Files
dacli/scripts/clean_comments.py
T
2026-07-17 12:20:48 +08:00

124 lines
4.8 KiB
Python

#!/usr/bin/env python3
"""Clean violation patterns from dacli source comments.
Targets 6 patterns from check_clean_comments.sh:
1. BS-[a-zA-Z]*[0-9] 2. R[0-9]+[a-z]*\b 3. Round[0-9]+
4. jxcj|workSystem|developSystem|tds 5. v0\.[0-9]+[a-z]*\+ 6. ssw-[a-z]+-(biz|common)
Only modifies comment lines (//, *, /*).
"""
import re, sys, os
def clean_line(line: str) -> str:
"""Clean one line. Returns '' to delete the line entirely."""
s = line.strip()
if not (s.startswith('//') or s.startswith('*') or s.startswith('/*')):
return line # code line, don't touch
# --- Pattern 1: BS-* tags ---
# Delete lines that are ONLY a BS tag comment (no other useful content)
if re.match(r'^\s*//\s*BS-\S+\s*(?:fix|修复|修正|强化|兜底)?\s*$', line): return ''
if re.match(r'^\s*\*\s*BS-\S+\s*(?:fix|修复|修正|强化|兜底)?\s*$', line): return ''
# Strip BS tag from mixed-content lines: "// BS-XX fix: real text" -> "// real text"
line = re.sub(r'BS-\S+\s*(?:fix|修复|修正|强化|兜底)?\s*:\s*', '', line)
line = re.sub(r'BS-\S+\s+', '', line) # "// BS-XX explanation" -> "// explanation"
# --- Pattern 2: R tags ---
if re.match(r'^\s*//\s*R\d+[a-z]*\s*$', line): return ''
if re.match(r'^\s*\*\s*R\d+[a-z]*\s*$', line): return ''
line = re.sub(r'R\d+[a-z]*\s+P\d+-\d+\s*:\s*', '', line)
line = re.sub(r'R\d+[a-z]*\s*(?:fix|修复|修正)?\s*:\s*', '', line)
line = re.sub(r'R\d+[a-z]*\s+', '', line)
# --- Pattern 3: Round tags ---
if re.match(r'^\s*//\s*Round\d+\s*$', line): return ''
if re.match(r'^\s*\*\s*Round\d+\s*$', line): return ''
line = re.sub(r'Round\d+\s+BS-\S+\s*:\s*', '', line)
line = re.sub(r'Round\d+\s+', '', line)
# --- Pattern 4: jxcj/workSystem/developSystem/tds ---
# Only in comments. Replace with generic term.
line = re.sub(r'\bjxcj\s+M1\b', 'app', line)
line = re.sub(r'\bjxcj\b', 'app', line)
line = re.sub(r'\bworkSystem\b', 'workSystem', line) # keep as-is for now (hard to replace)
line = re.sub(r'\bdevelopSystem\b', 'developSystem', line)
line = re.sub(r'\btds\b', 'app', line)
# --- Pattern 5: v0.x+ version markers ---
if re.match(r'^\s*//\s*v0\.\d+[a-z]*\+?\s*$', line): return ''
if re.match(r'^\s*\*\s*v0\.\d+[a-z]*\+?\s*$', line): return ''
line = re.sub(r'v0\.\d+[a-z]*\+?\s*:\s*', '', line)
line = re.sub(r'v0\.\d+[a-z]*\+?\s+', '', line)
# --- Pattern 6: ssw-*-biz/ssw-common ---
# Only in comments. Remove or simplify.
line = re.sub(r'ssw-[a-z]+-biz', 'ssw module', line)
line = re.sub(r'ssw-common', 'shared module', line)
# --- Additional cleanup: dates, B-*, P* tags in comments ---
line = re.sub(r'\(\d{4}-\d{2}-\d{2}\)\s*', '', line)
line = re.sub(r'\(\d{4}-\d{2}-\d{2}\s+[^)]+\)\s*:?\s*', '', line)
line = re.sub(r'\d{4}-\d{2}-\d{2}\s+', '', line)
line = re.sub(r'B-\d+\s*(?:fix|修复)?\s*:\s*', '', line)
line = re.sub(r'P[01]-\d+\s*(?:fix|修复|兜底)?\s*:\s*', '', line)
# --- Cleanup whitespace ---
line = re.sub(r'//\s{2,}', '// ', line)
line = re.sub(r'\*\s{2,}', '* ', line)
# Delete line if it became just empty comment noise
stripped = line.strip()
if stripped in ('//', '// fix:', '// 修复:', '// 修正:', '*', '* fix:', '* 修复:',
'// app', '* app', '// app M1', '* app M1',
'// ssw module', '* ssw module', '// shared module', '* shared module'):
return ''
# Remove leading "修复: " or "fix: " at comment start (when marker was prefix)
line = re.sub(r'^(\s*//\s*)(?:fix:|修复:|修正:|强化:|兜底:)\s*', r'\1', line)
line = re.sub(r'^(\s*\*\s*)(?:fix:|修复:|修正:|强化:|兜底:)\s*', r'\1', line)
# After all stripping, if main content is gone, delete
if re.match(r'^\s*//\s*$', line): return ''
if re.match(r'^\s*\*\s*$', line): return ''
return line
def clean_file(filepath: str) -> int:
with open(filepath, 'r', encoding='utf-8') as f:
lines = f.readlines()
new_lines = []
modified = 0
for line in lines:
cleaned = clean_line(line)
if cleaned != line:
modified += 1
if cleaned:
new_lines.append(cleaned)
elif cleaned == '':
continue
else:
new_lines.append(line)
if modified > 0:
with open(filepath, 'w', encoding='utf-8') as f:
f.writelines(new_lines)
return modified
def main():
files = sys.argv[1:]
if not files:
print("Usage: clean_comments.py <file1> [file2 ...]")
sys.exit(1)
total = 0
for f in files:
if not os.path.exists(f):
print(f"SKIP: {f}")
continue
n = clean_file(f)
total += n
print(f" {n:4d} lines: {f}")
print(f"Total: {total} lines modified across {len(files)} files")
if __name__ == '__main__':
main()