updated logic to crop pdfs

.gitignore

@@ -159,3 +159,5 @@ cython_debug/
 #  and can be added to the global gitignore or merged into this file.  For a more nuclear
 #  option (not recommended) you can uncomment the following to ignore the entire idea folder.
 #.idea/
+
+.DS_Store

app.py

@@ -12,7 +12,7 @@ import pytesseract
 def save_pdf_file(file_name, page):
     pdf_writer = PyPDF2.PdfWriter()
     pdf_writer.add_page(page)
-
+    
     #return pdf stream
     with open(f"files/{file_name}", "wb") as out:
         pdf_writer.write(out)
@@ -24,12 +24,13 @@ def process_estafeta_pdf(file_name, page):
     y = page['/MediaBox'][3]
 
     if x > y:
-        page.cropbox.upper_left = (40, 60)
-        page.cropbox.lower_right = (340,520)
+        page.mediabox.lower_left = (40, 60)
+        page.mediabox.upper_right = (340,520)
     else:
-        page.cropbox.upper_left = (30,40)
-        page.cropbox.lower_right = (490,340)
+        page.mediabox.lower_left = (30,40)
+        page.mediabox.upper_right = (490,340)
     
+
     save_pdf_file(file_name, page)
 
 def process_estafeta_text(pdf_text):
@@ -37,21 +38,17 @@ def process_estafeta_text(pdf_text):
     match = re.search(r'CONFIRMACION (\d+-\d+\w+)', pdf_text)
     if match:
         extracted_text = match.group(1)
-        index = -1
-        for i, char in enumerate(reversed(extracted_text)):
-            if char.isalpha():
-                index = len(extracted_text) - i
-                break
-    
-    extracted_text = extracted_text[0:index].replace("-", "")
+        extracted_text = extracted_text[0:23].replace("-", "")
+    else:
+        raise Exception("Pattern not found in the text.")
+
     file_name = extracted_text + ".pdf"
     file_path = f"files/{file_name}"
     return file_name, file_path
 
-
 def process_dhl_pdf(file_name, page):
-    page.cropbox.upper_left = (92,20)
-    page.cropbox.lower_right = (360,560)
+    page.mediabox.lower_left = (92,20)
+    page.mediabox.upper_right = (360,560)
     save_pdf_file(file_name, page)
 
 def process_dhl_text(pdf_text):
@@ -63,7 +60,10 @@ def process_dhl_text(pdf_text):
     file_path = f"files/{file_name}"
     return file_name, file_path
 
-def process_ups_pdf(file_name, page):
+def process_ups_pdf(page):
+    page.rotate(90)
+  
+def process_pdf_from_image(file_name, page):
     page.rotate(90)
     save_pdf_file(file_name, page)
     file_path = f"files/{file_name}"
@@ -71,25 +71,35 @@ def process_ups_pdf(file_name, page):
     images = convert_from_path(file_path)
     image = images[0]
     image.save(f"{file_path}.png", "PNG")
+    
+    # open the image and extract the text
     loaded_image = Image.open(f"{file_path}.png")
     extracted_text = pytesseract.image_to_string(loaded_image)
+
     image.close()
-    loaded_image.close()
     os.remove(f"{file_path}.png")
-    match = re.search(r'TRACKING #:\s+([A-Z\d\s]+)', extracted_text)
-    if match:
-        extracted_text = match.group(1)
-        extracted_text = extracted_text.replace(" ", "").replace("BILLING", "").replace("\n", "")
-        file_name = extracted_text + ".pdf"
+
+    if re.search("Fed2x", extracted_text, re.IGNORECASE):
+        file_name, file_path = process_fedex_text(extracted_text)
+        process_fedex_pdf(file_name, page)
     else:
-        print("Pattern not found in the text.")
+        image_path = f"{file_path}-resized.pdf"
+        loaded_image.resize((400, 500)).save(image_path, "PDF")    
+        loaded_image.close()
+        match = re.search(r'TRACKING #:\s+([A-Z\d\s]+)', extracted_text)
+        file_path = image_path
+        if match:
+            extracted_text = match.group(1)
+            extracted_text = extracted_text.replace(" ", "").replace("BILLING", "").replace("\n", "")
+            file_name = extracted_text + ".pdf"
+        else:
+            print("Pattern not found in the text.")
 
     return file_name, file_path
 
-
 def process_coppel_pdf(file_name, page):
-    page.cropbox.upper_left = (0,150)
-    page.cropbox.lower_right = (400,520)
+    page.mediabox.lower_left = (0,150)
+    page.mediabox.upper_right = (290,520)
     save_pdf_file(file_name, page)
 
 def process_coppel_text(pdf_text):
@@ -103,6 +113,15 @@ def process_coppel_text(pdf_text):
     file_path = f"files/{file_name}"
     return file_name, file_path
 
+def process_fedex_pdf(file_name, page):
+    page.rotate(-90)
+    page.mediabox.lower_left = (0, 0)
+    page.mediabox.upper_right = (500,650)
+    save_pdf_file(file_name, page)
+
+def process_fedex_text(pdf_text):
+     return "fedex.pdf", "files/fedex.pdf"
+
 def process_pdf_file(file): 
     """
     This function processes the PDF file and returns the file name, file path and transport company
@@ -115,7 +134,6 @@ def process_pdf_file(file):
     file_path (str): The path of the file
     transport_company (str): The transport company
     """
-
     pdf_stream = BytesIO(file.content)
     pdf = PyPDF2.PdfReader(pdf_stream)
     page = pdf.pages[0]
@@ -139,7 +157,7 @@ def process_pdf_file(file):
         process_coppel_pdf(file_name, page)
     else:
         transport_company = "ups"
-        file_name, file_path = process_ups_pdf(file_name, page)
+        file_name, file_path = process_pdf_from_image(file_name, page)
 
     pdf_stream.close()
     return file_name, file_path, transport_company