[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [h-e-w] Processing chars above \200

From: Eli Zaretskii
Subject: Re: [h-e-w] Processing chars above \200
Date: Sun, 23 Sep 2018 23:03:29 +0300

> Date: Sun, 23 Sep 2018 15:26:08 -0400
> From: John J. Xenakis <address@hidden>
> Cc: address@hidden
> >   But in any case, there are commands to fix those errors right
> >   away, as soon as you realize something like that happens.  We will
> >   get to that, once I understand more about the problem.
> Could you tell me what those commands are?

"C-x RET r" is the most important one.  It prompts you for an encoding
(with completion), and then re-visits the file using that encoding.

> By the way, how do I encode that keyboard string in Lisp?  How does
> one use "(universal-coding-system-argument CODING-SYSTEM)" in a macro?

In a macro, you just type "C-x RET c WHATEVER".  In Lisp, you need to
let-bind coding-system-for-read to the appropriate coding-system
symbol, like this:

  (let ((coding-system-for-read 'windows-1250))

> OK, you can download the following:
> The enclosed .txt file causes all the issues that I've described.

This file has 2 problems:

  . It includes a null byte.  When Emacs sees a null byte, it treats
    the file as binary, and in particular inhibits decoding it.  If
    you need to work with such files, you can set
    inhibit-null-byte-detection to t.  But it is best to just delete
    such bytes and then re-visit the file, because I doubt that you
    need them there, they probably got there due to some snafu.

  . Few of the non-ASCII characters in the file cannot be encoded by
    your locale's default encoding.  What does the following command
    display in "emacs -Q":

      M-: (default-value 'buffer-file-coding-system) RET

    I'm guessing it says something like iso-latin-1-dos, and that
    encoding cannot decode characters whose codes are between \200 and
    \237.  If that is the case, the following setting in your .emacs
    should solve these problems:

      (prefer-coding-system 'windows-1252)

> To make is easy for you to find some of the 8-bit characters causing
> the problems, I inserted the string ">>>" in front of four lines
> containing them.

Thanks, but there's no need, it's easy to use this trick:

  M-: (skip-chars-forward "\000-\177") RET

reply via email to

[Prev in Thread] Current Thread [Next in Thread]