| Re: Coyote Library Updates [message #77196 is a reply to message #62777] |
Wed, 10 August 2011 04:41   |
Paul Van Delst[1]
Messages: 1157 Registered: April 2002
|
Senior Member |
|
|
Michael Galloy wrote:
> On 8/9/11 2:43 PM, David Fanning wrote:
>> I've often wondered why I have the zip file under version
>> control at all. Do you think it is necessary? My problem is
>> I need to make a zip file, and it has to have the same name
>> all the time so my web pages can point to it. How would you
>> handle this for people with no access to the SVN repository?
>
> I don't think that the zip file should be under version control, BUT
> there should be a SCRIPT (or some other automated, repeatable, reliable
> method) for generating the zip file. For some of the things I distribute
> this way, the script generates the documentation, builds the zip file,
> and uploads it to the right location on my server. I can type one
> command to update the server to the current version.
Seconded.
We have a sort-of similar problem for versioning testcases of the various forecast systems. The input files are either
ginormous, or there are lots and lots and lots of little ones (either case I have observed giving our version control
setup slight indigestion...I think due to network latencies, made worse by IT security restrictions that keep the server
response time cutoff very short). My proposed solution to users is to version the script that retrieves the necessary
files from our tape backup system. I think it probably takes less time to retrieve large (or lots of) files from the
tape silos, than it would to check them out from our VCS - network throughput being what it is.
I use a script to generate the release tag that I distribute (our development directory structure is vertical, whereas
our release directory structure is flat). My next addition to that script is to allow for that tag to be exported and a
tarball built -- including a text file containing the "svn info" output for the belts-n-braces effect :o)
cheers,
paulv
|
|
|
|