Using Sequel's support on DataObjects, and hacking DataObjects::Connection to make sure it is a singleton, they both play very nice toguether. I've uploaded a gist with the details: http://gist.github.com/644328
Regards, [] On Sep 20, 12:47 pm, Daniel Ribeiro <[email protected]> wrote: > Great points David. Your post makes this hack seems way less fun, and > yet, way more productive. > > About automigrate: Yes. We still use in memory databases for 140 of > the tests (more than 90%). I'll refrain from speculating how much less > time it would take without testing. The test setup gets a bit more > compicated and that would really need to pay off to make it. > > About the ramdisk: we don't touch files anywhere else, as good unit > tests do. But, the trick should really go a mile for NOSQL databases, > should > we need them (don't spect to be soon, but we rely on data, not > expectations). > > Regards, > [] > On Sep 20, 12:37 am, David Masover <[email protected]> wrote: > > > On Sunday, September 19, 2010 08:52:18 pm Daniel Ribeiro wrote: > > > > Yes, this is the ultimate hack. It is quite fragile, and setup > > > intensive, as all team members must have a ramdisk enabled, and have > > > the same path. > > > While true, it's also not as bad as you'd think. If you're running a modern > > Linux, it's as simple as: > > > sudo mount tmpfs /wherever/i/want/a/ramdisk > > > ...and done! You could easily put that in a script. > > > I have a tmpfs mounted somewhere convenient all the time, because I find it > > generally useful. > > > > Or, what is worse, configurable ones (breaking our > > > "don't make me think" test rule). > > > Symlinks. > > > > Hacking sequel seems a lot less > > > work. > > > The above was a one-liner. > > > Now, if you're on Windows or OS X, I don't know how you'd do that. It also > > may > > violate other rules, like "Don't make me run things as root". I just don't > > see > > it being harder than hacking sequel, certainly not if you're on a Linux. > > > > And it might pay off, as truncating all databases is far more > > > performatic than automigrate! (for clearing up tests). > > > Have you run automigrate on anything not disk-backed? > > > By all means, don't let me stop you from doing this a cleaner way, if you > > can. > > I just like the one-liner, dirt-simple, generic solution. -- You received this message because you are subscribed to the Google Groups "DataMapper" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/datamapper?hl=en.
