This commit does the following:
- Introduces the get-machine command. This command contacts the build farm server
gets back a machine reservation and then opens a secure shell to the machine (Alternatively,
you can ask for a json). This also involved factoring out some commands to deal with authenticated
ddp from package-client into a more general auth-client.
- No longer publish binary builds in publish or publish-release; instead give the user a warning
to run get-machine and then publish-for-arch. Someone could ignore this: --existing-version and
publish-for-arch both publish binary builds, but you need to be at least somewhat familiar with
what you are doing to run them. Hopefully, you are running them from a certified build machine, but
if you are not, then, well, it is your package.
Stuff remaining:
- We are going to have a url to external documentation, but I haven't written it yet.
- We are currently talking to the test-build server, instead of the build server, so mac doesn't
work.
(Neither of those changes require significant tool changes)
This reverts commit 89b394d974.
I cherry-picked the change onto devel, since this change isn't
even supposed to change the behavior of the tool. Why risk
it just before the release? Thanks @nim for the suggestion.
The test was failing because it was written to run against the real
local warehouse, instead of a stubbed one. When initializing the
Sandbox with an explicit warehouse, the test failed because of
incorrect logic piping the right path to the packages database
(in this case "test-packages.data.db" instead of "packages.data.db").
Unfortunately, the test still fails -- this time potentially due to a bug:
After publishing the package for the first time with `--create`,
the next step tries to update but gets the following error:
There is no package named test:i1iyhj. If you are creating a
new package, use the --create flag.
The test was failing because it was written to run against the real
local warehouse, instead of a stubbed one. When initializing the
Sandbox with an explicit warehouse, the test failed because of
incorrect logic piping the right path to the packages database
(in this case "test-packages.data.db" instead of "packages.data.db").
Unfortunately, the test still fails -- this time potentially due to a bug:
After publishing the package for the first time with `--create`,
the next step tries to update but gets the following error:
There is no package named test:i1iyhj. If you are creating a
new package, use the --create flag.
Should fix a bug that was kinda hard to reproduce but was noticed before:
- open the app that wasn't HCP'd
- change the current url path with HTML5 pushState
- call window.location.reload()
- observe the packages code not being loaded properly
This reverts commit 05d2c5edd7.
We haven't worked out all the backwards-compatiblity issues of
non-?close and ?close OAuth servers and clients yet, so let's leave this
as is now, and ensure that all the services that the tool talks to
accept ?close URLs.
This is a regression caused by the SQL merge. Sorry for not catching it earlier!
We should not be calling semver on meteor package versions outside of the package
version parser package.
MDA stuff configured with the new tool does not use ?close, and the one configured with
the old tool, does. We are not going to automatically try contacting the ?close anymore,
since that doesn't work in the future; for backwards compatibility, we went through package
server & stat server and added the close-less urls for the new tool to use.
This should mean that you can access the package server from behind a
corporate firewall.
I tested this by setting up a Linux machine that doesn't have access to
packages.meteor.com:
$ sudo iptables -A OUTPUT -d 54.225.216.115 -j DROP
$ sudo iptables -A OUTPUT -d 184.72.252.20 -j DROP
$ sudo iptables -A OUTPUT -d 23.23.114.56 -j DROP
I confirmed that these commands both fail:
$ curl https://packages.meteor.com/
$ ./meteor search asdf # when it needs to sync first
I bought a proxy server from Proxy Bonanza and confirmed that setting
the environment variable HTTPS_PROXY to
https://PROXYUSERNAME:PROXYPASSWORD@PROXYIP:PROXYPORT/
made both of the commands above succeed.
Fixes#2515.
This commit introduces some conviluted logic to keep the old behavior of the
meteor bundle command. Hopefully we will drop it soon so we will simplify the
logic.
If we don't clean up the tarball, then the following command (`meteor
build`) fails ("Couldn't create tarball"), but for some reason still
exits with code 0. So we then go on to successfully untar the output of
`meteor build`. If we clean up the tarball from `meteor bundle`, we at
least catch the failure by trying to untar a nonexistent file.
