Re: Creating copies using "svn cp" preserving directory structure

Hi Ryan!

First of all, thanks a lot for yor quick and informative reply!

On Thu, 23 Jun 2005, Ryan Schmidt wrote:

> [...]
> On the other hand... since you're showing SQL files in these
> directories... perhaps the files in e.g. betra/1.0.2 are the SQL
> changes needed to get from version 1.0.1 to version 1.0.2 of "betra"?

Yes, that's what's meant. Sorry for not telling the entire story from the
very beginning. betra and las are in fact modules of a larger product having
their own version numbers (that is, those numbers are indepent of the
version number given to the entire product). So, what I want to achieve is
to create a dir structure representing the fact that e.g.

- the product is named "TEST1" (the tag name I used)
- it consists of
  - betra 1.0.2
  - las 2.8.7
  ... [additional modules and versions would follow]

the SQL scripts in these directories are in fact update scripts that perform
updates to an already existing database schema. It should also be noted that
not all modules may always require an update thus a tag subdir is supposed
to only contain the module and the version that actually require an update.

The same tag name (TEST1) would be used for the sources so that it's possible
to associate a certain revision of the sources with the revision of the SQL
scripts needed to perform the database update.

> If that's what you're doing, then I'd suggest that you do in fact tag
> the entire trunk directory.
> [...]

I'm afraid this won't work for my purposes since tagging the entire trunk
dir gives me all modules along with all versions below tags. That's not what
I need. I only need module/version combos of some modules, not all of them.

Does that mean I would have to create one subdir for each module, like this

betra-update
 |--> trunk
 |--> tags
       |--> 1.0.1
       |--> 1.0.2
       |...
 |--> branches

las-update
 |--> trunk
 |--> tags
       |--> 2.8.6
       |--> 2.8.7
 |--> branches

etc.

This would be sort of awkward since it requires one project per module just
for storing SQL update scripts.

Or would it be better to simply store the update scripts in the same
projects that contain the sources (by simply using an "update-scripts"
directory), like this

las
 |--> trunk
        |--> src
        |--> ooad
        |--> test
 |--> tags
 |--> branches
 |--> update-scripts

Any advice will be greatly appreciated! Thanks in advance!

Greetings,

        Holger

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Received on Thu Jun 23 16:19:29 2005