This was added earlier in an attempt to tolerate CRLF and convert CRLF
line endings on Windows, but it causes issues where vendored files
(which could be using either LF or CRLF depending on the original author's
preference) get permanent diffs when inconsistent with the platform's
preference.
The goal of this change, therefore, is to treat all of the files as binary,
with the standard that all of Terraform's own files will use Unix-style
LF endings and the vendor stuff will just be verbatim, byte-for-byte
copies of what's upstream.
This will apparently cause some difficulty for people hacking on Terraform
on Windows machines, because gofmt on Windows reportedly wants to convert
all files to CRLF endings. Unfortunately we're forced to compromise here
and treat development on Windows as an edge case in order to avoid the
weirdness with inconsistent endings in the vendor tree.
It's possible to not change the backend config, but require updating the
stored backend state by moving init options from the config file to the
`-backend-config` flag. If the config is the same, but the hash doesn't
match, update the stored state.
This method mirrors that of config.Backend, so we can compare the
configration of a backend read from a config vs that of a backend read
from a state. This will prevent init from reinitializing when using
`-backend-config` options that match the existing state.
The plug-in crashes if "localNetworkGateways" is not found in the id parameter. The fix is to verify the parameter before proceeding.
Also the "import" example in the documentation is wrong, "localNetworkGateways" should be case sensitive. The current example actually causes the plugin to crash due to the bug.
Most of the bios have five lines of text, so that's the driver for the
layout except for @grubernaut's and @mbfrahry's where there's only three
lines. This makes the pictures following the short bios look misaligned.
This quick fix just always leaves enough room for five lines of bio text,
ensuring that the photos appear at a consistent vertical rhythm.
It's annoying to hard-code a particular value here, since this value won't
survive e.g. a change to the typesetting, but a more involved fix here
(using flexbox layout, or something else complicated) doesn't seem
warranted.
The variable validator assumes that any AST node it gets from an
interpolation walk is an indicator of an interpolation. Unfortunately,
back in f223be15 we changed the interpolation walker to emit a LiteralNode
as a way to signal that the result is a literal but not identical to the
input due to escapes.
The existence of this issue suggests a bit of a design smell in that the
interpolation walker interface at first glance appears to skip over all
literals, but it actually emits them in this one situation. In the long
run we should perhaps think about whether the abstraction is right here,
but this is a shallow, tactical change that fixes#13001.
Jake Champlin