[Lazarus] other replies to "Development of other revision control"

[Graeme Geldenhuys]

On 13 April 2012 20:43, Dawson  wrote:
>>
> Sorry for the foolish sounding question, but who would make this kind of
> decision?

Normally the person that implements it - but this is not always a
guarantee either.

> feedback from the transactions as possible. For example as it is right
> now I am completely unable to get lazsvnpkg to do anything useful for


I had the same experience the last time I looked at lazsvnpkg. It was
very good at crashing though. And now that I switched to git I need
more fine grained commit control too. eg: I never commit whole files
because I often code for a hour or so making lots of changes, then
break my changes into lots of small atomic commits (committing a few
lines of code here and there in a unit, until all changes in that unit
are committed). Not even many GUI git frontends can do this, but git
command line client and the git-gui (included with git can) - and this
is vital to me and my workflow.

But I guess SCM + IDE integration is maybe meant for the simpler folk,
thus whole file commits will be okay in their case.


> me. There seems to be no mechanism for username / password challenging,

Username and passwords are so old-school. ;-) Can't SVN use a SSH or
GPG or DSA key pair to do the validation?




-- 
Regards,
  - Graeme -


_______________________________________________
fpGUI - a cross-platform Free Pascal GUI toolkit
http://fpgui.sourceforge.net