Respect argline passed into jinja renderer#55126
Conversation
|
I just realized that I opened this PR against master. The goal is to have this fix in 2019.2.3, does it need to be rebased against a different branch for this? |
0831c6f to
78ec844
Compare
|
Looks like there are some possibly related test failures? |
001d8f4 to
2d0ebc7
Compare
|
The tests that were failing were failing because they were not designed properly. |
|
@terminalmage @waynew the windows failures look legit. |
820273c to
74b25d0
Compare
|
@dwoz I pushed an update to the tests, and rebased to bring it up to date with master. |
|
@terminalmage looks like there are pending conflicts when you get time :) I'm updating the labels to next release so that it isn't confusing |
10a7ad1 to
8bed23a
Compare
|
Rebased again. |
This was being ignored previously, despite being checked for in the beginning of the render() function. This caused any renderer passing string data, as opposed to a StringIO or other file-like object (e.g. the gpg renderer) to fail jinja rendering, as the salt.utils.templates was not being properly informed that the input was a string.
|
Rebased yet again. It has been 5 1/2 months since this PR was opened. It's a one-line code change, and was submitted with tests. Can we please get this re-re-re-reviewed and merged? |
Resolves #55124.