On Thu, 1 Feb 2024 12:39:40 +0100 (CET), "Carlos E. R." <robin.listas@telefonica.net> wrote:
On Thursday, 2024-02-01 at 11:16 -0000, Robert Webb via openSUSE Users wrote:
On Thu, 1 Feb 2024 09:12:45 +0000, Dave Howorth <dave@howorth.org.uk> wrote:
On Thu, 1 Feb 2024 00:53:24 -0600 -pj via openSUSE Users <users@lists.opensuse.org> wrote:
On 02-01-2024 12:31AM, David T-G wrote:
Carlos & PJ, et al -- [...]
Why is this still a question??!? Does nobody else, really, see that PJ is using the wrong arg flag (something like an underscore but it's an extended char that I don't know what it is)? davidtg@jpo:~> date | susepaste _t "with underscore" openSUSE Paste script
usage: susepaste [-f format] [-n nick] [-t title] [-e expire] [-s schema] [file] davidtg@jpo:~> date | susepaste -t "with hyphen" Paste failed :-(
Ignore the "failed" error for the moment; it doesn't matter. The important thing is that the script doesn't barf on the illegal character. Note that it takes a hyphen to indicate an argument. I don't know what kind of man page y'all are reading, but mine shows a minus sign.
I'm surprised nobody has confirmed what David is saying. He's right. For whatever reason it seems pj isn't using a normal hyphen character to introduce some of the arguments. Fix that and the commands should work.
Sorry, no, I don't see it. You must have very good eyesight.
But you are right. Good catch.
cer@Telcontar:~> hexdump -C Downloads/Thunderbird_downloads/Re\:\ susepasting\ -\ -pj\ via\ openSUSE\ Users\ \<users@lists.opensuse.org\>\ -\ 2024-02-01\ 0239.eml | less
00002820 74 6d 70 20 23 20 73 75 73 65 70 61 73 74 65 20 |tmp # susepaste | 00002830 3d 45 32 3d 38 30 3d 39 30 74 20 22 2d 70 6a 20 |=E2=80=90t "-pj | 00002840 73 74 72 61 63 65 2e 74 78 74 22 20 3d 45 32 3d |strace.txt" =E2=| 00002850 38 30 3d 0d 0a 3d 39 30 65 20 22 31 30 38 30 22 |80=..=90e "1080"| 00002860 20 2d 6e 0d 0a 22 70 6a 22 20 27 73 74 72 61 63 | -n.."pj" 'strac| 00002870 65 2e 74 78 74 27 0d 0a 6f 70 65 6e 53 55 53 45 |e.txt'..openSUSE| 00002880 20 50 61 73 74 65 20 73 63 72 69 70 74 0d 0a 0d | Paste script...|
For purposes of discussion, we need to have common terminology. The character used to introduce command-line options is what Wikipedia calls "Hyphen-minus". [1] Other characters that look like it are the Hyphen, Non-breaking hyphen, Minus sign, En dash, and Em dash.
I know that it is the ascii-7 "hyphen", char(0x2d) :-)
cer@Telcontar:~> echo "-" | hexdump -C 00000000 2d 0a |-.| 00000002 cer@Telcontar:~>
If PJ is using something that isn't the proper hyphen-minus character, then maybe it is because he copy-pasted it right from the f'ed-up susepaste man page.
Some of you will see that none of the "hyphen-whatevers" shown for the option descriptions and the examples in the man page are the actual hyphen-minus. Others won't. It depends on your locale settings.
cer@Telcontar:~> man susepaste | hexdump -C | less
000002e0 20 20 20 2d 6e 20 4e 49 43 4b 0a 20 20 20 20 20 | -n NICK. |
00000320 20 20 20 20 20 2d 74 20 54 49 54 4c 45 0a 20 20 | -t TITLE. |
00002310 0a 0a 20 20 20 20 20 20 20 2d 65 20 45 58 50 49 |.. -e EXPI|
00002580 65 72 0a 0a 20 20 20 20 20 20 20 2d 73 20 53 43 |er.. -s SC|
I don't see the error in "my" man page, sorry.
Right. Instead of hex "2d", as you properly have, I see the UTF-8 sequence "e2 80 90" (Unicode U+2010 Hyphen) in "my" man page.
What do you think of this?: $ man susepaste |col -bx |grep -o -e '-' |wc -l 1
Yes, there is just one hyphen-minus in the whole man page. It is in the title line.
For me: $ locale LANG=en_US.UTF-8 LC_CTYPE="en_US.UTF-8" LC_NUMERIC="en_US.UTF-8" LC_TIME="en_US.UTF-8" LC_COLLATE=C LC_MONETARY="en_US.UTF-8" LC_MESSAGES="en_US.UTF-8" LC_PAPER="en_US.UTF-8" LC_NAME="en_US.UTF-8" LC_ADDRESS="en_US.UTF-8" LC_TELEPHONE="en_US.UTF-8" LC_MEASUREMENT="en_US.UTF-8" LC_IDENTIFICATION="en_US.UTF-8" LC_ALL=
cer@Telcontar:~> locale LANG=en_US.UTF-8 LC_CTYPE=en_US.UTF-8 LC_NUMERIC=es_ES.utf8 LC_TIME=en_DK.UTF-8 LC_COLLATE=POSIX LC_MONETARY=es_ES.utf8 LC_MESSAGES="en_US.UTF-8" LC_PAPER=es_ES.utf8 LC_NAME=es_ES.utf8 LC_ADDRESS="en_US.UTF-8" LC_TELEPHONE=es_ES.utf8 LC_MEASUREMENT=es_ES.utf8 LC_IDENTIFICATION="en_US.UTF-8" LC_ALL= cer@Telcontar:~>
I'm surprised that your LC_CTYPE value is the same as mine. Isn't that the one that should determine character encodings? When I tried setting LC_ALL=C, the man page was OK (using hex 2d). OK also with LC_CTYPE=C . I only noticed the problem recently, with not just the susepaste man page BTW, so maybe there is a difference between Leap and Tumbleweed (which I am using). Some man pages are OK, though, and some use mixed hyphen-minus characters in some places, and then semi-randomly other "hyphens" elsewhere (man rsync).
-- Robert Webb