Re: Why isn't it part of mutt "proper"

[Thomas Glanzmann <sithglan@xxxxxxxxxxxxxxxxxxxx>]

Dear <hcache user>,

[ This was a message I just sent to a hcache user to the question:
                 'Why isn't it part of mutt "proper"?'
]

[ From my website: http://wwwcip.informatik.uni-erlangen.de/~sithglan/mutt/

        I asked Thomas Roessler direct and via mutt-dev often to include
        the header cache into upstream. The only feedback I got so far
        from him regarding the header cache is this posting[1]. I
        disproved his reason for not including the header cache into
        mutt in some of the following postings in this thread[2]. But I
        did not receive a statement from him so far. So it's going to
        take some while until it gets into upstream. The crowd who want
        this patch getting upstream gets constantly bigger. So it's a
        matter of time. At htis point some distributors have added the
        maildir header cache into there packages which is a first step
        into the right direction.

[1] http://does-not-exist.org/mail-archives/mutt-dev/msg01613.html
[2] http://does-not-exist.org/mail-archives/mutt-dev/thrd51.html#01613

]

Thomas Roesslers point was:

        Don't add locking mechanisms to Maildir, it was designed to work
        without locking mechanisms via NFS.

My answer was:

        I don't add locking mechanisms to Maildir, but to the database
        where the cached header information are stored in order to
        prevent corruption due to multiple writers.

        bottom line: Maildir stays NFS safe *without locking* it.

        I also mentioned that the locking mechanims for the header cache
        database are NFS safe and *nonblocking*. Read: It tries to open
        the cache rw, if that fails ro, if that fails not. *Without* any
        blocking.

        It seems that this isn't enough for him.

Honestly,
        Thomas