feat: allow terramate and stack config from multiple files

[katcipis]

Since the terramate block can include a different range of configurations, like:

• version
• git
• code generation configs

It seemed useful to allow it to be defined on different files, multiple times. So this PR introduces that. For the stack block it seemed less useful, so it remains as it was before (only one block allowed).

Design wise the idea was to replace the function that would parse specific files/data for a function that given a dir parses the whole dir (all files matching *.tm or *tm.hcl). I kept the original logic as much as I could to avoid regressions + also did some piggybacking on the current tests to test with different filenames.

Globals and code generation still only loads the default file, so there is still more work to do.

As an example, this should work and list both stacks:

#!/bin/bash

set -o nounset
set -o errexit

basedir=$(mktemp -d)

cd "${basedir}"

# We need the project root config since it is not git
cat > terramate.tm.hcl <<- EOM
terramate {
  config {
  }
}
EOM

mkdir -p stacks/stack-{1,2}
terramate stacks init stacks/stack-1

cat > stacks/stack-2/stack.tm.hcl <<- EOM
stack {}
EOM

cat > stacks/stack-2/version.tm.hcl <<- EOM
terramate {
  required_version = "0.0.10"
}
EOM

cat > stacks/stack-2/config.tm.hcl <<- EOM
terramate {
  config {
    // stack specific config here
  }
}
EOM

terramate stacks list

[codecov-commenter]

Codecov Report

Merging #216 (0c6a8ba) into main (7905b9e) will increase coverage by 0.27%.
The diff coverage is 86.50%.

@@            Coverage Diff             @@
##             main     #216      +/-   ##
==========================================
+ Coverage   63.36%   63.64%   +0.27%     
==========================================
  Files          30       30              
  Lines        5408     5391      -17     
==========================================
+ Hits         3427     3431       +4     
+ Misses       1797     1777      -20     
+ Partials      184      183       -1     

Flag	Coverage Δ
tests	63.64% <86.50%> (+0.27%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
init.go	44.00% <20.00%> (+4.86%)	⬆️
generate/generate.go	88.98% <80.00%> (+0.09%)	⬆️
hcl/hcl.go	81.72% <90.62%> (+1.36%)	⬆️
config/config.go	96.15% <100.00%> (+3.99%)	⬆️
list.go	92.50% <100.00%> (+0.60%)	⬆️
stack/stack.go	75.90% <100.00%> (-1.88%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 7905b9e...0c6a8ba. Read the comment docs.

[i4ki]

lgtm but left some improvement comments, nothing serious.

[i4ki]

❤️

[i4ki]

@katcipis

Globals are still loaded from terramate.tm.hcl: https://github.com/mineiros-io/terramate/blob/ff66ae70f2175ac9b5557a3d3c6f9ba6bee566fb/globals.go#L233

so having a separate globals.tm.hcl does not work.

$ ls apps/a
globals.tm.hcl  terramate.tm.hcl
$ cat apps/a/globals.tm.hcl
globals {
        test = 1
        something = "else"
}

$ terramate stacks globals
$

[katcipis]

@katcipis

Globals are still loaded from terramate.tm.hcl:

https://github.com/mineiros-io/terramate/blob/ff66ae70f2175ac9b5557a3d3c6f9ba6bee566fb/globals.go#L233

so having a separate globals.tm.hcl does not work.

$ ls apps/a
globals.tm.hcl  terramate.tm.hcl
$ cat apps/a/globals.tm.hcl
globals {
        test = 1
        something = "else"
}

$ terramate stacks globals
$

Ah yes, Im aware, I also added a comment on the PR description:

Globals and code generation still only loads the default file, so there is still more work to do.

The idea is to incrementally bring all the code generation features to using multiple config files, but right now export_as_locals/globals and generate_hcl wont work yet. Since they would still work on terramate.tm.hcl and nothing would break I thought it would be OK to work on them incrementaly in a different PR. WDYT ?

[i4ki]

I thought it would be OK to work on them incrementaly in a different PR. WDYT ?

I guess it's ok.

[i4ki]

LGTM 🚀