tech-repository archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: Proposed conversion strategy



Hi,

On Mon, Oct 27, 2014 at 03:47:54PM -0400, Eric S. Raymond wrote:
> Jeff Rizzo <riz%tastylime.net@localhost>: That you're seeing flakiness now suggests
> the tools are badly maintained.  Which is unfortunare news, but not
> irretrievable.  I have infrastructure inside reposurgeon to build exporters
> (fast-import-stream -> target-system) based only on access to the target
> system's CLI.  If I have to write one for Mercurial I will.
> 
> > Thus far, the fastest and most reliable tool we have is for conversion
> > from CVS to fossil (written by J?rg Sonnenberger, who maintains the github
> > NetBSD mirror) - unfortunately, fossil itself has issues working with a
> > dataset the size of the NetBSD repo, and is not currently a good
> > end-choice.
> 
> Agreed.  I understand why exactly - mentioned it in previous email here.  I
> hope the Fossil fans in NetBSD-land are realistic about this.

I'm realistic on this; i know fossil needs some work and if i had time i'd
volunteer for it. Considering fossil's mission-statement i always wondered why
their main developers didn't take it hands-on directly since its a good `huge
repo' check :)

As for other repo types, i'm curious about Hg but git is not really an option
IMHO; its far too easy to screw up a git repo. I once thought i had it too
with fossil i have to admit but it turned out to be fixable and has never
repeated itself; and i am not that good a power-user of fossil ;)

Another big NONO for git is its half-baken support for sub-repos. Sure one can
get links to other repos in a tree but it fails to have a decent recursive
update and recursive diff .... not very usefull thus.

Cheers,
Reinoud

Attachment: pgpPW5qlEN431.pgp
Description: PGP signature



Home | Main Index | Thread Index | Old Index