"""Redaction for job errors and log lines: no password, key or token ever reaches job_runs or a log.""" from __future__ import annotations import os import re MAX_ERRORS = 20 MAX_ERROR_CHARS = 500 # Environment variables whose values are secrets when set (MFilesKit reads MFILES_PASSWORD; VisualKit may read # VISUAL_PASSWORD). They are read only to strip them from text, never logged. SECRET_ENV = ("VISUAL_PASSWORD", "MFILES_PASSWORD", "PGPASSWORD", "LDAP_PASSWORD") SECRET_FILE_ENV = ("VISUAL_PASSWORD_FILE", "MFILES_PASSWORD_FILE") _PATTERNS = ( re.compile(r"(?i)\b(password|passwd|pwd|secret|token|api[_-]?key)\s*[=:]\s*(\"[^\"]*\"|'[^']*'|[^\s;,&]+)"), re.compile(r"(?i)\bauthorization\s*:\s*[^\r\n]*"), re.compile(r"(?i)\bx-authentication\s*:\s*[^\r\n]*"), re.compile(r"(?i)(\w+://[^:/\s@]+:)([^@\s]+)(@)"), ) def _secrets() -> list[str]: out = [] for name in SECRET_ENV: v = os.environ.get(name) if v and len(v) >= 4: out.append(v) for name in SECRET_FILE_ENV: path = os.environ.get(name) if not path: continue try: with open(path, encoding="utf-8") as f: v = f.read().rstrip("\r\n") except OSError: continue if v and len(v) >= 4: out.append(v) return out def redact(text: str) -> str: """Strip anything that looks like a credential, plus the VISUAL and M-Files passwords if set.""" s = str(text) for secret in _secrets(): s = s.replace(secret, "[redacted]") s = _PATTERNS[0].sub(lambda m: f"{m.group(1)}=[redacted]", s) s = _PATTERNS[1].sub("Authorization: [redacted]", s) s = _PATTERNS[2].sub("X-Authentication: [redacted]", s) s = _PATTERNS[3].sub(lambda m: f"{m.group(1)}[redacted]{m.group(3)}", s) return s def clip(errors: list[str]) -> list[str]: """At most MAX_ERRORS redacted strings of at most MAX_ERROR_CHARS each.""" out = [] for e in errors[:MAX_ERRORS]: r = redact(e) out.append(r if len(r) <= MAX_ERROR_CHARS else r[: MAX_ERROR_CHARS - 3] + "...") return out