Christian Bartolomäus

Just for the record: Looking at the links of the CI runs I'd say the problem is not related to the ```master``` -> ```main``` rename. (That one probably happened on...

Back in 2015 @FROGGS and me used a hackathon to work on adding the ability to fudge tests for a specific version of Perl 6 (as it was called then)....

Basically I'm agreeing with @vrurg (from far above): > Moreover, _each_ test must have an explicit version. Since roast *defines* Raku, it has to be clear, which test is part...

Hi @XnS, yeah, this seems to happen because of ```node``` being treated as a special keyword by puppet-lint. Actually, I'm seeing the same error with your second code example. There...

Something like the following would probably help: ``` diff --git a/lib/puppet-lint/lexer.rb b/lib/puppet-lint/lexer.rb index e9d1696..7cf3fbe 100644 --- a/lib/puppet-lint/lexer.rb +++ b/lib/puppet-lint/lexer.rb @@ -211,6 +211,8 @@ class PuppetLint value = chunk[regex, 1] next...

I think I've discovered what caused the slowdown for large manifests: The regex that is used to match a function name (cmp. PR #854) becomes very slow for large inputs....

At least for Perl 5 I think I (kind of) understand now why the second regex is much slower. Looking at https://perldoc.perl.org/perlreguts.html, running the code in debug mode (https://perldoc.perl.org/re.html#%27debug%27-mode) and...

Parsing large manifests should be much faster on HEAD, since https://github.com/rodjek/puppet-lint/pull/854 has been merged. This issue could probably be closed after the next release.

I don't know for sure, but it seems like @rodjek was quite active the last days and the [milestone for 2.4.0](https://github.com/rodjek/puppet-lint/milestone/19) looks promising. So if I had to guess, I'd...

That's interesting. It seems that now some of the checks become the bottlenecks. When I looked into this issue, I identified parsing as the problematic part. Running something like this...