Finished adding metadata based groups

README.md

@@ -19,7 +19,7 @@ Features
 * User defined document checksum algorithm
 * Previews for a great deal of image formats, including PDF
 * Document OCR and searching
-
+* Group documents by metadata automatically
 
 Requirements
 ---

apps/common/templates/generic_list_subtemplate.html

@@ -6,7 +6,7 @@
 {% if side_bar %}
     <div class="block">
     <h3>
-        {{ title }}
+        {{ title|capfirst }}
     </h3>
     <div class="content">
         <p>

apps/documents/conf/settings.py

@@ -39,6 +39,10 @@ PREVIEW_SIZE = getattr(settings, 'DOCUMENTS_PREVIEW_SIZE', '640x480')
 THUMBNAIL_SIZE = getattr(settings, 'DOCUMENTS_THUMBNAIL_SIZE', '50x50')
 DISPLAY_SIZE = getattr(settings, 'DOCUMENTS_DISPLAY_SIZE', '1024x768')
 
+#Groups
+GROUP_MAX_RESULTS = getattr(settings, 'DOCUMENTS_GROUP_MAX_RESULTS', 20)
+GROUP_SHOW_EMPTY = getattr(settings, 'DOCUMENTS_GROUP_SHOW_EMPTY', True)
+
 # Serving
 FILESYSTEM_FILESERVING_ENABLE = getattr(settings, 'DOCUMENTS_FILESYSTEM_FILESERVING_ENABLE', True)
 FILESYSTEM_FILESERVING_PATH = getattr(settings, 'DOCUMENTS_FILESYSTEM_FILESERVING_PATH', u'/tmp/mayan/documents')

apps/documents/models.py

@@ -10,6 +10,7 @@ from django.db import models
 from django.template.defaultfilters import slugify
 from django.utils.translation import ugettext_lazy as _
 from django.utils.translation import ugettext
+from django.db.models import Q
 
 from dynamic_search.api import register
 
@@ -111,6 +112,36 @@ class Document(models.Model):
         self.delete_fs_links()
         super(Document, self).delete(*args, **kwargs)
 
+    def get_metadata_groups(self):
+        errors = []
+        metadata_groups = {}
+        if MetadataGroup.objects.all().count():
+            metadata_dict = {}
+            for document_metadata in self.documentmetadata_set.all():
+                metadata_dict['metadata_%s' % document_metadata.metadata_type.name] = document_metadata.value
+                
+            for group in MetadataGroup.objects.filter((Q(document_type=self.document_type) | Q(document_type=None)) & Q(enabled=True)):
+                total_query = Q()
+                for item in group.metadatagroupitem_set.filter(enabled=True):
+                    try:
+                        value_query = Q(**{'value__%s' % item.operator: eval(item.expression, metadata_dict)})
+                    except Exception, e:
+                        errors.append(e)
+                        value_query = Q()
+                        
+                    if item.negated:
+                        query = (Q(metadata_type__id=item.metadata_type.id) & ~value_query)
+                    else:
+                        query = (Q(metadata_type__id=item.metadata_type.id) & value_query)
+                        
+                    if item.inclusion == INCLUSION_AND:
+                        total_query &= query
+                    elif item.inclusion == INCLUSION_OR:
+                        total_query |= query
+                document_id_list = DocumentMetadata.objects.filter(query).values_list('document', flat=True)
+                metadata_groups[group] = Document.objects.filter(Q(id__in=document_id_list) & ~Q(id=self.id)) or []
+        return metadata_groups, errors
+        
     def create_fs_links(self):
         if FILESYSTEM_FILESERVING_ENABLE:
             if not self.exists():
@@ -326,6 +357,7 @@ class MetadataGroup(models.Model):
         verbose_name=_(u'document type'), help_text=_(u'If left blank, all document types will be matched.'))
     name = models.CharField(max_length=32, verbose_name=_(u'name'))
     label = models.CharField(max_length=32, verbose_name=_(u'label'))
+    enabled = models.BooleanField(default=True, verbose_name=_(u'enabled'))
     
     def __unicode__(self):
         return self.label if self.label else self.name
@@ -335,7 +367,6 @@ class MetadataGroup(models.Model):
         verbose_name_plural = _(u'metadata document groups')    
 
 
-
 INCLUSION_AND = '&'
 INCLUSION_OR = '|'
 
@@ -344,24 +375,36 @@ INCLUSION_CHOICES = (
     (INCLUSION_OR, _(u'or')),
 )
 
-OPERATOR_EQUAL = ' '
-OPERATOR_IS_NOT_EQUAL = '~'
-
 OPERATOR_CHOCIES = (
-    (OPERATOR_EQUAL, _(u'is equal')),
+    ('exact', _(u'is equal')),
-    (OPERATOR_IS_NOT_EQUAL, _(u'is not equal')),
+    ('iexact', _(u'is equal (case insensitive)')),
+    ('contains', _(u'contains')),
+    ('icontains', _(u'contains (case insensitive)')),
+    ('in', _(u'is in')),
+    ('gt', _(u'is greater than')),
+    ('gte', _(u'is greater than or equal')),
+    ('lt', _(u'is less than')),
+    ('lte', _(u'is less than or equal')),
+    ('startswith', _(u'starts with')),
+    ('istartswith', _(u'starts with (case insensitive)')),
+    ('endswith', _(u'ends with')),
+    ('iendswith', _(u'ends with (case insensitive)')),
+    ('regex', _(u'is in regular expression')),
+    ('iregex', _(u'is in regular expression (case insensitive)')),
 )
     
 class MetadataGroupItem(models.Model):
     metadata_group = models.ForeignKey(MetadataGroup, verbose_name=_(u'metadata group'))
-    inclusion = models.CharField(default=INCLUSION_AND, max_length=16, choices=INCLUSION_CHOICES)
+    inclusion = models.CharField(default=INCLUSION_AND, max_length=16, choices=INCLUSION_CHOICES, help_text=_(u'The inclusion is ignored for the first item.'))
     metadata_type = models.ForeignKey(MetadataType, verbose_name=_(u'metadata type'), help_text=_(u'This represents the metadata of all other documents.'))
     operator = models.CharField(max_length=16, choices=OPERATOR_CHOCIES)
-    expression = models.CharField(max_length=64,
+    expression = models.CharField(max_length=128,
         verbose_name=_(u'expression'), help_text=_(u'This expression will be evaluated against the current seleted document.  The document metadata is available as variables of the same name but with the "metadata_" prefix added their name.'))
+    negated = models.BooleanField(default=False, verbose_name=_(u'negated'), help_text=_(u'Inverts the logic of the operator.'))
+    enabled = models.BooleanField(default=True, verbose_name=_(u'enabled'))
     
     def __unicode__(self):
-        return '%s %s %s %s' % (self.get_inclusion_display(), self.metadata_type, self.get_operator_display(), self.expression)
+        return '[%s] %s %s %s %s %s' % ('x' if self.enabled else ' ', self.get_inclusion_display(), self.metadata_type, _(u'not') if self.negated else '', self.get_operator_display(), self.expression)
 
     class Meta:
         verbose_name = _(u'metadata group item')

apps/documents/views.py

@@ -32,6 +32,8 @@ from documents.conf.settings import FILESYSTEM_FILESERVING_ENABLE
 from documents.conf.settings import STAGING_FILES_PREVIEW_SIZE
 from documents.conf.settings import PREVIEW_SIZE
 from documents.conf.settings import THUMBNAIL_SIZE
+from documents.conf.settings import GROUP_MAX_RESULTS
+from documents.conf.settings import GROUP_SHOW_EMPTY
 
 from utils import save_metadata, save_metadata_list, decode_metadata_from_url
 
@@ -192,12 +194,6 @@ def upload_document_with_type(request, document_type_id, multiple=True):
     return render_to_response('generic_form.html', context,
         context_instance=RequestContext(request))
         
-from django.db.models import Q
-        
-from models import MetadataGroup
-
-from models import INCLUSION_AND, INCLUSION_OR, OPERATOR_EQUAL, OPERATOR_IS_NOT_EQUAL
-
 def document_view(request, document_id):
     document = get_object_or_404(Document, pk=document_id)
     form = DocumentForm_view(instance=document, extra_fields=[
@@ -213,42 +209,11 @@ def document_view(request, document_id):
         {'label':_(u'UUID'), 'field':'uuid'},
     ])
 
-    metadata_groups = {}
-    if MetadataGroup.objects.all().count():
-        metadata_dict = {}
-        for document_metadata in document.documentmetadata_set.all():
-            metadata_dict['metadata_%s' % document_metadata.metadata_type.name] = document_metadata.value
         
-        for group in MetadataGroup.objects.filter(Q(document_type=document.document_type) | Q(document_type=None)):
+    metadata_groups, errors = document.get_metadata_groups()
-            total_query = None
+    if request.user.is_staff and errors:
-            for count, item in enumerate(group.metadatagroupitem_set.all()):
+        for error in errors:
-                try:
+            messages.warning(request, _(u'Metadata group query error: %s' % error))
-                    expression_result = eval(item.expression, metadata_dict)
-                
-                    if item.operator == OPERATOR_EQUAL:
-                        value_query = Q(documentmetadata__value=expression_result)
-                    elif item.operator == OPERATOR_IS_NOT_EQUAL:
-                        value_query = ~Q(documentmetadata__value=expression_result)
-                        
-                    query = (Q(documentmetadata__metadata_type__id=item.metadata_type.id) & value_query)
-                    if count == 0:
-                        total_query = query
-                    else:
-                        if item.inclusion == INCLUSION_AND:
-                            total_query &= query
-                        elif item.inclusion == INCLUSION_AND:
-                            total_query |= query
-                except Exception, e:
-                    if request.user.is_staff:
-                        messages.warning(request, _(u'Metadata group query error: %s' % e))
-                    else:
-                        pass
-                        
-                    
-            if total_query:
-                print 'total_query',total_query
-                metadata_groups[group] = Document.objects.filter(total_query)
-                print 'documents',Document.objects.filter(total_query)
 
     preview_form = DocumentPreviewForm(document=document)
     form_list = [
@@ -284,10 +249,18 @@ def document_view(request, document_id):
 
     sidebar_groups = []
     for group, data in metadata_groups.items():
+        if len(data) or GROUP_SHOW_EMPTY:
+            if len(data):
+                if len(data) > GROUP_MAX_RESULTS:
+                    total_string = '(%s out of %s)' % (GROUP_MAX_RESULTS, len(data))
+                else:
+                    total_string = '(%s)' % len(data)
+            else:
+                total_string = ''
             sidebar_groups.append({
-            'title':group.label,
+                'title':'%s %s' % (group.label, total_string),
                 'name':'generic_list_subtemplate.html',
-            'object_list':data,
+                'object_list':data[:GROUP_MAX_RESULTS],
                 'hide_columns':True,
                 'hide_header':True,
                 })

docs/Changelog.txt

@@ -3,3 +3,4 @@
 * Show only document metadata in document list view.
 * If one document type exists, the create document wizard skips the first step.
 * Changed to a liquid css grid
+* Added the ability to group documents by their metadata

docs/TODO

@@ -25,6 +25,7 @@
 * Add css grids                                                        - DONE
 * If theres only one document type on db skip step 1 of wizard         - DONE
 * Be able to delete staging file                                       - DONE
+* Group documents by metadata                                          - DONE
 * Document list filtering by metadata
 * Filterform date filtering widget
 * Validate GET data before saving file
@@ -56,7 +57,6 @@
 * Add unpaper to pre OCR document cleanup
 * Support distributed OCR queues (RabbitMQ & Celery?)
 * DXF viewer - http://code.google.com/p/dxf-reader/source/browse/#svn%2Ftrunk
-* Group documents by metadata
 * Support spreadsheets, wordprocessing docs using openoffice in server mode
 * WebDAV support
 * Handle ziped or rar archives

settings.py

@@ -191,6 +191,10 @@ LOGIN_EXEMPT_URLS = (
 #DOCUMENTS_THUMBNAIL_SIZE = '50x50'
 #DOCUMENTS_DISPLAY_SIZE = '1024x768'
 
+# Groups
+#DOCUMENTS_GROUP_MAX_RESULTS = 20
+#DOCUMENTS_GROUP_SHOW_EMPTY = True
+
 # Serving
 #DOCUMENTS_FILESYSTEM_FILESERVING_ENABLE = True
 #DOCUMENTS_FILESYSTEM_FILESERVING_PATH = u'/tmp/mayan/documents'
@@ -202,6 +206,7 @@ LOGIN_EXEMPT_URLS = (
 #CONVERTER_CONVERT_PATH = u'/usr/bin/convert'
 #CONVERTER_OCR_OPTIONS = u'-colorspace Gray -depth 8 -resample 200x200'
 #OCR_TESSERACT_PATH = u'/usr/bin/tesseract'
+
 # Override
 SEARCH_SHOW_OBJECT_TYPE = False
 #======== End of configuration options =======