fix: not classifying .pdf files

main.py

@@ -12,13 +12,8 @@ from transformers import pipeline
 
 app = FastAPI()
 
-pipe_flan = pipeline("text2text-generation", model="google/flan-t5-small")
-
-
-@app.get("/infer_t5")
-def t5(input):
-    output = pipe_flan(input)
-    return {"output": output[0]["generated_text"]}
+"""## Poppler dir"""
+poppler_path = "poppler-23.11.0/Library/bin"
 
 
 @app.post("/classify")
@@ -32,22 +27,22 @@ async def classify_doc(files: List[UploadFile] = File(...)):
 
             if filename.endswith('.pdf'):
                 try:
-                    pages = convert_from_bytes(open(file, 'rb').read())
-
+                    pages = convert_from_bytes(contents, poppler_path = poppler_path)
+                    print(pages)
                     for pagenum, image in enumerate(pages):
-                        # if pagenum != 0 and pagenum < len(pages):
-                        # if pagenum == 0:
-                            classificationRes =  classify_acct_dtype_str(image, filename)
-                            
+                            classificationRes, dtype_conf = doctype_classify(image.convert('RGB'), filename)
+
                             # add/update classification result dictionary
                             if (classificationRes in classificationResults):
                                 classificationResults.update({classificationRes : classificationResults[classificationRes] + 1})
                             else:
                                 classificationResults.update({classificationRes : 1})
-                        # break
-                except:
-                    return f"Error in opening {filename}"
+
+                except Exception as err:
+                    print(err)
+                    return f"Error in opening {filename}, {err}"
             
+            # png, jpg, jpeg files
             else:
                 classificationRes = classify_acct_dtype_str(contents, filename)
 
@@ -281,8 +276,8 @@ def classify_acct_dtype_str(content, filename):
     try:
         ipt = Image.open(BytesIO(content))
         dtype_inf, dtype_conf = doctype_classify(ipt, filename)
-    except:
-        return f"Error in opening {filename}"
+    except Exception as err:
+        return f"Error in opening {filename}, {err}"
 
     return dtype_inf