fix: inconsistent evaluation of tm_try() #432
Conversation
Codecov Report
@@ Coverage Diff @@
## main #432 +/- ##
==========================================
- Coverage 69.70% 69.67% -0.04%
==========================================
Files 42 42
Lines 7745 7736 -9
==========================================
- Hits 5399 5390 -9
Misses 2094 2094
Partials 252 252
Flags with carried forward coverage won't be shown. Click here to find out more.
Continue to review full report at Codecov.
|
| want: map[string]*hclwrite.Block{ | ||
| "/stack": globals( | ||
| str("a", "value"), | ||
| str("b", "value"), | ||
| str("c", "value"), |
| { | ||
| path: "/stack", | ||
| add: globals( | ||
| str("a_interpolated", "${global.undefined}-something"), |
| @@ -111,18 +111,15 @@ func (ge *globalsExpr) eval(rootdir string, meta Metadata) (Globals, error) { | |||
| // This is relative only to root since meta.Path will look | |||
| // like: /some/path/relative/project/root | |||
| logger := log.With(). | |||
| Str("action", "eval()"). | |||
| Str("action", "globals.eval()"). | |||
There was a problem hiding this comment.
Nice!
The action metadata is currently very inconsistent.
There was a problem hiding this comment.
Yeah it is, I tried to check if the logger had some easy way to get method/function names but I didn't find anything at the time (wanted to remove actions in general), only a way to always log filename + line etc. The problem with using things like just "eval" is that we have a bunch of eval related stuff..so it can get confusing to understand the logs, would be cool to be consistent in a way to provide more context on the actions.
It should not be a breaking change, the only change in behavior is that code that used to fail before will now work, if you are pedantic it could be considered a breaking change, because maybe some people depended on the breaking/failing behavior, but I don't think we need to be that pedantic at this point =P |
No description provided.