Makes boolean options falsible

[igitur]

As discussed in issue #115 this makes a boolean option falsible. You can still optionally omit the 'true' or 'false' part.

[tbroust-trepia]

Hi!
Your PR breaks the unit tests
Could you fix it please ? I would integrate it in my fork

gsscoder doesn't show any sign of activity anywhere, so I took it upon myself to fix some stuff and integrate the pending PRs. I also created new Nuget packages with the PRs.

Thanks ! :)

[igitur]

Oh, there are unit tests! I'm on it.

[igitur]

@tbroust-trepia Success?

[cosmo0]

Oh sorry! I switched accounts, and did not notice you had made fixes. I'll try it out ASAP.

[cosmo0]

Yep it seems to be working (at least the unit tests pass, I must admit I haven't tested it live).

I've updated my nuget package : https://www.nuget.org/packages/CommandLineParser19/

Thanks :)

[igitur]

Cool. But why did you fork it? Is this project dead? @gsscoder You still alive?

[cosmo0]

Well, I'm guessing that yes, the project is dead. @gsscoder hasn't made a commit in over a year, as well as a post in his twitter and blog. So I'm guessing he's not "there" anymore (hoping nothing serious happened to him). See #132 .

Since I had to make a fix in the library anyway (PR #136), I thought that it would be better to just merge all PRs from this project into one, this way everyone can enjoy the work of others.

[gsscoder]

heartbeat, brain activity, reactivity to pain, yes, I think I'm alive... :)

this PR is against stable branch, so sorry, but for the moment I'm working on 2.0 pre-rel

can we resolve this problem with 2.0?

[gsscoder]

@igitur this kind of enhancement is discussed also here #158 (comment) (discussion started by @sushihangover)

beyond the version (stable or pre-release) I'd like hear more opinions if

• we need a means of add custom data validation and transformation machinery
• we need more control over boolean switch, how that should be implemented (I'm agree with @sushihangover that various applications use +\- tokens to allow more fine grained control over switches along the possibility of specify an option more than once)

[tsalomak]

I found this library and it was great, until I needed a boolean values, that are (almost) every time true, but for an exceptional cases where it's changed to false. Now I have to rename everyone to a negative way, to have true as the case when it's actually turned off.

At least for me this is really annoying. Now I either have to have code full of negative rules, or make some strange negation between the options and actual values being used. If I wouldn't have implemented everything else already, I would have already changed the library that I use. Is there any chance that someone would actually do something, since this has now taken 3 years?

[nemec]

This pull request is for version 1.9.x which is no longer receiving updates. I am happy to merge a PR updated to fit the new 2.0 codebase.

[igitur]

So to be clear, against which branch should I base the PR?

[nemec]

Use the Master branch