- Move all per-module security audit reports to docs/security/
- Move code review logs to docs/reviews/, analysis reports to docs/reports/
- Add docs/INDEX.md as single entry point for all project documentation
- Add documentation management conventions section to README.md
- Clean up empty per-module docs/ directories
Follow the same convention as the static module: remove local config.ini
fallback and --config/-i command-line argument entirely. All configuration
is now managed exclusively through confs service.
- Delete gateway/example.ini (moved to confs/configs/gateway.ini).
- Remove -i/--input option from boost::program_options; help (-h) and
version (-v) are the only CLI flags remaining.
- Simplify main.cpp config loading: remove local-file branch, always fetch
gateway.ini from confs with retry_on_refused=true waiting for confs ready.
- Update _run.sh: no longer passes any arguments; config comes from confs.
- Update README: remove example.ini references, -i flag docs, local file
mode description, and example.ini from directory structure tree.
- Clarify changelog entry for v0.7.0.
Migrate gateway from local -i ini file to the unified confs configuration
service, following the same two-layer pattern as Python services (dnfs/db/static):
- Add confs/configs/gateway.ini with all existing settings plus new
[Downstream] section for db/captcha/smtp host:port (previously hardcoded).
- Conf.hpp: add fromString() to parse ini from text (confs response),
env()/env_int() static helpers for environment variables, public default
constructor, and empty-filename guard in file constructor.
- DbC.hpp: add init(host, port) static method and static _host/_port members,
replacing hardcoded "127.0.0.1":8005. Defaults preserved for backward compat.
- main.cpp:
* Rewrite config loading: bootstrap CONFS_HOST/CONFS_PORT from env
(default 127.0.0.1:8010), fetch gateway.ini via pymsg from confs
(retry_on_refused=true), parse with ConfIni::fromString().
* -i flag is now optional; when provided, reads local ini (dev/debug mode).
* Add downstream host/port globals (_DB_HOST/_DB_PORT, _CAPTCHA_HOST/_CAPTCHA_PORT,
_SMTP_HOST/_SMTP_PORT, _CONFS_HOST/_CONFS_PORT), populated from [Downstream].
* Environment variable overrides (highest priority): LISTEN_ADDR/LISTEN_PORT,
DB_HOST/DB_PORT, CAPTCHA_HOST/CAPTCHA_PORT, SMTP_HOST/SMTP_PORT.
* Config integrity validation: required fields (Server.address/port,
Proxy.default, Downstream.db_host/db_port) cause startup failure if missing.
* Initialize DbC::init() and logs().set_endpoint() after config load.
* Replace all hardcoded "127.0.0.1":8005/8008/8009 in captcha/smtp/db calls
with configurable globals.
* Add downstream service info to startup log output.
* Fix: -v/--version flag now actually prints version and exits.
* Bump version to 0.7.0.
- _run.sh: remove -i example.ini default; now starts without args (confs mode),
passes "$@" through for optional -i override.
- Update README with new config loading docs, Downstream config section,
updated directory structure, and v0.7.0 changelog.
- Add MsgConnectionRefusedException to distinguish ECONNREFUSED from other
connection errors; openConnection() now throws it when connect() returns
ECONNREFUSED.
- MsgClient::call() adds retry_on_refused/retry_interval parameters (defaults
false/1.0s), matching Python pymsg.MsgClient.call(retry_on_refused=True).
- Retry loop: only retries on ECONNREFUSED; other exceptions (timeout,
connection reset, business errors) propagate immediately. Throttled logging:
first failure prints immediately, subsequent logs every 5 seconds.
- DbC::call_msg() enables retry_on_refused=true (db is a hard dependency;
gateway blocks waiting for db on startup instead of crashing).
- LogC::_send() explicitly passes retry_on_refused=false (fail-safe; never
block the background log thread when db is unavailable).
- captcha/smtp auxiliary services keep default retry_on_refused=false
(fail-fast; blocking in request threads would exhaust the thread pool).
- Update README with pymsg protocol documentation, downstream retry strategy
table, and v0.6.1 changelog entry.
- POST /refresh_token: accept current Bearer token (or Cookie fallback),
verify RSA signature, check grace window (iat + keep_time * multiplier)
and refresh count limit, then issue a new token.
- Token payload extended with iat, refresh_count, max_refresh fields.
- iat preserves original issue time; created_at updates on each refresh.
- Rate limited: 5 req / 5 min / IP via RateLimiter.
- Config: Token.refresh_grace_multiplier (default 2), Token.max_refresh (default 10).
- Backward compatible: old tokens without new fields still pass proxy checks.
qing
shufeng2012