comparison mercurial/keepalive.py @ 49542:8251f7cc787d stable

keepalive: ensure `close_all()` actually closes all cached connections

While debugging why LFS blob downloads are getting corrupted with workers, I
noticed that prior to spinning up the workers, the ConnectionManager has 2
connections to the server and calling `KeepAliveHandler.close_all()` left one
behind.  The reason is the value component of `self._cm.get_all().items()` is a
list, and `self._cm.remove()` modifies said list while the caller is iterating
over it.  Now `get_all()` is a deep copy of both the dict and lists in all
cases.

49540:bc2ecf08ae04

    def get_all(self, host=None):
        if host:
            return list(self._hostmap[host])
        else:
            return dict(self._hostmap)


class KeepAliveHandler:
    def __init__(self, timeout=None):
        self._cm = ConnectionManager()

49542:8251f7cc787d

    def get_all(self, host=None):
        if host:
            return list(self._hostmap[host])
        else:
            return dict(
                {h: list(conns) for (h, conns) in self._hostmap.items()}
            )


class KeepAliveHandler:
    def __init__(self, timeout=None):
        self._cm = ConnectionManager()