A new model for seeding the database that makes it easier to seed many entities. new_seed_many is a function that seeds in a different way. Instead of using redirects, it figures out what the high-water mark of seeding is and proceeds from there. This is obviously a half-way step. I've talked about it with Sverre; I think I've mentioned it to others. If we like it, I volunteer to convert everything to this model (I think there will be savings and simplification when we're all the way). If we don't like it, I'll back out this code. But I'll also be sad, because this makes it a lot easier to add many entities (which I care about because I'm trying to fix list view for that many entities). Patch by: Dan Bentley

     1

Third-party code that is *not* part of the application image (the Google App

     2

Engine SDK in trunk/thirdparty/google_appengine, for example) belongs in

     3

trunk/thirdparty.  For third-party code that *is* part of the application

     4

image, see the notes in trunk/app/README.

     5

     6

== THIRD PARTY LICENSES ==

     7

     8

Third-party code must also include a license text file, named LICENSE or

     9

COPYING, in the package subdirectory.  If the third-party package subdirectory

    10

is maintained automatically with a tool such as svn_load_dirs.pl (which

    11

removes files, such as license text files, that were added later to the

    12

"target" directory), place the license text file in the trunk/app directory

    13

and add the package name as a suffix.

    14