bug#40407: [PATCH] slow ENCODE_FILE and DECODE_FILE

[OGAWA Hirofumi]

Eli Zaretskii <eliz@gnu.org> writes:

>> > I prefer to wait for Handa-san's response, and meanwhile install the
>> > least disruptive change, which just fixes the one aspect that got
>> > broken.  Call me a coward, if you wish.
>> 
>> If so, the least disruptive change by far would be the two calls to 
>> coding-system-put in my patch (along with the tests, of course).
>
> Fine with me, but please say something about why we use 'put' instead
> of specifying :ascii-compatible-p directly in the coding-system's
> definition, and also add a FIXME there, since I hope this is not the
> final word on that matter.

BTW, 

Mattias Engdegård <mattiase@acm.org> writes:

>  (define-coding-system 'utf-7-imap
>    "UTF-7 encoding of Unicode, IMAP version (RFC 2060)"
> -  :coding-type 'utf-8
> +  :coding-type 'charset
> +  :charset-list '(ascii)
>    :mnemonic ?u
> -  :charset-list '(unicode)
>    :pre-write-conversion 'utf-7-imap-pre-write-conversion
>    :post-read-conversion 'utf-7-imap-post-read-conversion)
> +;; See comment for utf-7 above.
> +(coding-system-put 'utf-7-imap :ascii-compatible-p nil)

(check-coding-systems-region "あ" nil '(utf-7-imap))
=> ((utf-7-imap 0))

It says "cannot encodable by utf-7-imap", so looks like ":charset-list
'(ascii)" doesn't work at least.

Thanks.
-- 
OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>