On Wed, Nov 16, 2011 at 1:05 AM, David Lutterkort <[email protected]> wrote:
> On Mon, 2011-11-14 at 13:05 +0000, Matthew Booth wrote: > > On 14/11/11 03:52, David Lutterkort wrote: > > > I committed the lens as-is. When I fix the bug described above, I'll > fix > > > up the lens to use the more readable regexp syntax. > > > > I'll be interested in that fix. > > I just pushed a fix for the libfa problem (commit f591ddde), and also > converted the Mdadm lens to use the shorter syntax (commit 7f92c1be) > > > I was actually trying to do something a > > bit more ambitious with this lens: write a lens which would parse *any* > > file, but which put invalid content into an #error node in the > > appropriate context. > > That would be very cool; it will be pretty hard on the typechecker, > though, but that shouldn't deter anybody. > I think the idea is really nice. One problem would be that currently, users expect to find /augeas//error nodes when the syntax is wrong. Adding invalid content to #error nodes would remove this, and users might not be aware that the file is wrong. This would be a problem for me for example, as I use /augeas//error to check for invalid syntax in augeas-validator. Raphaël
_______________________________________________ augeas-devel mailing list [email protected] https://www.redhat.com/mailman/listinfo/augeas-devel
