remove unused syntax_identifier and syntax version overriding

fixes #99

Author: nerdoc

CHANGELOG.md

@@ -11,6 +11,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - BREAKING CHANGE: Remove SegmentCollection
 - BREAKING CHANGE: Remove FileSourcableMixin
 - BREAKING CHANGE: Segment() calls *must* provide the tag name now as the first parameter.
+- BREAKING CHANGE: Parser cannot override syntax_version and syntax_identifier anymore. It is determined from the UNB header.
 - change plugin system from `metaclass` to `__init_subclass__`
 - raise ValidationError, not AssertionError, when ServiceAdviceString is not 6 chars long
 - Drop support for Python versions < 3.10 (3.8 was EOL 2024-10-07, 3.9 not much used in Distros.)

pydifact/parser.py

@@ -34,6 +34,9 @@
 from pydifact.token import Token
 from pydifact.segments import Segment, SegmentFactory
 from pydifact.control import Characters
+import logging
+
+logger = logging.getLogger(__name__)
 
 
 class Parser:
@@ -52,16 +55,15 @@ def __init__(
         self,
         factory: SegmentFactory | None = None,
         characters: Characters | None = None,
-        version: str = "",
         directory: str = "",
-        syntax_identifier: str = "",
     ) -> None:
         """Initializes parser with segment factory and control characters"""
         self.factory = factory or SegmentFactory()
         self.characters = characters or Characters()
-        self.version = version
         self.directory = directory
-        self.syntax_identifier = syntax_identifier
+
+        self.syntax_identifier = ""
+        self.version = ""
 
     def parse(
         self,
@@ -121,7 +123,7 @@ def parse(
 
         for raw_segment in raw_segments:
             yield self.convert_raw_segment_to_segment(
-                raw_segment, version=self.version, directory=self.directory
+                raw_segment, directory=self.directory
             )
 
     @staticmethod
@@ -252,7 +254,6 @@ def convert_tokens_to_raw_segments(
     def convert_raw_segment_to_segment(
         self,
         raw_segment: Elements,
-        version: str = EDI_DEFAULT_VERSION,
         directory: str = EDI_DEFAULT_DIRECTORY,
     ) -> Segment:
         name = raw_segment.pop(0)
@@ -270,23 +271,13 @@ def convert_raw_segment_to_segment(
             # then we don't override it here, even if the UNB segment has another
             # value. The user might want to override this manually.
 
-            print(f"Found edifact syntax '{raw_segment[0][0]}' in UNB header", end="")
-            if self.syntax_identifier:
-                print(", but using override syntax '{self.syntax_identifier}'")
-            else:
-                print(".")
-                self.syntax_identifier = raw_segment[0][0]
-
-            print(
-                f"Found edifact version '{int(raw_segment[0][1])}' in UNB header",
-                end="",
+            self.syntax_identifier = raw_segment[0][0]
+            self.version = raw_segment[0][1]
+            logger.info(
+                f"Using edifact syntax identifier '{self.syntax_identifier}' with  "
+                f"syntax version {self.version} in UNB header.",
             )
-            if self.version:
-                print(f", but using override version '{self.version}'.")
 
-            else:
-                self.version = int(raw_segment[0][1])
-                print(".")
         return self.factory.create_segment(
             name,
             *raw_segment,

pydifact/segmentcollection.py

@@ -525,6 +525,7 @@ def from_segments(
         ):
             raise EDISyntaxError("Syntax identifier malformed.")
 
+        syntax_identifier = unb.elements[0][0]
         syntax_version = unb.elements[0][1]
 
         # Validate UNB segment according to the applicable syntax version
@@ -582,10 +583,10 @@ def from_segments(
                 )
         else:
             raise EDISyntaxError("Timestamp of file-creation malformed.")
-        syntax_identifier = (unb.elements[0][0], int(unb.elements[0][1]))
+
         timestamp = datetime.datetime.strptime(datetime_str, datetime_fmt)
         interchange = Interchange(
-            syntax_identifier=syntax_identifier,
+            syntax_identifier=(syntax_identifier, int(syntax_version)),
             sender=unb.elements[1],
             recipient=unb.elements[2],
             timestamp=timestamp,

tests/data/order2.edi

@@ -10,4 +10,3 @@ PRI+AAA:25.00'
 UNS+S'
 CNT+2:1'
 UNT+11+1'
-UNB+1'

tests/test_input_output.py

@@ -96,12 +96,12 @@ def test_order2_file():
 
     # UNH+1+ORDERS:D:01B:UN:EAN010'
     assert message.get_segment("UNH") == Segment(
-        "UNH", "1", ["ORDERS", "D", "01B", "UN", "EAN010"]
+        "UNH", "1", ["ORDERS", "D", "96A", "UN"]
     )
-    assert len(message.segments) == 24
-    assert message.recipient == ["4019315000007", "14"]
+    assert len(message.segments) == 11
+    assert message.recipient == "RECEIVERID"
     # Sender identification and Partner identification code qualifier
-    assert message.sender == ["4000000000000", "14"]
+    assert message.sender == "SENDERID"
 
 
 def test_patient1_file():