Re: 1.23 prints some strange error

[Walter Alejandro Iglesias]

On Thu, Oct 05, 2023 at 10:08:50AM +0200, Walter Alejandro Iglesias wrote:
> On Wed, Oct 04, 2023 at 10:20:57PM +0000, Bjarni Ingi Gislason wrote:
> >   Latin1 iacute has the utf8 code 'í'
> > and the hexadecimal code is C3AD which is "LATIN CAPITAL LETTER A WITH
> > TILDE" and "SOFT HYPHEN"
> > 
> > "groff" turns "soft hyphen" into "HYPHEN-MINUS" (0x2D)
> > 
> > More is in the attachment.
> 
> > file list.tr
> > 
> > .hw a-hÃ-
> > .hw a-ño
> > .hw ár-bol
> > .hw cu-brÃ--a
> > .hw e-té-re-o
> > .hw ca-mión
> > .hw ú-te-ro
> > .hw pin-güi-no
> > 
> > Output from "preconv -e utf8  list.tr
> > 
> > .lf 1 list.tr
> > .hw a-h\[uFFFD]-
> > .hw a-\[u00F1]o
> > .hw \[u00E1]r-bol
> > .hw cu-br\[uFFFD]--a
> > .hw e-t\[u00E9]-re-o
> > .hw ca-mi\[u00F3]n
> > .hw \[u00FA]-te-ro
> > .hw pin-g\[u00FC]i-no
> > 
> > Translate "list.tr" to latin1
> > 
> > iconv -f utf8 -f latin1 list.tr
> > 
> > .hw a-h
> > 
> > iconv: illegal input sequence at position 7
> 
> The list.tr you're using in this example is already latin1.
> Repeat what you did but this time with a UTF-8 list.tr. :-)

Now I realize you used a latin1 list.tr on purpose, right?  If that's
the case, sorry!

If I feed preconv with a file already in latin1 (using UTF-8 locales
here) ...

  $ preconv -e utf8 list_in_latin1.tr

... *all* non ASCII characters in the output are replaced by \[uFFFD].

It seems that when I call the file as a macro (using .mso list.tr) that
particular utf8 character (iacute) is read wrongly by groff, presumably
because of what you noticed.


> 
> 
> > 
> > \[uFFFD] is called "replacement character".
> > 
> > -.-
> > 
> >   Latin1 iacute has the utf8 code 'í'
> > and the hexadecimal code is C3AD which is "LATIN CAPITAL LETTER A WITH
> > TILDE" and "SOFT HYPHEN"
> > 
> > "groff" turns "soft hyphen" into "HYPHEN-MINUS" (0x2D)
> 


-- 
Walter