From 8d90a81c62b9a3f9b525a459b743fee1a4f9544a Mon Sep 17 00:00:00 2001
From: Gabe Fierro <gtfierro@mines.edu>
Date: Wed, 29 Nov 2023 11:22:19 -0700
Subject: [PATCH] capture errors in the token stream

---
 buildingmotif/label_parsing.py   | 19 ++++++++++++++++++-
 tests/unit/test_label_parsing.py | 24 ++++++++++++++++++++----
 2 files changed, 38 insertions(+), 5 deletions(-)

diff --git a/buildingmotif/label_parsing.py b/buildingmotif/label_parsing.py
index 8c5b0ad1d..9a13c078e 100644
--- a/buildingmotif/label_parsing.py
+++ b/buildingmotif/label_parsing.py
@@ -59,6 +59,23 @@ class TokenResult:
     length: int
     error: Optional[str] = None
 
+    def __eq__(self, other):
+        """
+        Compare two token results on every
+        field except for the error field.
+        """
+        if not isinstance(other, TokenResult):
+            return False
+        return (
+            self.value == other.value
+            and self.token == other.token
+            and self.length == other.length
+        )
+
+
+# null token result
+ErrorTokenResult = TokenResult(None, Null(), 0, None)
+
 
 # type definition for the output of a parser function
 @dataclass(frozen=True)
@@ -189,7 +206,7 @@ def parser(target):
         results = []
         while True:
             part = seq_parser(target)
-            if not part:
+            if not part or part[0].value is None:
                 break
             results.extend(part)
             # add up the length of all the tokens
diff --git a/tests/unit/test_label_parsing.py b/tests/unit/test_label_parsing.py
index c58913de4..877b96e16 100644
--- a/tests/unit/test_label_parsing.py
+++ b/tests/unit/test_label_parsing.py
@@ -1,8 +1,11 @@
+from typing import List
+
 from buildingmotif.label_parsing import (
     COMMON_ABBREVIATIONS,
     COMMON_EQUIP_ABBREVIATIONS_BRICK,
     Constant,
     Delimiter,
+    ErrorTokenResult,
     Identifier,
     Null,
     TokenResult,
@@ -23,6 +26,13 @@
 from buildingmotif.namespaces import BRICK
 
 
+def _parse_result_empty(result: List[TokenResult]):
+    if len(result) == 0:
+        return True
+    if len(result) == 1 and result[0].value is None:
+        return True
+
+
 def test_ensure_token():
     assert ensure_token(Identifier, "abc") == Identifier(
         "abc"
@@ -43,7 +53,9 @@ def test_string_parser():
         TokenResult("abc", Identifier("abc"), 3)
     ], "Should parse the string"
     # test that it does not parse the string if it is not at the beginning
-    assert parser("0abc") == [], "Should not parse the string"
+    assert parser("0abc") == [
+        ErrorTokenResult
+    ], f"Should not parse the string {parser('0abc')}"
 
 
 def test_substring_n_parser():
@@ -53,7 +65,7 @@ def test_substring_n_parser():
         TokenResult("ab", Identifier("ab"), 2)
     ], "Should parse the string"
     # test that it pulls the correct number of characters when there are less than it needs
-    assert parser("a") == [], "Should not parse the string"
+    assert parser("a") == [ErrorTokenResult], "Should not parse the string"
     # test that it pulls the correct number of characters when there are exactly the number it needs
     assert parser("ab") == [
         TokenResult("ab", Identifier("ab"), 2)
@@ -159,7 +171,8 @@ def test_sequence():
         TokenResult("1", Identifier("1"), 1),
     ], "Should parse the string"
     assert parser("AHU1-") == [
-        TokenResult("AHU", Constant(BRICK.Air_Handling_Unit), 3)
+        TokenResult("AHU", Constant(BRICK.Air_Handling_Unit), 3),
+        ErrorTokenResult,
     ], "Should not parse all of the string"
 
 
@@ -176,6 +189,7 @@ def test_many():
     assert parser("AHU1") == [
         TokenResult("AHU", Constant(BRICK.Air_Handling_Unit), 3),
         TokenResult("1", Identifier("1"), 1),
+        ErrorTokenResult,
     ]
     # no delimiter at the end, so it should not parse the whole thing
     assert parser("AHU1/SP2") == [
@@ -184,6 +198,7 @@ def test_many():
         TokenResult("/", Delimiter("/"), 1),
         TokenResult("SP", Constant(BRICK.Setpoint), 2),
         TokenResult("2", Identifier("2"), 1),
+        ErrorTokenResult,
     ]
     # test that it parses multiple sequences
     assert parser("AHU1/SP2/") == [
@@ -273,7 +288,8 @@ def test_parse():
 
     result = parse(parser, "AHU1-")
     assert result.tokens == [
-        TokenResult("AHU", Constant(BRICK.Air_Handling_Unit), 3)
+        TokenResult("AHU", Constant(BRICK.Air_Handling_Unit), 3),
+        ErrorTokenResult,
     ], "Should not parse all of the string"
     assert not result.success, "Should not parse the string"