ignore token like text

[rbalet]

Describe the solution you'd like
I assume it will be hard to achieve if not impossible

Some time we have to add token in our app, like the google api one or such.

here, for example, we got 3 error when it should just be ignored.

Describe alternatives you've considered
I know it does exists ways to ignore specific text (in case we would like to use the same token every where) but ignoring all of them would be easier.

[Jason3S]

@rbalet,

I have thought about this issue, but have not come up with a good detection solution. I think it is a fun problem.

It is a form of Automatic Language Identification and Detection. If the score is too low, then it should be considered a token.

That challenge here is that it needs to be:

1. fast
2. low memory usage.

It is an entropy measurement.

[rbalet]

@Jason3S Yes, I wanted to add an Idea but could figure it out. Maybe someone out there have it.

[Jason3S]

@rbalet,

I have had some ideas, but haven't had the time to try them. out.

[max-carroll-sky]

One idea I had is if we could have a comment annotation that switched on an ignore beased on a regex pattern, one of our devs has come up with a regex that picks up tokens really well, but it also picks up other things like strings in enums in typescript if we could have something like

// cspell ignore pattern start '[\w_=:-.{}[]“,?]{15,}'

"SOME_OBSURE_KEY_TOKEN" : "asjdflkasdjfa;osdkfj;orgijeaopijergfo"

// cspell end ignore pattern

then we could at least constrain it to certain bits of files

[ssbarnea]

I want to configure cspell to ignore spelling sequences like this as this is causing lots of false positives and noise, especially when using cspell vscode extension.

[Jason3S]

@ssbarnea,

#6615 should help address this issue. But, there needs to be a bit more tuning before it lands.