On Sun, 17 Jan 2010 16:54:30 +0100% Enrico Tröger enrico.troeger@uvena.de wrote:
On Tue, 12 Jan 2010 22:19:33 +0300, Eugene wrote:
<...>
Thank you, Dominic and Enrico, for your opinions. I have a little patch to fix this issue (attached).
As an alternative, I have a completely rewritten `load_startup_files' function. I wrote it to understand all the conditions which were in the original implementation. My function is *MUCH* more verbose, but I think it can be easier to understand and modify. I *DID NOT TEST* it, so consider it to be a sketch. Enrico, if you like this implementation, I can test and update it, so it can be safely committed.
Cool!
I need to know which way (the little patch or the rewritten function) is accepted as my future questions may require further modifications of `load_startup_files'.
I do like your rewritten function more as it is more clear and more structured. I didn't test it as well yet but I think we should use it. If you don't have more changes it for now, I'll test and commit it then.Thanks!!
No, I don't have more changes for now. I tested the function a bit today, it works as expected (like the version in current trunk except the little bug with -s option fixed). The only change I made is that I merged two comment lines into one :) So you can use either the function code from my previous post or the attached patch.
Before you commit, I'd like to know how the changes would go to "sm" branch. It is not so important for me (as I use a local Git repository anyway), but it could be important for the one who will commit my patches (e.g., probably Nick). Will it be done using SVN merge, or will there be separate commits (same patches applied to different branches)?
Best regards, Eugene.