ispell doesn’t seem to deal with hunspell reporting byte offsets

ispell doesn’t seem to deal with hunspell reporting byte offsets

[Nikolai Weibull]

Hi!

It seems that ispell doesn’t deal with hunspell reporting byte
offsets.  If I try to ispell a buffer containing

A “mispelled” word

I’ll get a

Ispell misalignment: word `mispelled' point 6; probably incompatible versions

error.  I’m guessing this is due to hunspell reporting the error at
byte 6, not 4 as ispell.el seems to want.

Any way of getting around this?

Re: ispell doesn’t seem to deal with hunspell reporting byte offsets

[Eli Zaretskii]

> Date: Tue, 23 Apr 2013 11:21:36 +0200
> From: Nikolai Weibull <now@disu.se>
> 
> It seems that ispell doesn’t deal with hunspell reporting byte
> offsets.

Emacs cannot "deal" with that, because the encoding emitted by
hunspell can be anything (it depends on the encoding used by the
dictionary, among other things).

The fact that byte offsets are reported is a bug in hunspell (for
which you can find a patch in the hunspell bug tracker).  You can also
find the patch here:

  http://debbugs.gnu.org/cgi/bugreport.cgi?bug=7781#31

> Any way of getting around this?

Fix hunspell with one of the patches for this bug.  That's the only
way that will work reliably.

Re: ispell doesn’t seem to deal with hunspell reporting byte offsets

[Nikolai Weibull]

On Tue, Apr 23, 2013 at 6:33 PM, Eli Zaretskii <eliz@gnu.org> wrote:
>> Date: Tue, 23 Apr 2013 11:21:36 +0200
>> From: Nikolai Weibull <now@disu.se>

>> It seems that ispell doesn’t deal with hunspell reporting byte
>> offsets.

>> Any way of getting around this?

> Fix hunspell with one of the patches for this bug.  That's the only
> way that will work reliably.

Thanks, Eli!  It now works fine.