[nycbug-talk] svnup(1) - worthy of promotion to base?

[Isaac (.ike) Levy]

On Mar 11, 2013, at 10:17 AM, George Rosamond wrote:
> On 03/11/13 09:54, Isaac (.ike) Levy wrote:
>> Hi All,
>> 
>> From prior threads, the FreeBSD svnup(1) utility is alive, all
>> buttoned up, man page and all!
>> 
>> It's currently in ports, but it's not over- I believe it would be
>> EXTREMELY advantageous for this utility to make it's way into the
>> base system (in time for 10x?).  If anyone has any info (or sees any
>> commits), please shout back in this thread!
> 
> and backported into base all the way to 8.x would be nice too.

Agreed, but personally I'd rather see the near future sured up.

> 
>> 
>> -- If you believe svnup(1) in base is a good idea, and want to help,
>> I wanted to ask folks for some user testing?  Break it.  Read it's
>> short source, (1310 lines- one file).  Anything you can think of to
>> make it tight as it can be, (worthy of base).
>> 
>> -- My observations so far:
>> 
>> svnup(1) speed "feel"
>> 
>> Without opening up any cans of worms on SCM tools, I believe the 'git
>> effect' creates some unrealistic expectations for the utility.
>> (Git's impact is so huge on development, it's become so popular, it
>> affects/warps perception of other tools.)  git(1) has spectacular
>> indexing/hashing, it's implementation is really thoughtful compared
>> to svn(1).  So, remote fetching of deltas, on massive codebases, is
>> extremely fast. svn(1) itself- (and svnup(1)), is not quite as slick
>> in this area.  To this end, svnup(1) *feels* surprisingly "slow" when
>> fetching deltas.  Yet, for what it's doing, (comparing every file),
>> it's pretty darned fast.
> 
> So I don't "feel" it's slower than csup/cvsup.. but haven't tested.

Compared to csup/cvsup, it does indeed feel similar- (which is totally acceptable to me).

>> Because it appears to walk every file, I'm curious to know how much
>> impact (if any) svnup(1) has when many clients are simultaneously
>> hitting a single repo?  Does anyone have local subversion mirrors
>> setup who can test this?
>> 
>> I'm wondering if anyone can point out where the bottlenecks are, so
>> "best practices" can emerge quickly for setting up local svn repos?
> 
> Breaking would be interesting.  

Are there any NYC*BUG resources to try testing this?  (e.g. set up an SVN source mirror and hammer it like crazy)  If not, I may be able to get hold of a box to test from…  Not sure if it's worth the effort(?)

> So I didn't blow away the svn created
> /usr/ports when I ran it, and nothing broke, and of course the .svn
> directory is gone.
> 
> g

Yeah- the thing certainly behaves within the expectations of c[v]sup!  Really exciting, really simple.

Rocket-
.ike