xmputils.py

   1 # -*- coding: utf-8 -*-
   2 #######################################################################################
   3 #   Photo is a part of Plinn - http://plinn.org                                       #
   4 #   Copyright (C) 2008 Benoît PIN <benoit.pin@ensmp.fr>                               #
   5 #                                                                                     #
   6 #   This program is free software; you can redistribute it and/or                     #
   7 #   modify it under the terms of the GNU General Public License                       #
   8 #   as published by the Free Software Foundation; either version 2                    #
   9 #   of the License, or (at your option) any later version.                            #
  10 #                                                                                     #
  11 #   This program is distributed in the hope that it will be useful,                   #
  12 #   but WITHOUT ANY WARRANTY; without even the implied warranty of                    #
  13 #   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the                     #
  14 #   GNU General Public License for more details.                                      #
  15 #                                                                                     #
  16 #   You should have received a copy of the GNU General Public License                 #
  17 #   along with this program; if not, write to the Free Software                       #
  18 #   Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.   #
  19 #######################################################################################
  20 """ XMP generation utilities.
  21
  22
  23
  24 """
  25
  26 from libxml2 import newNode, parseDoc, treeError
  27 # prefix <-> namespaces mappings as defined in the official xmp documentation
  28 from standards.xmp import namespaces as xmpNs2Prefix
  29 from standards.xmp import prefix2Ns as xmpPrefix2Ns
  30
  31 TIFF_ORIENTATIONS = {1 : (0, False)
  32                                         ,2 : (0, True)
  33                                         ,3 : (180, False)
  34                                         ,4 : (180, True)
  35                                         ,5 : (90, True)
  36                                         ,6 : (90, False)
  37                                         ,7 : (270, True)
  38                                         ,8 : (270, False)}
  39
  40 def _getRDFArrayValues(node, arrayType):
  41         values = []
  42         for element in iterElementChilds(node):
  43                 if element.name == arrayType and element.ns().content == 'http://www.w3.org/1999/02/22-rdf-syntax-ns#' :
  44                         for value in iterElementChilds(element):
  45                                 if value.name == 'li':
  46                                         values.append(value.content)
  47                         return tuple(values)
  48         else :
  49                 raise ValueError("No %s found" % arrayType )
  50
  51 def getBagValues(node):
  52         return _getRDFArrayValues(node, 'Bag')
  53
  54 def getSeqValues(node):
  55         return _getRDFArrayValues(node, 'Seq')
  56
  57
  58 def createRDFAlt(surrounded, defaultText, rootIndex):
  59         """
  60         returns (as libxml2 node):
  61         <surrounded>
  62            <rdf:Alt>
  63               <rdf:li xml:lang="x-default">defaultText</rdf:li>
  64            </rdf:Alt>
  65         <surrounded>
  66         """
  67         docNs = rootIndex.getDocumentNs()
  68         rdfPrefix = docNs['http://www.w3.org/1999/02/22-rdf-syntax-ns#']
  69         normalizedPrefix, name = surrounded.split(':')
  70         ns = xmpPrefix2Ns[normalizedPrefix]
  71         actualPrefix = docNs[ns]
  72
  73         surrounded = newNode('%s:%s' % (actualPrefix, name))
  74         alt = newNode('%s:Alt' % rdfPrefix)
  75         li = newNode('%s:li' % rdfPrefix)
  76         li.newProp('xml:lang', 'x-default')
  77         li.setContent(defaultText)
  78
  79         reduce(lambda a, b: a.addChild(b), (surrounded, alt, li))
  80
  81         return surrounded
  82
  83
  84 def createRDFBag(surrounded, values, rootIndex):
  85         """
  86         returns (as libxml2 node):
  87         <surrounded>
  88        <rdf:Bag>
  89           <rdf:li>values[0]</rdf:li>
  90                         ...
  91           <rdf:li>values[n]</rdf:li>
  92        </rdf:Bag>
  93     <surrounded>
  94     """
  95         return _createRDFArray(surrounded, values, False, rootIndex)
  96
  97 def createRDFSeq(surrounded, values, rootIndex):
  98         """
  99         returns (as libxml2 node):
 100         <surrounded>
 101        <rdf:Seq>
 102           <rdf:li>values[0]</rdf:li>
 103                         ...
 104           <rdf:li>values[n]</rdf:li>
 105        </rdf:Seq>
 106     <surrounded>
 107     """
 108         return _createRDFArray(surrounded, values, True, rootIndex)
 109
 110 def _createRDFArray(surrounded, values, ordered, rootIndex):
 111         docNs = rootIndex.getDocumentNs()
 112         rdfPrefix = docNs['http://www.w3.org/1999/02/22-rdf-syntax-ns#']
 113         normalizedPrefix, name = surrounded.split(':')
 114         ns = xmpPrefix2Ns[normalizedPrefix]
 115         actualPrefix = docNs[ns]
 116
 117
 118         surrounded = newNode('%s:%s' % (actualPrefix, name))
 119         if ordered is True :
 120                 array = newNode('%s:Seq' % rdfPrefix)
 121         elif ordered is False :
 122                 array = newNode('%s:Bag' % rdfPrefix)
 123         else :
 124                 raise ValueError("'ordered' parameter must be a boolean value")
 125
 126         surrounded.addChild(array)
 127
 128         for v in values :
 129                 li = newNode('%s:li' % rdfPrefix)
 130                 li.setContent(v)
 131                 array.addChild(li)
 132
 133         return surrounded
 134
 135 def createEmptyXmpDoc() :
 136         emptyDocument = """
 137 <x:xmpmeta xmlns:x='adobe:ns:meta/'>
 138   <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#">
 139     <rdf:Description/>
 140   </rdf:RDF>
 141 </x:xmpmeta>
 142         """
 143         d = parseDoc(emptyDocument)
 144         return d
 145
 146 def getPathIndex(doc) :
 147         root = doc.getRootElement()
 148         index = PathIndex(root)
 149         return index
 150
 151
 152 class PathIndex :
 153         """\
 154         Class used to provide a convenient tree access to xmp properties by paths.
 155         Issues about namespaces and prefixes are normalized during the object
 156         instanciation. Ns prefixes used to access elements are those recommended in the
 157         official xmp documentation from Adobe.
 158         """
 159
 160         def __init__(self, element, parent=None) :
 161                 self.unique = True
 162                 self.element = element
 163                 self.parent = parent
 164
 165                 elementNs = element.ns().content
 166                 elementPrefix = element.ns().name
 167                 recommendedPrefix = xmpNs2Prefix.get(elementNs, elementPrefix)
 168
 169                 self.name = '%s:%s' % (recommendedPrefix, element.name)
 170                 self.namespace = elementNs
 171                 self.prefix = elementPrefix
 172                 self._index = {}
 173
 174                 for prop in iterElementProperties(element) :
 175                         self.addChildIndex(prop)
 176
 177                 for child in iterElementChilds(element) :
 178                         self.addChildIndex(child)
 179
 180                 if self.parent is None:
 181                         self.nsDeclarations = self._namespaceDeclarations()
 182
 183         def addChildIndex(self, child) :
 184                 ns = child.ns()
 185                 if not ns :
 186                         return
 187
 188                 childNs = ns.content
 189                 childPrefix = ns.name
 190                 childRecommendedPrefix = xmpNs2Prefix.get(childNs, childPrefix)
 191                 childName = '%s:%s' % (childRecommendedPrefix, child.name)
 192
 193                 if not self._index.has_key(childName) :
 194                         self._index[childName] = PathIndex(child, parent=self)
 195                 else :
 196                         childIndex = self._index[childName]
 197                         childIndex.unique = False
 198                         for prop in iterElementProperties(child) :
 199                                 childIndex.addChildIndex(prop)
 200
 201                         for c in iterElementChilds(child) :
 202                                 childIndex.addChildIndex(c)
 203
 204                 self._index[childName].parent = self
 205                 return self._index[childName]
 206
 207         def _namespaceDeclarations(self) :
 208                 """\
 209                 returns ns / prefix pairs as found in xmp packet
 210                 """
 211                 namespaces = {}
 212                 namespaces[self.namespace] = self.prefix
 213                 for child in self._index.values() :
 214                         for namespace, prefix in child._namespaceDeclarations().items() :
 215                                 if namespaces.has_key(namespace) :
 216                                         assert namespaces[namespace] == prefix, \
 217                                                         "using several prefix for the same namespace is forbidden "\
 218                                                         "in this implementation"
 219                                 else :
 220                                         namespaces[namespace] = prefix
 221                 return namespaces
 222
 223         def getDocumentNs(self) :
 224                 root = self.getRootIndex()
 225                 return root.nsDeclarations
 226
 227         def exists(self, path) :
 228                 o = self
 229                 for part in path.split('/') :
 230                         if o._index.has_key(part) :
 231                                 o = o._index[part]
 232                         else :
 233                                 return False
 234                 return True
 235
 236         def __getitem__(self, path) :
 237                 o = self
 238                 try :
 239                         for part in path.split('/') :
 240                                 if part == '.' :
 241                                         continue
 242                                 elif part == '..' :
 243                                         o = o.parent
 244                                 o = o._index[part]
 245                 except ValueError :
 246                         raise KeyError, path
 247                 return o
 248
 249         def get(self, path, default=None) :
 250                 try :
 251                         return self[path]
 252                 except KeyError :
 253                         return default
 254
 255         def getRootIndex(self) :
 256                 root = self
 257                 while root.parent is not None :
 258                         root = root.parent
 259                 return root
 260
 261         def createChildAndIndex(self, name, rdfType, nsDeclarationElement) :
 262                 recommandedPrefix, name = name.split(':', 1)
 263
 264                 if rdfType == 'prop' :
 265                         try :
 266                                 node = self.element.newProp(name, '')
 267                         except treeError :
 268                                 raise ValueError, (self.element, name)
 269                 else :
 270                         node = newNode(name)
 271                         self.element.addChild(node)
 272
 273                 # bind namespace to new node
 274                 uri = xmpPrefix2Ns[recommandedPrefix]
 275                 docNamespaces = self.getDocumentNs()
 276                 if not docNamespaces.has_key(uri) :
 277                         try :
 278                                 ns = nsDeclarationElement.newNs(uri, recommandedPrefix)
 279                         except treeError :
 280                                 raise ValueError, (uri, prefix, self.element, list(nsDeclarationElement.nsDefs()))
 281                         docNamespaces[uri] = recommandedPrefix
 282                 else :
 283                         actualPrefix = docNamespaces[uri]
 284                         try :
 285                                 ns = self.element.searchNs(None, actualPrefix)
 286                         except treeError:
 287                                 # cas d'un xmp verbeux : le nouvel élément n'est pas ajouté
 288                                 # dans le rdf:Description du ns correspondant
 289                                 # (après tout, ce n'est pas une obligation)
 290                                 # => on ajoute le ns
 291                                 ns = nsDeclarationElement.newNs(uri, actualPrefix)
 292
 293
 294                 node.setNs(ns)
 295                 return self.addChildIndex(node)
 296
 297         def getOrCreate(self, path, rdfType, preferedNsDeclaration='rdf:RDF/rdf:Description') :
 298                 parts = path.split('/')
 299
 300                 if not parts :
 301                         return self
 302
 303                 name = parts[-1]
 304                 parts = parts[:-1]
 305                 root = self.getRootIndex()
 306                 nsDeclarationElement = root[preferedNsDeclaration].element
 307
 308                 parent = self
 309                 for p in parts :
 310                         child = parent._index.get(p, None)
 311                         if child is None :
 312                                 child = parent.createChildAndIndex(p, None, nsDeclarationElement)
 313                         parent = child
 314
 315                 child = parent._index.get(name, None)
 316                 if child is None :
 317                         child = parent.createChildAndIndex(name, rdfType, nsDeclarationElement)
 318
 319                 return child
 320
 321         def __str__(self) :
 322                 out = []
 323                 pr = out.append
 324                 path = [self.name]
 325                 parent = self.parent
 326                 while parent :
 327                         path.append(parent.name)
 328                         parent = parent.parent
 329                 path.reverse()
 330                 path = '/'.join(path)
 331                 pr(path)
 332                 pr(self.name)
 333                 pr(self.namespace)
 334                 pr(str(self.unique))
 335                 pr('-------')
 336
 337                 for child in self._index.values() :
 338                         pr(str(child))
 339
 340                 return '\n'.join(out)
 341
 342 def iterElementChilds(parent) :
 343         child  = parent.children
 344         while child :
 345                 if child.type == 'element' :
 346                         yield child
 347                 child = child.next
 348
 349 def iterElementProperties(element) :
 350         prop = element.properties
 351         while prop :
 352                 if prop.type == 'attribute' :
 353                         yield prop
 354                 prop = prop.next