[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [bug-gnu-libiconv] CP 850
|
From: |
Bruno Haible |
|
Subject: |
Re: [bug-gnu-libiconv] CP 850 |
|
Date: |
Sat, 01 Apr 2017 20:08:05 +0200 |
|
User-agent: |
KMail/5.1.3 (Linux/4.4.0-70-generic; KDE/5.18.0; x86_64; ; ) |
Hi,
> I'm looking at the file tests/CP850.TXT and get the impression that the
> codes 0-31 are mapped to the same character in ASCII/UTF8 ?
Yes.
> If this is the case, then that is not correct. Except for tab, linefeed,
> carriage return, those codes are graphical characters in CP 850.
Quoting Wikipedia [1]:
"code points 1–31 and 127 (01–1Fhex and 7Fhex) may be either ASCII control
characters or code page 437 graphics, depending on context."
See also [2].
In software that accesses fonts, it may be preferable to consider these
code points as graphic characters.
In software that converts text files, it is preferable to assume ASCII
compatibility.
GNU libiconv is in the latter camp.
Bruno
[1] https://en.wikipedia.org/wiki/Code_page_850
[2] https://en.wikipedia.org/wiki/Code_page_437#Characters