Loading ...
Sorry, an error occurred while loading the content.

Re: [linux] Re: Does anyone have a distro in which the seq command/program works?

Expand Messages
  • ed
    ... Around 2004-ish when I was building OpenBSD firewalls I had something similar to emulate the seq program in sh. The correct thing to do would have been to
    Message 1 of 9 , Jul 9 1:01 AM
    • 0 Attachment
      On Tue, Jul 09, 2013 at 07:48:17AM -0000, thad_floryan wrote:
      > Arggh! Facepalm and headslap. I'm removing those lines ASAP from
      > all systems (sigh, maybe tomorrow after some sleep) since that repeat
      > function is of NO use to me whatsoever. Actually I'll comment them
      > out with a note they hose the seq progrsm. This is gonna take awhile
      > to boot up some 30 systems and comment-out that code everywhere.

      Around 2004-ish when I was building OpenBSD firewalls I had something
      similar to emulate the seq program in sh. The correct thing to do would
      have been to locate packages or build coreutils, but due to laziness I
      emulated it in sh one way or another. It bit me though somehow, I can't
      remember the details but it was non-trivial and ended up producing
      duff output.

      > Thanks again!

      No problem! Glad it's sorted.

      --
      Best regards,
      Ed http://www.s5h.net/
    • thad_floryan
      ... Hi Ed, When I run seq 10 , seq 1 10 or seq 1 1 10 I m now getting one number per line over 10 lines which makes perfect sense for use in shell scripts
      Message 2 of 9 , Jul 9 2:42 AM
      • 0 Attachment
        --- In linux@yahoogroups.com, ed <ed@...> wrote:
        > On Tue, Jul 09, 2013 at 07:48:17AM -0000, thad_floryan wrote:
        > > Arggh! Facepalm and headslap. I'm removing those lines ASAP from
        > > [...]
        > > Thanks again!
        >
        > No problem! Glad it's sorted.

        Hi Ed,

        When I run 'seq 10', 'seq 1 10' or 'seq 1 1 10' I'm now getting one
        number per line over 10 lines which makes perfect sense for use in
        shell scripts and the floating point capability works fine too after
        commenting the errant lines in the .bash_aliases file(s):

        $ seq 1 .5 5
        1.0
        1.5
        2.0
        2.5
        3.0
        3.5
        4.0
        4.5
        5.0

        Hard to believe I haven't used seq in many years since I first found
        it circa the late 1980s -- I wonder if it was in the Usenet archives
        for comp.sources.unix? The only extant list of archives is here:

        http://ftp.sunet.se/pub/usenet/ftp.uu.net/comp.sources.unix/

        and that doesn't even include, for example, my tprobe program which
        is in volume 26 (1992) per:

        http://ae-www.technion.ac.il/pkgs/g-k/in/tprobe/tprobe

        and 'seq' is not even in this list:

        http://ae-www.technion.ac.il/pkgs/o-s/

        Most of the archive sites have gone belly-up at a great loss of good
        software. I'm going to need to check the Wayback Machine later this
        week. They have their own problems with the US Government's National
        Security Letters as Brewster Kahle (founder of the nonprofit
        Internet Archive, perhaps the greatest of our digital libraries, and
        of the Wayback Machine, which allows you to browse an archive of the
        Web that reaches back to 1996) relates in this article:

        "What It's Like to Get a National-Security Letter"

        http://www.newyorker.com/online/blogs/elements/2013/06/what-its-like-to-get-a-national-security-letter.html

        The US government is clearly going ape-/bat-s*** crazy nowadays.

        Thad
      • Doug
        ... Ran the commands on pclos KDE 32-bit. All commands work, but print the output one line at a time. 1 2 3 and so on. --doug -- Blessed are the
        Message 3 of 9 , Jul 9 10:09 AM
        • 0 Attachment
          On 07/09/2013 03:28 AM, ed wrote:
          >
          >
          > On Tue, Jul 09, 2013 at 12:22:47AM -0700, Thad Floryan wrote:
          >> [...]
          >> 1st: $ seq 10
          >> bash: [: 10: unary operator expected
          >>
          >> 2nd: $ seq 1 10
          >> 1 2 3 4 5 6 7 8 9 10
          >>
          >> 3rd: $ seq 1 1 10
          >> 1
          >> [...]
          >> If you would please perform the 1st, 2nd and 3rd examples and report
          >> back if the output of all 3 is identical, we can pin this down and
          >> hopefully get it fixed.

          Ran the commands on pclos KDE 32-bit. All commands work, but print the
          output one line at a time.
          1
          2
          3
          and so on.
          --doug
          --
          Blessed are the peacemakers..for they shall be shot at from both sides.
          --A.M.Greeley
        • J
          ... I must admit, sheepishly, to laughing hysterically at Thad s misfortune. Thanks for brightening my day :) I laugh mainly because I ve done the same thing
          Message 4 of 9 , Jul 9 11:54 AM
          • 0 Attachment
            On Tue, Jul 9, 2013 at 3:48 AM, thad_floryan <thad@...> wrote:
            > Arggh! Facepalm and headslap. I'm removing those lines ASAP from
            > all systems (sigh, maybe tomorrow after some sleep) since that repeat
            > function is of NO use to me whatsoever. Actually I'll comment them
            > out with a note they hose the seq progrsm. This is gonna take awhile
            > to boot up some 30 systems and comment-out that code everywhere.

            I must admit, sheepishly, to laughing hysterically at Thad's
            misfortune. Thanks for brightening my day :)

            I laugh mainly because I've done the same thing in the past by writing
            functions or specialized aliases into .bashrc and then forgetting
            about them down the road. Thankfully, I never had the amount of
            hardware Thad has so fixing the issue when it bit me was a matter of
            only three or four systems. But I do feel your pain and sympathise :)

            Jeff
          • thad_floryan
            ... Hi Jeff, You re very welcome! This is how we learn. For the life of me I don t know for how many years/decades those lines in .bash_aliases have been
            Message 5 of 9 , Jul 10 2:54 AM
            • 0 Attachment
              --- In linux@yahoogroups.com, J <dreadpiratejeff@...> wrote:
              > On Tue, Jul 9, 2013 at 3:48 AM, thad_floryan <thad@...> wrote:
              > > Arggh! Facepalm and headslap. I'm removing those lines ASAP from
              > > all systems (sigh, maybe tomorrow after some sleep) since that
              > > repeat function is of NO use to me whatsoever. Actually I'll
              > > comment them out with a note they hose the seq progrsm. This is
              > > gonna take awhile to boot up some 30 systems and comment-out that
              > > code everywhere.
              >
              > I must admit, sheepishly, to laughing hysterically at Thad's
              > misfortune. Thanks for brightening my day :)

              Hi Jeff,

              You're very welcome! This is how we learn. For the life of me I
              don't know for how many years/decades those lines in .bash_aliases
              have been copied from system to system much like my .appt, .emacs,
              .less, .xfigrc, .xinitrc, and other files have been.

              The .bash_aliases on one of the first systems I just "fixed" was
              dated 1-October-2008 and I know other sytems' files are w-a-y older
              than that one.

              The important thing to learn from this is better programming and the
              use of variables that absolutely and positively can/should not have
              any conflicts with other things on a system.

              I've written 10,000+ line scripts for clients to do special things,
              one of the most interesting was one for Sigaba to create installable
              packages for Solaris systems. Key points are these:

              1. explicitly define any program that is used in a script with all
              those definitions near the script's beginning, and

              2. variables and such used in a script must be unique to the script

              A quick example of (1) is the following:

              ECHO=/usr/bin/echo
              SEQ=/usr/bin/seq
              ...

              which should ONLY be used in the script in this fashion:

              $SEQ 10

              Several quick examples of (2) are the following to forms with the
              name of the script being the first part of the variable name or the
              second form with a "_" prefix to all variable names to help assure
              there's no conflicts with anything else on the system:

              SCRIPT_I=0
              _foo=123
              SCRIPT_I=$_foo
              $ECHO $SCRIPT_I
              123

              That may seem like extra and unnecessary typing, but when you look
              at large scripts in toto it's clear what's being done where and the
              items are also easier to search for in an editor when they cannot be
              confused with anything else (e.g., comments and other text).

              Yes, I understand that's a "style", but it's one that's served me well
              for many years.

              > I laugh mainly because I've done the same thing in the past by
              > writing functions or specialized aliases into .bashrc and then
              > forgetting about them down the road. Thankfully, I never had the
              > amount of hardware Thad has so fixing the issue when it bit me was
              > a matter of only three or four systems. But I do feel your pain
              > and sympathise :)

              It's not that bad and gives me an opportunity to verify some of the
              older systems still boot. I just had a problem with one having its
              CMOS/RTC battery die so I need to check all systems and replace with
              fresh batteries as required. Seems to be a mix of CR2032 and CR1220
              (all the same voltage but different diameters and thicknesses). This
              is a good reference:

              https://en.wikipedia.org/wiki/List_of_battery_sizes

              as is this (all of which are PDFs on the following page):

              http://www.digikey.com/catalog/en/partgroup/cr-coin-series/3437

              Thad
            • thad_floryan
              ... Hi Doug, Thank you for checking. And I should have caught the fact the test I ran had the numbers 1-10 horizontally which isn t very useful in a script
              Message 6 of 9 , Jul 10 3:02 AM
              • 0 Attachment
                --- In linux@yahoogroups.com, Doug <dmcgarrett@...> wrote:
                > On 07/09/2013 03:28 AM, ed wrote:
                > >
                > > On Tue, Jul 09, 2013 at 12:22:47AM -0700, Thad Floryan wrote:
                > >> [...]
                > >> 1st: $ seq 10
                > >> bash: [: 10: unary operator expected
                > >>
                > >> 2nd: $ seq 1 10
                > >> 1 2 3 4 5 6 7 8 9 10
                > >>
                > >> 3rd: $ seq 1 1 10
                > >> 1
                > >> [...]
                > >> If you would please perform the 1st, 2nd and 3rd examples and
                > >> report back if the output of all 3 is identical, we can pin this
                > >> down and hopefully get it fixed.
                >
                > Ran the commands on pclos KDE 32-bit. All commands work, but print
                > the output one line at a time.
                > 1
                > 2
                > 3
                > and so on.

                Hi Doug,

                Thank you for checking. And I should have "caught" the fact the test
                I ran had the numbers 1-10 horizontally which isn't very useful in a
                script needing numbers sequentially for, say, a loop's iteration.

                The intervention of bash in my test should also have alerted me to
                the source of the problem -- I was simply too weary to fathom that
                at the time.

                After several naps this afternoon I'm catching up on sleep so I should
                be back to normal Wednesday (today).

                Thad
              Your message has been successfully submitted and would be delivered to recipients shortly.