what exactly is wrong with the existing fixfile / fixfilediff?
​
nothing,

Its just nicer to be able to commit a patch without having to specify a
path to a patch file.  Simple is better, b/c I think people would want to
use commands prefixed with sandbox_ when they're in the debug environment.

> Now it takes an optional argument to spare all that typing to simply:
> fixfilediff hotfix

This solution works too, but it be better if we list current patches so the
user won't accidentally overwrite an existing patch.  it happens.



> and the in-tree rebuild may be fragile with complex packages, target
overlays and such, …?

I understand, but most packages are simple enough to rebuild within a
sandbox, on the fly.
Having just a function call to rebuild is worth space.  By the way, these
sandbox functions are only defined when in the sandbox environment, it
would not populate the SDE.

In the past, i had to patch aufs2 manually to linux26 within a sandbox.
 The patch was complicated that after everything is done, I can not just
use fixfilediff to generate a final patch file.  The only way to create a
patch after a-lot of work like that is to re-extract the original package,
and perform a directory diff from original source to the new source
directory.  Yet, the whole process can be automated if we were to create
another sandbox function, ie:  sandbox_create_full_patch.  IMHO, There are
more sandbox functions that we can create to make development/maintain
tasks to be a whole lot simpler.
----------------------------------------------------------- 
If you wish to unsubscribe from this mailing, send mail to
[email protected] with a subject of: unsubscribe t2

Reply via email to