fix validate credentials

idhub/admin/forms.py

@@ -1,7 +1,8 @@
 import csv
 import json
-import base64
 import copy
+import base64
+import jsonschema
 import pandas as pd
 
 from pyhanko.sign import signers
@@ -176,20 +177,10 @@ class ImportForm(forms.Form):
 
         self._schema = schema.first()
         try:
-            self.json_schema = json.loads(self._schema.data)
-            props = [x for x in self.json_schema["allOf"] if 'properties' in x.keys()]
-            prop = props[0]['properties']
-            self.properties = prop['credentialSubject']['properties']
+            self.json_schema = self._schema.get_credential_subject_schema()
         except Exception:
             raise ValidationError("Schema is not valid!")
 
-        if not self.properties:
-            raise ValidationError("Schema is not valid!")
-
-        # TODO we need filter "$ref" of schema for can validate a csv
-        self.json_schema_filtered = copy.copy(self.json_schema)
-        allOf = [x for x in self.json_schema["allOf"] if '$ref' not in x.keys()]
-        self.json_schema_filtered["allOf"] = allOf
         return data
 
     def clean_file_import(self):
@@ -197,17 +188,19 @@ class ImportForm(forms.Form):
         self.file_name = data.name
 
         df = pd.read_excel(data)
+        # convert dates to iso 8601
+        for col in df.select_dtypes(include='datetime').columns:
+            df[col] = df[col].dt.strftime("%Y-%m-%d")
+
         data_pd = df.fillna('').to_dict()
 
         if not data_pd:
             self.exception("This file is empty!")
 
-        head_row = {x: '' for x in self.properties.keys()}
         for n in range(df.last_valid_index()+1):
             row = {}
             for k in data_pd.keys():
-                if k in self.properties.keys():
-                    row[k] = data_pd[k][n] or ''
+                row[k] = data_pd[k][n] or ''
 
             user = self.validate_jsonld(n, row)
             self.rows[user] = row
@@ -229,13 +222,15 @@ class ImportForm(forms.Form):
 
     def validate_jsonld(self, line, row):
         try:
-            # check = credtools.validate_json(row, self.json_schema_filtered)
-            check = credtools.validate_json(row, self.json_schema)
-            if check is not True:
-                raise ValidationError("Not valid row")
-        except Exception as e:
-            msg = "line {}: {}".format(line+1, e)
+            jsonschema.validate(instance=row, schema=self.json_schema)
+        except jsonschema.exceptions.ValidationError as err:
+            msg = "line {}: {}".format(line+1, err)
             return self.exception(msg)
+        # try:
+        #     check = credtools.validate_json(row, self.json_schema)
+        #     if check is not True:
+        #         raise ValidationError("Not valid row")
+        # except Exception as e:
 
         user, new = User.objects.get_or_create(email=row.get('email'))
         if new:

idhub/models.py

@@ -541,6 +541,28 @@ class Schemas(models.Model):
     def description(self, value):
         self._description = value
 
+    def get_credential_subject_schema(self):
+        sc = self.get_data()
+        properties = sc["allOf"][1]["properties"]["credentialSubject"]["properties"]
+        required = sc["allOf"][1]["properties"]["credentialSubject"]["required"]
+
+        if "id" in required:
+            required.remove("id")
+
+        schema = {
+            "$schema": "https://json-schema.org/draft/2020-12/schema",
+            "type": "object",
+            "properties": properties,
+            "required": required,
+            "additionalProperties": False
+        }
+
+        return schema
+
+    def get_data(self):
+        return json.loads(self.data)
+
+
 class VerificableCredential(models.Model):
     """
         Definition of Verificable Credentials

utils/credtools.py

@@ -9,6 +9,7 @@ import requests
 from pyld import jsonld
 import jsonref
 from jsonpath_ng import jsonpath, parse
+from datetime import datetime
 
 
 # def remove_null_values(dictionary):
@@ -205,22 +206,75 @@ def schema_to_xls_comment(schema, xls_file_path):
 
   # Get the xlsxwriter workbook and worksheet objects
   workbook = writer.book
+  
+  matches_title = parse('$.title').find(schema)
+  title = matches_title[0].value if matches_title else 'no title'
+
+  matches_desc = parse('$.description').find(schema)
+  desc = matches_desc[0].value if matches_desc else 'no description'
+  
+  matches_id = parse("$['$id']").find(schema)
+  idschema = matches_id[0].value if matches_id else 'no schema'
+  
+  matches_subject_desc = parse('$..credentialSubject.description').find(schema)
+  subject_desc = matches_subject_desc[0].value if matches_subject_desc else 'no subject description'
+  
+  workbook.set_properties({
+    'title':    title,
+    'subject':  desc,
+    'author':   'IdHub Orchestral',
+    'category': subject_desc,
+    'keywords': 'schema, template, plantilla',
+    'created':  datetime.now().date(), #datetime.date(2018, 1, 1),
+    'comments': 'Created with Python for IdHub'})
+
+  workbook.set_custom_property('Schema', idschema)
+ 
   worksheet = writer.sheets['Full1']
 
   # Define a format for the required header cells
-  req_format = workbook.add_format({'border': 1})
-  # cell_format = workbook.add_format({'bold': True, 'font_color': 'red'})
+  req_f = workbook.add_format({'border': 1})
+  req_da = workbook.add_format({'border': 1, 'num_format': 'yyyy-mm-dd'})
+  req_in = workbook.add_format({'border': 1, 'num_format': '0'})
+  req_st = workbook.add_format({'border': 1, 'num_format': '@'})
+  opt_da = workbook.add_format({'num_format': 'yyyy-mm-dd'})
+  opt_in = workbook.add_format({'num_format': '0'})
+  opt_st = workbook.add_format({'num_format': '@'})
+  fmts = {
+    'string' : {True: req_st, False: opt_st},
+    'date' : {True: req_da, False: opt_da},
+    'integer' : {True: req_in, False: opt_in}
+  }
 
   # Write comments to the cells
   for i, header in enumerate(headers):
-    if header in req:
-        worksheet.set_column(i,i, None, req_format)
+    fmt = {}
+    #if header in req:
+    #    fmt = req_format
+    # worksheet.set_column(i,i, None, req_format)
+   
     # Get the description for the current field
     if 'description' in matches[0][header]:
       description = matches[0][header]['description']
       if description is not None:
         # Write the description as a comment to the corresponding cell
         worksheet.write_comment(0, i, description)
+   
+    # Get the type for the current field
+    if 'type' in matches[0][header]:
+      type_field = matches[0][header]['type']
+
+      format_field = None
+      if 'format' in matches[0][header]:
+        format_field = matches[0][header]['format']
+      
+      if type_field is not None:
+        if format_field is not None and format_field == 'date':
+            type_field = 'date'
+        fmt = fmts[type_field][header in req] # Add type format
+
+    print(f'header {header} with fmt {fmt}\n')
+    worksheet.set_column(i,i, None, fmt)    
         
   # Close the Pandas Excel writer and output the Excel file
   worksheet.autofit()