[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Using SVN for web applications made with PHP / MySQL?

From: Ryan Schmidt <subversion-2004_at_ryandesign.com>
Date: 2004-12-16 20:40:01 CET

Hello. My apologies if this has been addressed on the list before, but
I wasn't sure exactly what to search for, so I had trouble finding
anything relevant on Google or in the list archives.

I'm a programmer and system administrator for a web design company
specializing in large web-based applications with PHP and MySQL. We're
developing, for example, a tool that tracks sales and marketing
information for a large multinational client of ours. There are now 5
programmers working on this project, and although we're all in the same
small building, it's gotten to the point that we really need a version
control system to keep from stepping on each other's toes. But from
what I've read so far about such systems, it's unclear exactly how to
apply it to a web-based application, as opposed to a desktop
application. Some problems I think I've been conceptually able to
solve, but some I'm still having trouble with, for example:

* We're used to having a web URL that is "the development version" of
the project that we can look at. With a version control system, I'd
check out the project, and then I'd have my own private URL for it
where I could test the changes I'm making, but what happens when I
check it in? How does it get to the "development version" URL? Do we
have to have the equivalent of the "compile farm" which in our case
would just do a checkout of the most current version to the development
version htdocs folder every X minutes, or is there a better solution?
Is it for example possible to have the SVN server do something (execute
a shell script...) automatically on check-in, so that we could have the
copies made immediately instead of there being an X-minute delay?

* The application has a database behind it. How do I track changes to
that? If I need to make a change to the database that's incompatible
with existing code (like renaming a field) it would seem that I would
need to "check out" a copy of the database (the pristine copy) and then
a second copy that I could modify, so that later a diff could be done
between them. But I don't know of a tool to do diffs of a MySQL
database, and there's also the problem that the database in question is
almost half a gig in size and growing. Making even just one copy of
that will take a non-trivial amount of time, such that it would impede
someone's workflow to wait for such a copy to finish. We're considering
handling database updates the way we do now -- don't version control
them, and make potentially incompatible changes very quickly so nobody
notices. But since we're considering version control now for the PHP
files, I wanted to see if there's a solution for the database too.

I'd appreciate any insight you can offer, even pointers to existing web
resources on this topic if they exist. Thanks.

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Received on Thu Dec 16 20:43:41 2004

This is an archived mail posted to the Subversion Users mailing list.

This site is subject to the Apache Privacy Policy and the Apache Public Forum Archive Policy.