World: r4wp

Maxim, we are gaining experience with the git-svn bridget now. I 
think we will use it for a couple of weeks. Than the next move will 
be R3 and the View engine.

Thanks, Robert. Could you briefly describe somewhere (in just a few 
sentences) the workflow you use when working on/debugging r3-gui? 
i.e is it possible to stay in the same R3 instance and rebuild (do 
you even need to be using the monolithic r3-gui script when debugging?), 
reload the script safely, etc.

Robert, why not just dump the svn and create your own git server?

Is it just do %loader.r3 after a change?

yes

or you can use the build.r3 do get on-file version

(note Ladislav's INCLUDE is needed https://github.com/saphirion/include)

why not just dump the svn
 - not trivial :-)

our whole workflows etc. are based on SVN, so we need to change our 
toolchains etc.

step-by-step

So to work on / debug R3-GUI (Cyphre please correct me, if I'm wrong):

;; Load INCLUDE (once)
>>  do https://raw.github.com/saphirion/include/master/include.r


;; (Re-)load R3-GUI (multiple times; assumes you are in the R3-GUI 
toplevel directory)
>> include %loader/loader.r3

Include tips and tricks:


I run r3 with include and loader, the include.mdp documentation describes 
how it can be done.

Yes, the most efficient way(in Windows) is to setup INCLUDE+R3-GUI 
loader to be executed everytime you click on *.r3 script. I'm using 
that and every change to the R3GUI sources is then automatically 
"propagated" once I run any R3GUI script again.

Yes, the strategy Cyphre describes is ideal for immediate testing 
of any changes made to R3-GUI sources.

Thanks guys - maybe these brief notes could be added to the readme 
on GitHub.

Adrian, check the new wroding of README, and, eventually, submit 
a request, please.