openstack/swift - swift - OpenDev: Free Software Needs Free Tools

This patch ports the symlink middleware to py3.
The middleware itself seems to be mostly fine and most changes
are in the symlink unit tests.
Change-Id: I973c2e1bb8969cf6bffece8ce68881c393efbaef

Change-Id: Ie73f9df460f9014014168d6e60767e66b24d30b0

This adds wsgi_to_str(self.path_info) everywhere we forgot it,
not only in the slo module itself.
Dropping the body=''.join(body) after call_slo() is obvious:
the latter only returns strings of bytes, not lists of such.
Change-Id: I6b4d87e4cda4945bc128dbc9c1edd39e736a59d2

Change-Id: Ib5aeda94fe9b165f54ad7412d68d71b62e427a40

Related-Change: I3603247e0c36299de2107aa8e494a3f87647696f
Change-Id: Iaf14c1ab3e236e6774492ed711fccef066a16fca

This finishes the proxy unit tests.
Change-Id: I2452e4e4dabef071bfbf376d03ec63211f7c2906

Change-Id: I8287db75b4f19581203360c646e72f64fe45f170

Note that you need a pretty recent eventlet to pick up
https://github.com/eventlet/eventlet/commit/f0bc79e
Change-Id: I6b006b972e7431c406039f4e0f6890a8f74a4432

Change-Id: I378705a8d9ad51dbfa85287d44e6313aa7680c23

Change-Id: Iadb3bed811907740f85bd9277ba09a4ca17fc025

Drive-by: When passing a list or tuple to swob.Response as an app_iter,
check that it's full of byte strings.
Change-Id: Ifc35aacb2e45004f74c871f08ff3c52bc57c1463

Thorough replacement of git.openstack.org URLs with their opendev.org
counterparts.
Change-Id: I5e77307de6a484fd69b2a5863423ceb357e8601f

Change-Id: I7236ddea0acde93d0789ad8affa76df0097a86aa

Looking at https://governance.openstack.org/tc/reference/runtimes/train.html
OpenStack is no longer targetting Python 3.5, so it doesn't really make sense
to gate on it. Move to Python 3.7, just to make it so we can go longer before
having to do this shuffle again.
Presumably, once we actually declare support for Python 3.6, we should
make that job voting, too? We've had enough gate flaky-ness lately that
I'm not keen on it.
Change-Id: I9c9e5df76a6592ab9c512b7c1ba807f9a0053059

This patch is porting the staticweb middleware to py3.
Change-Id: I5d9a13baecedd13d2b7a8ae3dd639eaff0894441

Change-Id: Ibbff151703395d517452c3d6927652506a9eab66

Apparently some versions of py3 will drop symlinks in the venv.
Change-Id: I77c52b6ad776ab75c4f9713290cc42a1fcb6e0f7

Change-Id: Ib853f79bc062ec550627342d0fc5b2b92029f25c

...and fix this one warning-treated-as-error that crept in since
https://github.com/openstack/swift/commit/d185b60
Change-Id: Id46ee3ab23e2703170191528427aaa2788aba1ee

Drive-by: run api-ref tox env under py3.
Change-Id: Iebce47eabcb3b198d4aa7c1b8cfaf8f53300f7f9

When the lower-constraints tox target was added, it was assumed the
install_command was just running the install and that the dependencies
and constraints were being set using "deps = ".
This fixed the install_command and deps to follow the expected pattern
so the lower-constraints job actual does install the lower constraints.
Also:
 * raise the minimum for netifaces as the existing version would fail
 to load
 * pin oslo.log to the minimum required by our minimum
 keystonemiddleware
 * fix up some error handling for ancient versions of pastedeploy
 * specify that we shouldn't install xattr on a platform we don't
 support anyway, because reasons
Change-Id: Ie78c0dabe12e01109db2b6412166c3564b87ee96
Signed-off-by: Sean McGinnis <sean.mcginnis@gmail.com>

Change-Id: I83f8f59023eabc97386481c18ed8bbf8fab64fa8

Change-Id: I5a0d2108d01395aedb4ac18464ed2e38ad13ce84

Change-Id: Ib1cd0659d5085ff9f8555bc08bd17a8f02f2efc0

Change-Id: Ic00273e5d3120f7daba1d3cbf084159f1e7d8bbb

Change-Id: I8f3d4d5f6976ef5b63facd9b5723aac894066b74

* name_check
 * quotas
 * ratelimit
 * read_only
 * recon
Note that the middlewares themselves seem to be fine.
Fix proxy-logging for PEP-0479.
Change-Id: If47d09e8e8c09b9f50451f93d5b0803aa58743a7

Change-Id: Id74a93f10bc5c641d62141af33bef68e503f7e04

This started with ShardRanges and its CLI. The sharder is at the
bottom of the dependency chain. Even container backend needs it.
Once we started tinkering with the sharder, it all snowballed to
include the rest of the container services.
Beware, this does affect some of Python 2 code. Mostly it's trivial
and obviously correct, but needs checking by reviewers.
About killing the stray "from __future__ import unicode_literals":
we do not do it in general. The specific problem it caused was
a failure of functional tests because unicode leaked into a field
that was supposed to be encoded. It is just too hard to track the
types when rules change from file to file, so off with its head.
Change-Id: Iba4e65d0e46d8c1f5a91feb96c2c07f99ca7c666

Change-Id: I1095262563eff92d6d0a399eb0d5f9d3045becf3

Change-Id: I2701358f9f01f368c892a010a543ecc1f17684a3

Change-Id: I83aa5d72084ca183e00dc6d69e417fba310a2c59

When looking at the tuple thing, remember that tuples are comparable
with ints in py2, but they do not compare like (100)[0]. Instead, they
are always greater, acting almost like MAX_INT. No wonder py3 banned
that practice.
We had a much of other noncomparables, but we dealt with them in
obvious ways, like adding key= to sorted().
Change-Id: I52e96406c3c1f39b98c1d81bdc057805cd1a6278