Add load_file and load_file! methods, with tests. Fixes issue #386.

[keithrbennett]

Adds load_file methods analogous to YAML's.

[rafaltrojanowski]

@keithrbennett I'm just wondering, shouldn't we pass options for 'File.read'? It might be useful for supporting different encodings, etc. It's not related but I have a case where I struggle with BOM character in the beginning of the CSV file and by passing encoding I can simply remove those additional bytes. Maybe that would work similarly in JSON.

[keithrbennett]

First I want to thank you for pointing out to me that I was failing to pass the passed opts along to the called methods. I've fixed that and pushed the fix.

Regarding your point about File.read options, do I understand you correctly that you would want to specify options to File.read, not JSON.parse? If so, I'm not sure how we could do that, since we are already using the Ruby hash last parameter feature for the JSON.parse options.

Any strategy to handle options to File.read would need to be balanced with the cost of the complexity added to support it. Can you suggest a way to accommodate this need in the least complex way?

[rafaltrojanowski]

Yes, but I'm happy with current solution. I don't have idea how to solve it in a elegant way. Maybe maintainer can help. And sorry for the oftop, anyway it's better than before :)

[marcandre]

Best not mix options. If a user need options, they can do the File.read. FWIW, YAML.load_file has no file option either.

[rafaltrojanowski]

Thanks @keithrbennett!

[marcandre]

Please add a test with options...

[keithrbennett]

Sorry about that. I've added this (and an equivalent function for load_file!), would it suffice?:

  def test_load_file_with_option
    temp_file_containing(@json) do |filespec|
      key_classes = load_file(filespec, symbolize_names: true).keys.map(&:class)
      assert_true (key_classes.include?(Symbol) && (! key_classes.include?(String)))
    end
  end

Also, in order to get the tests to pass I had to rebase master my feature branch. Is it ok to git push -f this branch?

[marcandre]

Is it ok to git push -f this branch?

Yes, that's the best thing to do 😄

[marcandre]

Very good. Only thing remaining would be to squash your commits (although it can be done by maintainers with 'squash and merge')

[marcandre]

method(method_name).call(filespec) => public_send(method_name, filespec). I think it's more idiomatic.
Same for test_load_file_with_option_shared

[keithrbennett]

I've made this change and squashed the commits. Thanks for your guidance, I agree that these are both improvements.

[marcandre]

Thanks for the impressively quick turnaround and good PR!

[marcandre]

Good PR 👍
LGTM

[hsbt]

@keithrbennett Thanks. Could you rebase from this from master branch?

I will merge this targeted json-2.4.0 after releasing json-2.3.1.

[keithrbennett]

Done, thanks. - Keith

…

On Jun 30, 2020, at 07:32, Hiroshi SHIBATA ***@***.***> wrote: @keithrbennett <https://github.com/keithrbennett> Thanks. Could you rebase from this from master branch? I will merge this targeted json-2.4.0 after releasing json-2.3.1. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#387 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAAG56SOJBP7OXTPANNUXJDRZHEOFANCNFSM4JDKZJMA>.

[hsbt]

Thanks!