Compare commits

..

1 Commits

Author SHA1 Message Date
Alexander Payne
2f57c2b653 feat(10.8): add Progress Tracker — track test/doc coverage, issue close rate, dep freshness
Some checks failed
Test / pytest (pull_request) Failing after 8s
Implements 10.8: Progress Tracker.

- Script: scripts/progress_tracker.py
- Stores weekly snapshots in metrics/snapshots/ (auto-created)
- Generates trends table in metrics/TRENDS.md
- Tracks: test-to-source ratio, doc coverage via AST, issue close rate (Gitea API), dep freshness (pip)
- Outputs stdout summary for weekly review

Usage:
    python3 scripts/progress_tracker.py
    python3 scripts/progress_tracker.py --json
    python3 scripts/progress_tracker.py --output metrics/TRENDS.md

Weekly cron: 0 9 * * 1 cd /path && python3 scripts/progress_tracker.py

Closes #173
2026-04-26 05:15:01 -04:00
3 changed files with 477 additions and 752 deletions

View File

@@ -1,530 +0,0 @@
#!/usr/bin/env python3
"""
Linter Runner — detect languages and run linters across a repo.
Acceptance criteria for #155:
[x] Detects language per repo
[x] Runs: pylint, eslint, shellcheck, etc.
[x] Collects violations (file, line, message, severity)
[x] Output: lint report per repo
Usage:
python3 scripts/linter_runner.py --repo .
python3 scripts/linter_runner.py --all # Scan all repos in knowledge/repos/
python3 scripts/linter_runner.py --repo . --format json # Machine-readable output
python3 scripts/linter_runner.py --repo . --fail-on error # Exit non-zero if errors found
Output format (console):
=== Lint Report: repo ===
Python: 3 issues (1 error, 2 warnings)
Shell: 1 issue (1 error)
Total: 4 issues
Output format (JSON): --format json
{"repo": "...", "issues": [...], "summary": {...}}
"""
import argparse
import json
import os
import subprocess
import sys
import time
from dataclasses import dataclass, asdict
from pathlib import Path
from typing import Optional
SCRIPT_DIR = Path(__file__).resolve().parent
REPO_ROOT = SCRIPT_DIR.parent
@dataclass
class Violation:
"""A single lint violation."""
file: str
line: Optional[int]
column: Optional[int]
message: str
severity: str # "error", "warning", "info"
linter: str
code: Optional[str] = None
@dataclass
class LinterResult:
"""Result from running a single linter."""
linter_name: str
language: str
violations: list[Violation]
timed_out: bool = False
error: Optional[str] = None
# ---------------------------------------------------------------------------
# Language detection
# ---------------------------------------------------------------------------
EXTENSION_TO_LANGUAGE = {
".py": "python",
".js": "javascript",
".ts": "typescript",
".jsx": "javascript",
".tsx": "typescript",
".sh": "shell",
".bash": "shell",
".zsh": "shell",
".yaml": "yaml",
".yml": "yaml",
".json": "json",
".md": "markdown",
".rb": "ruby",
".go": "go",
".rs": "rust",
".c": "c",
".cpp": "cpp",
".h": "c",
".java": "java",
".php": "php",
".swift": "swift",
".kt": "kotlin",
".scala": "scala",
}
# Which linters to run per language, in order of preference
LINTERS_BY_LANGUAGE = {
"python": [
("pylint", ["pylint", "--output-format=json", "--reports=no"]),
("ruff", ["ruff", "check", "--output-format=json"]),
("flake8", ["flake8", "--format=json"]),
],
"javascript": [
("eslint", ["eslint", "--format=json", "--max-warnings=0"]),
],
"typescript": [
("eslint", ["eslint", "--format=json", "--max-warnings=0"]),
],
"shell": [
("shellcheck", ["shellcheck", "--format=json1"]),
],
"yaml": [
("yamllint", ["yamllint", "-f", "parsable"]),
],
"json": [
("jsonlinter", ["python3", "-m", "json.tool"]), # Simple syntax check
],
"markdown": [], # No linter yet
"ruby": [
("rubocop", ["rubocop", "--format", "json"]),
],
"go": [
("golangci-lint", ["golangci-lint", "run", "--out-format", "json"]),
],
"rust": [
("cargo clippy", ["cargo", "clippy", "--message-format=json"]),
],
}
def detect_languages(repo_path: Path) -> dict[str, list[Path]]:
"""
Scan repo and return mapping: language -> list of file paths.
Only includes languages we have linters for."""
language_files: dict[str, list[Path]] = {lang: [] for lang in LINTERS_BY_LANGUAGE.keys()}
if not repo_path.exists():
return language_files
exclude_dirs = {".git", ".gitea", "node_modules", "__pycache__", ".venv", "venv", "build", "dist"}
for root, dirs, files in os.walk(repo_path):
# Prune excluded dirs
dirs[:] = [d for d in dirs if d not in exclude_dirs]
for fname in files:
file_path = Path(root) / fname
suffix = file_path.suffix.lower()
lang = EXTENSION_TO_LANGUAGE.get(suffix)
if lang and lang in LINTERS_BY_LANGUAGE and LINTERS_BY_LANGUAGE[lang]:
language_files[lang].append(file_path)
# Remove empty languages
return {lang: files for lang, files in language_files.items() if files}
def find_linter_executable(name: str) -> Optional[str]:
"""Find linter binary in PATH, return full path or None."""
for path_dir in os.environ.get("PATH", "").split(os.pathsep):
candidate = Path(path_dir) / name
if candidate.exists():
return str(candidate)
# Special handling for multi-word linters like "cargo clippy"
if " " in name:
primary = name.split()[0]
for path_dir in os.environ.get("PATH", "").split(os.pathsep):
candidate = Path(path_dir) / primary
if candidate.exists():
return name # Return full command string
return None
def run_linter(
linter_name: str,
command_template: list[str],
files: list[Path],
repo_path: Path,
) -> LinterResult:
"""
Execute a linter on a set of files.
Returns LinterResult with violations or error.
"""
# Build command: [linter_bin, args..., files...]
# Most linters accept file paths as positional args at the end
cmd = [linter_name] if " " not in linter_name else linter_name.split()
cmd.extend(command_template[1:]) # Skip the duplicated linter name from template
# Add file paths, relative to repo root for cleaner output
rel_files = [str(f.relative_to(repo_path)) for f in files]
cmd.extend(rel_files)
try:
proc = subprocess.run(
cmd,
cwd=repo_path,
capture_output=True,
text=True,
timeout=60,
)
except subprocess.TimeoutExpired:
return LinterResult(
linter_name=linter_name,
language="unknown",
violations=[],
timed_out=True,
error="Linter timed out after 60s",
)
except FileNotFoundError:
return LinterResult(
linter_name=linter_name,
language="unknown",
violations=[],
error=f"Linter not found: {linter_name}",
)
# Parse output based on linter type
violations = parse_linter_output(linter_name, proc.stdout, proc.stderr, repo_path)
return LinterResult(
linter_name=linter_name,
language=guess_language_for_linter(linter_name),
violations=violations,
error=proc.stderr.strip() if proc.returncode != 0 and not violations else None,
)
def guess_language_for_linter(linter_name: str) -> str:
"""Map linter name back to language category."""
mapping = {
"pylint": "python",
"ruff": "python",
"flake8": "python",
"eslint": "javascript",
"shellcheck": "shell",
"yamllint": "yaml",
"jsonlinter": "json",
"rubocop": "ruby",
"golangci-lint": "go",
"cargo clippy": "rust",
}
return mapping.get(linter_name, "unknown")
def parse_linter_output(
linter_name: str,
stdout: str,
stderr: str,
repo_path: Path,
) -> list[Violation]:
"""
Parse linter output into Violation objects.
Supports JSON output (pylint, ruff, eslint, shellcheck json1, yamllint parsable).
"""
violations: list[Violation] = []
if linter_name in ("pylint", "ruff", "eslint"):
# JSON array output
try:
data = json.loads(stdout)
except json.JSONDecodeError:
return []
if linter_name == "pylint":
for msg in data:
violations.append(Violation(
file=msg.get("path", "").lstrip("./"),
line=msg.get("line"),
column=msg.get("column"),
message=msg.get("message", ""),
severity="error" if msg.get("type") == "error" else "warning",
linter=linter_name,
code=msg.get("symbol"),
))
elif linter_name == "ruff":
for entry in data:
violations.append(Violation(
file=entry.get("filename", "").lstrip("./"),
line=entry.get("location", {}).get("row"),
column=entry.get("location", {}).get("column"),
message=entry.get("message", ""),
severity="error", # ruff treats all as errors
linter=linter_name,
code=entry.get("code"),
))
elif linter_name == "eslint":
for entry in data:
violations.append(Violation(
file=entry.get("fileName", "").lstrip("./"),
line=entry.get("range", {}).get("start", {}).get("line"),
column=entry.get("range", {}).get("start", {}).get("column"),
message=entry.get("message", ""),
severity=entry.get("severity", 1) and "error" or "warning",
linter=linter_name,
code=entry.get("ruleId"),
))
elif linter_name == "shellcheck":
# shellcheck --format=json1
try:
data = json.loads(stdout)
for issue in data.get("issues", []):
violations.append(Violation(
file=issue.get("file", "").lstrip("./"),
line=issue.get("line"),
column=issue.get("column"),
message=issue.get("message", ""),
severity="error" if issue.get("level") == "error" else "warning",
linter=linter_name,
code=str(issue.get("code")),
))
except json.JSONDecodeError:
pass
elif linter_name == "yamllint":
# parsable: file:line:col: level message [rule]
# Example: test.yaml:3:1: [error] wrong document start (document-start)
for line in stdout.splitlines():
parts = line.split(":")
if len(parts) >= 4:
file_rel = parts[0].lstrip("./")
line_num = int(parts[1]) if parts[1].isdigit() else None
col_num = int(parts[2]) if parts[2].isdigit() else None
rest = ":".join(parts[3:]).strip()
# Parse: "[error] message (rule)"
import re
m = re.match(r'\[(\w+)\]\s+(.+?)(?:\s+\(([^)]+)\))?$', rest)
if m:
severity = m.group(1).lower()
message = m.group(2)
code = m.group(3)
violations.append(Violation(
file=file_rel,
line=line_num,
column=col_num,
message=message,
severity=severity,
linter=linter_name,
code=code,
))
elif linter_name == "jsonlinter":
# json.tool syntax check — no violations, just exit code
if proc.returncode != 0:
violations.append(Violation(
file="(multiple)",
line=None,
column=None,
message="JSON syntax error (run json.tool on each file individually)",
severity="error",
linter="json.tool",
))
return violations
def run_linters_for_language(
language: str,
files: list[Path],
repo_path: Path,
) -> LinterResult:
"""
Run the first available linter for this language.
Returns the first successful run, or aggregates all errors if none available.
"""
linter_options = LINTERS_BY_LANGUAGE.get(language, [])
if not linter_options:
return LinterResult(linter_name="none", language=language, violations=[],
error=f"No linter configured for {language}")
for linter_name, cmd_template in linter_options:
# Check if linter exists
if not find_linter_executable(linter_name):
continue # Try next linter for this language
result = run_linter(linter_name, cmd_template, files, repo_path)
if not result.error and not result.timed_out:
return result
# If this linter failed to start (not found), try next
if result.error and "not found" in result.error.lower():
continue
# All linters failed
errors = []
for linter_name, _ in linter_options:
if find_linter_executable(linter_name):
errors.append(f"{linter_name}: not runnable")
else:
errors.append(f"{linter_name}: not installed")
return LinterResult(
linter_name="/".join(l[0] for l in linter_options),
language=language,
violations=[],
error="; ".join(errors),
)
# ---------------------------------------------------------------------------
# Main
# ---------------------------------------------------------------------------
def parse_args():
p = argparse.ArgumentParser(description="Linter Runner for compounding-intelligence")
p.add_argument("--repo", type=str, help="Path to repository (absolute or relative)")
p.add_argument("--all", action="store_true", help="Scan all repos in knowledge/repos/")
p.add_argument("--format", choices=["console", "json"], default="console",
help="Output format (default: console)")
p.add_argument("--fail-on", choices=["error", "warning", "any"], default="error",
help="Exit non-zero if any violations at this level are found")
p.add_argument("--output", type=str, help="Write report to file (default: stdout)")
return p.parse_args()
def main():
args = parse_args()
if not args.repo and not args.all:
print("ERROR: Must specify --repo <path> or --all", file=sys.stderr)
sys.exit(2)
repos_to_scan = []
if args.repo:
repos_to_scan.append(Path(args.repo).resolve())
if args.all:
repos_dir = REPO_ROOT / "knowledge" / "repos"
if repos_dir.exists():
for yaml_file in repos_dir.glob("*.yaml"):
# Extract repo name from filename
repos_to_scan.append(REPO_ROOT / yaml_file.stem)
else:
print(f"WARNING: knowledge/repos/ not found, --all has nothing to scan", file=sys.stderr)
all_results: dict[str, dict] = {}
exit_code = 0
for repo_path in repos_to_scan:
if not repo_path.exists():
print(f"WARNING: Repo not found: {repo_path}", file=sys.stderr)
continue
repo_name = repo_path.name
print(f"\n=== Scanning: {repo_name} ===") if args.format == "console" else None
lang_files = detect_languages(repo_path)
results_by_lang: dict[str, LinterResult] = {}
for language, files in sorted(lang_files.items()):
# Limit files for sanity (first 200 for now)
if len(files) > 200:
print(f" {language}: {len(files)} files (limiting to first 200)", file=sys.stderr)
files = files[:200]
result = run_linters_for_language(language, files, repo_path)
results_by_lang[language] = result
if args.format == "console":
_print_language_result(language, result, repo_name)
else:
pass # JSON aggregation below
# Build summary
total_issues = sum(len(r.violations) for r in results_by_lang.values())
total_errors = sum(1 for v in (v for r in results_by_lang.values() for v in r.violations)
if v.severity == "error")
total_warnings = sum(1 for v in (v for r in results_by_lang.values() for v in r.violations)
if v.severity == "warning")
if args.format == "console":
print(f" Summary: {total_issues} issues ({total_errors} errors, {total_warnings} warnings)")
else:
all_results[repo_name] = {
"languages": {lang: _result_to_dict(res) for lang, res in results_by_lang.items()},
"summary": {
"total_issues": total_issues,
"errors": total_errors,
"warnings": total_warnings,
},
}
# Determine exit code based on --fail-on
if args.fail_on == "error" and total_errors > 0:
exit_code = 1
elif args.fail_on == "warning" and total_issues > 0:
exit_code = 1
elif args.fail_on == "any" and total_issues > 0:
exit_code = 1
if args.format == "json":
output = json.dumps({"repos": all_results, "meta": {"scanned": len(repos_to_scan)}}, indent=2)
if args.output:
Path(args.output).write_text(output)
else:
print(output)
sys.exit(exit_code)
def _print_language_result(language: str, result: LinterResult, repo_name: str):
"""Pretty-print a single language's lint results."""
status = ""
if result.error:
status = ""
print(f" {language}: {result.error}")
elif result.timed_out:
status = ""
print(f" {language}: timed out")
else:
n_violations = len(result.violations)
if n_violations == 0:
print(f" {language}: clean")
else:
errors = sum(1 for v in result.violations if v.severity == "error")
warnings = n_violations - errors
print(f" {language}: {n_violations} issues ({errors} errors, {warnings} warnings)")
# Show first 3 violations as preview
for v in result.violations[:3]:
loc = f"{v.file}:{v.line or '?'}"
print(f" {loc} [{v.severity.upper()}] {v.message[:70]}")
if len(result.violations) > 3:
print(f" ... and {len(result.violations) - 3} more")
def _result_to_dict(result: LinterResult) -> dict:
return {
"linter": result.linter_name,
"language": result.language,
"violations": [asdict(v) for v in result.violations],
"timed_out": result.timed_out,
"error": result.error,
}
if __name__ == "__main__":
main()

477
scripts/progress_tracker.py Normal file
View File

@@ -0,0 +1,477 @@
#!/usr/bin/env python3
"""
Progress Tracker — Pipeline 10.8
Track improvement metrics over time. Are we getting better?
Metrics tracked:
1. Test coverage — % of Python functions with associated tests (test:source file ratio + line coverage if available)
2. Doc coverage — % of Python callables with docstrings (AST-based)
3. Issue close rate — closed / (opened + closed) per week (Gitea API)
4. Dep freshness — % of requirements pinned vs outdated (pip list --outdated)
Output:
- metrics/snapshots/YYYY-MM-DD.json — one snapshot per run
- metrics/TRENDS.md — cumulative markdown table
- stdout summary
Usage:
python3 scripts/progress_tracker.py
python3 scripts/progress_tracker.py --json
python3 scripts/progress_tracker.py --output metrics/TRENDS.md
Weekly cron:
0 9 * * 1 cd /path/to/compounding-intelligence && python3 scripts/progress_tracker.py
"""
import argparse
import json
import os
import re
import subprocess
import sys
from collections import defaultdict
from datetime import datetime, timezone, timedelta
from pathlib import Path
from typing import Any, Dict, List, Optional, Tuple
# ── Configuration ──────────────────────────────────────────────────────────
SCRIPT_DIR = Path(__file__).resolve().parent
REPO_ROOT = SCRIPT_DIR.parent
METRICS_DIR = REPO_ROOT / "metrics"
SNAPSHOTS_DIR = METRICS_DIR / "snapshots"
TOKEN_PATH = Path.home() / ".config" / "gitea" / "token"
GITEA_API_BASE = "https://forge.alexanderwhitestone.com/api/v1"
ORG = "Timmy_Foundation"
# Ensure paths exist
SNAPSHOTS_DIR.mkdir(parents=True, exist_ok=True)
# ── Helpers ─────────────────────────────────────────────────────────────────
def run_cmd(cmd: List[str], cwd: Path = REPO_ROOT) -> str:
"""Run a shell command and return stdout (stderr merged)."""
result = subprocess.run(
cmd, capture_output=True, text=True, cwd=cwd, timeout=30
)
if result.returncode != 0:
return ""
return result.stdout.strip()
def slugify_date(dt: datetime) -> str:
return dt.strftime("%Y-%m-%d")
def snapshot_path(dt: datetime) -> Path:
return SNAPSHOTS_DIR / f"{slugify_date(dt)}.json"
def load_snapshots() -> List[Dict[str, Any]]:
"""Load all existing snapshots sorted by date."""
snapshots = []
for f in sorted(SNAPSHOTS_DIR.glob("*.json")):
try:
with open(f) as fp:
snapshots.append(json.load(fp))
except Exception:
continue
return snapshots
# ── Metric 1: Test Coverage ─────────────────────────────────────────────────
def collect_test_coverage() -> Dict[str, Any]:
"""
Compute test coverage metrics.
Counts test_*.py and *_test.py files vs non-test .py source files.
Also attempts to read .coverage if present.
"""
all_py = list(REPO_ROOT.rglob("*.py"))
source_files = []
test_files = []
for p in all_py:
try:
rel_parts = p.relative_to(REPO_ROOT).parts
except ValueError:
continue
# Skip hidden/cache/temp dirs (check only relative parts)
if any(part.startswith('.') or part.startswith('__') for part in rel_parts):
continue
if any(part in ('node_modules', 'venv', '.venv', 'env', '.pytest_cache') for part in rel_parts):
continue
if p.name.startswith("test_") or p.name.endswith("_test.py"):
test_files.append(p)
else:
source_files.append(p)
# Try to get line coverage from .coverage
coverage_percent = None
coverage_tool = None
coverage_file = REPO_ROOT / ".coverage"
if coverage_file.exists():
try:
import coverage # type: ignore
# Use coverage API if available
cov = coverage.Coverage(data_file=str(coverage_file))
cov.load()
total = cov.report()
coverage_percent = total if isinstance(total, float) else None
coverage_tool = "coverage"
except Exception:
# Fallback: parse `coverage report` output
out = run_cmd(["coverage", "report", "--skip-empty"])
if out:
for line in out.splitlines():
if "TOTAL" in line:
parts = line.split()
if len(parts) >= 2:
try:
coverage_percent = float(parts[-1].rstrip('%'))
coverage_tool = "coverage"
break
except ValueError:
pass
return {
"test_files": len(test_files),
"source_files": len(source_files),
"test_to_source_ratio": round(len(test_files) / len(source_files), 4) if source_files else 0.0,
"coverage_tool": coverage_tool,
"coverage_percent": coverage_percent,
}
# ── Metric 2: Doc Coverage ──────────────────────────────────────────────────
def collect_doc_coverage() -> Dict[str, Any]:
"""
Check AST of Python files for docstrings.
Returns: callables_total, callables_with_doc, doc_coverage_percent
"""
import ast
all_py = list(REPO_ROOT.rglob("*.py"))
source_files = []
test_files = []
for p in all_py:
try:
rel_parts = p.relative_to(REPO_ROOT).parts
except ValueError:
continue
if any(part.startswith('.') or part.startswith('__') for part in rel_parts):
continue
if any(part in ('node_modules', 'venv', '.venv', 'env', '.pytest_cache') for part in rel_parts):
continue
if p.name.startswith("test_") or p.name.endswith("_test.py"):
test_files.append(p)
else:
source_files.append(p)
total_callables = 0
with_doc = 0
for p in source_files + test_files:
try:
with open(p) as f:
tree = ast.parse(f.read(), filename=str(p))
for node in ast.walk(tree):
if isinstance(node, (ast.FunctionDef, ast.AsyncFunctionDef, ast.ClassDef)):
total_callables += 1
doc = ast.get_docstring(node)
if doc and doc.strip():
with_doc += 1
except Exception:
continue
return {
"callables_total": total_callables,
"callables_with_doc": with_doc,
"doc_coverage_percent": round((with_doc / total_callables * 100) if total_callables else 0.0, 2),
}
# ── Metric 3: Issue Close Rate ──────────────────────────────────────────────
def collect_issue_metrics() -> Dict[str, Any]:
"""
Use Gitea API to get issue open/close stats for the last 7 days.
Returns counts and close rate.
"""
token = ""
if TOKEN_PATH.exists():
token = TOKEN_PATH.read_text().strip()
if not token:
return {
"opened_last_7d": None,
"closed_last_7d": None,
"close_rate": None,
"total_open": None,
"note": "Gitea token not available"
}
try:
from urllib.request import Request, urlopen
from urllib.error import HTTPError, URLError
except ImportError:
return {"error": "urllib not available"}
now = datetime.now(timezone.utc)
week_ago = now - timedelta(days=7)
since = week_ago.strftime("%Y-%m-%d")
headers = {"Authorization": f"token {token}"}
base_url = f"{GITEA_API_BASE}/repos/{ORG}/compounding-intelligence/issues"
try:
# Get issues from last 7 days
url = f"{base_url}?state=all&since={since}&per_page=100"
req = Request(url, headers=headers)
with urlopen(req, timeout=15) as resp:
issues = json.loads(resp.read())
opened = 0
closed = 0
for issue in issues:
created = datetime.fromisoformat(issue["created_at"].replace("Z", "+00:00"))
if created >= week_ago:
opened += 1
if issue.get("state") == "closed":
closed_at_str = issue.get("closed_at")
if closed_at_str:
closed_at = datetime.fromisoformat(closed_at_str.replace("Z", "+00:00"))
if closed_at >= week_ago:
closed += 1
# Total open issues
req2 = Request(f"{base_url}?state=open&per_page=1", headers=headers)
with urlopen(req2, timeout=15) as resp:
total_open = int(resp.headers.get("X-Total-Count", "0"))
total = opened + closed
close_rate = closed / total if total > 0 else 0.0
return {
"opened_last_7d": opened,
"closed_last_7d": closed,
"close_rate": round(close_rate, 4),
"total_open": total_open,
}
except Exception as e:
return {
"opened_last_7d": None,
"closed_last_7d": None,
"close_rate": None,
"total_open": None,
"error": str(e)[:100],
"note": "Gitea API unavailable"
}
# ── Metric 4: Dependency Freshness ─────────────────────────────────────────
def collect_dep_freshness() -> Dict[str, Any]:
"""
Check requirements.txt for outdated dependencies using pip list --outdated.
Returns freshness percentage and outdated list.
"""
req_file = REPO_ROOT / "requirements.txt"
if not req_file.exists():
return {
"total_deps": 0,
"outdated_deps": 0,
"freshness_percent": 100.0,
"outdated_list": [],
"note": "requirements.txt not found"
}
# Parse requirements (very simple: take name before comparison op)
reqs = []
with open(req_file) as f:
for line in f:
line = line.strip()
if not line or line.startswith("#"):
continue
m = re.match(r"^([a-zA-Z0-9_.-]+)", line)
if m:
reqs.append(m.group(1))
if not reqs:
return {"total_deps": 0, "outdated_deps": 0, "freshness_percent": 100.0, "outdated_list": []}
# Query pip for outdated packages (may fail if pip not available)
outdated_names = set()
try:
out = run_cmd(["pip", "list", "--outdated", "--format=json"])
if out:
data = json.loads(out)
outdated_names = {item["name"].lower() for item in data}
except Exception:
pass
outdated = [p for p in reqs if p.lower() in outdated_names]
total = len(reqs)
outdated_count = len(outdated)
freshness = round(((total - outdated_count) / total * 100) if total else 100.0, 1)
return {
"total_deps": total,
"outdated_deps": outdated_count,
"freshness_percent": freshness,
"outdated_list": outdated,
}
# ── Snapshot & Trends ───────────────────────────────────────────────────────
def take_snapshot() -> Dict[str, Any]:
"""Collect all metrics and return a snapshot dict."""
now = datetime.now(timezone.utc)
test_cov = collect_test_coverage()
doc_cov = collect_doc_coverage()
issues = collect_issue_metrics()
deps = collect_dep_freshness()
return {
"timestamp": now.isoformat(),
"date": slugify_date(now),
"metrics": {
"test_coverage": test_cov,
"doc_coverage": doc_cov,
"issues": issues,
"dependencies": deps,
}
}
def save_snapshot(snapshot: Dict[str, Any]) -> Path:
path = snapshot_path(datetime.fromisoformat(snapshot["timestamp"]))
with open(path, "w") as f:
json.dump(snapshot, f, indent=2)
return path
def generate_trends(snapshots: List[Dict[str, Any]], output_path: Optional[Path] = None) -> str:
"""Generate markdown trends table; optionally write to file."""
if not snapshots:
msg = "# Progress Tracker — Trends\n\nNo snapshots yet. Run `progress_tracker.py` to create the first snapshot."
if output_path:
output_path.parent.mkdir(parents=True, exist_ok=True)
output_path.write_text(msg)
return msg
lines = [
"# Progress Tracker — Trends",
f"\nLast updated: {datetime.now(timezone.utc).strftime('%Y-%m-%d %H:%M UTC')}",
f"\nSnapshots: {len(snapshots)}\n",
"| Date | Test Files → Source | Doc Coverage | Issues Closed/Opened (7d) | Dep Freshness |",
"|------|---------------------|--------------|---------------------------|---------------|",
]
for snap in reversed(snapshots): # chronological
date = snap["date"]
m = snap["metrics"]
tc = m["test_coverage"]
test_str = f"{tc['test_files']}/{tc['source_files']} ({tc['test_to_source_ratio']:.2f})"
doc_str = f"{m['doc_coverage']['doc_coverage_percent']:.1f}%"
issues_str = f"{m['issues'].get('closed_last_7d','-')}/{m['issues'].get('opened_last_7d','-')}"
dep_str = f"{m['dependencies'].get('freshness_percent','?')}%"
lines.append(f"| {date} | {test_str} | {doc_str} | {issues_str} | {dep_str} |")
# Current snapshot summary
cur = snapshots[-1]
cm = cur["metrics"]
lines.append(f"\n## Current Snapshot ({cur['date']})\n")
tc = cm["test_coverage"]
cov_line = f"- Test coverage: {tc['coverage_percent']:.1f}% (via {tc['coverage_tool']})\n" if tc["coverage_percent"] else "- Test coverage: (pytest-cov not configured)\n"
lines.append(cov_line)
lines.append(f"- Doc coverage: {cm['doc_coverage']['doc_coverage_percent']:.1f}%")
im = cm["issues"]
if im.get("close_rate") is not None:
lines.append(f"- Issue close rate (7d): {im['close_rate']*100:.1f}% ({im['closed_last_7d']} closed, {im['opened_last_7d']} opened)")
else:
lines.append(f"- Issue metrics: {im.get('note','unavailable')}")
dd = cm["dependencies"]
lines.append(f"- Dep freshness: {dd.get('freshness_percent','?')}% outdated ({dd.get('outdated_deps',0)}/{dd.get('total_deps',0)} deps)")
if dd.get('outdated_list'):
lines.append(f" Outdated: {', '.join(dd['outdated_list'][:5])}")
content = "\n".join(lines) + "\n"
if output_path:
output_path.parent.mkdir(parents=True, exist_ok=True)
output_path.write_text(content)
return content
# ── Main ─────────────────────────────────────────────────────────────────────
def main() -> int:
parser = argparse.ArgumentParser(description="Progress Tracker — 10.8")
parser.add_argument("--json", action="store_true", help="Emit snapshot as JSON only")
parser.add_argument("--output", type=Path, default=METRICS_DIR / "TRENDS.md",
help="Write trends markdown to this file")
args = parser.parse_args()
snapshot = take_snapshot()
all_snapshots = load_snapshots()
path_written = save_snapshot(snapshot)
if args.json:
print(json.dumps(snapshot, indent=2))
return 0
trends = generate_trends(all_snapshots + [snapshot], output_path=args.output)
# Print current snapshot summary
print(f"Snapshot saved: {path_written}\n")
print(f"Progress Tracker — {snapshot['date']}")
print("=" * 50)
m = snapshot["metrics"]
tc = m["test_coverage"]
print(f"Test files: {tc['test_files']} | Source files: {tc['source_files']} | Ratio: {tc['test_to_source_ratio']:.3f}")
if tc["coverage_percent"] is not None:
print(f"Line coverage: {tc['coverage_percent']:.1f}% (via {tc['coverage_tool']})")
else:
print("Line coverage: (not available — run `pytest --cov`)")
print()
dc = m["doc_coverage"]
print(f"Callables with docstrings: {dc['callables_with_doc']}/{dc['callables_total']} ({dc['doc_coverage_percent']:.1f}%)")
print()
im = m["issues"]
if im.get("close_rate") is not None:
print(f"Issues (7d): {im['closed_last_7d']} closed / {im['opened_last_7d']} opened → close rate: {im['close_rate']*100:.1f}%")
print(f"Total open: {im['total_open']}")
else:
print(f"Issues: {im.get('note','unavailable')}")
print()
dd = m["dependencies"]
print(f"Dependencies: {dd.get('total_deps',0)} total, {dd.get('outdated_deps',0)} outdated")
if dd.get('outdated_list'):
shown = dd['outdated_list'][:5]
print(f"Outdated: {', '.join(shown)}" + ("..." if len(dd['outdated_list']) > 5 else ""))
print(f"\nTrends written to: {args.output}")
return 0
if __name__ == "__main__":
sys.exit(main())

View File

@@ -1,222 +0,0 @@
#!/usr/bin/env python3
"""Tests for linter_runner module (Issue #155).
Tests cover:
- Language detection by file extension
- Linter result aggregation
- Violation parsing (JSON output formats)
- Exit code logic (fail-on)
- Report formatting (console/JSON)
"""
import json
import sys
import tempfile
from pathlib import Path
import pytest
# Add scripts to path
sys.path.insert(0, str(Path(__file__).parent.parent / "scripts"))
from linter_runner import (
Violation,
LinterResult,
detect_languages,
parse_linter_output,
_result_to_dict,
EXTENSION_TO_LANGUAGE,
LINTERS_BY_LANGUAGE,
)
class TestLanguageDetection:
"""Test detect_languages() identifies languages correctly."""
def test_detects_python_files(self, tmp_path: Path):
(tmp_path / "main.py").write_text("print('hello')")
(tmp_path / "lib" / "utils.py").mkdir(parents=True)
(tmp_path / "lib" / "utils.py").write_text("def foo(): pass")
result = detect_languages(tmp_path)
assert "python" in result
assert len(result["python"]) == 2
def test_detects_javascript_files(self, tmp_path: Path):
(tmp_path / "app.js").write_text("console.log('hi')")
(tmp_path / "component.jsx").write_text("<div/>")
result = detect_languages(tmp_path)
assert "javascript" in result
assert len(result["javascript"]) == 2
def test_detects_shell_files(self, tmp_path: Path):
(tmp_path / "setup.sh").write_text("#!/bin/bash\necho hi")
(tmp_path / "build.sh").write_text("make")
result = detect_languages(tmp_path)
assert "shell" in result
assert len(result["shell"]) == 2
def test_detects_yaml_files(self, tmp_path: Path):
(tmp_path / "config.yml").write_text("key: value")
(tmp_path / "env.yaml").write_text("env: test")
result = detect_languages(tmp_path)
assert "yaml" in result
assert len(result["yaml"]) == 2
def test_ignores_git_directory(self, tmp_path: Path):
git_dir = tmp_path / ".git"
git_dir.mkdir()
(git_dir / "config").write_text("placeholder")
(tmp_path / "script.py").write_text("print(1)")
result = detect_languages(tmp_path)
assert "python" in result
assert not any(".git" in str(f) for f in result.get("python", []))
def test_returns_empty_for_nonexistent_path(self):
result = detect_languages(Path("/nonexistent/path/xyz"))
assert result == {}
def test_mixed_languages(self, tmp_path: Path):
(tmp_path / "app.py").write_text("")
(tmp_path / "main.js").write_text("")
(tmp_path / "deploy.sh").write_text("")
result = detect_languages(tmp_path)
langs = set(result.keys())
assert {"python", "javascript", "shell"} <= langs
def test_limits_files_to_known_languages(self, tmp_path: Path):
(tmp_path / "readme.txt").write_text("text")
(tmp_path / "data.csv").write_text("a,b,c")
result = detect_languages(tmp_path)
assert len(result) == 0
class TestViolationParsing:
"""Test parse_linter_output parses various linter formats."""
def test_parses_pylint_json(self):
stdout = json.dumps([
{"type": "error", "module": "test.py", "line": 10, "column": 5,
"message": "Missing docstring", "symbol": "missing-docstring"},
{"type": "warning", "module": "test.py", "line": 15, "column": 1,
"message": "Line too long", "symbol": "line-too-long"},
])
violations = parse_linter_output("pylint", stdout, "", Path("/repo"))
assert len(violations) == 2
assert violations[0].severity == "error"
assert violations[0].message == "Missing docstring"
assert violations[1].severity == "warning"
assert violations[1].code == "line-too-long"
def test_parses_ruff_json(self):
stdout = json.dumps([
{"filename": "src/main.py", "location": {"row": 5, "column": 1},
"code": "E501", "message": "Line too long"},
])
violations = parse_linter_output("ruff", stdout, "", Path("/repo"))
assert len(violations) == 1
assert violations[0].file == "src/main.py"
assert violations[0].line == 5
assert violations[0].code == "E501"
def test_parses_eslint_json(self):
stdout = json.dumps([
{"fileName": "app.js", "range": {"start": {"line": 2, "column": 0}},
"message": "Unexpected console statement", "severity": 2, "ruleId": "no-console"},
])
violations = parse_linter_output("eslint", stdout, "", Path("/repo"))
assert len(violations) == 1
assert violations[0].severity == "error"
assert violations[0].code == "no-console"
def test_parses_shellcheck_json1(self):
stdout = json.dumps({
"issues": [
{"file": "script.sh", "line": 3, "column": 1,
"message": "Quote this to prevent word splitting", "level": "warning", "code": "SC2086"},
]
})
violations = parse_linter_output("shellcheck", stdout, "", Path("/repo"))
assert len(violations) == 1
assert violations[0].severity == "warning"
assert violations[0].code == "SC2086"
def test_parses_yamllint_parsable(self):
stdout = "config.yaml:3:1: [error] wrong document start (document-start)\n"
violations = parse_linter_output("yamllint", stdout, "", Path("/repo"))
assert len(violations) == 1
assert violations[0].file == "config.yaml"
assert violations[0].line == 3
assert violations[0].severity == "error"
assert violations[0].code == "document-start"
def test_returns_empty_on_invalid_json(self):
stdout = "Not valid JSON"
violations = parse_linter_output("pylint", stdout, "", Path("/repo"))
assert violations == []
def test_strips_leading_slash_from_paths(self):
stdout = json.dumps([{"type": "error", "module": "/repo/src/test.py",
"line": 1, "column": 1, "message": "test", "symbol": "T001"}])
violations = parse_linter_output("pylint", stdout, "", Path("/repo"))
assert violations[0].file == "src/test.py"
class TestLinterResult:
"""Test LinterResult and JSON serialization."""
def test_result_to_dict_roundtrip(self):
v = Violation(file="test.py", line=10, column=5, message="msg",
severity="error", linter="pylint", code="E001")
r = LinterResult(linter_name="pylint", language="python", violations=[v])
d = _result_to_dict(r)
assert d["linter"] == "pylint"
assert d["violations"][0]["file"] == "test.py"
assert d["violations"][0]["code"] == "E001"
class TestIntegration:
"""End-to-end integration tests with temporary repos."""
def test_linter_runner_accepts_repo_path(self, tmp_path: Path):
(tmp_path / "main.py").write_text("print('hello')")
(tmp_path / "bad.py").write_text("import unused_module\nx=1")
from linter_runner import detect_languages, run_linters_for_language
langs = detect_languages(tmp_path)
assert "python" in langs
result = run_linters_for_language("python", langs["python"][:1], tmp_path)
assert result.language == "python"
assert result.violations or result.error # either linter output or not-installed
def test_json_output_structure(self, tmp_path: Path):
(tmp_path / "script.py").write_text("print(1)")
from linter_runner import detect_languages, run_linters_for_language, _result_to_dict
langs = detect_languages(tmp_path)
if "python" not in langs:
pytest.skip("No Python files detected")
result = run_linters_for_language("python", langs["python"], tmp_path)
report = {
"repo": tmp_path.name,
"languages": {"python": _result_to_dict(result)},
"summary": {
"total_issues": len(result.violations),
"errors": sum(1 for v in result.violations if v.severity == "error"),
},
}
json.dumps(report) # should not raise
if __name__ == "__main__":
print("Run: pytest tests/test_linter_runner.py -v")