.dockerignore

@@ -1,10 +0,0 @@
-media/
-static/
-documentation/
-.env
-.venv/
-.vscode/
-.git/
-.gitignore
-.woodpecker/
-**/__pycache__/

Dockerfile

@@ -10,7 +10,7 @@ COPY scripts/ /app/scripts/
 RUN chmod +x /app/scripts/start.sh
 
 # Install packages
-RUN apt update && apt install -y graphviz libgraphviz-dev graphviz-dev tesseract-ocr
+RUN apt update && apt install -y graphviz libgraphviz-dev graphviz-dev
 RUN pip3 install --upgrade pip && pip3 install --no-cache-dir -r requirements.txt
 
 # Expose port 8000 for the web server

docmanager_backend/documents/apps.py

@@ -4,7 +4,3 @@ from django.apps import AppConfig
 class DocumentsConfig(AppConfig):
     default_auto_field = "django.db.models.BigAutoField"
     name = "documents"
-
-    def ready(self) -> None:
-        import documents.signals
-        return super().ready()

docmanager_backend/documents/migrations/0002_document_metadata_alter_document_document_type.py

@@ -1,25 +0,0 @@
-# Generated by Django 5.1.3 on 2024-11-24 05:17
-
-from django.db import migrations, models
-
-
-class Migration(migrations.Migration):
-
-    dependencies = [
-        ("documents", "0001_initial"),
-    ]
-
-    operations = [
-        migrations.AddField(
-            model_name="document",
-            name="metadata",
-            field=models.TextField(null=True),
-        ),
-        migrations.AlterField(
-            model_name="document",
-            name="document_type",
-            field=models.CharField(
-                choices=[("memorandum", "Memorandum"), ("hoa", "HOA")], max_length=32
-            ),
-        ),
-    ]

docmanager_backend/documents/migrations/0003_remove_document_metadata_document_ocr_metadata.py

@@ -1,22 +0,0 @@
-# Generated by Django 5.1.3 on 2024-11-24 06:04
-
-from django.db import migrations, models
-
-
-class Migration(migrations.Migration):
-
-    dependencies = [
-        ("documents", "0002_document_metadata_alter_document_document_type"),
-    ]
-
-    operations = [
-        migrations.RemoveField(
-            model_name="document",
-            name="metadata",
-        ),
-        migrations.AddField(
-            model_name="document",
-            name="ocr_metadata",
-            field=models.TextField(blank=True, null=True),
-        ),
-    ]

docmanager_backend/documents/models.py

@@ -1,4 +1,3 @@
-
 from django.db import models
 from django.utils.timezone import now
 import uuid
@@ -8,15 +7,17 @@ class Document(models.Model):
     name = models.CharField(max_length=100)
 
     DOCUMENT_TYPE_CHOICES = (
-        ("memorandum", "Memorandum"),
+        ("pdf", "PDF"),
-        ("hoa", "HOA"),
+        ("image", "Image"),
-        # TODO: Update this list on types of documents
+        ("video", "Video"),
+        ("doc", "Word Document"),
+        ("excel", "Excel Document"),
+        ("ppt", "Powerpoint Document"),
     )
     document_type = models.CharField(
         max_length=32, choices=DOCUMENT_TYPE_CHOICES, null=False, blank=False
     )
     number_pages = models.IntegerField(null=False, blank=False)
-    ocr_metadata = models.TextField(null=True, blank=True)
 
     def upload_to(instance, filename):
         _, extension = filename.split(".")

docmanager_backend/documents/serializers.py

@@ -36,13 +36,12 @@ class DocumentSerializer(serializers.ModelSerializer):
     class Meta:
         model = Document
         fields = ["id", "name", "document_type",
-                  "number_pages", "ocr_metadata", "date_uploaded"]
+                  "number_pages", "date_uploaded"]
         read_only_fields = [
             "id",
             "name",
             "document_type",
             "number_pages",
-            "ocr_metadata",
             "date_uploaded",
         ]
 
@@ -60,17 +59,15 @@ class DocumentFileSerializer(serializers.ModelSerializer):
             "id",
             "name",
             "document_type",
-            "number_pages",
-            "ocr_metadata",
-            "date_uploaded",
             "file",
+            "number_pages",
+            "date_uploaded",
         ]
         read_only_fields = [
             "id",
             "name",
             "document_type",
             "number_pages",
-            "ocr_metadata",
             "date_uploaded",
             "file",
         ]

docmanager_backend/documents/signals.py

@@ -1,26 +0,0 @@
-from django.db.models.signals import post_save
-from django.dispatch import receiver
-from config.settings import MEDIA_ROOT
-import os
-import fitz
-import pytesseract
-from PIL import Image
-from .models import Document
-
-
-@receiver(post_save, sender=Document)
-def domain_post_save(sender, instance, **kwargs):
-    if not instance.ocr_metadata:
-        metadata = ""
-        with fitz.open(os.path.join(MEDIA_ROOT, instance.file.name)) as doc:
-            mat = fitz.Matrix(1.2, 1.2)
-            for page in doc:
-                pix = page.get_pixmap(matrix=mat)
-                output = f'{page.number}.jpg'
-                pix.save(output)
-                res = str(pytesseract.image_to_string(Image.open(output)))
-                os.remove(output)
-                metadata += res
-
-        instance.ocr_metadata = metadata
-        instance.save()

requirements.txt

@@ -11,7 +11,6 @@ defusedxml==0.8.0rc2
 Django==5.1.3
 django-cleanup==9.0.0
 django-cors-headers==4.6.0
-django-extensions==3.2.3
 django-rest-framework==0.1.0
 django-unfold==0.41.0
 djangorestframework==3.15.2
@@ -19,28 +18,18 @@ djangorestframework-simplejwt==5.3.1
 djoser==2.3.1
 drf-spectacular==0.27.2
 drf-spectacular-sidecar==2024.11.1
-filelock==3.16.1
-fsspec==2024.10.0
 gunicorn==23.0.0
 idna==3.10
 inflection==0.5.1
-Jinja2==3.1.4
 jsonschema==4.23.0
 jsonschema-specifications==2024.10.1
-MarkupSafe==2.1.5
-mpmath==1.3.0
 mypy-extensions==1.0.0
-networkx==3.4.2
 oauthlib==3.2.2
 packaging==24.2
 pathspec==0.12.1
-pillow==11.0.0
 platformdirs==4.3.6
 pycparser==2.22
-pyflakes==3.2.0
 PyJWT==2.10.0
-PyMuPDF==1.24.14
-pytesseract==0.3.13
 python-dotenv==1.0.1
 python3-openid==3.2.0
 PyYAML==6.0.2
@@ -48,14 +37,12 @@ referencing==0.35.1
 requests==2.32.3
 requests-oauthlib==2.0.0
 rpds-py==0.21.0
-setuptools==70.2.0
 social-auth-app-django==5.4.2
 social-auth-core==4.5.4
 sqlparse==0.5.2
-sympy==1.13.1
-typing_extensions==4.12.2
 tzdata==2024.2
 uritemplate==4.1.1
 urllib3==2.2.3
 whitenoise==6.8.2
+django-extensions==3.2.3
 pygraphviz==1.14; platform_system == 'Linux'