add category_matching_thresh config

mrzaizai2k · Nov 13, 2024 · 240e360 · 240e360
1 parent 99a485e
commit 240e360
Show file tree

Hide file tree

Showing 3 changed files with 5 additions and 3 deletions.
diff --git a/config/config.yaml b/config/config.yaml
@@ -83,6 +83,7 @@ excel:
 
 egw:
   output_path: output
+  category_matching_thresh: 0.6
 
 
 

diff --git a/config/data1.json b/config/data1.json
@@ -42,7 +42,7 @@
                 "start_time": "06:00:00",
                 "end_time": "16:00:00",
                 "break_time": 0.5,
-                "description": "reisezeit ",
+                "description": "Reisezeit Auftrag",
                 "has_customer_signature": false
             },
             {

diff --git a/src/egw_export.py b/src/egw_export.py
@@ -86,7 +86,7 @@ def calculate_similarity(a: str, b: str) -> float:
     """
     return SequenceMatcher(None, a, b).ratio()
 
-def handle_kategorie(invoice_info: dict, line: dict, threshold: float = 0.7) -> str:
+def handle_kategorie(invoice_info: dict, line: dict, threshold: float = 0.6) -> str:
     """
     Determine 'Kategorie' based on 'is_without_measuring_technology' and 'description'.
     If 'description' closely matches "reisezeit auftrag" based on the similarity threshold, 
@@ -132,7 +132,8 @@ def export_egw_file(config: dict, invoice_lists: list) -> str:
         for line in invoice_info["lines"]:
             title = handle_title(invoice_info, line)
             dauer = handle_dauer(line)
-            kategorie = handle_kategorie(invoice_info, line)  # Now passing `line` for description check
+            kategorie = handle_kategorie(invoice_info, line, 
+                                         threshold=config['category_matching_thresh'])  # Now passing `line` for description check
 
             row = {
                 "Stundenzettel ID": None,
Original file line number	Diff line number	Diff line change
Expand Up		@@ -83,6 +83,7 @@ excel:

		egw:
		output_path: output
		category_matching_thresh: 0.6



Expand Down