hp-terminals on linux - HP UX

This is a discussion on hp-terminals on linux - HP UX ; Hi, is there anybody out there using hp-terminals (hpterm emulation our hw-terminals) together with ncurses on linux. If so please give me feedback or just another newsgroup if I'm off topic here. We do have problems with video enhancements. On ...

+ Reply to Thread
Results 1 to 19 of 19

Thread: hp-terminals on linux

  1. hp-terminals on linux

    Hi,

    is there anybody out there using hp-terminals (hpterm emulation our
    hw-terminals) together with ncurses on linux. If so please give me
    feedback or just another newsgroup if I'm off topic here.

    We do have problems with video enhancements. On HP systems anything
    works fine but they are using a different version of curses. If there
    is sombody using such a configuration I will discuss the problem with
    him in details.

    Thanks in advance,
    Bernd Rieke

  2. Re: hp-terminals on linux

    Bernd Rieke wrote:
    > Hi,
    >
    > is there anybody out there using hp-terminals (hpterm emulation our
    > hw-terminals) together with ncurses on linux. If so please give me
    > feedback or just another newsgroup if I'm off topic here.
    >
    > We do have problems with video enhancements. On HP systems anything
    > works fine but they are using a different version of curses. If there
    > is sombody using such a configuration I will discuss the problem with
    > him in details.


    It's more likely a difference in the terminfo entries. The HP escape
    sequences for video enhancement can have all kinds of timing subtleties
    which few people get right. So I suggest you take a working terminfo(4)
    entry from HP-UX and try it on Linux (perhaps you have to untic(1M) it
    on HP-UX and on Linux).

  3. Re: hp-terminals on linux

    Frank Slootweg schrieb am 18.01.2006 20:21:
    > Bernd Rieke wrote:
    >
    >>Hi,
    >>
    >>is there anybody out there using hp-terminals (hpterm emulation our
    >>hw-terminals) together with ncurses on linux. If so please give me
    >>feedback or just another newsgroup if I'm off topic here.
    >>
    >>We do have problems with video enhancements. On HP systems anything
    >>works fine but they are using a different version of curses. If there
    >>is sombody using such a configuration I will discuss the problem with
    >>him in details.

    >
    >
    > It's more likely a difference in the terminfo entries. The HP escape
    > sequences for video enhancement can have all kinds of timing subtleties
    > which few people get right. So I suggest you take a working terminfo(4)
    > entry from HP-UX and try it on Linux (perhaps you have to untic(1M) it
    > on HP-UX and on Linux).


    Frank,

    thank you very much for your answer. Since 30 years with HP (HP1000,
    HP3000, any HPUX) we have to go the Linux way, sorry HPUX. We have
    thousands of blockmode programs (starting with them on HP2640 terminals
    in 1974) and curses programs running without any problems on HPUX. So we
    took the program xhpterm (this emulator written for 3k) and after some
    fixes it runs well on Linux for all our blockmode applications. Even the
    curses programs run ok if startet on HPUX (telnet from Linux to HPUX).
    But when these programs (after they are ported to Linux) run on Linux
    the problem comes up. The termcap and everything around this is ok, we
    made any investigations even together with ncurses to bring this part
    down to the basics. It's the behaviour of ncurses on Linux which brings
    the problem. I'll try to explain it.

    Assume an input-field of 20 characters shown invers. If the letter x is
    typed in it's echoed back to the screen by ncurses as \e&dBx\e&db@ which
    turns of the invers enhancement just after the x till the end of the
    field. On HPUX with Hcurses the \e&d@ is not xmitted by Hcurses, only
    when the end of the field is reached which is ok (after typing in the
    20'th character). So the user of the program is confused because parts
    of the input field seem to disappear while typing (yes, only the
    enhancement disappers but for the user it's the same).

    We tried to debug the source code of ncurses but it's hard because these
    things are handled within the output optimizer of ncurses. Before we
    start to understand all these things I had the idea to ask the world if
    anybody has the same problem and probably there might be a solution.

    Seems ncurses thinks different while dealing with video enhancements. It
    assumes, while writing out an enhanced character, that it can switch the
    enhancement on for this character and switch it off just after it not
    affecting the rest of the field's enhancement. But on HP terminals the
    enhancement sequences work from current position to the end of the
    line/field resulting in \e&d@ switching off the invers enhancement just
    after the echoed character. This is different on emulations like xterms
    which are working fine because one can switch enhancements on/off for a
    single character not affecting things around (but we can't use xterms
    because they are missing the fkey handling and they can't handle roman8
    characters:-)).

    It's a really big issue for us so I have to find a fix for this. Ok, one
    can asked if it's time to bring these programs to the end. But the users
    love them because they can work very fast with them compared to a modern
    software (X11, web-based). And they are proofed and stable, an important
    feature of software even in these days...

    Thanks again for your reply,

    Greetings Bernd Rieke



  4. Re: hp-terminals on linux

    Bernd Rieke wrote:

    > Seems ncurses thinks different while dealing with video enhancements.
    > [...]
    > on HP terminals the enhancement sequences work from current position
    > to the end of the line/field resulting in \e&d@ switching off the
    > invers enhancement just after the echoed character. This is different
    > on emulations like xterms which are working fine because one can
    > switch enhancements on/off for a single character not affecting things
    > around (but we can't use xterms because they are missing the fkey
    > handling and they can't handle roman8 characters:-)).


    Both ncurses and (xfree86-)xterm are actively maintained by T.Dickey,
    you really might want to contact him about this issue.

  5. Re: hp-terminals on linux

    Sven Mascheck wrote:
    > Bernd Rieke wrote:


    >> Seems ncurses thinks different while dealing with video enhancements.
    >> [...]
    >> on HP terminals the enhancement sequences work from current position
    >> to the end of the line/field resulting in \e&d@ switching off the
    >> invers enhancement just after the echoed character. This is different
    >> on emulations like xterms which are working fine because one can
    >> switch enhancements on/off for a single character not affecting things
    >> around (but we can't use xterms because they are missing the fkey
    >> handling and they can't handle roman8 characters:-)).


    > Both ncurses and (xfree86-)xterm are actively maintained by T.Dickey,
    > you really might want to contact him about this issue.


    yes. He posted first to a newsgroup that's not carried on this provider (so
    google told me). google would have provided him with a contact address.

    There might be some problem with the terminfo (some of the function key
    bindings may/may not have a \r which I'm not sure is needed).

    But the main problem is the support for "magic cookies", for which Bernd Reike
    gave a description. A few implementations of curses provide this, but probably
    none except for HP's curses have been tested in the past ten years.

    ncurses has a configure script option to provide limited support for this
    feature. It is certainly possible to improve it, but I've had no feedback on
    the topic since making it workable several years ago.

    --
    Thomas E. Dickey
    http://invisible-island.net
    ftp://invisible-island.net

  6. Re: hp-terminals on linux

    Bernd Rieke writes:

    [...]
    > ok if startet on HPUX (telnet from Linux to HPUX). But when these programs
    > (after they are ported to Linux) run on Linux the problem comes up. The
    > termcap and everything around this is ok, we made any investigations even
    > together with ncurses to bring this part down to the basics. It's the
    > behaviour of ncurses on Linux which brings the problem. I'll try to explain
    > it.


    Depending on the version of Linux you have, [n]curses may have severe
    bugs. Recently I tried to port an application that was developed on a CADMUS
    (no idea which UNIX that was), and later ported to SunOS. Linux has incomplete
    documentation, implementation, and libraries it seems. Finally I quit.

    Back at the time when I had written that program, I also had ported it to
    MS-DOS, hacking the curses library that was available then. Actually, I'm
    really surprised how terrible current Linux curses libraries actually are. I'm
    only glad that both, vi and emacs work fine on almost every terminal.

    Regards,
    Ulrich
    [...]

  7. Re: hp-terminals on linux

    Ulrich Windl wrote:
    > Bernd Rieke writes:


    > [...]
    >> ok if startet on HPUX (telnet from Linux to HPUX). But when these programs
    >> (after they are ported to Linux) run on Linux the problem comes up. The
    >> termcap and everything around this is ok, we made any investigations even
    >> together with ncurses to bring this part down to the basics. It's the
    >> behaviour of ncurses on Linux which brings the problem. I'll try to explain
    >> it.


    > Depending on the version of Linux you have, [n]curses may have severe


    it's really amazing how these people pop up citing severe bugs,
    but never report them to me.

    regards.

    --
    Thomas E. Dickey
    http://invisible-island.net
    ftp://invisible-island.net

  8. Re: hp-terminals on linux

    Thomas Dickey wrote:
    > Ulrich Windl wrote:
    >> Depending on the version of Linux you have, [n]curses may have severe

    > it's really amazing how these people pop up citing severe bugs,
    > but never report them to me.


    aha. googling around, it's apparent that Ulrich is reporting his experiences
    in 1993, and doesn't have any relevant recent experience.

    (looking forward to bug reports)

    --
    Thomas E. Dickey
    http://invisible-island.net
    ftp://invisible-island.net

  9. Re: hp-terminals on linux

    Ulrich Windl wrote:

    > Recently I tried to port an application that was developed on a CADMUS
    > (no idea which UNIX that was),


    Offtopic:

    It was called "MUNIX", AFAIK almost completely a vanilla SVR3 (apart
    from the extensions for becoming CAD capable). That's a huge step...

  10. Re: hp-terminals on linux

    Thomas Dickey schrieb am 19.01.2006 12:02:
    > Sven Mascheck wrote:
    >
    >>Bernd Rieke wrote:

    >
    >
    >>>Seems ncurses thinks different while dealing with video enhancements.
    >>>[...]
    >>>on HP terminals the enhancement sequences work from current position
    >>>to the end of the line/field resulting in \e&d@ switching off the
    >>>invers enhancement just after the echoed character. This is different
    >>>on emulations like xterms which are working fine because one can
    >>>switch enhancements on/off for a single character not affecting things
    >>>around (but we can't use xterms because they are missing the fkey
    >>>handling and they can't handle roman8 characters:-)).

    >
    >
    >>Both ncurses and (xfree86-)xterm are actively maintained by T.Dickey,
    >>you really might want to contact him about this issue.

    >
    >
    > yes. He posted first to a newsgroup that's not carried on this provider (so
    > google told me). google would have provided him with a contact address.
    >
    > There might be some problem with the terminfo (some of the function key
    > bindings may/may not have a \r which I'm not sure is needed).
    >
    > But the main problem is the support for "magic cookies", for which Bernd Reike
    > gave a description. A few implementations of curses provide this, but probably
    > none except for HP's curses have been tested in the past ten years.
    >
    > ncurses has a configure script option to provide limited support for this
    > feature. It is certainly possible to improve it, but I've had no feedback on
    > the topic since making it workable several years ago.


    Hi Thomas,

    thank you for answering. I played around with that "magic-cookies" flag
    but had no success. I ran ./configure with the --enable-xmc-glitch
    switch and establihed a symlink hpterm-mc -> hpterm but the behaviour
    is still the same. Yes, I checked for having the right library bound to
    my test-program by inserting temporary some fprintf's in the curses
    modules behind the ifdefs of USE_XMC_SUPPORT.

    What else can I do? Is there a bug in my program where I create the pad?
    I can send you the source if you want to have a look at. Can I provide
    you with HP hardware, I have some old HP2648 graphical terminals, the
    first PC's on earth :-).

    Do you think there's a chance to fix it?

    Greetings Bernd Rieke
    PS: Sorry, forgot, I'm speaking about ncurses 5.4-68 which comes with
    SuSE-Linux 9.3.

  11. Re: hp-terminals on linux

    Bernd Rieke wrote:
    > Thomas Dickey schrieb am 19.01.2006 12:02:
    >> Sven Mascheck wrote:
    >>
    >>>Bernd Rieke wrote:

    >>
    >>
    >>>>Seems ncurses thinks different while dealing with video enhancements.
    >>>>[...]
    >>>>on HP terminals the enhancement sequences work from current position
    >>>>to the end of the line/field resulting in \e&d@ switching off the
    >>>>invers enhancement just after the echoed character. This is different
    >>>>on emulations like xterms which are working fine because one can
    >>>>switch enhancements on/off for a single character not affecting things
    >>>>around (but we can't use xterms because they are missing the fkey
    >>>>handling and they can't handle roman8 characters:-)).

    >>
    >>
    >>>Both ncurses and (xfree86-)xterm are actively maintained by T.Dickey,
    >>>you really might want to contact him about this issue.

    >>
    >>
    >> yes. He posted first to a newsgroup that's not carried on this provider (so
    >> google told me). google would have provided him with a contact address.
    >>
    >> There might be some problem with the terminfo (some of the function key
    >> bindings may/may not have a \r which I'm not sure is needed).
    >>
    >> But the main problem is the support for "magic cookies", for which Bernd Reike
    >> gave a description. A few implementations of curses provide this, but probably
    >> none except for HP's curses have been tested in the past ten years.
    >>
    >> ncurses has a configure script option to provide limited support for this
    >> feature. It is certainly possible to improve it, but I've had no feedback on
    >> the topic since making it workable several years ago.


    > Hi Thomas,


    > thank you for answering. I played around with that "magic-cookies" flag
    > but had no success. I ran ./configure with the --enable-xmc-glitch
    > switch and establihed a symlink hpterm-mc -> hpterm but the behaviour
    > is still the same. Yes, I checked for having the right library bound to
    > my test-program by inserting temporary some fprintf's in the curses
    > modules behind the ifdefs of USE_XMC_SUPPORT.


    I usually use the trace feature. The configure option --with-trace defines
    TRACE at compile-time, and tells the configure script to add calls to
    the _tracef() and related functions. Then I can set NCURSES_TRACE and
    see detailed results. (fprintf's are ok, but then you have to remove them).

    > What else can I do? Is there a bug in my program where I create the pad?
    > I can send you the source if you want to have a look at. Can I provide
    > you with HP hardware, I have some old HP2648 graphical terminals, the
    > first PC's on earth :-).


    that would be hard (to find places to put things, etc.). For my own testing,
    (since it's been quite a while since I had X on an HPUX), I used an
    experimental feature that I made in xterm. It may not be exact, but I can
    probably understand bug reports well enough to improve the cookie feature in
    xterm and provide better testing (and it's more compact than additional
    hardware).

    > Do you think there's a chance to fix it?


    yes. I spent some time yesterday reviewing what the magic cookie code does,
    and made some fixes to it.

    It "should" work properly for applications that are designed to work with magic
    cookies (that is, they have blanks in the proper places).

    ncurses' cookie support tries to find blanks for putting cookies in, and will
    also try to move things on the screen if no blanks exist. That movement is not
    entirely satisfactory - but let's start by assuming that it is not a problem.

    A copy of your program that I could use for testing also would be helpful.
    For my own testing, I used ncurses' test programs.

    > Greetings Bernd Rieke
    > PS: Sorry, forgot, I'm speaking about ncurses 5.4-68 which comes with
    > SuSE-Linux 9.3.


    well, for testing and fixing, you really need to work from source.
    The current release is 5.5.

    I store incremental patches from that in

    ftp://invisible-island.net/ncurses/5.5/

    Of course for testing purposes, it is best to install local libraries rather
    than try to reinstall the system libraries. For instance, I install
    static test-libraries for ncurses in /usr/local/ncurses

    --
    Thomas E. Dickey
    http://invisible-island.net
    ftp://invisible-island.net

  12. Re: hp-terminals on linux

    Thomas Dickey schrieb am 22.01.2006 21:46:
    > Bernd Rieke wrote:
    >
    >>Thomas Dickey schrieb am 19.01.2006 12:02:
    >>
    >>>Sven Mascheck wrote:
    >>>
    >>>
    >>>>Bernd Rieke wrote:
    >>>
    >>>
    >>>>>Seems ncurses thinks different while dealing with video enhancements.
    >>>>>[...]
    >>>>>on HP terminals the enhancement sequences work from current position
    >>>>>to the end of the line/field resulting in \e&d@ switching off the
    >>>>>invers enhancement just after the echoed character. This is different
    >>>>>on emulations like xterms which are working fine because one can
    >>>>>switch enhancements on/off for a single character not affecting things
    >>>>>around (but we can't use xterms because they are missing the fkey
    >>>>>handling and they can't handle roman8 characters:-)).
    >>>
    >>>
    >>>>Both ncurses and (xfree86-)xterm are actively maintained by T.Dickey,
    >>>>you really might want to contact him about this issue.
    >>>
    >>>
    >>>yes. He posted first to a newsgroup that's not carried on this provider (so
    >>>google told me). google would have provided him with a contact address.
    >>>
    >>>There might be some problem with the terminfo (some of the function key
    >>>bindings may/may not have a \r which I'm not sure is needed).
    >>>
    >>>But the main problem is the support for "magic cookies", for which Bernd Reike
    >>>gave a description. A few implementations of curses provide this, but probably
    >>>none except for HP's curses have been tested in the past ten years.
    >>>
    >>>ncurses has a configure script option to provide limited support for this
    >>>feature. It is certainly possible to improve it, but I've had no feedback on
    >>>the topic since making it workable several years ago.

    >
    >
    >>Hi Thomas,

    >
    >
    >>thank you for answering. I played around with that "magic-cookies" flag
    >>but had no success. I ran ./configure with the --enable-xmc-glitch
    >>switch and establihed a symlink hpterm-mc -> hpterm but the behaviour
    >>is still the same. Yes, I checked for having the right library bound to
    >>my test-program by inserting temporary some fprintf's in the curses
    >>modules behind the ifdefs of USE_XMC_SUPPORT.

    >
    >
    > I usually use the trace feature. The configure option --with-trace defines
    > TRACE at compile-time, and tells the configure script to add calls to
    > the _tracef() and related functions. Then I can set NCURSES_TRACE and
    > see detailed results. (fprintf's are ok, but then you have to remove them).
    >
    >
    >>What else can I do? Is there a bug in my program where I create the pad?
    >>I can send you the source if you want to have a look at. Can I provide
    >>you with HP hardware, I have some old HP2648 graphical terminals, the
    >>first PC's on earth :-).

    >
    >
    > that would be hard (to find places to put things, etc.). For my own testing,
    > (since it's been quite a while since I had X on an HPUX), I used an
    > experimental feature that I made in xterm. It may not be exact, but I can
    > probably understand bug reports well enough to improve the cookie feature in
    > xterm and provide better testing (and it's more compact than additional
    > hardware).
    >
    >
    >>Do you think there's a chance to fix it?

    >
    >
    > yes. I spent some time yesterday reviewing what the magic cookie code does,
    > and made some fixes to it.
    >
    > It "should" work properly for applications that are designed to work with magic
    > cookies (that is, they have blanks in the proper places).
    >
    > ncurses' cookie support tries to find blanks for putting cookies in, and will
    > also try to move things on the screen if no blanks exist. That movement is not
    > entirely satisfactory - but let's start by assuming that it is not a problem.
    >
    > A copy of your program that I could use for testing also would be helpful.
    > For my own testing, I used ncurses' test programs.
    >
    >
    >>Greetings Bernd Rieke
    >>PS: Sorry, forgot, I'm speaking about ncurses 5.4-68 which comes with
    >>SuSE-Linux 9.3.

    >
    >
    > well, for testing and fixing, you really need to work from source.
    > The current release is 5.5.
    >
    > I store incremental patches from that in
    >
    > ftp://invisible-island.net/ncurses/5.5/
    >
    > Of course for testing purposes, it is best to install local libraries rather
    > than try to reinstall the system libraries. For instance, I install
    > static test-libraries for ncurses in /usr/local/ncurses


    Thomas, just for clearness. Currently I don't know what that
    "magic-cookie" means. I saw this "token" in your thread and I played
    around with it within the ncurses source tree. The goal was to enable it
    and then to see if the behaviour changed. But at this time I have no
    idea what ist meant by "magic-cookies" (although I'm rather familiar
    with (HP) terminals because I had to fix all these bugs related to
    special HP features (blockmode. memory-lock, cursor addressing) within
    this freeware X-hpterm emulator we are running on linux).

    To make sure that we are speaking about the same problem I'll send you
    an email with that part of our program which is responsible for the
    handling of the input fields.

    Bernd Rieke




  13. Re: hp-terminals on linux

    Bernd Rieke wrote:

    > Thomas, just for clearness. Currently I don't know what that
    > "magic-cookie" means. I saw this "token" in your thread and I played
    > around with it within the ncurses source tree. The goal was to enable it
    > and then to see if the behaviour changed. But at this time I have no
    > idea what ist meant by "magic-cookies" (although I'm rather familiar
    > with (HP) terminals because I had to fix all these bugs related to
    > special HP features (blockmode. memory-lock, cursor addressing) within
    > this freeware X-hpterm emulator we are running on linux).


    I had encountered an xhpterm source, but (looking at it now) am not
    sure that I actually saw it running.

    > To make sure that we are speaking about the same problem I'll send you
    > an email with that part of our program which is responsible for the
    > handling of the input fields.


    thanks

    --
    Thomas E. Dickey
    http://invisible-island.net
    ftp://invisible-island.net

  14. Re: hp-terminals on linux

    Thomas Dickey schrieb am 22.01.2006 23:18:
    > Bernd Rieke wrote:
    >
    >
    >>Thomas, just for clearness. Currently I don't know what that
    >>"magic-cookie" means. I saw this "token" in your thread and I played
    >>around with it within the ncurses source tree. The goal was to enable it
    >>and then to see if the behaviour changed. But at this time I have no
    >>idea what ist meant by "magic-cookies" (although I'm rather familiar
    >>with (HP) terminals because I had to fix all these bugs related to
    >>special HP features (blockmode. memory-lock, cursor addressing) within
    >>this freeware X-hpterm emulator we are running on linux).

    >
    >
    > I had encountered an xhpterm source, but (looking at it now) am not
    > sure that I actually saw it running.


    Thomas, if you want I can make a tar from the source tree of our program
    and send it to you by mail (1 mbyte). May be you can bring it to work on
    Linux. For us it works fine. It handles all specialities of the HP
    terminals. Even the bugs of the hardware-terminals are emulated well...

    Bernd Rieke

  15. Re: hp-terminals on linux

    Bernd Rieke wrote:
    > Thomas Dickey schrieb am 22.01.2006 23:18:
    >> I had encountered an xhpterm source, but (looking at it now) am not
    >> sure that I actually saw it running.


    > Thomas, if you want I can make a tar from the source tree of our program
    > and send it to you by mail (1 mbyte). May be you can bring it to work on
    > Linux. For us it works fine. It handles all specialities of the HP
    > terminals. Even the bugs of the hardware-terminals are emulated well...


    yes, that would be useful (thanks).

    --
    Thomas E. Dickey
    http://invisible-island.net
    ftp://invisible-island.net

  16. Re: hp-terminals on linux

    Thomas Dickey writes:

    > Thomas Dickey wrote:
    > > Ulrich Windl wrote:
    > >> Depending on the version of Linux you have, [n]curses may have severe

    > > it's really amazing how these people pop up citing severe bugs,
    > > but never report them to me.

    >
    > aha. googling around, it's apparent that Ulrich is reporting his experiences
    > in 1993, and doesn't have any relevant recent experience.


    You must know. My experience was for SuSE Linux 9.3 (just to let you know).

    >
    > (looking forward to bug reports)


    I don't know by heart, but I think all the subwindow stuff is absent or
    broken, would you deny?

    Regards,
    Ulrich

    >
    > --
    > Thomas E. Dickey
    > http://invisible-island.net
    > ftp://invisible-island.net


  17. Re: hp-terminals on linux

    Thomas Dickey writes:

    > Bernd Rieke wrote:

    [...]

    > > Greetings Bernd Rieke
    > > PS: Sorry, forgot, I'm speaking about ncurses 5.4-68 which comes with
    > > SuSE-Linux 9.3.

    >
    > well, for testing and fixing, you really need to work from source.
    > The current release is 5.5.


    SuSE 10.0 (latest release) is shipping ncurses 5.4 just for your interest.
    I couldn't find out what the next release will ship.

    [...]
    Regards,
    Ulrich

  18. Re: hp-terminals on linux

    Ulrich Windl wrote:
    > Thomas Dickey writes:


    >> Thomas Dickey wrote:
    >> > Ulrich Windl wrote:
    >> >> Depending on the version of Linux you have, [n]curses may have severe
    >> > it's really amazing how these people pop up citing severe bugs,
    >> > but never report them to me.

    >>
    >> aha. googling around, it's apparent that Ulrich is reporting his experiences
    >> in 1993, and doesn't have any relevant recent experience.


    > You must know. My experience was for SuSE Linux 9.3 (just to let you know).


    that's a small datapoint...

    >> (looking forward to bug reports)


    > I don't know by heart, but I think all the subwindow stuff is absent or
    > broken, would you deny?


    without a test-case, I don't have any information.

    --
    Thomas E. Dickey
    http://invisible-island.net
    ftp://invisible-island.net

  19. Re: hp-terminals on linux

    Ulrich Windl wrote:
    > Thomas Dickey writes:


    >> Bernd Rieke wrote:

    > [...]


    >> > Greetings Bernd Rieke
    >> > PS: Sorry, forgot, I'm speaking about ncurses 5.4-68 which comes with
    >> > SuSE-Linux 9.3.

    >>
    >> well, for testing and fixing, you really need to work from source.
    >> The current release is 5.5.


    > SuSE 10.0 (latest release) is shipping ncurses 5.4 just for your interest.
    > I couldn't find out what the next release will ship.


    generally the rpm packagers do not ship the release version, but some
    development patch. What does "tic -V" say?

    --
    Thomas E. Dickey
    http://invisible-island.net
    ftp://invisible-island.net

+ Reply to Thread