import/pdf: Read text from all pages

import/pdf.py

@@ -1,20 +1,16 @@
-from pypdf import PdfReader
-from beangulp import mimetypes, Importer
-from beangulp.cache import cache
 import re
 from datetime import date
-from beancount.core.data import Transaction, Posting, Document
+
-from beancount.core import flags
+from beancount.core import amount, data, flags
-from beancount.core import data
+from beancount.core.data import Document, Posting, Transaction
-from beancount.core import amount
 from beancount.core.number import D
+from beangulp import Importer, mimetypes
+from pypdf import PdfReader
 
 
-@cache
+def pdf_lines(filename):
-def pdf_to_text(filename):
     reader = PdfReader(filename)
-    page1 = reader.pages[0]
+    return [line for page in reader.pages for line in page.extract_text().split("\n")]
-    return page1.extract_text()
 
 
 class MollieInvoiceImporter(Importer):
@@ -23,15 +19,13 @@ class MollieInvoiceImporter(Importer):
         if mimetype != "application/pdf":
             return False
 
-        lines = pdf_to_text(filepath).split("\n")
+        return any(line.startswith("Mollie B.V.") for line in pdf_lines(filepath))
-        return any(line.startswith("Mollie B.V.") for line in lines)
 
     def account(self, filepath):
         return "Uitgaven:Bankkosten"
 
     def tx_ref(self, filepath):
-        lines = pdf_to_text(filepath).split("\n")
+        for line in pdf_lines(filepath):
-        for line in lines:
             if m := re.search(r"^Invoice reference (MOL-.+)$", line):
                 return m[1]
         raise Exception("Mollie invoice reference not found")
@@ -40,15 +34,13 @@ class MollieInvoiceImporter(Importer):
         return f"{self.tx_ref(filepath)}.pdf"
 
     def date(self, filepath):
-        lines = pdf_to_text(filepath).split("\n")
+        for line in pdf_lines(filepath):
-        for line in lines:
             if m := re.search(r"^Invoice date (\d{2})-(\d{2})-(\d{4})$", line):
                 return date(int(m[3]), int(m[2]), int(m[1]))
         raise Exception("Mollie invoice date not found")
 
     def extract(self, filepath, existing):
-        lines = pdf_to_text(filepath).split("\n")
+        for line in pdf_lines(filepath):
-        for line in lines:
             if m := re.search(r"^Total including VAT ([\d.]+) EUR$", line):
                 total = D(m[1])
                 break
@@ -92,15 +84,15 @@ class StatiegeldImporter(Importer):
         if mimetype != "application/pdf":
             return False
 
-        lines = pdf_to_text(filepath).split("\n")
+        return any(
-        return any(line.startswith("Statiegeld Nederland") for line in lines)
+            line.startswith("Statiegeld Nederland") for line in pdf_lines(filepath)
+        )
 
     def account(self, filepath):
         return "Inkomsten:Statiegeld"
 
     def tx_ref(self, filepath):
-        lines = pdf_to_text(filepath).split("\n")
+        for line in pdf_lines(filepath):
-        for line in lines:
             if m := re.search(r"^Factuurnr : (RP\d+)$", line):
                 return f"SNL-{m[1]}"
         raise Exception("Mollie invoice reference not found")
@@ -109,8 +101,7 @@ class StatiegeldImporter(Importer):
         return f"{self.tx_ref(filepath)}.pdf"
 
     def date(self, filepath):
-        lines = pdf_to_text(filepath).split("\n")
+        for line in pdf_lines(filepath):
-        for line in lines:
             if m := re.search(r"^Datum : (\d{2})/(\d{2})/(\d{4})$", line):
                 return date(int(m[3]), int(m[2]), int(m[1]))
         raise Exception("Date not found")

scripts/import-generic.py

@@ -22,12 +22,9 @@ set_cachier_params(
 )
 
 
-@cachier()
 def pdf_lines(filename):
     reader = PdfReader(filename)
-    return list(
+    return [line for page in reader.pages for line in page.extract_text().split("\n")]
-        line for page in reader.pages for line in page.extract_text().split("\n")
-    )
 
 
 @cachier()