feat(tooling): rewrite merge_spec.py — batch spec merging + new package support

Rewrite merge_spec.py to merge one or many feature spec packages into a
single review file.

Batch modes:
- single number: python merge_spec.py 038
- inclusive range: python merge_spec.py 036-041
- explicit list: python merge_spec.py 036 038 044
- by dir name: python merge_spec.py 042-dashboard-scenario-registry
- all: python merge_spec.py all
- custom output: python merge_spec.py 036-041 -o out.md

Handles the new spec package structure that plain *.md merging missed:
- includes contracts/openapi.yaml (YAML), contracts/ux/* (decisions.md),
  prototype/index.html + prototype/manifest.md
- skips .json/.py/.zip/.pyc and __pycache__ (fixtures/code/binaries)
- canonical per-feature order: spec -> ux_reference -> checklists ->
  UX contracts -> plan -> research -> data-model -> modules -> openapi ->
  quickstart -> traceability -> tasks -> prototype
- missing numbers warn+skip; dedup; per-feature grouping in one output

Verified: 043 (14 files), 036-041 (6 features/104 files), 036-047 (12/186),
all (50/572) with no .json/.zip/.pyc leakage.
This commit is contained in:
2026-08-09 11:06:36 +07:00
parent 4858992e15
commit f7e539440e

View File

@@ -1,9 +1,19 @@
# #region Tooling.MergeSpecModule [C:2] [TYPE Module]
# #region Tooling.MergeSpecModule [C:3] [TYPE Module]
# @LAYER Infra
# @BRIEF Merge one or many feature spec packages into a single review file.
# @RELATION DEPENDS_ON -> [Doc.Specify.Templates]
# @RATIONALE A reviewer LLM consumes the whole package as one artifact; batch mode
# (ranges, multiple numbers, or "all") produces a single merged file per feature group.
# @REJECTED Merging only *.md — rejected because new packages carry contracts/openapi.yaml,
# contracts/ux/*, and prototype/index.html that must be reviewed; a reviewer that never
# sees the OpenAPI or prototype cannot judge contract closure.
# #region Tooling.Merge.Spec [C:3] [TYPE Module]
# @LAYER Infra
# #region Tooling.Merge.Spec [C:2] [TYPE Module]
# @LAYER Infra
from __future__ import annotations
import re
import sys
from datetime import datetime
from pathlib import Path
@@ -15,42 +25,58 @@ REVIEW_PROMPT = (
"Focus on spec review, not rewriting the implementation."
)
# Canonical artifact order for the merged output.
# Each stage: (match_type, value, section_label)
# match_type "exact" = exact filename match
# match_type "prefix" = directory prefix — matches all files under that dir
CANONICAL_MD_STAGES = (
# Extensions that are safe to inline into a merged markdown review file.
MERGEABLE_EXTENSIONS = {".md", ".yaml", ".yml", ".html"}
# Extensions that are skipped (binary or non-review data).
SKIP_SUFFIXES = (
".json", ".py", ".pyc", ".zip", ".png", ".jpg", ".jpeg", ".webp", ".pdf",
)
# Canonical artifact order for each feature's merged output.
# Each stage: (match_type, path_value, section_label)
# exact = exact relative-path match (single file)
# prefix = all files under that directory prefix
# glob = fnmatch-style pattern against relative path
CANONICAL_STAGES = (
# Layer 1: Requirements & UX narrative
("exact", "spec.md", "SPEC — Feature Specification"),
("exact", "ux_reference.md", "UX REFERENCE — Interaction Narrative"),
("prefix", "checklists/", "CHECKLISTS — Requirements Quality"),
("exact", "spec.md", "SPEC — Feature Specification"),
("exact", "ux_reference.md", "UX REFERENCE — Interaction Narrative"),
("prefix", "checklists/", "CHECKLISTS — Requirements Quality"),
# Layer 2: UX Design (from /speckit.ux)
("exact", "contracts/ux/alternatives.md", "UX ALTERNATIVES — Design Space Explored"),
("exact", "contracts/ux/decisions.md", "UX DECISIONS — Final Choices"),
("exact", "contracts/ux/screen-models.md", "UX SCREEN MODELS — Model Inventory"),
("exact", "contracts/ux/api-ux.md", "UX API CONTRACT — Endpoints & Shapes"),
("prefix", "contracts/ux/", "UX DESIGN — Per-Screen Contracts"),
("exact", "contracts/ux/design-tokens.md", None), # force-first within prefix
# Layer 2: UX Design contracts
("exact", "contracts/ux/alternatives.md", "UX ALTERNATIVES — Design Space Explored"),
("exact", "contracts/ux/decisions.md", "UX DECISIONS — Final Choices"),
("exact", "contracts/ux/screen-models.md", "UX SCREEN MODELS — Model Inventory"),
("exact", "contracts/ux/api-ux.md", "UX API CONTRACT — Endpoints & Shapes"),
("exact", "contracts/ux/design-tokens.md", None), # force-first within prefix below
("prefix", "contracts/ux/", "UX DESIGN — Per-Screen Contracts"),
# Layer 3: Implementation Plan
("exact", "plan.md", "PLAN — Implementation Plan"),
("exact", "research.md", "RESEARCH — Technical Decisions"),
("exact", "data-model.md", "DATA MODEL — Entities & Relations"),
("prefix", "contracts/", "CONTRACTS — Module & Function Contracts"),
("exact", "quickstart.md", "QUICKSTART — Dev Onboarding"),
# Layer 3: Implementation plan + API
("exact", "plan.md", "PLAN — Implementation Plan"),
("exact", "research.md", "RESEARCH — Technical Decisions"),
("exact", "data-model.md", "DATA MODEL — Entities & Relations"),
("exact", "contracts/modules.md", "CONTRACTS — Module & Function Contracts"),
("exact", "contracts/openapi.yaml", "OPENAPI — REST/Event API Contract"),
("prefix", "contracts/", "CONTRACTS — Remaining"),
("exact", "quickstart.md", "QUICKSTART — Dev Onboarding"),
# Layer 4: Traceability
("exact", "traceability.md", "TRACEABILITY — Requirements Matrix"),
("exact", "traceability.md", "TRACEABILITY — Requirements Matrix"),
# Layer 5: Execution
("exact", "tasks.md", "TASKS — Implementation Tasks"),
("exact", "tasks.md", "TASKS — Implementation Tasks"),
# Layer 6: Prototype
("exact", "prototype/manifest.md", "PROTOTYPE — State/Manifest"),
("exact", "prototype/index.html", "PROTOTYPE — Interactive HTML"),
("glob", "prototype/*", "PROTOTYPE — Other"),
)
ORDER_HINT = (
"Artifact order: spec → ux_reference → checklists → "
"UX (alternatives → decisions → screen-models → api-ux → per-screen → design-tokens) → "
"plan → research → data-model → contracts → quickstart → traceability → tasks → remaining."
"plan → research → data-model → modules → openapi → contracts → quickstart → "
"traceability → tasks → prototype. Non-mergeable files (.json/.py/binary) are skipped."
)
@@ -58,137 +84,240 @@ def relative_key(path: Path, root: Path) -> str:
return path.relative_to(root).as_posix()
def ordered_markdown_files(target_dir: Path) -> list[Path]:
"""Order markdown files according to CANONICAL_MD_STAGES."""
markdown_files = [path for path in target_dir.rglob("*.md") if path.is_file()]
remaining = {relative_key(path, target_dir): path for path in markdown_files}
def is_mergeable(path: Path) -> bool:
if path.suffix in SKIP_SUFFIXES:
return False
if any(part in path.parts for part in ("__pycache__", ".git")):
return False
return path.suffix in MERGEABLE_EXTENSIONS
def ordered_artifacts(target_dir: Path) -> list[Path]:
"""Order reviewable files in a feature dir per CANONICAL_STAGES."""
files = [p for p in target_dir.rglob("*") if p.is_file() and is_mergeable(p)]
remaining = {relative_key(p, target_dir): p for p in files}
ordered: list[Path] = []
# Track which stages have already been matched to avoid duplicates
seen = set()
seen: set[str] = set()
for stage_type, stage_value, _section_label in CANONICAL_MD_STAGES:
def take(path: Path) -> None:
rel = relative_key(path, target_dir)
if rel in seen:
return
ordered.append(path)
seen.add(rel)
remaining.pop(rel, None)
for stage_type, stage_value, _label in CANONICAL_STAGES:
if stage_type == "exact":
# Exact: only match if the file exists and hasn't been collected yet
rel = f"contracts/ux/{stage_value}" if stage_value.startswith("contracts/ux/") else stage_value
if stage_value.startswith("contracts/ux/"):
path = remaining.pop(stage_value, None)
else:
path = remaining.pop(stage_value, None)
if path is not None and path not in seen:
ordered.append(path)
seen.add(path)
continue
path = remaining.pop(stage_value, None)
if path is not None:
take(path)
elif stage_type == "prefix":
matches = sorted(
(p for rel, p in remaining.items() if rel.startswith(stage_value)),
key=lambda p: relative_key(p, target_dir),
)
for p in matches:
take(p)
elif stage_type == "glob":
import fnmatch
matches = sorted(
(p for rel, p in remaining.items() if fnmatch.fnmatch(rel, stage_value)),
key=lambda p: relative_key(p, target_dir),
)
for p in matches:
take(p)
# Prefix: match all files whose relative path starts with stage_value
stage_matches = sorted(
[
path
for rel_path, path in remaining.items()
if rel_path.startswith(stage_value)
],
key=lambda p: relative_key(p, target_dir),
)
for path in stage_matches:
if path not in seen:
ordered.append(path)
seen.add(path)
rel = relative_key(path, target_dir)
remaining.pop(rel, None)
# Remaining files — append at end
remaining_paths = sorted(remaining.values(), key=lambda p: relative_key(p, target_dir))
for path in remaining_paths:
if path not in seen:
ordered.append(path)
# Any remaining reviewable files — deterministic order at the end.
for rel in sorted(remaining):
take(remaining[rel])
return ordered
def merge_specs(feature_number: str) -> str | None:
"""Merge all spec artifacts for a feature into one review file."""
def resolve_targets(specs_dir: Path, tokens: list[str]) -> list[Path]:
"""Resolve CLI tokens into concrete feature directories.
Supports:
- single number: "038"
- number range: "036-041" (inclusive)
- explicit list: "036 038 044"
- dir name: "042-dashboard-scenario-registry"
- all: "all"
"""
targets: list[Path] = []
def find_by_number(num: str) -> Path | None:
for item in specs_dir.iterdir():
if item.is_dir() and item.name.startswith(f"{num}-"):
return item
return None
for token in tokens:
token = token.strip()
if not token:
continue
if token.lower() == "all":
for item in sorted(specs_dir.iterdir(), key=lambda p: p.name):
if item.is_dir() and re.match(r"^\d{3}-", item.name):
targets.append(item)
continue
range_match = re.fullmatch(r"(\d{3})-(\d{3})", token)
if range_match:
lo, hi = int(range_match.group(1)), int(range_match.group(2))
if lo > hi:
lo, hi = hi, lo
for num in range(lo, hi + 1):
d = find_by_number(f"{num:03d}")
if d is not None:
targets.append(d)
else:
print(f" [warn] no spec for number {num:03d} — skipped")
continue
if re.fullmatch(r"\d{3}", token):
d = find_by_number(token)
if d is not None:
targets.append(d)
else:
print(f" [warn] no spec for number {token} — skipped")
continue
# Treat as an explicit relative path / dir name.
cand = specs_dir / token
if cand.is_dir():
targets.append(cand)
else:
print(f" [warn] unrecognized target '{token}' — skipped")
# De-duplicate while preserving order.
seen: set[Path] = set()
unique: list[Path] = []
for t in targets:
if t.resolve() not in seen:
seen.add(t.resolve())
unique.append(t)
return unique
def _section_label(rel: str) -> str:
for stage_type, stage_value, label in CANONICAL_STAGES:
if label is None:
continue
if stage_type == "exact" and rel == stage_value:
return label
if stage_type == "prefix" and rel.startswith(stage_value):
return f"{label}{Path(rel).name}"
if stage_type == "glob":
import fnmatch
if fnmatch.fnmatch(rel, stage_value):
return f"{label}{Path(rel).name}"
return rel
def merge_one_feature(target_dir: Path, blocks: list[str]) -> int:
"""Append one feature's merged content to `blocks`. Returns file count."""
feature_name = target_dir.name
artifacts = ordered_artifacts(target_dir)
blocks.append("")
blocks.append("=" * 80)
blocks.append(f"FEATURE: {feature_name}")
blocks.append(f"Files: {len(artifacts)}")
blocks.append("=" * 80)
blocks.append("")
for file_path in artifacts:
rel = relative_key(file_path, target_dir)
try:
content = file_path.read_text(encoding="utf-8")
except Exception as e: # noqa: BLE001
print(f" [warn] skipping {rel}: {e}")
continue
blocks.append("")
blocks.append("-" * 60)
blocks.append(f"## {_section_label(rel)}")
blocks.append(f"Source: {rel}")
blocks.append("-" * 60)
blocks.append("")
blocks.append(content)
blocks.append("")
return len(artifacts)
def merge_specs(tokens: list[str], output: str | None = None) -> str | None:
"""Merge all requested spec packages into one review file."""
specs_dir = Path("specs")
if not specs_dir.exists():
print("Error: 'specs' directory not found.")
return None
# Find directory starting with feature number
target_dir = None
for item in specs_dir.iterdir():
if item.is_dir() and item.name.startswith(f"{feature_number}-"):
target_dir = item
break
if not target_dir:
print(f"Error: No directory found for feature number '{feature_number}' in 'specs/'.")
targets = resolve_targets(specs_dir, tokens)
if not targets:
print("Error: no specs matched the given arguments.")
return None
feature_name = target_dir.name
now = datetime.now().strftime("%Y%m%d-%H%M%S")
output_filename = f"{feature_name}-{now}.md"
if output is None:
if len(targets) == 1:
output = f"{targets[0].name}-{now}.md"
else:
first, last = targets[0].name[:3], targets[-1].name[:3]
output = f"specs-{first}-{last}-{now}.md"
content_blocks = [
blocks = [
REVIEW_PROMPT,
"",
"=" * 80,
ORDER_HINT,
f"Feature: {feature_name}",
f"Features: {', '.join(t.name for t in targets)}",
f"Generated: {datetime.now().isoformat()}",
"=" * 80,
"",
]
files_to_merge = ordered_markdown_files(target_dir)
total_files = 0
for target in targets:
total_files += merge_one_feature(target, blocks)
for file_path in files_to_merge:
relative_path = file_path.relative_to(target_dir)
try:
content = file_path.read_text(encoding="utf-8")
except Exception as e:
print(f"Skipping {file_path}: {e}")
continue
blocks.append("")
blocks.append("=" * 80)
blocks.append(f"Features merged: {len(targets)} | Files merged: {total_files}")
blocks.append("=" * 80)
blocks.append("")
# Determine section label
section_label = str(relative_path)
for stage_type, stage_value, label in CANONICAL_MD_STAGES:
if label is None:
continue
if stage_type == "exact" and str(relative_path) == stage_value:
section_label = label
break
if stage_type == "prefix" and str(relative_path).startswith(stage_value):
section_label = f"{label}{relative_path.name}"
break
merged = "\n".join(blocks)
Path(output).write_text(merged, encoding="utf-8")
print(f"Created: {output} ({len(targets)} features, {total_files} files)")
return output
content_blocks.append("")
content_blocks.append("-" * 60)
content_blocks.append(f"## {section_label}")
content_blocks.append(f"Source: {relative_path}")
content_blocks.append("-" * 60)
content_blocks.append("")
content_blocks.append(content)
content_blocks.append("")
# Summary
content_blocks.append("")
content_blocks.append("=" * 80)
content_blocks.append(f"Files merged: {len(files_to_merge)}")
content_blocks.append("=" * 80)
def main(argv: list[str]) -> int:
if len(argv) < 2:
print("Usage:")
print(" python merge_spec.py <number> # single spec, e.g. 038")
print(" python merge_spec.py <lo>-<hi> # range, e.g. 036-041")
print(" python merge_spec.py <n> <n> ... # explicit list, e.g. 036 040 044")
print(" python merge_spec.py all # every spec package")
print(" python merge_spec.py <dir-name> # by directory name")
print(" python merge_spec.py <n>-<n> -o out.md # custom output file")
return 1
merged = "\n".join(content_blocks)
output_filename = output_filename
Path(output_filename).write_text(merged, encoding="utf-8")
tokens = argv[1:]
out_path = None
if "-o" in tokens or "--output" in tokens:
flag_idx = tokens.index("-o") if "-o" in tokens else tokens.index("--output")
if flag_idx + 1 >= len(tokens):
print("Error: -o/--output requires a filename")
return 1
out_path = tokens[flag_idx + 1]
tokens = tokens[:flag_idx] + tokens[flag_idx + 2:]
print(f"Successfully created: {output_filename} ({len(files_to_merge)} files)")
return output_filename
result = merge_specs(tokens, output=out_path)
return 0 if result else 1
if __name__ == "__main__":
if len(sys.argv) < 2:
print("Usage: python merge_spec.py <feature_number>")
sys.exit(1)
merge_specs(sys.argv[1])
sys.exit(main(sys.argv))
# #endregion Tooling.Merge.Spec