1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
# -*- coding: utf-8 -*-
##############################################################################
#
# Copyright (c) 2009 Nexedi SA and Contributors. All Rights Reserved.
# Jean-Paul Smets-Solanes <jp@nexedi.com>
#
# WARNING: This program as such is intended to be used by professional
# programmers who take the whole responsibility of assessing all potential
# consequences resulting from its eventual inadequacies and bugs
# End users who are looking for a ready-to-use solution with commercial
# guarantees and support are strongly adviced to contract a Free Software
# Service Company
#
# This program is Free Software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
#
##############################################################################
import re, sys, os
from operator import add
from zLOG import LOG
from AccessControl import ClassSecurityInfo, getSecurityManager
from AccessControl.SecurityManagement import newSecurityManager, setSecurityManager
from Acquisition import aq_base
from Products.ERP5Type.Accessor.Constant import PropertyGetter as ConstantGetter
from Products.ERP5Type.Globals import get_request
from Products.CMFCore.utils import getToolByName, _checkPermission
from Products.ERP5Type import Permissions, PropertySheet, interfaces
from Products.ERP5Type.XMLObject import XMLObject
from Products.ERP5Type.DateUtils import convertDateToHour,\
number_of_hours_in_day, number_of_hours_in_year
from Products.ERP5Type.Utils import convertToUpperCase, fill_args_from_request
from Products.ERP5Type.TransactionalVariable import getTransactionalVariable
from Products.ERP5Type.ExtensibleTraversable import ExtensibleTraversableMixIn
from Products.ERP5Type.Cache import getReadOnlyTransactionCache
from Products.ERP5.Document.Url import UrlMixIn
from Products.ERP5.Tool.ContributionTool import MAX_REPEAT
from Products.ERP5Type.UnrestrictedMethod import unrestricted_apply
from Products.ZSQLCatalog.SQLCatalog import SQLQuery
from AccessControl import Unauthorized
import zope.interface
from Products.PythonScripts.Utility import allow_class
import tempfile
from subprocess import Popen, PIPE
# Mixin Import
from Products.ERP5.mixin.cached_convertable import CachedConvertableMixin
from Products.ERP5.mixin.text_convertable import TextConvertableMixin
from Products.ERP5.mixin.downloadable import DownloadableMixin
from Products.ERP5.mixin.document import DocumentMixin
from Products.ERP5.mixin.extensible_traversable import DocumentExtensibleTraversableMixin
from Products.ERP5.mixin.crawable import CrawableMixin
_MARKER = []
VALID_ORDER_KEY_LIST = ('user_login', 'content', 'file_name', 'input')
# these property ids are unchangable
FIXED_PROPERTY_IDS = ('id', 'uid', 'rid', 'sid')
# XXX: move to an easier to configure place (System Preference ?)
VALID_TEXT_FORMAT_LIST = ('text', 'txt', 'html', 'base_html',
'stripped-html')
VALID_IMAGE_FORMAT_LIST = ('jpg', 'jpeg', 'png', 'gif', 'pnm', 'ppm', 'tiff')
DEFAULT_DISPLAY_ID_LIST = ('nano', 'micro', 'thumbnail',
'xsmall', 'small', 'medium',
'large', 'large', 'xlarge',)
# default image quality
DEFAULT_IMAGE_QUALITY = 75
DEFAULT_CONTENT_TYPE = 'text/html'
class ConversionError(Exception):pass
class DocumentProxyError(Exception):pass
class NotConvertedError(Exception):pass
allow_class(NotConvertedError)
class Document(DocumentExtensibleTraversableMixin, XMLObject, UrlMixIn, CachedConvertableMixin,
CrawableMixin, TextConvertableMixin, DownloadableMixin, DocumentMixin):
"""Document is an abstract class with all methods related to document
management in ERP5. This includes searchable text, explicit relations,
implicit relations, metadata, versions, languages, etc.
Documents may either store their content directly or cache content
which is retrieved from a specified URL. The second case if often
referred as "External Document". Standalone "External Documents" may
be created by specifying a URL to the contribution tool which is in
charge of initiating the download process and selecting the appropriate
document type. Groups of "External Documents" may also be generated from
so-called "External Source" (refer to ExternalSource class for more
information).
External Documents may be downloaded once or updated at regular interval.
The later can be useful to update the content of an external source.
Previous versions may be stored in place or kept in a separate file.
This feature is known as the crawling API. It is mostly implemented
in ContributionTool with wrappers in the Document class. It can be useful
for create a small search engine.
There are currently two types of Document subclasses:
* File for binary file based documents. File has subclasses such as Image,
OOoDocument, PDFDocument, etc. to implement specific conversion methods.
* TextDocument for text based documents. TextDocument has subclasses such
as Wiki to implement specific methods.
TextDocument itself has a subclass (XSLTDocument) which provides
XSLT based analysis and transformation of XML content based on XSLT
templates.
Conversion should be achieved through the convert method and other methods
of the conversion API (convertToBaseFormat, etc.).
Moreover, any Document subclass must ne able to convert documents to text
(asText method) and HTML (asHTML method). Text is required for full text
indexing. HTML is required for crawling.
Instances can be created directly, or via portal_contributions tool which
manages document ingestion process whereby a file can be uploaded by http
or sent in by email or dropped in by webdav or in some other way as yet
unknown. The ingestion process has the following steps:
(1) portal type detection
(2) object creation and upload of data
(3) metadata discovery (optionally with conversion of data to another format)
(4) other possible actions to finalise the ingestion (ex. by assigning
a reference)
This class handles (3) and calls a ZMI script to do (4).
Metadata can be drawn from various sources:
input - data supplied with http request or set on the object during (2) (e.g.
discovered from email text)
file_name - data which might be encoded in file name
user_login - information about user who is contributing the file
content - data which might be derived from document content
If a certain property is defined in more than one source, it is set according to
preference order returned by a script
Document_getPreferredDocumentMetadataDiscoveryOrderList
(or any type-based version since discovery is type dependent)
Methods for discovering metadata are:
getPropertyDictFromInput
getPropertyDictFromFileName
getPropertyDictFromUserLogin
getPropertyDictFromContent
Methods for processing content are implemented either in Document class
or in Base class:
getSearchableReferenceList (Base)
getSearchableText (Base)
index_html (overriden in Document subclasses)
Methods for handling relations are implemented either in Document class
or in Base class:
getImplicitSuccessorValueList (Base)
getImplicitPredecessorValueList (Base)
getImplicitSimilarValueList (Base)
getSimilarCloudValueList (Document)
Implicit relations consist in finding document references inside
searchable text (ex. INV-23456) and deducting relations from that.
Two customisable methods required. One to find a list of implicit references
inside the content (getSearchableReferenceList) and one to convert a given
document reference into a list of reference strings which could be present
in other content (asSearchableReferenceList).
document.getSearchableReferenceList() returns
[
{'reference':' INV-12367'},
{'reference': 'INV-1112', 'version':'012}',
{'reference': 'AB-CC-DRK', 'version':'011', 'language': 'en'}
]
The Document class behaviour can be extended / customized through scripts
(which are type-based so can be adjusted per portal type).
* Document_getPropertyDictFromUserLogin - finds a user (by user_login or
from session) and returns properties which should be set on the document
* Document_getPropertyDictFromContent - analyzes document content and returns
properties which should be set on the document
* Base_getImplicitSuccessorValueList - finds appropriate all documents
referenced in the current content
* Base_getImplicitPredecessorValueList - finds document predecessors based on
the document coordinates (can use only complete coordinates, or also partial)
* Document_getPreferredDocumentMetadataDiscoveryOrderList - returns an order
in which metadata should be set/overwritten
* Document_finishIngestion - called by portal_activities after all the ingestion
is completed (and after document has been converted, so text_content
is available if the document has it)
* Document_getNewRevision - calculates revision number which should be set
on this document. Implementation depends on revision numbering policy which
can be very different. Interaction workflow should call setNewRevision method.
* Document_populateContent - analyses the document content and produces
subcontent based on it (ex. images, news, etc.). This scripts can
involve for example an XSLT transformation to process XML.
Subcontent: documents may include subcontent (files, images, etc.)
so that publication of rich content can be path independent. Subcontent
can also be used to help the rendering in HTML of complex documents
such as ODF documents.
Consistency checking:
Default implementation uses DocumentReferenceConstraint to check if the
reference/language/version triplet is unique. Additional constraints
can be added if necessary.
NOTE: Document.py supports a notion of revision which is very specific.
The underlying concept is that, as soon as a document has a reference,
the association of (reference, version, language) must be unique accross
the whole system. This means that a given document in a given version in a
given language is unique. The underlying idea is similar to the one in a Wiki
system in which each page is unique and acts the the atom of collaboration.
In the case of ERP5, if a team collaborates on a Text document written with
an offline word processor, all updates should be placed inside the same object.
A Contribution will thus modify an existing document, if allowed from security
point of view, and increase the revision number. Same goes for properties
(title). Each change generates a new revision.
conversion API - not same as document - XXX BAD
XXX make multiple interfaces
TODO:
- move all implementation bits to MixIn classes
- in the end, Document class should have zero code
and only serve as a quick and easy way to create
new types of documents (one could even consider
that this class should be trashed)
-
"""
meta_type = 'ERP5 Document'
portal_type = 'Document'
add_permission = Permissions.AddPortalContent
isDocument = ConstantGetter('isDocument', value=True)
__dav_collection__=0
zope.interface.implements(interfaces.IConvertable,
interfaces.ITextConvertable,
interfaces.IHtmlConvertable,
interfaces.ICachedConvertable,
interfaces.IVersionable,
interfaces.IDownloadable,
interfaces.ICrawlable,
interfaces.IDocument
)
# Regular expressions
href_parser = re.compile('<a[^>]*href=[\'"](.*?)[\'"]',re.IGNORECASE)
body_parser = re.compile('<body[^>]*>(.*?)</body>', re.IGNORECASE + re.DOTALL)
title_parser = re.compile('<title[^>]*>(.*?)</title>', re.IGNORECASE + re.DOTALL)
base_parser = re.compile('<base[^>]*href=[\'"](.*?)[\'"][^>]*>', re.IGNORECASE + re.DOTALL)
charset_parser = re.compile('(?P<keyword>charset="?)(?P<charset>[a-z0-9\-]+)', re.IGNORECASE)
# Declarative security
security = ClassSecurityInfo()
security.declareObjectProtected(Permissions.AccessContentsInformation)
# Declarative properties
property_sheets = ( PropertySheet.Base
, PropertySheet.XMLObject
, PropertySheet.CategoryCore
, PropertySheet.DublinCore
, PropertySheet.Version
, PropertySheet.Document
, PropertySheet.ExternalDocument
, PropertySheet.Url
, PropertySheet.Periodicity
)
index_html = DownloadableMixin.index_html
security.declareProtected(Permissions.AccessContentsInformation, 'isExternalDocument')
def isExternalDocument(self):
"""
Return true if this document was obtained from an external source
"""
return bool(self.getUrlString())
### Relation getters
security.declareProtected(Permissions.AccessContentsInformation, 'getSearchableReferenceList')
def getSearchableReferenceList(self):
"""
This method returns a list of dictionaries which can
be used to find objects by reference. It uses for
that a regular expression defined at system level
preferences.
"""
text = self.getSearchableText() # XXX getSearchableText or asText ?
regexp = self.portal_preferences.getPreferredDocumentReferenceRegularExpression()
try:
rx_search = re.compile(regexp)
except TypeError: # no regexp in preference
LOG('ERP5/Document/Document.getSearchableReferenceList', 0,
'Document regular expression must be set in portal preferences')
return ()
result = []
tmp = {}
for match in rx_search.finditer(text):
group = match.group()
group_item_list = match.groupdict().items()
group_item_list.sort()
key = (group, tuple(group_item_list))
tmp[key] = None
for group, group_item_tuple in tmp.keys():
result.append((group, dict(group_item_tuple)))
return result
security.declareProtected(Permissions.AccessContentsInformation, 'getImplicitSuccessorValueList')
def getImplicitSuccessorValueList(self):
"""
Find objects which we are referencing (if our text_content contains
references of other documents). The whole implementation is delegated to
Base_getImplicitSuccessorValueList script.
The implementation goes in 2 steps:
- Step 1: extract with a regular expression
a list of distionaries with various parameters such as
reference, portal_type, language, version, user, etc. This
part is configured through a portal preference.
- Step 2: read the list of dictionaries
and build a list of values by calling portal_catalog
with appropriate parameters (and if possible build
a complex query whenever this becomes available in
portal catalog)
The script is reponsible for calling getSearchableReferenceList
so that it can use another approach if needed.
NOTE: passing a group_by parameter may be useful at a
later stage of the implementation.
"""
tv = getTransactionalVariable() # XXX Performance improvement required
cache_key = ('getImplicitSuccessorValueList', self.getPhysicalPath())
try:
return tv[cache_key]
except KeyError:
pass
reference_list = [r[1] for r in self.getSearchableReferenceList()]
result = self.Base_getImplicitSuccessorValueList(reference_list)
tv[cache_key] = result
return result
security.declareProtected(Permissions.AccessContentsInformation, 'getImplicitPredecessorValueList')
def getImplicitPredecessorValueList(self):
"""
This function tries to find document which are referencing us - by reference only, or
by reference/language etc. Implementation is passed to
Base_getImplicitPredecessorValueList
The script should proceed in two steps:
Step 1: build a list of references out of the context
(ex. INV-123456, 123456, etc.)
Step 2: search using the portal_catalog and use
priorities (ex. INV-123456 before 123456)
( if possible build a complex query whenever
this becomes available in portal catalog )
NOTE: passing a group_by parameter may be useful at a
later stage of the implementation.
"""
return self._getTypeBasedMethod('getImplicitPredecessorValueList')()
security.declareProtected(Permissions.AccessContentsInformation, 'getImplicitSimilarValueList')
def getImplicitSimilarValueList(self):
"""
Analyses content of documents to find out by the content which documents
are similar. Not implemented yet.
No cloud needed because transitive process
"""
return []
security.declareProtected(Permissions.AccessContentsInformation, 'getSimilarCloudValueList')
def getSimilarCloudValueList(self, depth=0):
"""
Returns all documents which are similar to us, directly or indirectly, and
in both directions. In other words, it is a transitive closure of similar
relation. Every document is returned in the latest version available.
"""
lista = {}
depth = int(depth)
#gettername = 'get%sValueList' % convertToUpperCase(category)
#relatedgettername = 'get%sRelatedValueList' % convertToUpperCase(category)
def getRelatedList(ob, level=0):
level += 1
#getter = getattr(self, gettername)
#relatedgetter = getattr(self, relatedgettername)
#res = getter() + relatedgetter()
res = ob.getSimilarValueList() + ob.getSimilarRelatedValueList()
for r in res:
if lista.get(r) is None:
lista[r] = True # we use dict keys to ensure uniqueness
if level != depth:
getRelatedList(r, level)
getRelatedList(self)
lista_latest = {}
for o in lista.keys():
lista_latest[o.getLatestVersionValue()] = True # get latest versions avoiding duplicates again
if lista_latest.has_key(self):
lista_latest.pop(self) # remove this document
if lista_latest.has_key(self.getLatestVersionValue()):
# remove last version of document itself from related documents
lista_latest.pop(self.getLatestVersionValue())
return lista_latest.keys()
### Version and language getters - might be moved one day to a mixin class in base
security.declareProtected(Permissions.View, 'getLatestVersionValue')
def getLatestVersionValue(self, language=None):
"""
Tries to find the latest version with the latest revision
of self which the current user is allowed to access.
If language is provided, return the latest document
in the language.
If language is not provided, return the latest version
in original language or in the user language if the version is
the same.
"""
if not self.getReference():
return self
catalog = getToolByName(self.getPortalObject(), 'portal_catalog')
kw = dict(reference=self.getReference(), sort_on=(('version','descending'),))
if language is not None:
kw['language'] = language
result_list = catalog(**kw)
original_language = self.getOriginalLanguage()
user_language = self.Localizer.get_selected_language()
# if language was given return it (if there are any docs in this language)
if language is not None:
try:
return result_list[0].getObject()
except IndexError:
return None
else:
first = result_list[0].getObject()
in_original = None
for record in result_list:
document = record.getObject()
if document.getVersion() != first.getVersion():
# we are out of the latest version - return in_original or first
if in_original is not None:
return in_original
else:
return first # this shouldn't happen in real life
if document.getLanguage() == user_language:
# we found it in the user language
return document
if document.getLanguage() == original_language:
# this is in original language
in_original = document
# this is the only doc in this version
return self
security.declareProtected(Permissions.View, 'getVersionValueList')
def getVersionValueList(self, version=None, language=None):
"""
Returns a list of documents with same reference, same portal_type
but different version and given language or any language if not given.
"""
catalog = getToolByName(self.getPortalObject(), 'portal_catalog')
kw = dict(portal_type=self.getPortalType(),
reference=self.getReference(),
sort_on=(('version', 'descending', 'SIGNED'),)
)
if version:
kw['version'] = version
if language:
kw['language'] = language
return catalog(**kw)
security.declareProtected(Permissions.AccessContentsInformation, 'isVersionUnique')
def isVersionUnique(self):
"""
Returns true if no other document exists with the same
reference, version and language, or if the current
document has no reference.
"""
if not self.getReference():
return True
kw = dict(portal_type=self.getPortalDocumentTypeList(),
reference=self.getReference(),
version=self.getVersion(),
language=self.getLanguage(),
validation_state="!=cancelled")
catalog = getToolByName(self.getPortalObject(), 'portal_catalog')
self_count = catalog.unrestrictedCountResults(uid=self.getUid(), **kw)[0][0]
count = catalog.unrestrictedCountResults(**kw)[0][0]
# If self is not indexed yet, then if count == 1, version is not unique
return count <= self_count
security.declareProtected(Permissions.AccessContentsInformation, 'getRevision')
def getRevision(self):
"""
Returns the current revision by analysing the change log
of the current object. The return value is a string
in order to be consistent with the property sheet
definition.
"""
getInfoFor = getToolByName(self.getPortalObject(),
'portal_workflow').getInfoFor
revision = len(getInfoFor(self, 'history', (), 'edit_workflow'))
# XXX Also look at processing_status_workflow for compatibility.
revision += len([history_item for history_item in\
getInfoFor(self, 'history', (), 'processing_status_workflow')\
if history_item.get('action') == 'edit'])
return str(revision)
security.declareProtected(Permissions.AccessContentsInformation, 'getRevisionList')
def getRevisionList(self):
"""
Returns the list of revision numbers of the current document
by by analysing the change log of the current object.
"""
return map(str, range(1, 1 + int(self.getRevision())))
security.declareProtected(Permissions.ModifyPortalContent, 'mergeRevision')
def mergeRevision(self):
"""
Merge the current document with any previous revision
or change its version to make sure it is still unique.
NOTE: revision support is implemented in the Document
class rather than within the ContributionTool
because the ingestion process requires to analyse the content
of the document first. Hence, it is not possible to
do any kind of update operation until the whole ingestion
process is completed, since update requires to know
reference, version, language, etc. In addition,
we have chosen to try to merge revisions after each
metadata discovery as a way to make sure that any
content added in the system through the ContributionTool
(ex. through webdav) will be merged if necessary.
It may be posssible though to split disoverMetadata and
finishIngestion.
"""
document = self
catalog = getToolByName(self.getPortalObject(), 'portal_catalog')
if self.getReference():
# Find all document with same (reference, version, language)
kw = dict(portal_type=self.getPortalDocumentTypeList(),
reference=self.getReference(),
where_expression=SQLQuery("validation_state NOT IN ('cancelled', 'deleted')"))
if self.getVersion():
kw['version'] = self.getVersion()
if self.getLanguage():
kw['language'] = self.getLanguage()
document_list = catalog.unrestrictedSearchResults(**kw)
existing_document = None
# Select the first one which is not self and which
# shares the same coordinates
document_list = list(document_list)
document_list.sort(key=lambda x: x.getId())
for o in document_list:
if o.getRelativeUrl() != self.getRelativeUrl() and\
o.getVersion() == self.getVersion() and\
o.getLanguage() == self.getLanguage():
existing_document = o.getObject()
break
# We found an existing document to update
if existing_document is not None:
document = existing_document
if existing_document.getPortalType() != self.getPortalType():
raise ValueError, "[DMS] Ingestion may not change the type of an existing document"
elif not _checkPermission(Permissions.ModifyPortalContent, existing_document):
raise Unauthorized, "[DMS] You are not allowed to update the existing document which has the same coordinates (id %s)" % existing_document.getId()
else:
update_kw = {}
for k in self.propertyIds():
if k not in FIXED_PROPERTY_IDS and self.hasProperty(k):
update_kw[k] = self.getProperty(k)
existing_document.edit(**update_kw)
# Erase self
self.delete() # XXX Do we want to delete by workflow or for real ?
return document
security.declareProtected(Permissions.AccessContentsInformation, 'getLanguageList')
def getLanguageList(self, version=None):
"""
Returns a list of languages which this document is available in
for the current user.
"""
if not self.getReference(): return []
catalog = getToolByName(self.getPortalObject(), 'portal_catalog')
kw = dict(portal_type=self.getPortalType(),
reference=self.getReference(),
group_by=('language',))
if version is not None:
kw['version'] = version
return map(lambda o:o.getLanguage(), catalog(**kw))
security.declareProtected(Permissions.AccessContentsInformation, 'getOriginalLanguage')
def getOriginalLanguage(self):
"""
Returns the original language of this document.
XXX-JPS not implemented yet ?
"""
# Approach 1: use portal_catalog and creation dates
# Approach 2: use workflow analysis (delegate to script if necessary)
# workflow analysis is the only way for multiple orginals
# XXX - cache or set?
reference = self.getReference()
if not reference:
return
catalog = getToolByName(self.getPortalObject(), 'portal_catalog')
res = catalog(reference=self.getReference(), sort_on=(('creation_date','ascending'),))
# XXX this should be security-unaware - delegate to script with proxy roles
return res[0].getLanguage() # XXX what happens if it is empty?
### Property getters
# Property Getters are document dependent so that we can
# handle the weird cases in which needed properties change with the type of document
# and the usual cases in which accessing content changes with the meta type
security.declareProtected(Permissions.ModifyPortalContent,'getPropertyDictFromUserLogin')
def getPropertyDictFromUserLogin(self, user_login=None):
"""
Based on the user_login, find out as many properties as needed.
returns properties which should be set on the document
"""
if user_login is None:
user_login = str(getSecurityManager().getUser())
method = self._getTypeBasedMethod('getPropertyDictFromUserLogin',
fallback_script_id='Document_getPropertyDictFromUserLogin')
return method(user_login)
security.declareProtected(Permissions.ModifyPortalContent,'getPropertyDictFromContent')
def getPropertyDictFromContent(self):
"""
Based on the document content, find out as many properties as needed.
returns properties which should be set on the document
"""
# accesss data through convert
mime, content = self.convert(None)
if not content:
# if document is empty, we will not find anything in its content
return {}
method = self._getTypeBasedMethod('getPropertyDictFromContent',
fallback_script_id='Document_getPropertyDictFromContent')
return method()
security.declareProtected(Permissions.ModifyPortalContent,'getPropertyDictFromFileName')
def getPropertyDictFromFileName(self, file_name):
"""
Based on the file name, find out as many properties as needed.
returns properties which should be set on the document
"""
return self.portal_contributions.getPropertyDictFromFileName(file_name)
security.declareProtected(Permissions.ModifyPortalContent,'getPropertyDictFromInput')
def getPropertyDictFromInput(self):
"""
Get properties which were supplied explicitly to the ingestion method
(discovered or supplied before the document was created).
The implementation consists in saving document properties
into _backup_input by supposing that original input parameters were
set on the document by ContributionTool.newContent as soon
as the document was created.
"""
kw = getattr(self, '_backup_input', {})
if kw:
return kw
for id in self.propertyIds():
# We should not consider file data
if id not in ('data', 'categories_list', 'uid', 'id',
'text_content', 'base_data',) \
and self.hasProperty(id):
kw[id] = self.getProperty(id)
self._backup_input = kw # We could use volatile and pass kw in activate
# if we are garanteed that _backup_input does not
# disappear within a given transaction
return kw
### Metadata disovery and ingestion methods
security.declareProtected(Permissions.ModifyPortalContent, 'discoverMetadata')
def discoverMetadata(self, file_name=None, user_login=None):
"""
This is the main metadata discovery function - controls the process
of discovering data from various sources. The discovery itself is
delegated to scripts or uses preference-configurable regexps. The
method returns either self or the document which has been
merged in the discovery process.
file_name - this parameter is a file name of the form "AA-BBB-CCC-223-en"
user_login - this is a login string of a person; can be None if the user is
currently logged in, then we'll get him from session
"""
# Preference is made of a sequence of 'user_login', 'content', 'file_name', 'input'
method = self._getTypeBasedMethod('getPreferredDocumentMetadataDiscoveryOrderList',
fallback_script_id = 'Document_getPreferredDocumentMetadataDiscoveryOrderList')
order_list = list(method())
order_list.reverse()
# build a dictionary according to the order
kw = {}
for order_id in order_list:
result = None
if order_id not in VALID_ORDER_KEY_LIST:
# Prevent security attack or bad preferences
raise AttributeError, "%s is not in valid order key list" % order_id
method_id = 'getPropertyDictFrom%s' % convertToUpperCase(order_id)
method = getattr(self, method_id)
if order_id == 'file_name':
if file_name is not None:
result = method(file_name)
elif order_id == 'user_login':
if user_login is not None:
result = method(user_login)
else:
result = method()
if result is not None:
for key, value in result.iteritems():
if value not in (None, ''):
kw[key]=value
if file_name is not None:
# filename is often undefined....
kw['source_reference'] = file_name
# Prepare the content edit parameters - portal_type should not be changed
kw.pop('portal_type', None)
# Try not to invoke an automatic transition here
self._edit(**kw)
# Finish ingestion by calling method
self.finishIngestion() # XXX - is this really the right place ?
self.reindexObject() # XXX - is this really the right place ?
# Revision merge is tightly coupled
# to metadata discovery - refer to the documentation of mergeRevision method
merged_doc = self.mergeRevision() # XXX - is this really the right place ?
merged_doc.reindexObject() # XXX - is this really the right place ?
return merged_doc # XXX - is this really the right place ?
security.declareProtected(Permissions.ModifyPortalContent, 'finishIngestion')
def finishIngestion(self):
"""
Finish the ingestion process by calling the appropriate script. This
script can for example allocate a reference number automatically if
no reference was defined.
"""
method = self._getTypeBasedMethod('finishIngestion', fallback_script_id='Document_finishIngestion')
return method()
security.declareProtected(Permissions.View, 'asSubjectText')
def asSubjectText(self, **kw):
"""
Converts the subject of the document to a textual representation.
"""
subject = self.getSubject('')
if not subject:
# XXX not sure if this fallback is a good idea.
subject = self.getTitle('')
return str(subject)
security.declareProtected(Permissions.View, 'asEntireHTML')
def asEntireHTML(self, **kw):
"""
Returns a complete HTML representation of the document
(with body tags, etc.). Adds if necessary a base
tag so that the document can be displayed in an iframe
or standalone.
Actual conversion is delegated to _asHTML
"""
html = self._asHTML(**kw)
if self.getUrlString():
# If a URL is defined, add the base tag
# if base is defined yet.
html = str(html)
if not html.find('<base') >= 0:
base = '<base href="%s"/>' % self.getContentBaseURL()
html = html.replace('<head>', '<head>%s' % base, 1)
self.setConversion(html, mime='text/html', format='base-html')
return html
security.declarePrivate('_asHTML')
def _asHTML(self, **kw):
"""
A private method which converts to HTML. This method
is the one to override in subclasses.
"""
kw['format'] = 'html'
mime, html = self.convert(**kw)
return html
security.declareProtected(Permissions.View, 'asStrippedHTML')
def asStrippedHTML(self, **kw):
"""
Returns a stripped HTML representation of the document
(without html and body tags, etc.) which can be used to inline
a preview of the document.
"""
return self._stripHTML(self._asHTML(**kw))
security.declarePrivate('_guessEncoding')
def _guessEncoding(self, string, mime='text/html'):
"""
Try to guess the encoding for this string.
Returns None if no encoding can be guessed.
"""
try:
import chardet
except ImportError:
chardet = None
if chardet is not None and (mime == 'text/html'\
or os.sys.platform != 'linux2'):
# chardet works fine on html document and its platform independent
return chardet.detect(string).get('encoding', None)
else:
# file command provide better result
# for text/plain documents
# store the content into tempfile
file_descriptor, path = tempfile.mkstemp()
file_object = os.fdopen(file_descriptor, 'w')
file_object.write(string)
file_object.close()
# run file command against tempfile to and read encoded
command_result = Popen(['file', '-b', '--mime-encoding', path],
stdout=PIPE).communicate()[0]
# return detected encoding
return command_result.strip()
def _stripHTML(self, html, charset=None):
"""
A private method which can be reused by subclasses
to strip HTML content
"""
body_list = re.findall(self.body_parser, str(html))
if len(body_list):
stripped_html = body_list[0]
else:
stripped_html = html
return stripped_html
security.declareProtected(Permissions.AccessContentsInformation, 'getContentInformation')
def getContentInformation(self):
"""
Returns the content information from the HTML conversion.
The default implementation tries to build a dictionnary
from the HTML conversion of the document and extract
the document title.
"""
result = {}
html = self.asEntireHTML()
if not html: return result
title_list = re.findall(self.title_parser, str(html))
if title_list:
result['title'] = title_list[0]
return result
security.declareProtected(Permissions.AccessContentsInformation,
'getMetadataMappingDict')
def getMetadataMappingDict(self):
"""
Return a dict of metadata mapping used to update base metadata of the
document
"""
try:
method = self._getTypeBasedMethod('getMetadataMappingDict')
except KeyError, AttributeError:
method = None
if method is not None:
return method()
else:
return {}
# Transformation API
security.declareProtected(Permissions.ModifyPortalContent, 'populateContent')
def populateContent(self):
"""
Populates the Document with subcontent based on the
document base data.
This can be used for example to transform the XML
of an RSS feed into a single piece per news or
to transform an XML export from a database into
individual records. Other application: populate
an HTML text document with its images, used in
conversion with convertToBaseFormat.
"""
try:
method = self._getTypeBasedMethod('populateContent')
except KeyError, AttributeError:
method = None
if method is not None: method()
# Crawling API
security.declareProtected(Permissions.AccessContentsInformation, 'getContentURLList')
def getContentURLList(self):
"""
Returns a list of URLs referenced by the content of this document.
Default implementation consists in analysing the document
converted to HTML. Subclasses may overload this method
if necessary. However, it is better to extend the conversion
methods in order to produce valid HTML, which is useful to
many people, rather than overload this method which is only
useful for crawling.
"""
html_content = self.asStrippedHTML()
return re.findall(self.href_parser, str(html_content))
security.declareProtected(Permissions.ModifyPortalContent, 'updateContentFromURL')
def updateContentFromURL(self, repeat=MAX_REPEAT, crawling_depth=0):
"""
Download and update content of this document from its source URL.
Implementation is handled by ContributionTool.
"""
self.portal_contributions.updateContentFromURL(self, repeat=repeat, crawling_depth=crawling_depth)
security.declareProtected(Permissions.ModifyPortalContent, 'crawlContent')
def crawlContent(self):
"""
Initialises the crawling process on the current document.
"""
self.portal_contributions.crawlContent(self)
security.declareProtected(Permissions.View, 'isIndexContent')
def isIndexContent(self, container=None):
"""
Ask container if we are and index, or a content.
In the vast majority of cases we are content.
This method is required in a crawling process to make
a difference between URLs which return an index (ex. the
list of files in remote server which is accessed through HTTP)
and the files themselves.
"""
if container is None:
container = self.getParentValue()
if hasattr(aq_base(container), 'isIndexContent'):
return container.isIndexContent(self)
return False
security.declareProtected(Permissions.AccessContentsInformation, 'getContentBaseURL')
def getContentBaseURL(self):
"""
Returns the content base URL based on the actual content or
on its URL.
"""
base_url = self.asURL()
base_url_list = base_url.split('/')
if len(base_url_list):
if base_url_list[-1] and base_url_list[-1].find('.') > 0:
# Cut the trailing part in http://www.some.site/at/trailing.html
# but not in http://www.some.site/at
base_url = '/'.join(base_url_list[:-1])
return base_url