From d9abf9f19e9e93a0e27b11c284872a57014a81dd Mon Sep 17 00:00:00 2001
From: Fred Boniface <fred@fjla.uk>
Date: Sun, 1 Dec 2024 12:32:02 +0000
Subject: [PATCH] Properly handle blank PIS codes and prevent matching next
 service incorrectly

---
 src/parse_pdf.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/src/parse_pdf.py b/src/parse_pdf.py
index b52d335..9c53624 100644
--- a/src/parse_pdf.py
+++ b/src/parse_pdf.py
@@ -5,7 +5,8 @@ import re
 schedule_pattern = r"(?<=NOTES)(.*?)(?=NOTES|$)" # Split Schedule Cards
 train_start_pattern = r"\b\d[A-Z]\d{2}\s+\d{5}\s+\d{2}\.\d{2}(?:\s+\d{2}\.\d{2})?\s+[A-Za-z &]+" # Defines start of train section
 train_first_line_pattern = r"(?P<headcode>\d[A-Z]\d{2})\s+(?P<gsmr_code>\d{5})\s+(?P<time0>\d{2}\.\d{2})(?:\s+(?P<time1>\d{2}\.\d{2}))?\s+(?P<location>[A-Za-z0-9& ]+)" # Extracts Train Data
-train_pis_line_pattern = r"\(PIS Code\s*:\s*(?P<pis_code>\d+)\s*\)" # Extracts PIS Code
+train_pis_line_pattern = r"\(PIS Code\s*:\s*(?P<pis_code>\d+|____)\s*\)" # Extracts PIS Code
+
 
 # Extract Schedule Cards to list of strings