S14: 58-program benchmark suite — Lark grammar fixes + external COBOL validation
Grammar fixes: 1. COPY regex: handle quoted names COPY "STD-REC.CPY" 2. Quoted name strip: remove quotes before file lookup 3. VALUE clause: support comma-separated 88-level values 4. PIC STRING: support decimal dot (ZZ9.99 -> PICTURE_STRING.99 + DOT) 5. LEVEL: use INT for level number (fixes 05/01/77 all levels) Results on 58 telecom billing COBOL programs: - Parse OK: 54/58 (93%) - Parse fail: 4 (special chars: TAB, X'01', U'NNNN', &) - Classification known issue: matching programs misclassified as '文件编成' because FILE-CONTROL keyword overrides matching signals (requires rule engine priority fix - separate issue) Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
NB-076
@@ -7,14 +7,15 @@ FD_SUFFIX: /(?:"[^"]*"|'[^']*'|[^.])*\./
|
||||
working_storage: "WORKING-STORAGE" "SECTION" DOT data_item*
|
||||
linkage: "LINKAGE" "SECTION" DOT data_item*
|
||||
data_item: level_num (NAME | "FILLER") clause* DOT
|
||||
level_num: LEVEL
|
||||
level_num: INT
|
||||
clause: pic_clause | value_clause | occurs_clause | redefines_clause | usage_clause
|
||||
| "SYNC" | "SYNCHRONIZED"
|
||||
| "JUSTIFIED" "RIGHT"?
|
||||
| "BLANK" "WHEN" "ZERO"
|
||||
| "GLOBAL" | "EXTERNAL"
|
||||
pic_clause: "PIC" "IS"? PICTURE_STRING
|
||||
value_clause: "VALUE" "IS"? value_literal+
|
||||
value_clause: "VALUE" "IS"? value_list
|
||||
value_list: value_literal (","? value_literal)*
|
||||
value_literal: INT | SIGNED_NUMBER | STRING | SQSTRING
|
||||
| "ZERO" | "ZEROS" | "ZEROES"
|
||||
| "SPACE" | "SPACES"
|
||||
@@ -27,9 +28,9 @@ key_clause: ("ASCENDING" | "DESCENDING") "KEY" "IS"? NAME (","? NAME)*
|
||||
indexed_clause: "INDEXED" "BY" NAME (","? NAME)*
|
||||
usage_clause: USAGE_VAL
|
||||
USAGE_VAL: "COMP" | "COMP-3" | "COMP-5" | "BINARY" | "PACKED-DECIMAL" | "DISPLAY"
|
||||
LEVEL: /0[1-9]|[1-4][0-9]|49|77|88/
|
||||
LEVEL: /0[1-9]|[0-4][0-9]|49|77|88|[0-9]+/
|
||||
NAME: /[A-Z][A-Z0-9-]*/i
|
||||
PICTURE_STRING: /[0-9A-Z()+,\-*\/V]+/i
|
||||
PICTURE_STRING: /[0-9A-Z()+,\-*\/V]+(?:\.[0-9A-Z()+,\-*\/V]+)?/i
|
||||
INT: /[0-9]+/
|
||||
DOT: /\./
|
||||
%import common.SIGNED_NUMBER
|
||||
|
||||
@@ -95,7 +95,7 @@ _COPYBOOK_EXTENSIONS = ['.cpy', '.cbl', '.cpb', '']
|
||||
def resolve_copybooks(source: str, source_dir: str, _recursion_depth: int = 0) -> str:
|
||||
"""Find COPY statements and replace with copybook content."""
|
||||
_RE_COPY = re.compile(
|
||||
r"^\s*COPY\s+(\w[\w-]*)(?:\s+REPLACING\s+(.+?))?\s*\.?\s*$",
|
||||
r"^\s*COPY\s+(\w[\w-]*|\"[^\"]*\"|\'[^\']*\')(?:\s+REPLACING\s+(.+?))?\s*\.?\s*$",
|
||||
re.IGNORECASE
|
||||
)
|
||||
_RE_PAIR = re.compile(r"==(.+?)==\s+BY\s+==(.+?)==", re.IGNORECASE)
|
||||
@@ -105,7 +105,8 @@ def resolve_copybooks(source: str, source_dir: str, _recursion_depth: int = 0) -
|
||||
for line in lines:
|
||||
m = _RE_COPY.match(line)
|
||||
if m:
|
||||
name = m.group(1).upper()
|
||||
raw_name = m.group(1)
|
||||
name = raw_name.strip('"').strip("'").upper()
|
||||
found = None
|
||||
for ext in _COPYBOOK_EXTENSIONS:
|
||||
p = Path(source_dir, name + ext)
|
||||
|
||||
@@ -0,0 +1,117 @@
|
||||
"""S14: External benchmark suite — 58 telecom billing COBOL programs"""
|
||||
import sys, os, time, json
|
||||
sys.path.insert(0, os.path.join(os.path.dirname(__file__), '..'))
|
||||
P=0;F=0;BUGS=[]
|
||||
def ck(v,m=""): global P,F; (P:=P+1) if v else (F:=F+1,BUGS.append(m))
|
||||
def sec(n): print(f"\n--- {n} ---")
|
||||
|
||||
ROOT = "D:/cobol-java/cobol-test-programs/"
|
||||
from cobol_testgen import extract_structure, generate_data
|
||||
from hina.pipeline.pipeline import classify_program
|
||||
from hina.classifier import detect_keyword
|
||||
|
||||
progs = []
|
||||
for d in sorted(os.listdir(ROOT)):
|
||||
dp = os.path.join(ROOT, d)
|
||||
if os.path.isdir(dp):
|
||||
for f in sorted(os.listdir(dp)):
|
||||
if f.endswith(".cbl"):
|
||||
progs.append(os.path.join(dp, f))
|
||||
print(f"Total: {len(progs)} programs")
|
||||
|
||||
sec("PARSE: Extract structure for all 58 programs")
|
||||
parse_ok=0; parse_fail=0
|
||||
for fp in progs:
|
||||
name = os.path.relpath(fp, ROOT).replace("\\","/")
|
||||
src = open(fp, encoding="utf-8-sig").read()
|
||||
try:
|
||||
st = extract_structure(src)
|
||||
parse_ok += 1
|
||||
except Exception as e:
|
||||
parse_fail += 1
|
||||
ck(False, f"PARSE: {name} -> {str(e)[:40]}")
|
||||
ck(parse_fail == 0, f"Parse: {parse_fail}/{len(progs)} FAIL")
|
||||
|
||||
sec("CLASSIFY: Directory name vs classification match")
|
||||
# Expected types from directory names
|
||||
expected_map = {
|
||||
"matching": ["01","02","03","16","17","18","19","20","22","32","33"],
|
||||
"keybreak": ["07","08","30"],
|
||||
"divide": ["10","11","12"],
|
||||
"validation": ["13","27","31"],
|
||||
"csv": ["15","21"],
|
||||
"select": ["23"],
|
||||
"search": ["24","26"],
|
||||
"subprogram": ["25"],
|
||||
"sort": ["34"],
|
||||
"merge": ["35"],
|
||||
"evaluate": ["06"],
|
||||
"branch": ["05"],
|
||||
"edit": ["04"],
|
||||
"cics": ["14"],
|
||||
"sysin": ["28"],
|
||||
"ascii": ["29"],
|
||||
"pipeline": ["pipeline"],
|
||||
}
|
||||
for fp in progs:
|
||||
name = os.path.relpath(fp, ROOT).replace("\\","/")
|
||||
src = open(fp, encoding="utf-8-sig").read()
|
||||
try:
|
||||
cp = classify_program(src)
|
||||
cat = cp.get("category", "?")
|
||||
except:
|
||||
cat = "ERROR"
|
||||
# Check if directory name indicates matching type
|
||||
dir_id = name.split("-")[0] if "-" in name else name[:2]
|
||||
# Matching programs should say マッチング
|
||||
if dir_id in ["01","02","03","16","17","18","19","20","22"]:
|
||||
is_matching = "マッチング" in str(cat) or "matching" in str(cat).lower()
|
||||
if not is_matching:
|
||||
BUGS.append(f"MISCLASSIFY: {name} -> {cat}")
|
||||
ck(False, f"CLASSIFY: {name} expected matching, got {cat}")
|
||||
# Division programs should say DIVIDE
|
||||
if dir_id in ["10","11","12"]:
|
||||
if "DIVIDE" not in str(cat).upper() and "divide" not in str(cat).lower():
|
||||
BUGS.append(f"MISCLASSIFY: {name} (divide) -> {cat}")
|
||||
# Sort programs should say SORT
|
||||
if dir_id == "34":
|
||||
if "SORT" not in str(cat).upper() and "sort" not in str(cat).lower():
|
||||
BUGS.append(f"MISCLASSIFY: {name} (sort) -> {cat}")
|
||||
|
||||
ck(len([b for b in BUGS if "MISCLASSIFY" in b]) <= 10, f"Classification mismatch count")
|
||||
|
||||
sec("GENERATE: Non-zero data produce")
|
||||
zero_data = 0
|
||||
max_recs = 0; max_name = ""
|
||||
for fp in progs:
|
||||
name = os.path.relpath(fp, ROOT).replace("\\","/")
|
||||
src = open(fp, encoding="utf-8-sig").read()
|
||||
try:
|
||||
st = extract_structure(src)
|
||||
recs = generate_data(src, st)
|
||||
if len(recs) == 0:
|
||||
zero_data += 1
|
||||
if len(recs) > max_recs:
|
||||
max_recs = len(recs); max_name = name
|
||||
except:
|
||||
zero_data += 1
|
||||
ck(zero_data <= len(progs) * 0.5, f"Generate: {zero_data}/{len(progs)} zero records")
|
||||
ck(max_recs < 10000, f"Max records: {max_recs} ({max_name}) - path explosion risk")
|
||||
|
||||
sec("PERF: Average performance")
|
||||
times = []
|
||||
for fp in progs[:10]:
|
||||
src = open(fp, encoding="utf-8-sig").read()
|
||||
t0=time.time(); st=extract_structure(src); t1=time.time()
|
||||
times.append(t1-t0)
|
||||
avg = sum(times)/len(times)
|
||||
ck(avg < 5.0, f"Avg extract time: {avg:.3f}s (max 5s)")
|
||||
|
||||
sec("SUMMARY")
|
||||
print(f"\n{'='*55}")
|
||||
print(f"S14: {P} PASS / {F} FAIL")
|
||||
print(f"Bugs found: {len(BUGS)}")
|
||||
for b in BUGS:
|
||||
print(f" {b}")
|
||||
print(f"{'='*55}")
|
||||
if F > 0: sys.exit(1)
|
||||
Reference in New Issue
Block a user