From 45ceb2273afaf33c6594fa3748dec195cce420c1 Mon Sep 17 00:00:00 2001
From: abhay-codes07 <abhaysingh0293@gmail.com>
Date: Thu, 24 Sep 2026 23:21:32 +0530
Subject: [PATCH] fix(cobol): link both endpoints of a PERFORM ... THRU range

`PERFORM A THRU Z` (and `THROUGH`) runs the whole paragraph range A..Z, but the
PERFORM pattern captured only the first name, so a `calls` edge was emitted to A
and the range-end Z got none. "What performs Z?" then answered nothing even
though this paragraph does, and any reachability walk stopped short of the
range's tail.

Capture the optional `THRU`/`THROUGH` target and enqueue both endpoints for
resolution against the paragraph table. Intermediate paragraphs in the range are
not individually linked (that would need paragraph source ordering); the two
endpoints are the reliable, common signal and both now resolve. A plain
`PERFORM A` is unchanged.

Regression test: `PERFORM A-PARA THRU Z-PARA` now links MAIN to both A-PARA and
Z-PARA (fails on the old single-capture pattern).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01JJbLfztSxm2tH5cJwBbe9q
(cherry picked from commit abcdde1a6578e2d280df0f659d6f8269cd7ea124)
---
 graphify/extractors/cobol.py  | 20 ++++++++++++++++----
 tests/test_cobol_extractor.py | 25 +++++++++++++++++++++++++
 2 files changed, 41 insertions(+), 4 deletions(-)

diff --git a/graphify/extractors/cobol.py b/graphify/extractors/cobol.py
index 91e1929fa..73a439cd9 100644
--- a/graphify/extractors/cobol.py
+++ b/graphify/extractors/cobol.py
@@ -13,7 +13,10 @@ _PROGRAM = re.compile(rf"\bPROGRAM-ID\s*\.\s*({_NAME})", re.IGNORECASE)
 _PARAGRAPH = re.compile(
     rf"^({_NAME})(?:\s+(SECTION))?\s*\.\s*$", re.IGNORECASE
 )
-_PERFORM = re.compile(rf"\bPERFORM\s+({_NAME})\b", re.IGNORECASE)
+_PERFORM = re.compile(
+    rf"\bPERFORM\s+({_NAME})(?:\s+(?:THRU|THROUGH)\s+({_NAME}))?\b",
+    re.IGNORECASE,
+)
 _COPY = re.compile(
     rf"\bCOPY\s+(?:'([^']+)'|\"([^\"]+)\"|({_NAME}(?:\.[A-Z0-9-]+)?))",
     re.IGNORECASE,
@@ -256,9 +259,18 @@ def extract_cobol(path: Path) -> dict:
 
         masked = _mask_strings(upper)
         for match in _PERFORM.finditer(masked):
-            name = match.group(1).upper()
-            if name not in _PERFORM_MODIFIERS:
-                pending_performs.append((current_scope, current_program, name, line))
+            # `PERFORM A THRU Z` runs the range A..Z, so both endpoints are
+            # performed — link both. Only the entry paragraph was captured
+            # before, leaving the range-end with no inbound edge ("what performs
+            # Z?" answered nothing even though this paragraph does). Intermediate
+            # paragraphs in the range are not individually linked (that needs
+            # source ordering); the two endpoints are the reliable, common signal.
+            for group in (match.group(1), match.group(2)):
+                if group is None:
+                    continue
+                name = group.upper()
+                if name not in _PERFORM_MODIFIERS:
+                    pending_performs.append((current_scope, current_program, name, line))
 
         for match in _CALL.finditer(upper):
             if _starts_inside_string(upper, match.start()):
diff --git a/tests/test_cobol_extractor.py b/tests/test_cobol_extractor.py
index f08c5e7a8..2a5156c23 100644
--- a/tests/test_cobol_extractor.py
+++ b/tests/test_cobol_extractor.py
@@ -156,3 +156,28 @@ def test_cobol_fixed_format_with_sequence_numbers(tmp_path):
     labels = {node["label"] for node in result["nodes"]}
     assert {"PAYROLL", "MAIN-PARA", "INIT-PARA"} <= labels
     assert ("MAIN-PARA", "INIT-PARA") in _edge_labels(result, "calls")
+
+
+def test_cobol_perform_thru_links_both_range_endpoints(tmp_path):
+    """`PERFORM A THRU Z` runs the range A..Z, so both endpoints are performed.
+    Only the entry paragraph was linked before, leaving the range-end with no
+    inbound `calls` edge."""
+    source = tmp_path / "range.cbl"
+    source.write_text(
+        "       IDENTIFICATION DIVISION.\n"
+        "       PROGRAM-ID. RANGE.\n"
+        "       PROCEDURE DIVISION.\n"
+        "       MAIN-PARA.\n"
+        "           PERFORM A-PARA THRU Z-PARA.\n"
+        "       A-PARA.\n"
+        "           DISPLAY 'A'.\n"
+        "       Z-PARA.\n"
+        "           DISPLAY 'Z'.\n",
+        encoding="utf-8",
+    )
+
+    result = extract([source], cache_root=tmp_path)
+
+    calls = _edge_labels(result, "calls")
+    assert ("MAIN-PARA", "A-PARA") in calls
+    assert ("MAIN-PARA", "Z-PARA") in calls