Update api/views.py

api/views.py

@@ -11,13 +11,15 @@ import threading
 import random
 import google.generativeai as genai
 import google.ai.generativelanguage as glm
+import io
+import base64
 import os
 from .models import UseCases, DocumentTypes
 
 
-
+os.environ["_BARD_API_KEY"] = "WwhMDr8qCuXIPs1pwR-lOgPsB51q86WuevPCH5VpcCcxVEEQqywHYShAfkE19lCRXUOQaQ."
 # host_url = "http://16.170.244.54"
-host_url = "https://thejagstudio-absoluteai.hf.space/"
+host_url = "https://qbh39rzw-8000.euw.devtunnels.ms"
 googleAPIKey = "AIzaSyBeo4NGA__U6Xxy-aBE6yFm19pgq8TY-TM"
 genai.configure(api_key='AIzaSyALFCivW9GP25mbxL3W7Fv6u7m2ZHVlC8w')
 
@@ -25,9 +27,19 @@ genai.configure(api_key='AIzaSyALFCivW9GP25mbxL3W7Fv6u7m2ZHVlC8w')
 def getAnswer(images):
     url = "https://content-vision.googleapis.com/v1/images:annotate?alt=json&key="+googleAPIKey
     payload = {"requests": []}
-    for i in images:
+    for img in images:
+        # temp = {
+        #     "image": {"source": {"imageUri": i}},
+        #     "features": [
+        #         {
+        #             "type": "DOCUMENT_TEXT_DETECTION",
+        #             "maxResults": 50,
+        #             "model": "builtin/latest",
+        #         }
+        #     ],
+        # }
         temp = {
-            "image": {"source": {"imageUri": i}},
+            "image": {"content": img},
             "features": [
                 {
                     "type": "DOCUMENT_TEXT_DETECTION",
@@ -54,7 +66,7 @@ def getAnswer(images):
         for i in range(len(response.json()["responses"])):
             OCRString += "\n\n\n"+response.json()["responses"][i]["fullTextAnnotation"]["text"]
     except Exception as e:
-        print(e)
+        print(e, response.text)
     return OCRString
 
 
@@ -65,27 +77,40 @@ def dataExtract(request, link):
         pdf_file = request.FILES["pdf"]
         randomUUID = str(uuid.uuid4())
         fs = FileSystemStorage(location="static/pdf/")
-        filename = fs.save(f"{randomUUID}.pdf", pdf_file)
-        os.mkdir(f"./static/pages/{randomUUID}")
+        # filename = fs.save(f"{randomUUID}.pdf", pdf_file)
+        # os.mkdir(f"./static/pages/{randomUUID}")
         image_list = []
-        images = convert_from_path(
-            f"./static/pdf/{randomUUID}.pdf",
+        images = convert_from_bytes(
+            pdf_file.read(),
             dpi=150,
-            output_folder=f"./static/pages/{randomUUID}",
             fmt="png",
             output_file=f"image",
             thread_count=5,
             poppler_path="./poppler-23.05.0/Library/bin/"
         )
-        for filename in os.listdir(f"./static/pages/{randomUUID}"):
-            image_list.append(f"/static/pages/{randomUUID}/{filename}")
-        image_Array = []
-        for i in range(len(image_list)):
-            image_Array.append(host_url + image_list[i])
-
-        OCRString = getAnswer(image_Array)
-        with open(f"./OCR.txt", "w", encoding="utf-8") as f:
-            f.write(OCRString)
+        for img in images:
+            buffer = io.BytesIO()
+            img.save(buffer, format='PNG')
+            img_bytes = buffer.getvalue()
+            # Encode the bytes to base64
+            img_base64 = base64.b64encode(img_bytes).decode()
+            image_list.append(img_base64)
+        # images = convert_from_path(
+        #     f"./static/pdf/{randomUUID}.pdf",
+        #     dpi=150,
+        #     output_folder=f"./static/pages/{randomUUID}",
+        #     fmt="png",
+        #     output_file=f"image",
+        #     thread_count=5,
+        #     poppler_path="./poppler-23.05.0/Library/bin/"
+        # )
+        # for filename in os.listdir(f"./static/pages/{randomUUID}"):
+        #     image_list.append(f"/static/pages/{randomUUID}/{filename}")
+        # image_Array = []
+        # for i in range(len(image_list)):
+        #     image_Array.append(host_url + image_list[i])
+
+        OCRString = getAnswer(image_list)
         fields = documentData.fields
         properties = {}
         for field in fields: