Add OCR metadata field processing for documents

docmanager_backend/documents/apps.py

@@ -4,3 +4,7 @@ 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

@@ -0,0 +1,25 @@
+# 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

@@ -0,0 +1,22 @@
+# 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

@@ -1,3 +1,4 @@
+
 from django.db import models
 from django.utils.timezone import now
 import uuid
@@ -7,17 +8,15 @@ class Document(models.Model):
     name = models.CharField(max_length=100)
 
     DOCUMENT_TYPE_CHOICES = (
-        ("pdf", "PDF"),
+        ("memorandum", "Memorandum"),
-        ("image", "Image"),
+        ("hoa", "HOA"),
-        ("video", "Video"),
+        # TODO: Update this list on types of documents
-        ("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

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

docmanager_backend/documents/signals.py

@@ -0,0 +1,26 @@
+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()