Re: Performance degradation from long lines

[Stefan Monnier]

>> Then I tried opening the 663K minified JSON file with 671209
>> characters on one line.
> Just for comparison, the test file I've mostly been using is
> the one from https://emacs.stackexchange.com/q/598 which is a
> 18M single-line JSON file with 18922366 characters on one line,
> which opens in about a second on my machine when so-long has
> been enabled.

I think Mithraeum's point is a good one: it would be worthwhile also to
see what it is that the user wants to do with those kinds of files,
because the slowness is but one of the problems.

E.g. if you take your above example, maybe it does take just a second to
open in so-long-mode, but what happens when you do M-> ?

Is the "show the first screen" use case the main one (after which the
user will see that the file format is not the right one, will change it
(e.g. by pretty-printing it), and try again), or do we also want to care
about what happens if the user then goes on to navigate in this buffer?


        Stefan