Can CPD output include the duplicate tokens?

[wesnolte]

I'm working on a very large codebase, and focussing on Apex at the moment. Love the CPD tool, and we've got a lot to work through but unfortunately the output doesn't tell me what the duplicates actually are. I can't manually go through file because there are hundreds, and some of them are very large (don't judge!). Is there a way to get CPD to tell me what the duplicates are? At the moment all I'm getting is:

=== 2428. DetectCopyPasteForApex
    severity:  5 (Info)
    engine:    cpd
    message:   Duplicate code detected for language 'apex'. Found 2 code locations containing the same block of code consisting of 176 tokens across 19 lines.
    locations:
        (main) force-app/main/default/classes/File.cls (444:29-462:54)
        force-app/main/default/classes/File.cls (479:10-497:54)
    resource:  https://docs.pmd-code.org/latest/pmd_userdocs_cpd.html#refactoring-duplicates

The HTML output (as expected I guess) has the same info.

[johnbelosf]

hi @wesnolte , hope you're doing well!

So are you looking to have code snippets as part of the output? Or something else?

[wesnolte]

Good question. I guess I need something to search for when a duplicate is detected e.g. in some paces I have "10 code locations containing the same block of code" and I know the file names but I don't have any clue as to what the duplicate code might be. CPD seems to have a verbose flag but I'm not sure if that'll do the trick, and I haven't found a way to pass that flag to the analysis run.

[wesnolte]

I believe I was on your CTA board a long, long time ago @johnbelosf !

[johnbelosf]

Thank you @wesnolte - and yes, you're correct - around 11 years ago now!

Have you considered using one of our machine-readable formats (e.g. JSON) as an output that can then be parsed by a script that reveals the relevant code blocks? This would be the most immediate solution. We can consider something on the Code Analyzer end in the future, but this is a very specific use case that we wouldn't be able to prioritize for now

[wesnolte]

I've taken a look at the JSON output and it's made me realise the info to find the duplicates is in the all formats, I just didn't understand what it meant i.e.

=== 2428. DetectCopyPasteForApex
    severity:  5 (Info)
    engine:    cpd
    message:   Duplicate code detected for language 'apex'. Found 2 code locations containing the same block of code consisting of 176 tokens across 19 lines.
    locations:
        (main) force-app/main/default/classes/ClassA.cls (444:29-462:54)
        force-app/main/default/classes/ClassA.cls (479:10-497:54)
    resource:  https://docs.pmd-code.org/latest/pmd_userdocs_cpd.html#refactoring-duplicates

Means the duplicate token is in ClassA at line 444, column 29 -> line 462, column 54 and line 479 column 10 -> line 497, column 54. Before I thought it meant something like there are 1 or more duplicate instances within that range.

The JSON is a little clearer in this messaging in my opinion. So, no problem except that the output (--view detail, HTML) could be a little clearer.