Text files not recognised by magic

[WalterS]

Somehow plaintext files are not recognised:

irb>> MimeMagic::VERSION
=> "0.3.2"
irb>> file = '/tmp/foo'
=> "/tmp/foo"
irb>> File.write(file, 'This is a text file')
=> 19
irb>> MimeMagic.by_magic(file)
=> nil

[georgechang0117]

I've got the same problem.

[jonnynux]

I've got the same problem.

[jonnynux]

I confirm the issue, every plain textual file like csv, json, css, txt are not recognized. The workaround is to return text/plain if the result of by_magic is nil, but this will make some side effects (example for application/octet-stream. Other suggestions?

[jamesremuscat]

It appears to only detect text files by magic if the files begin with the strings:

• This is TeX,; or
• This is METAFONT,

Needless to say, this would only appear to cover a minimal subset of text files.

[seabre]

I have a proposed PR for testing for CSVs: #75

[minad]

Is it possible to test via file extension and then via magic? This should handle the tex, json, css, csv cases.

[seabre]

If you want to be able to match and be mostly sure, you have to parse at least part of the file. I'm not sure with the current setup that you actually can correctly check for CSVs, JSON, CSS, and etc. You can check by extension, but I can do that without the library.

[kelly-croswell]

Has any movement been made on this? I just ran into this problem on a project I'm working on and could really use a solution.

[minad]

Invalid, this should be fixed in the shared-mime-info database upstream.