Fix unicode values in fields with STRICT validation

hl7apy/base_datatypes.py

@@ -35,6 +35,7 @@
 from __future__ import absolute_import
 import re
 import numbers
+import sys
 from datetime import datetime
 from decimal import Decimal
 from functools import cmp_to_key
@@ -44,6 +45,10 @@
     InvalidDateOffset, InvalidMicrosecondsPrecision
 from hl7apy.validation import Validator
 
+if sys.version_info.major == 2:
+    bytes, unicode = str, unicode
+else:
+    bytes, unicode = bytes, str
 
 class BaseDataType(object):
     """
@@ -71,8 +76,13 @@ def __init__(self, value, max_length=None, validation_level=None):
         self.validation_level = validation_level
         self.max_length = max_length
         if Validator.is_strict(self.validation_level):
-            if self.max_length is not None and len('{0}'.format(value)) > self.max_length:
-                raise MaxLengthReached(value, self.max_length)
+            if self.max_length is not None:
+                if isinstance(value, (bytes, unicode)):
+                    if len(value) > self.max_length:
+                        raise MaxLengthReached(value, self.max_length)
+                elif len("{0}".format(value)) > self.max_length:
+                    raise MaxLengthReached(value, self.max_length)
+
         self.value = value
 
     def to_er7(self, encoding_chars=None):

tests/test_to_string.py

@@ -266,6 +266,11 @@ def test_to_string_wd_field(self):
         parsed_s = parse_segment(s, version='2.7')
         self.assertEqual(parsed_s.to_er7(), 'EVN||20080115153000||AAA|AAA|20080114003000')
 
+    def test_to_string_unicode_field(self):
+        pid = Segment('PID', validation_level=VALIDATION_LEVEL.STRICT)
+        pid.pid_5.pid_5_1 = u'€'
+        self.assertEqual(pid.to_er7(), u'PID|||||€')
+
 
 if __name__ == '__main__':
     unittest.main()