add read and write file validators #249

[rafiw]

Will close #249.

[codecov]

Codecov Report

Merging #250 into master will decrease coverage by 0.03%.
The diff coverage is 83.33%.

@@            Coverage Diff             @@
##           master     #250      +/-   ##
==========================================
- Coverage     100%   99.96%   -0.04%     
==========================================
  Files          12       12              
  Lines        2566     2570       +4     
==========================================
+ Hits         2566     2569       +3     
- Misses          0        1       +1

Impacted Files	Coverage Δ
include/CLI/Validators.hpp	99.68% <83.33%> (-0.32%)	⬇️

---

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 be8a08f...6f88e6c. Read the comment docs.

[codecov]

Codecov Report

Merging #250 into master will not change coverage.
The diff coverage is 100%.

@@          Coverage Diff           @@
##           master   #250    +/-   ##
======================================
  Coverage     100%   100%            
======================================
  Files          12     12            
  Lines        3397   2786   -611     
======================================
- Hits         3397   2786   -611

Impacted Files	Coverage Δ
include/CLI/Validators.hpp	100% <100%> (ø)	⬆️
include/CLI/Formatter.hpp	100% <0%> (ø)	⬆️
include/CLI/Option.hpp	100% <0%> (ø)	⬆️
include/CLI/App.hpp	100% <0%> (ø)	⬆️
include/CLI/StringTools.hpp	100% <0%> (ø)	⬆️
include/CLI/TypeTools.hpp	100% <0%> (ø)	⬆️
include/CLI/Split.hpp	100% <0%> (ø)	⬆️
include/CLI/ConfigFwd.hpp	100% <0%> (ø)	⬆️
include/CLI/Error.hpp	100% <0%> (ø)	⬆️

---

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 d9379cc...bef2388. Read the comment docs.

[rafiw]

@henryiii if you agree with this approach i will add tests

[henryiii]

CLI::ExistingReadableFile And ...Writable... look like better names I think, but otherwise I like the way you’ve added it! +1

[henryiii]

Will close #249.

[rafiw]

@henryiii the permissions in windows is different then the linux once. that why the test fails.
How do you want to approach this?

[cbachhuber]

What is the plan with this PR? Continue working on it or close it? I can also try to pick up the work

[rafiw]

the issue is with windows since the permissions there is more complicated.
i can put the windows test under ifdef so it will pass.
is this ok?

[phlptp]

This would need to be made to work with the changes from #341. For C++17 that enables the use of filesystem so it might be a little easier in C++17 mode on windows now.

[cbachhuber]

Right, but we would still need to enable file handling (and therefore, this feature) for windows under C++11, correct? Consequently, I would prefer to implement it for Windows & C++11 and not put the test under ifdef. What do you think?

[rafiw]

i will take another look on this and add C++17 support as well

[phlptp]

@henryiii this and a few other issues about validators indicate that some additional validators are of interest. The balance here is that the Validators can take time to compile so for simple applications that don't use them that is a lot of non useful code that needs to compile that must be balanced with the high utility of the additional Validators for those that use them.

What I am wondering if it makes sense to split the Validators into two pieces. The basic definition is required for CLI11, but many of the specific validators and all the machinery that goes with them could be split into a separate file that possibly is even not included in the single file header. Then a separate Validators.hpp would define a library of additional validators, that if someone wanted it they could include that header as well. With the exception of the underlying filesystem operations none of the validators are required by the rest of CLI11. So apart from the basic definition and maybe a couple of the most common validators, the library could be split without issue for 2.0. Then we could get these validators in and some of the other issues without an increasing impact on the compile time.