Hi everybody,

I have few metaphysical problem with RB :

— Why project claim a reachable SCM to be created ?
I don't understand very well what is the usage of SCM for RB and why it is 
mandatory.
If SCM is available, RB can fetch the full original file like today, but if 
not, the review can always be performed only with diff context lines.
Today, a project must have a reachable SCM to be created and it's SCM URL 
must be the same on each developper workspace to easily use post-review.

— Why review with unknow file or revision in the current RB SCM state is 
forbidden ?
Ditto, why I can't post diff with files or revisions unknow from RB SCM 
point of view ?
With centralized SCM (SVN, CVS…), this is not very annoying, because all 
devs are closed to « official » repo.
But with new DSCM like Git and Mercurial, each dev can have personnal push 
URL, pull commits and branches from others which are not yet merge onto the 
official repo, etc.


In my case, those 2 points lead to very complicated review process.

First, Mercurial SSH support is broken on RB, but because SCM URL is 
mandatory and must point to real existing repo, I have to use filesystem 
path instead of ssh:// (fortunately, RB and mercurial-server are on the 
same server).
But in this case, the push URL of my repo is not the one configured in RB, 
so post-review is a very mess when I want to submit a new review.
I have to put my hands into post-review's code to see it looks for an 
undocumented « reviewboard »-named push URL (I preferred to patch 
post-review to look into mercurial hgrc instead).

Secondly, I'm the leader of my team, using mercurial, and when my co-dev 
want submit reviews, they have not existing files or revisions from my RB 
SCM point of view, pulled from other repos and not yet merge onto my 
official repo.
Posting corresponding diff leads to RB #207 error and review is impossible…


I also have the case of very punctal review on a specific code problem, 
related to any existing project, with or without reachable SCM.
A « misc » project will be appropriate for this purpose, with manual raw 
diff submission, but today, it's impossible with RB.


IMO, RB it's a *very good* review tool, but it must be possible to create 
project without SCM and create review from not-scm-related diff.
I look into RB code but this SCM behaviour seems very deeply integrated.
Does this feature has already been addressed or is there any plan to 
integrate it ?

Cheers,

-- 
Want to help the Review Board project? Donate today at 
http://www.reviewboard.org/donate/
Happy user? Let us know at http://www.reviewboard.org/users/
-~----------~----~----~----~------~----~------~--~---
To unsubscribe from this group, send email to 
reviewboard+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/reviewboard?hl=en

Reply via email to