From b1eaf9c7bca47b059a38242bf341836dc1d7d53d Mon Sep 17 00:00:00 2001
From: Tris Forster <tris@shoddynet.org>
Date: Thu, 2 Mar 2023 10:05:26 +1100
Subject: [PATCH] Added document extractor

---
 app/library/models.py                         | 17 +++++++---
 app/library/pdf_utils.py                      |  4 +--
 .../templates/library/document_annotate.html  |  2 +-
 .../templates/library/work_detail.html        | 21 +++++++-----
 app/library/tests.py                          | 10 +++---
 app/library/urls.py                           |  1 +
 app/library/views/__init__.py                 | 34 ++++++++++++++++++-
 app/library/views/api.py                      | 19 ++++++-----
 8 files changed, 77 insertions(+), 31 deletions(-)

diff --git a/app/library/models.py b/app/library/models.py
index 63b68f0..77cb216 100644
--- a/app/library/models.py
+++ b/app/library/models.py
@@ -202,13 +202,14 @@ class Work(models.Model):
     def folder(self):
         return f"{slugify(self.composer)}/{slugify(self.name)}-{self.pk:04d}"
 
-    def extract(self, *tags):
-
+    def tagged_sections(self, *tags):
         qs = self.docs.filter(sections__tag__in=tags)
         qs = qs.annotate(Count('sections'), end=Min('sections__end'), start=Max('sections__start')) \
             .filter(sections__count=len(tags))
-        
-        return list(qs.values_list('upload', 'start', 'end'))
+        return qs
+
+    def list_sections(self, *tags):
+        return list(self.tagged_sections(*tags).values_list('upload', 'start', 'end'))
 
     @property
     def digital_parts(self):
@@ -276,6 +277,14 @@ class Work(models.Model):
         assigned = set(self.assigned_instruments())
         return [ x for x in self.orchestration.as_list() if not x[0] in assigned ]
 
+    def music_tags(self):
+        tags = dict(self.orchestration.as_list())
+        for section in Section.objects.filter(doc__work_id=self.pk):
+            tags.setdefault(section.tag, section.name)
+
+        return tags.items()
+
+
     def __str__(self):
         return f"{self.name} ({self.composer})"
 
diff --git a/app/library/pdf_utils.py b/app/library/pdf_utils.py
index c0e5b67..a9eec64 100644
--- a/app/library/pdf_utils.py
+++ b/app/library/pdf_utils.py
@@ -24,12 +24,12 @@ def extract_and_concat(items):
         if count == 0:
             continue
 
-        if start is None:
+        if not start:
             sections.append(source)
 
         else:
 
-            if end is None:
+            if not end:
                 end = start
 
             dest = os.path.join(d.name, f'section_{i}.pdf')
diff --git a/app/library/templates/library/document_annotate.html b/app/library/templates/library/document_annotate.html
index 2cf3fba..78bd13b 100644
--- a/app/library/templates/library/document_annotate.html
+++ b/app/library/templates/library/document_annotate.html
@@ -75,7 +75,7 @@
     </div>
   </div>
   <ul id="unassigned-area">
-  {% for tag, inst in document.work.unassigned_instruments %}
+  {% for tag, inst in document.work.music_tags %}
     <li class="is-clickable" onclick="assignInstrument('{{tag}}', this)")>{{ inst }}</li>
   {% endfor %}
     <li><a onclick="document.getElementById('add-modal').classList.add('is-active')">Add instrument</a></li>
diff --git a/app/library/templates/library/work_detail.html b/app/library/templates/library/work_detail.html
index afc392e..ff49a66 100644
--- a/app/library/templates/library/work_detail.html
+++ b/app/library/templates/library/work_detail.html
@@ -87,15 +87,18 @@
                     Digital Parts
                 </h4>
                 <div class="tags">
-                    {% if work.digital_parts %}
-                    <a class="tag is-danger" href="{% url 'work_partset' collection.pk work.pk %}">Full Set</a>
-                    {% endif %}
-                    {% for section in work.digital_parts %}
-                    <a class="tag is-info" href="{% url 'part_download' collection.pk section.pk section.filename %}"
-                        target="section_{{ section.pk }}" rel="">{{ section.name }}</a>
-                    {% empty %}
-                    <p class="is-italic">No digital parts available</p>
-                    {% endfor %}
+                    {% with sections=work.digital_parts %}
+                        {% if sections %}
+                        <a class="tag is-danger" href="{% url 'work_partset' collection.pk work.pk %}">Full Set</a>
+                        {% endif %}
+                        {% regroup sections by tag as unique_sections %}
+                        {% for section in unique_sections %}
+                        <a class="tag is-info" href="{% url 'work_download' collection.pk work.pk %}?tag={{ section.list.0.tag }}"
+                            target="section_{{ section.list.0.pk }}" rel="">{{ section.list.0.name }}</a>
+                        {% empty %}
+                        <p class="is-italic">No digital parts available</p>
+                        {% endfor %}
+                    {% endwith %}
                 </div>
             </div>
         </div>
diff --git a/app/library/tests.py b/app/library/tests.py
index b6c6a1a..3063ca0 100644
--- a/app/library/tests.py
+++ b/app/library/tests.py
@@ -162,15 +162,15 @@ class LibraryTestCase(AccessTestCase):
             doc.sections.create(tag=g)
 
         # no tags - get nothing (should it be everything?)
-        self.assertEqual(work.extract(), [])
+        self.assertEqual(work.list_sections(), [])
 
         # single tag - should get just that range
-        self.assertEqual(work.extract('vl-1'), [('sel/beethoven/some_quartet/some_quartet_vl-1.pdf', None, None)])
+        self.assertEqual(work.list_sections('vl-1'), [('sel/beethoven/some_quartet/some_quartet_vl-1.pdf', None, None)])
 
         # single tag - returns all documents with that range
-        result = work.extract('mvmt-2')
+        result = work.list_sections('mvmt-2')
         self.assertEqual(len(result), 4)
 
         # multiple tags - returns the overlapping portion of all documents that have all tags
-        self.assertEqual(work.extract('vl-1', 'mvmt-2'), [('sel/beethoven/some_quartet/some_quartet_vl-1.pdf', 4, 8)])
-        self.assertEqual(work.extract('vl-1', 'vl-2'), [])
\ No newline at end of file
+        self.assertEqual(work.list_sections('vl-1', 'mvmt-2'), [('sel/beethoven/some_quartet/some_quartet_vl-1.pdf', 4, 8)])
+        self.assertEqual(work.list_sections('vl-1', 'vl-2'), [])
\ No newline at end of file
diff --git a/app/library/urls.py b/app/library/urls.py
index 431b577..f816add 100644
--- a/app/library/urls.py
+++ b/app/library/urls.py
@@ -27,6 +27,7 @@ urlpatterns = [
     path('collections/<int:collection>/works/<int:pk>/partset', views.WorkPartSetView.as_view(), name="work_partset"),
     path('collections/<int:collection>/works/<int:pk>/add_to_project', views.WorkAddToProject.as_view(), name="work_add_to_project"),
     path('collections/<int:collection>/works/<int:pk>/upload', views.WorkAddDocumentView.as_view(), name="document_add"),
+    path('collections/<int:collection>/works/<int:pk>/download', views.WorkDownloadView.as_view(), name="work_download"),
 
     path('collections/<int:collection>/docs/<int:pk>/delete', views.DocumentDeleteView.as_view(), name="document_delete"),
     path('collections/<int:collection>/docs/<int:pk>/download', views.DocumentDownloadView.as_view(), name="document_download"),
diff --git a/app/library/views/__init__.py b/app/library/views/__init__.py
index 7af3ddb..ed9dc15 100644
--- a/app/library/views/__init__.py
+++ b/app/library/views/__init__.py
@@ -19,7 +19,7 @@ import re
 from interface.views import EnsembleMixin, ProjectMixin, AuthorizedResourceMixin
 from interface.models import Project
 from library.models import Collection, Work, Document, Section
-from library.music_tags import MUSIC_TAGS, MUSIC_TAG_BY_NAME
+from library.music_tags import MUSIC_TAGS, MUSIC_TAG_BY_NAME, MusicTag
 from library import forms, models
 from library.pdf_utils import extract_pages, extract_and_concat
 
@@ -308,6 +308,38 @@ class WorkPartSetView(CollectionMixin, DetailView):
             works = works.filter(collection__allowed_ensembles__ensemble=self.request.ensemble_id)
         return works
 
+class WorkDownloadView(CollectionMixin, SingleObjectMixin, View):
+    model = models.Work
+
+    def get(self, request, *args, **kwargs):
+        self.object = self.get_object()
+
+        tags = request.GET.getlist('tag')
+
+        if not tags:
+            raise Http404("No tags given")
+        
+        sections = list(self.object.tagged_sections(*tags))
+        print(sections)
+
+        if len(sections) == 0:
+            raise Http404("No matching sections")
+
+        if len(sections) == 1 and sections[0].start == 0:
+            # bypass extraction and redirect to the url
+            logger.debug("Redirecting to url")
+            return redirect(sections[0].upload.url)
+
+        result = extract_and_concat([ (s.upload.path, s.upload.name, s.start, s.end, 1) for s in sections ])
+
+        tag_names = " - ".join([ str(MusicTag.from_tag(tag)) for tag in tags ])
+
+        download_name = f'{self.object.name} - {tag_names}.pdf'
+
+        response = FileResponse(result, content_type="application/pdf")
+        response['Content-Disposition'] = f'inline; filename="{download_name}"'
+        return response
+
 class WorkAddDocumentView(CollectionMixin, CreateView):
     template_name = "interface/default_form.html"
     model = Document
diff --git a/app/library/views/api.py b/app/library/views/api.py
index fa1b2b5..596a2d1 100644
--- a/app/library/views/api.py
+++ b/app/library/views/api.py
@@ -35,6 +35,7 @@ class WorkExportView(EnsembleMixin, WorkMixin, View):
 from interface.views import AuthorizedResourceMixin
 
 from rest_framework import routers, serializers, viewsets
+from rest_framework.exceptions import APIException
 
 from library.models import Collection, Work, Document, Section, WorkMeta
 
@@ -72,14 +73,12 @@ class SectionSerializer(serializers.ModelSerializer):
 
     def to_internal_value(self, data):
         tag, start, end = data.split(":")
-        try:
-            start = int(start)
-        except:
-            start = 0
-        try:
-            end = int(end)
-        except:
-            end = 0
+        start = int(start)
+        end = int(end)
+        if start < 1:
+            start = None
+        if end < 1:
+            end = None
         return super().to_internal_value({'tag': tag, 'start': start, 'end': end})
 
 class DocumentSerializer(serializers.ModelSerializer):
@@ -139,7 +138,9 @@ class WorkSerializer(serializers.ModelSerializer):
                 filename = os.path.basename(url.path)
 
                 r = requests.get(d['upload'], stream=True)
-                f = TemporaryUploadedFile(filename, r.headers['content-type'], r.headers['content-length'], r.encoding)
+                if r.status_code != 200:
+                    raise APIException("Failed to download file")
+                f = TemporaryUploadedFile(filename, r.headers['content-type'], r.headers.get('content-length'), r.encoding)
                 shutil.copyfileobj(r.raw, f.file)
                 r.close()
                 d['upload'] = f