|---|
| 182 | ancestor.remove(outermost_c_ancestor) |
|---|
| 183 | else: |
|---|
| 184 | # It's not a genshi statement it's a python expression |
|---|
| 185 | a_node.attrib['{%s}replace' % self.namespaces['py']] = expr |
|---|
| 186 | parent = a_node.getparent().getparent() |
|---|
| 187 | if parent is None or parent.tag != table_cell_tag: |
|---|
| 188 | continue |
|---|
| 189 | if parent.attrib.get(office_valuetype, 'string') != 'string': |
|---|
| 190 | # The grand-parent tag is a table cell we set the |
|---|
| 191 | # office:value attribute of this cell |
|---|
| 192 | dico = "{'%s': %s}" % (office_name, expr) |
|---|
| 193 | parent.attrib[attrib_name] = dico |
|---|
| 194 | parent.attrib.pop(office_name, None) |
|---|
| 195 | |
|---|
| 196 | def _handle_images(self, tree): |
|---|
| 197 | for draw in tree.xpath('//draw:frame', namespaces=self.namespaces): |
|---|
| 198 | d_name = draw.attrib['{%s}name' % self.namespaces['draw']] |
|---|
| 199 | if d_name.startswith('image: '): |
|---|
| 200 | attr_expr = "make_href(%s, %r)" % (d_name[7:], d_name[7:]) |
|---|
| 201 | attributes = {} |
|---|
| 202 | attributes['{%s}attrs' % self.namespaces['py']] = attr_expr |
|---|
| 203 | image_node = ETElement('{%s}image' % self.namespaces['draw'], |
|---|
| 204 | attrib=attributes, |
|---|
| 205 | nsmap=self.namespaces) |
|---|
| 206 | draw.replace(draw[0], image_node) |
|---|
| 207 | |
|---|
| 208 | |
|---|
| 209 | def generate(self, *args, **kwargs): |
|---|
| 210 | serializer = OOSerializer(self.filepath) |
|---|
| 211 | kwargs['make_href'] = ImageHref(serializer.outzip) |
|---|
| 212 | generate_all = super(Template, self).generate(*args, **kwargs) |
|---|
| 213 | |
|---|
| 214 | return OOStream(generate_all, serializer) |
|---|
| 215 | |
|---|
| 216 | |
|---|
| 217 | class OOStream(genshi.core.Stream): |
|---|
| 218 | |
|---|
| 219 | def __init__(self, content_stream, serializer): |
|---|
| 220 | self.events = content_stream |
|---|
| 221 | self.serializer = serializer |
|---|
| 222 | |
|---|
| 223 | def render(self, method=None, encoding='utf-8', out=None, **kwargs): |
|---|
| 224 | return self.serializer(self.events) |
|---|
| 225 | |
|---|
| 226 | def serialize(self, method, **kwargs): |
|---|
| 227 | return self.render(method, **kwargs) |
|---|
| 228 | |
|---|
| 229 | def __or__(self, function): |
|---|
| 230 | return OOStream(self.events | function, self.serializer) |
|---|
| 231 | |
|---|
| 232 | |
|---|
| 233 | class OOSerializer: |
|---|
| 234 | |
|---|
| 235 | def __init__(self, oo_path): |
|---|
| 236 | self.inzip = zipfile.ZipFile(oo_path) |
|---|
| 237 | self.new_oo = StringIO() |
|---|
| 238 | self.outzip = zipfile.ZipFile(self.new_oo, 'w') |
|---|
| 239 | self.xml_serializer = genshi.output.XMLSerializer() |
|---|
| 240 | |
|---|
| 241 | def __call__(self, stream): |
|---|
| 242 | files = {'styles.xml': [], 'content.xml': []} |
|---|
| 243 | for kind, data, pos in stream: |
|---|
| 244 | if kind == genshi.core.PI and data[0] == 'relatorio': |
|---|
| 245 | stream_for = data[1] |
|---|
| 246 | continue |
|---|
| 247 | files[stream_for].append((kind, data, pos)) |
|---|
| 248 | |
|---|
| 249 | for f in self.inzip.infolist(): |
|---|
| 250 | if f.filename in files: |
|---|
| 251 | stream = files[f.filename] |
|---|
| 252 | self.outzip.writestr(f.filename, |
|---|
| 253 | _encode(self.xml_serializer(stream))) |
|---|
| 254 | else: |
|---|
| 255 | self.outzip.writestr(f, self.inzip.read(f.filename)) |
|---|
| 256 | self.inzip.close() |
|---|
| 257 | self.outzip.close() |
|---|
| 258 | |
|---|
| 259 | return self.new_oo |
|---|