Files

328 lines
12 KiB
Python

"""Flat file I/O — write fixed-length records from COBOL FD definitions"""
import re, struct
from pathlib import Path
from typing import Any
def analyze_fd_layout(source_text: str, copybook_dirs: list[str] = None) -> dict[str, dict]:
"""From COBOL source, extract FD file layouts."""
from .read import preprocess, parse_file_control, parse_file_section, parse_data_division, extract_data_division, scan_open_statements
pp = preprocess(source_text, extra_search_paths=copybook_dirs)
fc = parse_file_control(pp) if pp else {}
fs = parse_file_section(pp) if pp else {}
ops = scan_open_statements(pp) if pp else {}
dd = extract_data_division(pp)
all_fields = parse_data_division(dd) if dd else []
layouts = {}
for fd_name, rec_names in fs.items():
records = []
for rec_name in rec_names:
children = []
found = False
rec_level = None
offset = 0
for f in all_fields:
if f.name == rec_name:
found = True
rec_level = f.level
rec_field_obj = f
continue
if found:
if f.level is not None and f.level <= rec_level:
break
if f.is_88 or f.is_filler:
continue
pi = f.pic_info
if pi:
length = (pi.digits + pi.decimal) if pi.type == "numeric" else (pi.length or 0)
else:
length = 0
ftype = pi.type if pi else "unknown"
usage = f.usage if f.usage else None
children.append({
"name": f.name, "pic": str(f.pic or ""),
"type": ftype, "length": length, "offset": offset,
"usage": usage,
"pic_info": {
"type": f.pic_info.type if f.pic_info else "unknown",
"digits": f.pic_info.digits if f.pic_info else 0,
"decimal": f.pic_info.decimal if f.pic_info else 0,
"length": f.pic_info.length if f.pic_info else 0,
"signed": f.pic_info.signed if f.pic_info else False,
} if f.pic_info else None,
})
offset += length
# If record has no elementary children but the 01-level has PIC info
# (e.g. 01 SYSINREC PIC X(080)), use it as a single opaque field
if not children and rec_field_obj and rec_field_obj.pic_info:
pi = rec_field_obj.pic_info
length = pi.length or 0
if length > 0:
children.append({
"name": rec_field_obj.name,
"pic": str(rec_field_obj.pic or ""),
"type": "alphanumeric",
"length": length,
"offset": 0,
"usage": None,
"pic_info": {
"type": "alphanumeric",
"digits": 0,
"decimal": 0,
"length": length,
"signed": False,
},
})
offset = length
records.append({"record_name": rec_name, "fields": children, "record_length": offset})
assign_to = fc.get(fd_name, {}).get("assign", fd_name)
layouts[assign_to] = {
"fd_name": fd_name, "records": records,
"direction": ops.get(fd_name, "INPUT"),
}
return layouts
def select_records_for_file(records: list[dict], layout: dict) -> list[dict]:
"""Extract and route only the fields belonging to this file layout."""
if not layout or not layout.get("records"):
return records
field_names = set()
for rec in layout["records"]:
for f in rec["fields"]:
field_names.add(f["name"])
result = []
for rec in records:
row = {k: v for k, v in rec.items() if k in field_names}
if row:
result.append(row)
return result if result else records
def _format_value(value: Any, field: dict) -> bytes:
"""Format a value for COBOL fixed-length storage."""
from . import file_io
usage = field.get("usage")
if usage in ("COMP", "COMP-3", "BINARY", "PACKED-DECIMAL"):
pic_info = field.get("pic_info") or {}
packed = file_io.pack_value(str(value) if value is not None else "", {
"usage": usage,
"pic_info": pic_info,
})
want_len = file_io.get_storage_length({
"usage": usage,
"pic_info": pic_info,
})
if len(packed) < want_len:
packed = packed.rjust(want_len, b'\x00')
return packed[:want_len]
ftype = field["type"]
length = field["length"]
val = str(value) if value is not None else ""
if ftype == "numeric":
sval = str(val) if val is not None else ""
# Preserve spaces for DP#7 (R01EMP-ID = SPACE) and similar COBOL checks
if sval and all(c == ' ' for c in sval):
return (' ' * length).encode("ascii")
try:
num = int(float(val)) if val else 0
except (ValueError, TypeError):
num = 0
num = abs(num)
max_val = 10 ** length - 1
if num > max_val:
num = max_val
s = str(num).zfill(length)
if len(s) > length:
s = s[-length:]
return s.encode("ascii")
else:
s = val.ljust(length)[:length]
return s.encode("ascii", errors="replace")
def write_flat_file(records: list[dict], layout: dict, outpath: Path, field_filter: set = None):
"""Write records as a COBOL-compatible fixed-length flat file.
Supports multi-record FDs: uses the longest record layout (most fields)
to maximize compatible field coverage.
"""
outpath = Path(outpath)
if not layout or not layout.get("records"):
return
# Pick the record with the most fields (best coverage for multi-record FDs)
rec = max(layout["records"], key=lambda r: (len(r["fields"]), r["record_length"]))
if rec["record_length"] == 0:
return
rec_fields = rec["fields"]
if field_filter:
rec_fields = [f for f in rec_fields if f["name"] in field_filter]
with open(outpath, "wb") as f:
for row in records:
buf = bytearray()
for field in rec_fields:
val = row.get(field["name"], "")
formatted = _format_value(val, field)
buf.extend(formatted)
f.write(buf)
def write_all_files(records: list[dict], source_text: str, outdir: Path, prefix: str = "", copybook_dirs: list[str] = None):
"""Analyze source, write flat files for all INPUT FDs."""
outdir = Path(outdir)
layouts = analyze_fd_layout(source_text, copybook_dirs=copybook_dirs)
written = []
for filename, layout in layouts.items():
if layout["direction"] == "OUTPUT":
continue
# Skip SYSIN files — handled separately by write_sysin_file
if layout["fd_name"] == "SYSINFILE":
continue
fnames = set()
for rec in layout["records"]:
for f in rec["fields"]:
fnames.add(f["name"])
if not fnames:
continue
# Filter generated records to only include fields from this FD
filtered = [{k: v for k, v in r.items() if k in fnames} for r in records]
has_data = any(v for row in filtered for v in row.values())
if not has_data:
# Fallback: one zero-filled record from FD layout
fallback = {}
for rec in layout["records"]:
for f in rec["fields"]:
fallback[f["name"]] = 0 if f["type"] == "numeric" else " "
filtered = [fallback] if fallback else []
if filtered:
outpath = outdir / (prefix + filename)
write_flat_file(filtered, layout, outpath)
written.append((filename, outpath, len(filtered)))
return written
def write_sysin_file(records: list[dict], source_text: str, outdir: Path, prefix: str = "", copybook_dirs: list[str] = None, run_cfg: dict | None = None):
"""Generate SYSIN configuration card file from FD layout + generated records.
Args:
run_cfg: Scenario sysin override dict, e.g. {"period": "202607", "include_invalid_period": True, "modes": ["NORMAL", "RESET"]}.
If None or empty, uses defaults (existing behavior).
"""
outdir = Path(outdir)
layouts = analyze_fd_layout(source_text, copybook_dirs=copybook_dirs)
sysin_filename = None
sysin_layout = None
for filename, layout in layouts.items():
if layout["fd_name"] == "SYSINFILE":
sysin_filename = filename
sysin_layout = layout
break
if not sysin_layout:
return None
# Determine record length from layout
rec_length = 0
for rec in sysin_layout["records"]:
if rec["record_length"] > rec_length:
rec_length = rec["record_length"]
if rec_length == 0:
rec_length = 80
# Extract employee IDs from the generated R01 flat file if it exists,
# falling back to JSON record fields for backward compatibility.
r01_path = outdir / "KIN08R01"
emp_ids_ordered = []
emp_set = set()
if r01_path.exists():
rec_size = 200
data = r01_path.read_bytes()
num_recs = len(data) // rec_size
for i in range(num_recs):
off = i * rec_size
eid = data[off:off+8].decode("ascii", errors="replace").strip()
if eid and eid != "00000000":
emp_ids_ordered.append(eid)
else:
for r in records:
eid = r.get("R01EMP-ID", "") or r.get("R01INNREC", {}).get("R01EMP-ID", "")
if not eid or eid == "00000000":
line = r.get("R01LINE", "")
if line:
eid = line.split(",")[0].strip()
if eid and eid != "00000000":
emp_ids_ordered.append(eid)
if eid not in emp_set:
emp_set.add(eid)
else:
emp_ids_ordered.append(eid)
# Build SYSIN card records
# Card format: position 1 = type, position 2 = space (ignored), position 3+ = data
if run_cfg is None:
run_cfg = {}
lines = ["* GENERATED TEST DATA"]
sysin = run_cfg
period = sysin.get("period", "202607")
if period is not None:
lines.append(f"P YEAR-MONTH={period}")
if sysin.get("include_invalid_period", True):
lines.append("P YEAR-MONTH=000000")
for mode in (sysin.get("modes") or ["NORMAL"]):
lines.append(f"M MODE={mode}")
# Detect duplicate EMP_IDs → need RESET mode
seen = set()
has_dups = False
for eid in emp_ids_ordered:
if eid in seen:
has_dups = True
break
seen.add(eid)
if has_dups:
lines.append("M MODE=RESET")
chunks = [emp_ids_ordered[i:i+8] for i in range(0, len(emp_ids_ordered), 8)]
for chunk in chunks:
lines.append(f"T {','.join(chunk)}")
lines.append("M MODE=NORMAL")
# Always add a RESET-mode batch with duplicate EMP-ID for UPDATE path
unique_ids = list(dict.fromkeys(emp_ids_ordered))
if not unique_ids:
unique_ids = ["EMP00001", "EMP00002", "EMP00003", "EMP00004", "EMP00005"]
# Inject duplicates: use first EMP-ID twice to trigger RESET/UPDATE
if len(unique_ids) >= 1:
dup_id = unique_ids[0]
reset_ids = [dup_id, dup_id]
lines.append("M MODE=RESET")
lines.append(f"T {','.join(reset_ids)}")
lines.append("M MODE=NORMAL")
# Unique T card for NORMAL mode
if unique_ids:
chunks = [unique_ids[i:i+8] for i in range(0, len(unique_ids), 8)]
for chunk in chunks:
lines.append(f"T {','.join(chunk)}")
# Unknown card type to cover IF WRK-CARD-TYPE '*' ELSE branch (DP#8 F)
lines.append("X UNKNOWN")
# End with RESET mode so 3000STPSOR runs the RESET path (DP#22-#23)
lines.append("M MODE=RESET")
# Write as fixed-length flat file
outpath = outdir / (prefix + sysin_filename)
with open(outpath, "wb") as f:
for line in lines:
buf = line.encode("ascii", errors="replace")
if len(buf) < rec_length:
buf = buf.ljust(rec_length, b" ")
f.write(buf[:rec_length])
return outpath