stimside: Refactor the leakage tag parsing logic

glue/stimside/src/stimside/op_handlers/leakage_handlers/common_parsing.py

@@ -0,0 +1,67 @@
+import re
+from typing import Literal
+
+# --- REGEXES ---
+
+# Matches "LEAKAGE_NAME: (arg1) (arg2)"
+LEAKAGE_TAG_MATCH = re.compile(r"(?P<name>LEAKAGE_\w+):(?P<args>.+)")
+
+# Matches "CONDITIONED_ON_NAME: args : targets"
+CONDITION_TAG_MATCH = re.compile(
+    r"(?P<name>CONDITIONED_ON_[A-Z]+)\s*:(?P<args>[^:]+):*(?P<targets>.+)*"
+)
+
+# Matches "LEAKAGE_MEASUREMENT: (prob, state), ... : target, ..."
+MEASURE_TAG_MATCH = re.compile(r"(?P<name>LEAKAGE_MEASUREMENT)\s*:(?P<args>.+):(?P<targets>.+)")
+
+# Matches "state-->state" or "state<->state"
+# Allows multi-digit states and 'U'
+LEAKAGE_TRANSITIONS_1_MATCH = re.compile(
+    r"(?P<s0>[\dU]+)(?P<direction>[-<]->)(?P<s1>[\dU]+)"
+)
+
+# Matches "state_state-->state_state"
+# Allows states: digits, U, V, D, X, Y, Z
+LEAKAGE_TRANSITIONS_2_MATCH = re.compile(
+    r"(?P<s0>[\dU])_(?P<s1>[\dU])(?P<direction>[-<]->)(?P<s2>[\dUVDXYZ])_(?P<s3>[\dUVDXYZ])"
+)
+
+
+# --- UTILITIES ---
+
+def try_as_integer(state: str) -> int | str:
+    """Try to parse a state as an integer, otherwise return the stripped string."""
+    s = state.strip()
+    if s.isdigit():
+        return int(s)
+    return s
+
+
+def split_arguments(args: str) -> list[tuple[str, ...]]:
+    """
+    Robustly split parenthesized arguments, e.g., "(0.1, 2) (0.2, 3)" -> [("0.1", "2"), ("0.2", "3")].
+    Handles optional commas and varied whitespace between parentheses.
+    """
+    stripped_args = args.strip()
+    if not stripped_args:
+        raise ValueError("Empty arguments")
+
+    if not (stripped_args.startswith("(") and stripped_args.endswith(")")):
+        raise ValueError(f"Arguments must be enclosed in parentheses: '{args}'")
+
+    # re.split(r"\)\s*,*\s*\(", ...) handles ") (", "), (", ") ,(", etc.
+    args_list = re.split(r"\)\s*,*\s*\(", stripped_args[1:-1])
+    return [tuple(b.strip() for b in a.split(",")) for a in args_list]
+
+
+def parse_list_of_targets(targets: str) -> list[int]:
+    """Parse a space-separated list of qubit targets."""
+    targets_set = []
+    for target in targets.split():
+        if target.isdigit():
+            targets_set.append(int(target))
+        else:
+            raise ValueError(
+                f"Targets in tag must be integers separated by spaces, got: '{targets}'"
+            )
+    return targets_set