[bug #52374] src: Migrate diagnostic messages away from `quotes like this'

[Bjarni Ingi Gislason]

Follow-up Comment #3, bug #52374 (project groff):


  I second that (with a comment):

1) it is in my list "It is time to modernize "groff""
[lists.gnu.org/archive/html/groff/2017-08/msg00054]

2) See the "GNU Coding Standards", paragraph 5.10 Quote Characters.

  Comment (UTF-8):

  I find that people too often _type_ (or copy) multibyte quotation marks
where the ASCII ones are totally sufficient.

  This caught my attention in "ChangeLogs" for Debian packages, where also
other multibyte punctuations were used.
As ASCII transliterations were used hitherto and still can be used.
Multibyte punctuations are not backward portable, but as people got the new
tool (toy) they forgot all about what they had used prior to that.

  Also the did not understand the demand to use UTF-8 encoding for
changelogs.
It is only needed to encode non-English (non-ASCII) letters.

  For examples of unnecessary (and not backwards portable characters) in the
Debian ChangeLogs see the last part of this writing.

  So please do not use multibyte characters for punctuations in patches
or changelogs.
The ASCII-ones are sufficient (and faster to type).

N.B.  I use an 8-bit locale as I do not need anything more.
And I see more that way!

###

punct.sed (with bytes both in hexadecimal and UTF-8 representation):

#!/bin/sed -f
s/\xE2\x80\x98/'/g  # u2018, left single quotation mark
s/\xE2\x80\x99/'/g  # u2019, right single quotation mark
#s/â^À^Ù/'/g  # u2019, right single quotation mark
s/\xE2\x80\x9C/"/g  # u201C, left double quotation mark
#s/â^À^Ü/"/g  # u201C, left double quotation mark
s/\xE2\x80\x9D/"/g  # u201C, left double quotation mark
#s/â^À^Ý/"/g  # u201D, right double quotation mark
s/\xE2\x80\xA2/*/g   # u2022, bullet, "iconv" uses 'o'
#s/â^À¢/*/g   # u2022, bullet, "iconv" uses 'o'
s/\xE2\x94\x81/-/g  # u2501, box drawings heavy horizontal
#s/â^Ô^Á/-/g  # u2501, box drawings heavy horizontal
s/\xC2\xA0/ /g     # u00A0, no-break space
#s/Â / /g     # u00A0, no-break space
s/\xE2\x80\x93/--/g # u2013, en-dash (as in TeX)
#s/â^À^Ó/--/g # u2013, en-dash (as in TeX)
s/\xE2\x80\x94/---/g # u2014, em-dash (as in TeX)
#s/â^À^Ô/---/g # u2014, em-dash (as in TeX)
s/\xE2\x96\xA1/*/g   # u25A1, square
#s/â^Ö¡/*/g   # u25A1, square
s/\xE2\x8B\xAF/.../g # u22EF, midline horizontal ellipsis (mathematical
operator) (wrong character)
#s/â^˯/.../g # u22EF, midline horizontal ellipsis (mathematical operator)
(wrong character)
s/\xE2\x80\xA6/.../g # u2026, horizontal ellipsis
#s/â^À¦/.../g # u2026, horizontal ellipsis
s/\xE2\x98\x86/*/g  # u2606, white star
#s/â^Ø^Æ/*/g  # u2606, white star
s/\xE2\x86\x92/->/g  # u2192, rightwards arrow
#s/â^Æ^Ò/->/g # u2192, rightwards arrow



    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?52374>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/