Louis Coilliot wrote:
Hello,
I've (mostly) fixed it.
The patch is here:
http://func.pastebin.com/m2bd4d
Unified diff are usually a little easier to deal with, if possible
("diff -u" or just "git diff")
The modified/improved test script is here:
http://func.pastebin.com/m25900459
Modifications on the module:
- fix for trailing /
- new function=backup
(because in func+augeas we can't use /augeas/save; each action is a new
augeas instance)
- new function=getenv
Modifications on the test script:
- the tests are now 'chrooted' with AUGEAS_ROOT
(but you need to put something like export AUGEAS_ROOT='/tmp'
in /etc/init.d/funcd, I don't know yet of a better way to pass this
variable to func)
- new tests, in particular for extended paths
Not sure the best way to test it. Maybe the test scripts could just
install some example config
files to /tmp and not have to set the an AUGEAS_ROOT at all?
Todo:
- fix for some extended path failures
- find a way to use /augeas/save in a persistent way
- ability to work on any single conf. file when it will be available in
Augeas
I merged and pushed the latest versions. I tested against latest
upstream augeas and python-augeas, so
don't know how well it works against say, the F10 augeas.
Whats you'r thoughts on the module name? we could sub module it and make
it config.augeas.*. or just
augeas.*. Not sure myself, just augeus.* is short and namespace is
probably fine, but it is a little obscure for
someone looking for stuff to change config files if they didn't already
know what augeas was. Though, they
kind of need to know what augeas to really make much sense of the
module, so maybe thats not that bad.
Adrian
_______________________________________________
Func-list mailing list
Func-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/func-list
