Added document type

src/documents/actions.py

@@ -4,7 +4,7 @@
 from django.core.exceptions import PermissionDenied
 from django.template.response import TemplateResponse
 
-from documents.models import Tag, Correspondent
+from documents.models import Tag, Correspondent, DocumentType
 
 
 def add_tag_to_selected(modeladmin, request, queryset):
@@ -159,3 +159,67 @@ def remove_correspondent_from_selected(modeladmin, request, queryset):
 
 
 remove_correspondent_from_selected.short_description = "Remove correspondent from selected documents"
+
+
+def set_document_type_on_selected(modeladmin, request, queryset):
+    opts = modeladmin.model._meta
+    app_label = opts.app_label
+
+    if not modeladmin.has_change_permission(request):
+        raise PermissionDenied
+
+    if request.POST.get('post'):
+        n = queryset.count()
+        document_type = DocumentType.objects.get(id=request.POST.get('document_type_id'))
+        if n:
+            for obj in queryset:
+                obj_display = str(obj)
+                modeladmin.log_change(request, obj, obj_display)
+            queryset.update(document_type=document_type)
+            modeladmin.message_user(request, "Successfully set document type %(document_type)s on %(count)d %(items)s." % {
+                "document_type": document_type.name, "count": n, "items": model_ngettext(modeladmin.opts, n)
+            }, messages.SUCCESS)
+
+        # Return None to display the change list page again.
+        return None
+
+    title = "Set document type on multiple documents"
+
+    context = dict(
+        modeladmin.admin_site.each_context(request),
+        title=title,
+        queryset=queryset,
+        opts=opts,
+        action_checkbox_name=helpers.ACTION_CHECKBOX_NAME,
+        media=modeladmin.media,
+        document_types=DocumentType.objects.all()
+    )
+
+    request.current_app = modeladmin.admin_site.name
+
+    return TemplateResponse(request,
+        "admin/%s/%s/set_document_type.html" % (app_label, opts.model_name)
+    , context)
+
+
+set_document_type_on_selected.short_description = "Set document type on selected documents"
+
+
+def remove_document_type_from_selected(modeladmin, request, queryset):
+    if not modeladmin.has_change_permission(request):
+        raise PermissionDenied
+
+    n = queryset.count()
+    if n:
+        for obj in queryset:
+            obj_display = str(obj)
+            modeladmin.log_change(request, obj, obj_display)
+        queryset.update(document_type=None)
+        modeladmin.message_user(request, "Successfully removed document type from %(count)d %(items)s." % {
+            "count": n, "items": model_ngettext(modeladmin.opts, n)
+        }, messages.SUCCESS)
+
+    return None
+
+
+remove_document_type_from_selected.short_description = "Remove document type from selected documents"

src/documents/admin.py

@@ -12,8 +12,8 @@
 from django.utils.safestring import mark_safe
 
 from documents.actions import add_tag_to_selected, remove_tag_from_selected, set_correspondent_on_selected, \
-    remove_correspondent_from_selected
-from .models import Correspondent, Tag, Document, Log
+    remove_correspondent_from_selected, set_document_type_on_selected, remove_document_type_from_selected
+from .models import Correspondent, Tag, Document, Log, DocumentType
 
 
 class FinancialYearFilter(admin.SimpleListFilter):
@@ -120,6 +120,22 @@ def save_model(self, request, obj, form, change):
     def document_count(self, obj):
         return obj.documents.count()
 
+class DocumentTypeAdmin(CommonAdmin):
+
+    list_display = ("name", "match", "matching_algorithm", "document_count")
+    list_filter = ("matching_algorithm",)
+    list_editable = ("match", "matching_algorithm")
+
+    def save_model(self, request, obj, form, change):
+        super().save_model(request, obj, form, change)
+
+        for document in Document.objects.filter(document_type__isnull=True).exclude(tags__is_archived_tag=True):
+            if obj.matches(document.content):
+                document.document_type = obj
+                document.save(update_fields=("document_type",))
+
+    def document_count(self, obj):
+        return obj.documents.count()
 
 class DocumentAdmin(CommonAdmin):
 
@@ -132,12 +148,12 @@ class Media:
     search_fields = ("correspondent__name", "title", "content", "tags__name")
     readonly_fields = ("added",)
     list_display = ("title", "created", "added", "thumbnail", "correspondent",
-                    "tags_", "archive_serial_number")
-    list_filter = ("tags", "correspondent", FinancialYearFilter)
+                    "tags_", "archive_serial_number", "document_type")
+    list_filter = ("document_type", "tags", "correspondent", FinancialYearFilter)
 
     ordering = ["-created", "correspondent"]
 
-    actions = [add_tag_to_selected, remove_tag_from_selected, set_correspondent_on_selected, remove_correspondent_from_selected]
+    actions = [add_tag_to_selected, remove_tag_from_selected, set_correspondent_on_selected, remove_correspondent_from_selected, set_document_type_on_selected, remove_document_type_from_selected]
 
     date_hierarchy = 'created'
 
@@ -273,6 +289,7 @@ class LogAdmin(CommonAdmin):
 
 admin.site.register(Correspondent, CorrespondentAdmin)
 admin.site.register(Tag, TagAdmin)
+admin.site.register(DocumentType, DocumentTypeAdmin)
 admin.site.register(Document, DocumentAdmin)
 admin.site.register(Log, LogAdmin)
 

src/documents/apps.py

@@ -13,6 +13,7 @@ def ready(self):
         from .signals.handlers import (
             set_correspondent,
             set_tags,
+            set_document_type,
             run_pre_consume_script,
             run_post_consume_script,
             cleanup_document_deletion,
@@ -23,6 +24,7 @@ def ready(self):
 
         document_consumption_finished.connect(set_tags)
         document_consumption_finished.connect(set_correspondent)
+        document_consumption_finished.connect(set_document_type)
         document_consumption_finished.connect(set_log_entry)
         document_consumption_finished.connect(run_post_consume_script)
 

src/documents/migrations/0023_auto_20180823_1155.py

@@ -0,0 +1,33 @@
+# Generated by Django 2.0.7 on 2018-08-23 11:55
+
+from django.db import migrations, models
+import django.db.models.deletion
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ('documents', '0022_workflow_improvements'),
+    ]
+
+    operations = [
+        migrations.CreateModel(
+            name='DocumentType',
+            fields=[
+                ('id', models.AutoField(auto_created=True, primary_key=True, serialize=False, verbose_name='ID')),
+                ('name', models.CharField(max_length=128, unique=True)),
+                ('slug', models.SlugField(blank=True)),
+                ('match', models.CharField(blank=True, max_length=256)),
+                ('matching_algorithm', models.PositiveIntegerField(choices=[(1, 'Any'), (2, 'All'), (3, 'Literal'), (4, 'Regular Expression'), (5, 'Fuzzy Match')], default=1, help_text='Which algorithm you want to use when matching text to the OCR\'d PDF.  Here, "any" looks for any occurrence of any word provided in the PDF, while "all" requires that every word provided appear in the PDF, albeit not in the order provided.  A "literal" match means that the text you enter must appear in the PDF exactly as you\'ve entered it, and "regular expression" uses a regex to match the PDF.  (If you don\'t know what a regex is, you probably don\'t want this option.)  Finally, a "fuzzy match" looks for words or phrases that are mostly—but not exactly—the same, which can be useful for matching against documents containg imperfections that foil accurate OCR.')),
+                ('is_insensitive', models.BooleanField(default=True)),
+            ],
+            options={
+                'abstract': False,
+            },
+        ),
+        migrations.AddField(
+            model_name='document',
+            name='document_type',
+            field=models.ForeignKey(blank=True, null=True, on_delete=django.db.models.deletion.SET_NULL, related_name='documents', to='documents.DocumentType'),
+        ),
+    ]

src/documents/models.py

@@ -15,6 +15,7 @@
 from django.template.defaultfilters import slugify
 from django.utils import timezone
 
+from reminders.models import Reminder
 from .managers import LogManager
 
 
@@ -189,6 +190,11 @@ class Tag(MatchingModel):
         help_text="Marks this tag as an archive tag: All documents tagged with archive tags will never be modified automatically (i.e., modifying tags by matching rules)")
 
 
+class DocumentType(MatchingModel):
+
+    pass
+
+
 class Document(models.Model):
 
     TYPE_PDF = "pdf"
@@ -215,6 +221,14 @@ class Document(models.Model):
 
     title = models.CharField(max_length=128, blank=True, db_index=True)
 
+    document_type = models.ForeignKey(
+        DocumentType,
+        blank=True,
+        null=True,
+        related_name="documents",
+        on_delete=models.SET_NULL
+    )
+
     content = models.TextField(
         db_index=True,
         blank=True,

src/documents/signals/handlers.py

@@ -8,7 +8,7 @@
 from django.contrib.contenttypes.models import ContentType
 from django.utils import timezone
 
-from ..models import Correspondent, Document, Tag
+from ..models import Correspondent, Document, Tag, DocumentType
 
 
 def logger(message, group):
@@ -44,6 +44,35 @@ def set_correspondent(sender, document=None, logging_group=None, **kwargs):
     document.save(update_fields=("correspondent",))
 
 
+def set_document_type(sender, document=None, logging_group=None, **kwargs):
+
+    # No sense in assigning a correspondent when one is already set.
+    if document.document_type:
+        return
+
+    # No matching document types, so no need to continue
+    potential_document_types = list(DocumentType.match_all(document.content))
+    if not potential_document_types:
+        return
+
+    potential_count = len(potential_document_types)
+    selected = potential_document_types[0]
+    if potential_count > 1:
+        message = "Detected {} potential document types, so we've opted for {}"
+        logger(
+            message.format(potential_count, selected),
+            logging_group
+        )
+
+    logger(
+        'Assigning document type "{}" to "{}" '.format(selected, document),
+        logging_group
+    )
+
+    document.document_type = selected
+    document.save(update_fields=("document_type",))
+
+
 def set_tags(sender, document=None, logging_group=None, **kwargs):
 
     current_tags = set(document.tags.all())

src/documents/templates/admin/documents/document/change_list_results.html

@@ -24,7 +24,7 @@
     border: 1px solid #cccccc;
     border-radius: 2%;
     overflow: hidden;
-    height: 300px;
+    height: 350px;
     position: relative;
   }
   .result .header {
@@ -61,6 +61,11 @@
   .result a.tag {
     color: #ffffff;
   }
+  .result .documentType {
+    padding: 5px;
+    background-color: #eeeeee;
+    text-align: center;
+  }
   .result .date {
     padding: 5px;
   }
@@ -163,7 +168,8 @@
     {# 5: Correspondent #}
     {# 6: Tags #}
     {# 7: Archive serial number #}
-    {# 8: Document edit url #}
+    {# 8: Document type #}
+    {# 9: Document edit url #}
     <div class="box">
       <div class="result">
         <div class="header">
@@ -177,14 +183,15 @@
             selection would not be possible with mouse click + drag. Instead,
             the underlying link would be dragged.
           {% endcomment %}
-          <div class="headerLink" onclick="location.href='{{ result.8 }}';"></div>
+          <div class="headerLink" onclick="location.href='{{ result.9 }}';"></div>
           <div class="checkbox">{{ result.0 }}</div>
           <div class="info">
             {{ result.5 }}
           </div>
           {{ result.1 }}
           <div style="clear: both;"></div>
         </div>
+        {% if '>-<' not in result.8 %}<div class="documentType">{{ result.8 }}</div>{% endif %}
         <div class="tags">{{ result.6 }}</div>
         <div class="date">{{ result.2 }}</div>
         <div style="clear: both;"></div>

src/documents/templates/admin/documents/document/set_document_type.html

@@ -0,0 +1,46 @@
+{% extends "admin/base_site.html" %}
+{% load i18n l10n admin_urls static %}
+{% load staticfiles %}
+
+{% block extrahead %}
+{{ block.super }}
+{{ media }}
+<script type="text/javascript" src="{% static 'admin/js/cancel.js' %}"></script>
+
+{% endblock %}
+
+{% block bodyclass %}{{ block.super }} app-{{ opts.app_label }} model-{{ opts.model_name }} delete-confirmation delete-selected-confirmation{% endblock %}
+
+{% block breadcrumbs %}
+<div class="breadcrumbs">
+    <a href="{% url 'admin:index' %}">{% trans 'Home' %}</a>
+    &rsaquo; <a href="{% url 'admin:app_list' app_label=opts.app_label %}">{{ opts.app_config.verbose_name }}</a>
+    &rsaquo; <a href="{% url opts|admin_urlname:'changelist' %}">{{ opts.verbose_name_plural|capfirst }}</a>
+    &rsaquo; {{title}}
+</div>
+{% endblock %}
+
+{% block content %}
+<p>Please select the document type.</p>
+<form method="post">{% csrf_token %}
+    <div>
+        {% for obj in queryset %}
+        <input type="hidden" name="{{ action_checkbox_name }}" value="{{ obj.pk|unlocalize }}"/>
+        {% endfor %}
+        <p>
+            <select name="document_type_id">
+                {% for document_type in document_types %}
+                <option value="{{document_type.id}}">{{document_type.name}}</option>
+                {% endfor %}
+            </select>
+        </p>
+
+        <input type="hidden" name="action" value="set_document_type_on_selected"/>
+        <input type="hidden" name="post" value="yes"/>
+        <p>
+            <input type="submit" value="{% trans " Confirm" %}" />
+            <a href="#" class="button cancel-link">{% trans "Go back" %}</a>
+        </p>
+    </div>
+</form>
+{% endblock %}