On Tue, 10 Feb 2015 20:32:09 +0100, John Found <johnfo...@asm32.info>
wrote:
I was talking about the CLI of fossil, not the web interface.
It doesn't matter. It is even more simple, just detect the first hex
number, enclosed in square brackets and you will be fine, notice,
without assuming any length at all. If you want to use finfo with -b
option, simply scan to the first not-hex-number.
see somewhere further up in this thread: the problem was/is not to
_identify_
the sha1 strings but to map some "place holder" of full sha1 to incremental
check in numbers in a "globally" correct way (i.e. across fossil
sub-commands).
the pitfall was to assume that the sha1 strings displayed by `timeline' are
already such place holders (they are not since `finfo -b' does it
differently).
Here I have to add one important note: The output of fossil does not
seems to be designed for machine processing. It is for human processing.
So, if you need to process it by program, you have to make some effort
to make this program flexible enough.
sure. and such scripts necessarily remain fragile (e.g. are very
susceptible to formatting
changes of the `fossil' output). my point was that the different
conventions
used by `timeline' and `finfo -b' regarding display of sha1 are simply not
desirable
in my view (and that the variable lengths sha1 strings might cause more
problems
than they solve in the first place...)
--
Using Opera's revolutionary email client: http://www.opera.com/mail/
_______________________________________________
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
