Merge pull request #94 from akretion/feat/geodis_no_accents

Geodis Don't send utf8 in EDI anymore
akretion · Mar 7, 2018 · b840d2f · b840d2f
2 parents 331ca1f + a758d0a
commit b840d2f
Show file tree

Hide file tree

Showing 3 changed files with 33 additions and 3 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -9,6 +9,8 @@ Roadmap / TODO:
 	- Support additionnal methods of api
 	- Write tests
 
+# 0.3.3 2018-02-19
+	- Geodis force ASCII conversion for EDI.
 
 # 0.3.2 2018-02-06
 	- Geodis fix missing CTA segment

diff --git a/VERSION b/VERSION
@@ -1 +1 @@
-0.3.2
+0.3.3
diff --git a/roulier/carriers/geodis/geodis_transport_edi.py b/roulier/carriers/geodis/geodis_transport_edi.py
@@ -53,7 +53,7 @@ def parse_lines(lines):
             return "\n".join([parse_segment(segment) for segment in lines])
 
         def sanitize(token):
-            # replace escapable chars by space
+            # remove accents and replace escapable chars by space
             # because, replacing "'" by "?'"
             # increase length of the token
             # which is limited by 35
@@ -63,7 +63,35 @@ def sanitize(token):
                 .replace("'", " ")
                 .replace("+", " ")
                 .replace(":", " ")
-            )
+                # quick and dirty replacement
+                # of common accentued chars in french
+                # because geodis don't handle well utf8
+                # TODO: put it in api coerce
+                .replace(u"é", "e")
+                .replace(u"è", "e")
+                .replace(u"ë", "e")
+                .replace(u"ê", "e")
+                .replace(u"ô", "o")
+                .replace(u"ï", "i")
+                .replace(u"ö", "o")
+                .replace(u"à", "a")
+                .replace(u"â", "a")
+                .replace(u"ç", "c")
+                .replace(u"û", "u")
+                .replace(u"ù", "u")
+                .replace(u"É", "E")
+                .replace(u"È", "E")
+                .replace(u"Ë", "E")
+                .replace(u"Ê", "E")
+                .replace(u"Ô", "O")
+                .replace(u"Ï", "I")
+                .replace(u"Ö", "O")
+                .replace(u"À", "A")
+                .replace(u"Â", "A")
+                .replace(u"Ç", "C")
+                .replace(u"Û", "U")
+                .replace(u"Ù", "U")
+            ).encode('ascii', 'ignore') # cut remaining chars
             return sanitized
 
         return parse_lines(arr)