Getting an error:

[srcrip]

I get this error:

E5108: Lua: Vim:E900: Invalid channel id                                                                                                                           
stack traceback:
        [C]: in function 'jobpid'                                                                                                                                  
        ...e/pack/core/opt/pathfinder.nvim/lua/pathfinder/utils.lua:68: in function 'get_terminal_cwd'                                                             
        ...e/pack/core/opt/pathfinder.nvim/lua/pathfinder/utils.lua:235: in function 'get_absolute_path'                                                           
        ...te/pack/core/opt/pathfinder.nvim/lua/pathfinder/core.lua:325: in function 'process_cursor_file'                                                         
        ...te/pack/core/opt/pathfinder.nvim/lua/pathfinder/core.lua:355: in function 'custom_gf'                                                                   
        ...te/pack/core/opt/pathfinder.nvim/lua/pathfinder/core.lua:563: in function <...te/pack/core/opt/pathfinder.nvim/lua/pathfinder/core.lua:562> 

when trying to do gf on something like this:

lib/example/record_types.ex
206:      "paymentFrequency" => "Monthly",
533:      "paymentFrequency",
549:      "paymentFrequency",
562:      "paymentFrequency",
577:      "paymentFrequency", <------------------------------ hit gf here
1203:      "paymentFrequency",
1204:      "paymentPerFrequency",
1229:      "paymentFrequency",
1230:      "paymentPerFrequency",
1257:      "paymentFrequency",
1258:      "paymentPerFrequency",
1282:      "paymentFrequency",
1283:      "paymentPerFrequency",

[HawkinsT]

Thanks for the report. Just to clarify, this is a grep output where by hitting gf you're wanting to open record_types.ex at line 577?

[srcrip]

correct. This is the result of running ripgrep <whatever>. By default you get this output where theres filenames, then a bunch of line numbers, then the line. I love the idea of your plugin and if you could make this scenario work I would absolutely loveeeee it.

[srcrip]

I can also try and submit a PR if your open to contributions!

[HawkinsT]

Thank you. Sorry for the slow response, I've been rather busy with work the past couple of weeks. I've just pushed a commit that should stop the error from occurring (which was due to the plugin being unable to retrieve the CWD of an exited terminal buffer). Please let me know if you still see the error at all.

In terms of getting the functionality you're after (instead of just getting rid of this error), I'll have to think about a good way to implement it as I'd only want to implement something that's going to be more generally useful to people instead of just targeting unique use-cases.

Here, I'm concerned that just grabbing the number at the beginning of the line and the file name from the top could lead to many false positives in other instances. I'm very open to discussing this and taking suggestions though, and if you want to submit a PR that you think solves this, even better!

[srcrip]

I think at the end of the day there’s probably going to be no one size fits all thing here… ripgrep is pretty popular though so it would be nice to support this one. But if you don’t want to have custom implementations for different tools perhaps you could add an escape hatch for people to add their own matching functions or something.