David Glasser wrote:
> On Sun, Oct 18, 2009 at 1:26 PM, Jan Horak <horak.honza_at_gmail.com> wrote:
>> Hi everybody,
>> I'm writing first time to Subversion Mailing List with the aim of
>> getting involved in the project. I will shortly present my intention:
>> I'm studying at Brno University of Technology (Czech Republic, Europe)
>> and I'm just starting to work on my dimploma thesis. The goal of the
>> thesis is to design and implement an SQL backend for the Subversion
>> (specificly MySQL).
>> As I've read some messages in the mailing list archive, there had been
>> some attempts to implement this extension and I belive there are enough
>> reasons to spend a time to do that. But as far as I know, nobody has got
>> any satisfactory results yet.
>> If there is anybody, who has considered with this topic, I would very
>> appreciate every relevant information - for example some existing
>> materials, analyses, desings, parts of code, .. simply anything.
> I guess the first question is: why?
> Not that this is a bad idea or anything. But what goal are you trying
> to accomplish by implementing a new Subversion backend? Is it just an
> exercise in "to see if I can", or will it offer benefits over the
> current backends? (eg performance, maintainability, interesting new
> queries, etc.) If nothing else, knowing what problem you're trying to
> solve will help guide your schema design.
Well, the first impuls to implement new backend didn't come up from me,
but from my diploma thesis tutor, so I can say it is kind of exercise
Detailed analyse of advantages and disadvantages of SQL backend
maintainability, etc.) in comparison to BDB and FSFS is one part of my next
For the answer to your question "why?" I have currently only some notes
the mailing list archive, that suggest some noticeable features of SQL:
Received on 2009-10-20 16:51:01 CEST