setup: bundle setuptools-0.6c12dev (our own toothpick of setuptools) this version completes my patch to fix http://bugs.python.org/setuptools/issue54 , which is necessary for tahoe to build with --prefix=support without doing a lot of PYTHONPATH gymnastics around the call to setup.py
roll back the upgrade of the bundled setuptools_trial from 0.5 to 0.5.1
Upgrading setuptools trial in that way leads to a very tricky problem in setuptools's handling of recursive installation of the build-time dependencies of build-time dependencies...
rolling back:
Tue Feb 3 22:17:18 MST 2009 zooko@zooko.com
* setup: bundle new setuptools_trial-0.5.1
A ./misc/dependencies/setuptools_trial-0.5.1.tar
Tue Feb 3 22:17:32 MST 2009 zooko@zooko.com
* setup: remove old bundled setuptools_trial
setup: remove a couple of horrible work-arounds in setup.py now that we rely on our own toothpick of setuptools which fixes those issues
also specify that we need our latest revision (11dev) of our toothpick of setuptools
also *always* setup_require setuptools_darcs at module import time. Formerly we added setup_require setuptools_darcs only if the PKG-INFO file were not already created. There is some weird, irreproducible bug to do with setuptool_darcs, and I guess that the fact that whether it is required or not depends on that autogenerated file might have something to do with it. Anyway, this is simpler.
setup: bundle setuptools-0.6c11dev (our own toothpick of setuptools)
Hopefully this one fixes the issue with easy_install not searching the sys.path for packages that were requested to be installed, (setuptools #17), thus allowing us to do away with the "--site-dirs=" kludges, which are currently breaking some of our buildbots.
Brian Warner [Sat, 31 Jan 2009 02:32:05 +0000 (19:32 -0700)]
web/directory: add a link from readwrite directories to a read-only version, and fix the 'SI=xxx' header to actually use the storage index, not the writekey
setup: make sure you use darcsver whenever you are going to run trial
This fixes the bug Brian had where he ran "python ./setup.py trial" and the allmydata-tahoe version number came out as 0.0.0.
Brian Warner [Fri, 30 Jan 2009 11:25:36 +0000 (04:25 -0700)]
Makefile: use 'setup.py test' for test/quicktest targets (instead of
'setup.py trial'). 'setup.py trial' clobbers the tahoe .egg's PKG-INFO
"Version:" field (resetting it to 0.0.0), possibly because it isn't invoking
the darcsver subcommand that 'setup.py test' does before it runs the 'trial'
subcommand.
This slows down quicktest by another couple of seconds (see #591) and adds
more noise to its output, but without this change, 'make test' and 'make
quicktest' fail on test_runner (which spawns bin/tahoe as a subprocess, and
with a mangled Version:, the setuptools-based entry point script refuses to
recognize our source tree as a usable version of Tahoe).
setup: subclass setuptools.Command instead of distutils Command
There's almost no difference between them, but let's be consistent try to use the setuptools API as it was intended.
setup: always create a support dir and populate it with a site-packages and add same to the PYTHONPATH, just in case someone is going to do "build", "develop", or "test" or something else which triggers a build
I think there must be a much better solution for this -- probably to fix setuptools #54 and ship our own fork of setuptools and rely on it.
setup: if any of "build", "develop", or "test" appear in the sys.argv then that means we'll be doing a develop, so add the workarounds for setuptools #17 in any case
I think there must be a much better solution for this -- probably to fix setuptools #17 and ship our own fork of setuptools and rely on it.
setup: a new improved way to create tahoe executables
Create the 'tahoe-script.py' file under the 'bin' directory. The 'tahoe-script.py' file is exactly the same as the 'tahoe-script.template' script except that the shebang line is rewritten to use our sys.executable for the interpreter. On Windows, create a tahoe.exe will execute it. On non-Windows, make a symlink to it from 'tahoe'. The tahoe.exe will be copied from the setuptools egg's cli.exe and this will work from a zip-safe and non-zip-safe setuptools egg.
storage: enable the test of a share larger than 2 GiB; this will probably be too expensive on Mac OS X, but I think it won't be on all the other platforms ; I will probably set it to SkipTest if the sys.platform is Mac after seeing the results of this buildbot run
setup: fix test_runner to assert that lines which are output to stderr must end with a punctuation mark (apparently re.search("x$", "x\r\n") does not match. :-()
setup: hack the sys.argv to set poll reactor if "python ./setup.py test" in addition to if "python ./setup.py trial"; remove another hack which has been moved setup.cfg; remove setup_requires Twisted since now we actually setup_require setuptools_trial and it install_requires Twisted.
setup: sys.exit() raises a SystemExit exception on Python 2.4 on Solaris -- fix it so that bin/tahoe doesn't interpret this as its brother having been non-executable
setup: find a "bin/tahoe" executable to test based on allmydata.__file__ instead of based on the CWD
This means that the tests still work if you are executing them from a CWD other than the src dir -- *if* the "bin/tahoe" is found at os.path.dirname(os.path.dirname(os.path.dirname(allmydata.__file__))).
If no file is found at that location, then just skip the tests of executing the "tahoe" executable, because we don't want to accidentally run those tests against an executable from a different version of tahoe.
setup: use "trial" executable instead of the setuptools_trial plugin for "make quicktest"
This is to fix #591 ("make quicktest" could be quicker and less noisy). This means that "make quicktest" won't work unless the user has manually installed Twisted already such that the "trial" executable is on their PATH and the Twisted package is on their PYTHONPATH. This bypasses the behavior of setuptools_trial which builds and checks package dependencies before executing the tests.
setup: change test_runner to invoke "bin/tahoe" in a subprocess instead of executing runner.runner()
This is necessary because loading allmydata code now depends on PYTHONPATH manipulation which is done in the "bin/tahoe" script. Unfortunately it makes test_runner slower since it launches and waits for many subprocesses.
rollback the feature of making "ambient upload authority" configurable
This reverses some, but not all, of the changes that were committed in the following set of patches.
rolling back:
Sun Jan 18 09:54:30 MST 2009 toby.murray
* add 'web.ambient_upload_authority' as a paramater to tahoe.cfg
M ./src/allmydata/client.py -1 +3
M ./src/allmydata/test/common.py -7 +9
A ./src/allmydata/test/test_ambient_upload_authority.py
M ./src/allmydata/web/root.py +12
M ./src/allmydata/webish.py -1 +4
Sun Jan 18 09:56:08 MST 2009 zooko@zooko.com
* trivial: whitespace
I ran emacs's "M-x whitespace-cleanup" on the files that Toby's recent patch had touched that had trailing whitespace on some lines.
M ./src/allmydata/test/test_ambient_upload_authority.py -9 +8
M ./src/allmydata/web/root.py -2 +1
M ./src/allmydata/webish.py -2 +1
Mon Jan 19 14:16:19 MST 2009 zooko@zooko.com
* trivial: remove unused import noticed by pyflakes
M ./src/allmydata/test/test_ambient_upload_authority.py -1
Mon Jan 19 21:38:35 MST 2009 toby.murray
* doc: describe web.ambient_upload_authority
M ./docs/configuration.txt +14
M ./docs/frontends/webapi.txt +11
Mon Jan 19 21:38:57 MST 2009 zooko@zooko.com
* doc: add Toby Murray to the CREDITS
setup: require darcsver always, and not just when we see the string "darcsver" in sys.argv, because the new aliases hack means that the string might not appear in sys.argv
setup: use setup.cfg aliases to map "setup.py test" to "setup.py trial" and "setup.py build" to "setup.py darcsver --count-all-patches build_tahoe"
Thanks to dpeterson for the suggestion.