Add first failing test for #551

[triceo]

The test for #551 is failing on List-based arguments. Before I move on with writing extra coverage, we should probably agree on how to deal with this. WDYT @remkop?

[triceo]

I can do this here, because of the change above.

[triceo]

Also added a test to see if the simplified files work with quote trimming.

[triceo]

@remkop Any chance this could also be set through a system property? This way, it prevents Picocli from being used in the simple way: CommandLine.call(...)

[remkop]

Yes sure. Do you want to provide a separate pull request for that?

[triceo]

I'll take a look in another PR.

[remkop]

Thanks for the PR! It’s my bedtime tonight, I’ll take a look tomorrow.

[remkop]

I took a look just now. The approach looks good, but the tests fail in the Travis CI build. Can you make the tests pass?

picocli.CommandLineTest > testAtFileSimplified FAILED
    picocli.CommandLine$UnmatchedArgumentException: Unmatched argument: yet something else
        at picocli.CommandLine$Interpreter.parse(CommandLine.java:6759)
        at picocli.CommandLine$Interpreter.parse(CommandLine.java:6615)
        at picocli.CommandLine.parse(CommandLine.java:722)
        at picocli.CommandLine.populateCommand(CommandLine.java:675)
        at picocli.CommandLineTest.testAtFileSimplified(CommandLineTest.java:3649)

picocli.CommandLineTest > testAtFileSimplifiedWithQuotesTrimmed FAILED
    java.lang.AssertionError: expected: java.lang.String<https://picocli.info/> but was: java.net.URL<https://picocli.info/>
        at org.junit.Assert.fail(Assert.java:88)
        at org.junit.Assert.failNotEquals(Assert.java:834)
        at org.junit.Assert.assertEquals(Assert.java:118)
        at org.junit.Assert.assertEquals(Assert.java:144)
        at picocli.CommandLineTest.testAtFileSimplifiedWithQuotesTrimmed(CommandLineTest.java:3677)

[remkop]

Sorry, I just re-read and saw your first comment. So, you are asking, if I have an option like this:

@Option(names = "-x") List<String> values;

and I give it multiple values, like -x a b c, why does picocli not add all values "a", "b" and "c" to the list but gives an 'Unmatched argument: b" error?

The answer is that default arity for arrays and collection options is 1. If an option needs to consume multiple arguments, you can specify the exact desired arity:

// consume 1 or 2 args for each -y option
// for example: <cmd> -y A -y A B -y E -y F G
// gives y = [A, A, B, E, F, G]
@Option(names = "-y", arity = "1..2") List<String> y;

// consume as many args as possible until the next option
// for example: <cmd> -z A B C D E -y Y
// gives z = [A, B, C, D, E] and y= [Y]
@Option(names = "-z", arity = "1..*") List<String> z;

[triceo]

@remkop Thanks, I completely forgot about the arity. The test is fixed now.

Did you have any more tests in mind? If so, please share - otherwise I'll move on to creating the PR wrt. the quote-trimming property.

[remkop]

I’m happy to merge this PR when the tests pass.

[codecov-io]

Codecov Report

Merging #558 into master will increase coverage by 0.09%.
The diff coverage is n/a.

@@             Coverage Diff              @@
##             master     #558      +/-   ##
============================================
+ Coverage     89.04%   89.13%   +0.09%     
- Complexity      285      286       +1     
============================================
  Files             4        4              
  Lines          3961     3967       +6     
  Branches        974      976       +2     
============================================
+ Hits           3527     3536       +9     
+ Misses          216      212       -4     
- Partials        218      219       +1

Impacted Files	Coverage Δ	Complexity Δ
src/main/java/picocli/CommandLine.java	89.17% <0%> (+0.1%)	161% <0%> (+1%)	⬆️

---

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 95c9b99...2df98ba. Read the comment docs.

[remkop]

Merged. Thanks for the PR!
I’ll update the release notes later.

[remkop]

FYI: I created #561 as a more strategic solution for improving the convenience methods.

[triceo]

@remkop Thanks for merging this PR, I also submitted #562 in the meantime. (Although #561 will definitely be useful and will supersede my solution.)