Re: [bug-recutils] GSoC: Ideas for Recutils

[Jose E. Marchesi]

    >     The ideas page mentions determining if the index is up to date, I 
don't
    >     see other practical solutions than using filesystem metadata of the
    >     database file (checksumming the file contents should be much slower 
than
    >     doing a simple query using a tree index).
    >
    > We could have a "checksum" comment at the end of the rec file, which
    > would be generated by recfix when creating the index file.  The problem
    > with this approach is that the creation of the index wont be completely
    > decoupled from the recfile itself, but that may not be really a problem.
    
    Another problem is that users editing recfiles with a text editor might
    forget to change the comment, leading to incorrect query results (while
    other solutions would give correct results slowly).

Hm, yes.  Using hashes in comments is not a good idea.

    Python and Mercurial uses modification timestamps to avoid reading
    files (modules to compile into cached bytecode or versioned files to
    diff), I haven't observed any problems with reliability of this
    solution.

What about using a heuristic based on several properties of the file
such as its size, last modification date, etc?  That could provide
a rate of success big enough to be a practical solution.

-- 
Jose E. Marchesi         http://www.jemarch.net
GNU Project              http://www.gnu.org