temp

api/views/purchaseimport.py

@@ -39,8 +39,11 @@ def __init__(self, **kwargs):
         self.is_duplicate_file = False
         self.duplicate_purchases = []
         self.duplicate_purchase_count = 0
-        self.data_schema = json.load(open("data/schemas/imports/achats.json"))
-        self.expected_header = [field["name"] for field in self.data_schema["fields"]]
+        self.schema_url = (
+            "https://raw.githubusercontent.com/betagouv/ma-cantine/refs/heads/staging/data/schemas/imports/achats.json"
+        )
+        self.schema_json = json.load(open("data/schemas/imports/achats.json"))
+        self.expected_header = [field["name"] for field in self.schema_json["fields"]]
         super().__init__(**kwargs)
 
     def post(self, request):
@@ -54,16 +57,22 @@ def post(self, request):
             self.file_digest = super()._get_file_digest()
             self._check_duplication()
 
+            self.dialect = super()._get_file_dialect()
+            print(self.dialect)
+            super()._verify_first_line_is_header()
+
             with transaction.atomic():
-                self._process_file()
+                res = super()._send_to_validata()
+                print(res)
+                # self._process_file()
 
-                # If at least an error has been detected, we raise an error to interrupt the
-                # transaction and rollback the insertion of any data
-                if self.errors:
-                    raise IntegrityError()
+                # # If at least an error has been detected, we raise an error to interrupt the
+                # # transaction and rollback the insertion of any data
+                # if self.errors:
+                #     raise IntegrityError()
 
-                # Update all purchases's import source with file digest
-                Purchase.objects.filter(import_source=self.tmp_id).update(import_source=self.file_digest)
+                # # Update all purchases's import source with file digest
+                # Purchase.objects.filter(import_source=self.tmp_id).update(import_source=self.file_digest)
 
             return self._get_success_response()
 

api/views/utils.py

@@ -1,8 +1,11 @@
+import csv
 import hashlib
+import io
 import json
 import logging
 
 import chardet
+import requests
 from django.conf import settings
 from django.core.exceptions import ValidationError
 from djangorestframework_camel_case.render import CamelCaseJSONRenderer
@@ -31,6 +34,33 @@ def _get_file_digest(self):
             file_hash.update(row)
         return file_hash.hexdigest()
 
+    def _get_file_dialect(self):
+        self.file.seek(0)
+        row_1 = self.file.readline()
+        (decoded_row, _) = decode_bytes(row_1)
+        self.dialect = csv.Sniffer().sniff(decoded_row)
+
+    def _verify_first_line_is_header(self):
+        self.file.seek(0)
+        row_1 = self.file.readline()
+        (decoded_row, _) = decode_bytes(row_1)
+        csvreader = csv.reader(io.StringIO("".join(decoded_row)), self.dialect)
+        for header in csvreader:
+            if header != self.expected_header:
+                raise ValidationError("La première ligne du fichier doit contenir les bon noms de colonnes")
+
+    def _send_to_validata(self):
+        # Reset the file pointer to the beginning
+        self.file.seek(0)
+        response = requests.post(
+            "https://api.validata.etalab.studio/validate",
+            files={
+                "file": ("file.csv", self.file.read(), self.file.content_type),
+            },
+            data={"schema": self.schema_url, "header_case": True},
+        )
+        return response.json()["report"]
+
 
 def camelize(data):
     camel_case_bytes = CamelCaseJSONRenderer().render(data)