Changeset 0331028 in rattail

Timestamp:

07/25/2022 11:13:55 AM (6 months ago)

Author:

Lance Edgar <lance@…>

Branches:

master

Children:

9820a0f

Parents:

d094569

Message:

Update invoice parser for KeHE, per new format

File:

• rattail/contrib/vendors/invoices/kehe.py (modified) (8 diffs)

rattail/contrib/vendors/invoices/kehe.py

@@ -3 +3 @@
 #
 #  Rattail -- Retail Software Framework
-#  Copyright © 2010-2021 Lance Edgar
+#  Copyright © 2010-2022 Lance Edgar
 #
 #  This file is part of Rattail.
@@ -33 +33 @@
 import six
 
-from rattail.db import model
-from rattail.gpc import GPC
 from rattail.vendors.invoices import InvoiceParser
 
@@ -42 +40 @@
     Vendor invoice parser for KeHE Distributors.
 
-    This parser is actually capable of handling two different but similar
-    formats.  CSV data is expected in both cases.  Basically the KeHE portal
-    changed the format available for download, so the parser now needs to
-    support both the "old" as well as "new" formats.  Note that the newer
-    formats appeared circa 2020-02-11.
+    KeHE has changed their format a few times; hence this parser is
+    capable of handling a few similar but different formats.  All are
+    essentially text/csv.
 
-    See the :meth:`detect_version()` method for details of how the parser
-    decides which "version" (format) of file it's dealing with.
+    Format #1 is the original and is tab-separated.
+
+    Format #2 came circa 2020-02-11 and is comma-separated.
+
+    Format #3 is like #1 but changed a column header.
+
+    Format #4 is like #2 but changed column headers and data formats.
+
+    See the :meth:`detect_version()` method for details of how the
+    parser decides which "version" (format) of file it's dealing with.
     """
     key = 'rattail.contrib.kehe'
@@ -65 +69 @@
             line = f.readline()
 
-        # if line has no tab characters, we can safely assume new version 2
-        if '\t' not in line:
-            return 2
+        delimiter = str('\t') if '\t' in line else ','
 
-        # the file *does* have tab characters, so we must check a column header
         if six.PY3:
             csv_file = open(path, 'rt')
-            reader = csv.DictReader(csv_file, delimiter='\t')
+            reader = csv.DictReader(csv_file, delimiter=delimiter)
         else: # PY2
             csv_file = open(path, 'rb')
-            reader = csv.DictReader(csv_file, delimiter=b'\t')
+            reader = csv.DictReader(csv_file, delimiter=delimiter)
         data = next(reader)
         csv_file.close()
 
-        # check for old version 1 column header
-        if 'Invoice Date' in data:
+        if delimiter == ',': # comma-separated
+            if 'OrderQuatity' in data:
+                return 4
+            return 2
+
+        else: # tab-separated
+            if 'InvoiceDate' in data:
+                return 3
             return 1
-
-        # okay, assume new version 3 then
-        # assert 'InvoiceDate' in data
-        return 3
 
     def parse_invoice_date(self, path):
@@ -104 +107 @@
         if version == 1:
             return datetime.datetime.strptime(data['Invoice Date'], '%Y-%m-%d').date()
-        else: # version in (2, 3)
+        elif version in (2, 3):
             return datetime.datetime.strptime(data['InvoiceDate'], '%m/%d/%Y %I:%M:%S %p').date()
+        elif version == 4:
+            return datetime.datetime.strptime(data['InvoiceDate'], '%m/%d/%Y').date()
 
     def parse_rows(self, path):
@@ -126 +131 @@
         }
 
-        if version in (2, 3):
+        if version > 1:
             fields.update({
                 'upc': 'Upc',
@@ -135 +140 @@
                 'net_billable': 'NetBillable',
                 'pack_size': 'PackSize',
+            })
+
+        if version == 4:
+            fields.update({
+                'order_quantity': 'OrderQuatity',
             })
 
@@ -146 +156 @@
         for data in reader:
 
-            row = model.VendorInvoiceBatchRow()
+            row = self.make_row()
             row.item_entry = data[fields['upc']]
-            row.upc = GPC(row.item_entry)
+            row.upc = self.app.make_gpc(row.item_entry)
             row.vendor_code = data[fields['ship_item']]
             row.brand_name = data[fields['brand']]