Implement proper handling of dirent in FormulationManager, fixes #517

[hellkite500]

Fixes a bug that prevents forcing files from being properly discovered/validated on certain systems/file-systems.

Changes

• Refactor directory iteration to fallback to stat when appropriate.

Testing

1. All unit tests tested locall

Todos

• Should check any other places in the code potentially using dirent and readdir for similar issues.

Checklist

• PR has an informative and human-readable title
• Changes are limited to a single goal (no scope creep)
• Code can be automatically merged (no conflicts)
• Code follows project standards (link if applicable)
• Passes all existing automated tests
• Any change in functionality is tested
• New functions are documented (with a description, list of inputs, and expected output)
• Placeholder code is flagged / future todos are captured in comments
• Project documentation has been updated (including the "Unreleased" section of the CHANGELOG)
• Reviewers requested with the Reviewers tool ➡️

Target Environment support

• Linux
• MacOS

[christophertubbs]

This is throwing me off - if the _DIRENT_HAVE_D_TYPE condition doesn't pass, does this yield:

if( match ) {
    {
        struct stat st;
        if( stat((path+entry->d_name)>c_str(), &st) != 0) {...
    }
}

[hellkite500]

Yes, it might be a little easier to see from this SS where the IDE does some nice high/low lighting

[donaldwj]

Where is _DIRENT_HAVE_D_TYPE defined and is it a gcc extension? Ie will this work with intel compilers

[donaldwj]

So it looks like type information can also be present with the preprocessor directives of _DEFAULT_SOURCE glibc > 2.19 or _BSD_SOURCE for earlier library versions. Even in this case the d_type could be set to unknown based on the file system being read.

[hellkite500]

Where is _DIRENT_HAVE_D_TYPE defined and is it a gcc extension? Ie will this work with intel compilers

This is in dirent.h (see the documentation about the additional fields.)

[hellkite500]

So it looks like type information can also be present with the preprocessor directives of _DEFAULT_SOURCE glibc > 2.19 or _BSD_SOURCE for earlier library versions. Even in this case the d_type could be set to unknown based on the file system being read.

I think the fallback to stat.h is sufficient in either case. I don't think the _DIRENT_HAVE_D_TYPE is gcc specific, but is GNU system specific. It may be that the macro isn't defined on all systems that are actually capable of providing d_type from a dirent struct, but then we can just fall back to stat and be covered on any POSIX system.