Check namespaces and class imports for leading backslash

[andrewnicols]

Fixes #20

[codecov]

Codecov Report

Merging #61 (e3dae34) into main (5f76734) will increase coverage by 0.08%.
The diff coverage is 100.00%.

@@             Coverage Diff              @@
##               main      #61      +/-   ##
============================================
+ Coverage     95.53%   95.62%   +0.08%     
- Complexity      440      445       +5     
============================================
  Files            19       20       +1     
  Lines          1165     1188      +23     
============================================
+ Hits           1113     1136      +23     
  Misses           52       52              

Files Changed	Coverage Δ
...odle/Sniffs/Namespaces/NamespaceStatementSniff.php	100.00% <100.00%> (ø)

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[stronk7]

Q: Shouldn't we also add PSR12.Files.ImportStatement.LeadingSlash to get imports also covered?

And little detail, it looks "strange" that our class is named "NoLeadingSlash", shouldn't it be better named "LeadingSlash".

Similar sentiment with the problem name, we are naming it "NamespaceStartsWithSlash", that is redundant ("namespace") and uses different wording. Why not, also for the problem, just use "LeadingSlash".

FYC, ciao :-)

PS, I've seen that there are a few tests failing right now...

[andrewnicols]

Q: Shouldn't we also add PSR12.Files.ImportStatement.LeadingSlash to get imports also covered?
Bah - I originally applied this to the extra ruleset which already has it. Updated to include it/.

And little detail, it looks "strange" that our class is named "NoLeadingSlash", shouldn't it be better named "LeadingSlash".

Similar sentiment with the problem name, we are naming it "NamespaceStartsWithSlash", that is redundant ("namespace") and uses different wording. Why not, also for the problem, just use "LeadingSlash".

If I make this LeadingSlash and name the problem LeadingSlash it becomes moodle.Namespace.LeadingSlash.LeadingSlash which is also redundant.

Rather than that, I've tried to take inspiration from the PSR-12 one and now have:
moodle.Namespace.NamespaceStatement.LeadingSlash

I think the duplication of Namespace in both the Sniff type, and the sniff name is okay because it's a check of the Namespace statement and naming it simply Statement could be confusing - moodle.Namespace.Statement is fine, but Statement on its own may confuse.

Thoughts?

PS, I've seen that there are a few tests failing right now...

phplint was actually picking this up already so I think I've got that now!

[stronk7]

I like making it NamespaceStatement so we can add later more checks there.

Note: I'm completing the API checks for PHPUnit classes (that already have Namespace checks). Once working I plan to move them from being "phpunit only" to be "whole codebase" ones. Surely then we'll be adding more checks to the NamespaceStatement sniff added here.

[andrewnicols]

Note: I'm completing the API checks for PHPUnit classes (that already have Namespace checks). Once working I plan to move them from being "phpunit only" to be "whole codebase" ones. Surely then we'll be adding more checks to the NamespaceStatement sniff added here.

Perfect - that was about what I had in mind too!