aboutsummaryrefslogtreecommitdiff
path: root/COPYING
diff options
context:
space:
mode:
authorStephen Finucane <stephen@that.guru>2020-04-10 12:08:32 +0100
committerStephen Finucane <stephen@that.guru>2020-04-26 13:56:38 +0100
commit9bb5caec4180f70f0d941580553994d7e36ca6f9 (patch)
treeebfbc1b97d50d1a4d26ae2a10bf9e72fafc6287d /COPYING
parent0910f9771113b7b6228b745ba9a74bd77f9b68ea (diff)
downloadpatchwork-9bb5caec4180f70f0d941580553994d7e36ca6f9.tar
patchwork-9bb5caec4180f70f0d941580553994d7e36ca6f9.tar.gz
migrations: Squash 0001 to 0040
Now that we're moving a new major version, we can squash the migrations we have to date. This coexists alongside the old migrations, which we can remove and replace entirely when we release 4.0, per the advice of the 'squashmigrations' tool. You should commit this migration but leave the old ones in place; the new migration will be used for new installs. Once you are sure all instances of the codebase have applied the migrations you squashed, you can delete them. The 'squashmigrations' tool can't parse the output of 'RunPython' blocks so the output of the tool was less optimized than it could be. As a result, we've manually modified this change to remove the 'RunPython' block and unnecessary 'AlterField' entries. This was done by removing all migrations and generating a new "initial" migration, which was then modified to mark all strings as byte strings (as they were when we were using Python 2 to generate these migrations) so that 0041 would apply cleanly. The main benefit of this change is that it significantly reduces the startup time for unit tests. Executed on my host, the run time for a single test goes from ~ 22 seconds to ~ 14 seconds. This is obviously reduced for additional tests. Signed-off-by: Stephen Finucane <stephen@that.guru>
Diffstat (limited to 'COPYING')
0 files changed, 0 insertions, 0 deletions