disabling or overriding is_buf_binary_pre_read()

[bsturk]

I edit a lot of code for vintage computers, and sometimes there are character codes in the files that trigger this plugin to assume it is a binary.

I've tried the normal ways of trying to override the behavior (setting nobinary, etc in an au), but it keeps getting filtered. Is there an option to say for a particular filetype to ignore?

[RaafatTurki]

Hmm, yes the current binary buffer detection is not too sophisticated (the correct way would be to lookup magic bytes but that's a bit too computationally expensive and requires file streaming for it to be feasible)

As for your issue I believe your files aren't utf8 encoded which trips up hex.nvim into thinking it's binary.

You could disable the utf8 check by setting the following in the configs

  is_file_binary_post_read = function() return false end

[bsturk]

Hi, thanks for that!

Here's what I have now, but unfortunately, it is still happening (commenting it out I can open the file as text, but with all of the below it opens up w/ xxd etc)

{
    'RaafatTurki/hex.nvim',
    config = true,
    is_file_binary_pre_read = function() return false end,
    is_file_binary_post_read = function() return false end
},

I've attached a file which exhibits the issue for me. rebound.txt

[RaafatTurki]

Ahaa, as I suspected it's not utf8.

You seem to be using lazy to manage your plugins, you should put the options in an opts table like so:

you only need the post_read hook disabled since that's the one that does the utf8 check

{
  'RaafatTurki/hex.nvim',
  config = true,
  opts = {
    is_file_binary_post_read = function() return false end
  }
},

I've tried it on your file and it works.