• binkd 1.0.4 vs 1.1a

    From Carlos Navarro@2:341/66.3463 to All on Fri May 21 21:30:54 2021
    Hi,

    I am using BinkD 1.0.4 because AFAIK it is the latest stable version. However, most nodes I connect to are using 1.1a-112.

    Isn't that an alpha version? Is it recommended to use that one instead of stable?

    What's new in 1.1a since 1.0?

    Carlos

    --- GoldED+/W32-MSVC 1.1.5-b20180707
    * Origin: (2:341/66.3463)
  • From Björn Felten@2:203/2 to Carlos Navarro on Fri May 21 22:39:50 2021
    Isn't that an alpha version? Is it recommended to use that one instead
    of stable?

    I've been using this version for almost seven years now without any problems:

    R:\binkd>binkd -v
    Binkd 1.1a-65 (Sep 21 2014 18:59:33/Win32)

    Stable enough for me... 8-)




    ..

    --- Mozilla/5.0 (Windows; U; Windows NT 5.1; sv-SE; rv:1.9.1.16) Gecko/20101125
    * Origin: news://eljaco.se (2:203/2)
  • From Alan Ianson@1:153/757.2 to Carlos Navarro on Fri May 21 18:40:44 2021
    Re: binkd 1.0.4 vs 1.1a
    By: Carlos Navarro to All on Fri May 21 2021 09:30 pm

    I am using BinkD 1.0.4 because AFAIK it is the latest stable version. However, most nodes I connect to are using 1.1a-112.

    I was using that for a long time too.. but I am not sure the release version is getting any love anymore so I moved to the git 1.1a version.

    Isn't that an alpha version? Is it recommended to use that one instead of stable?

    I find it good and stable.

    What's new in 1.1a since 1.0?

    Here's the newest that I see in the HISTORY file..


    2021/01/19 11:07:20 1.1a-112 git
    breaksig.c,2.26,2.27
    Graceful exit when idle for Windows builds, for Nick Andre.
    Set the binkd_exit flag to TRUE when signaled to exit and return TRUE (indicating that the signal was handled by the process) thus mimicking the behavior of the *nix builds' handling of termination signals.

    2021/01/19 11:07:11 git
    Makefile,2.55,2.56
    Define HAS_SNPRINTF for modern MSVC build compatibility (e.g. MSVC2019)
    Fix build error using "nmake MSVC_VER=10" and MSVC2019:

    Linking binkd.exe...
    snprintf.obj : error LNK2005: _snprintf already defined in readcfg.obj bin\msvc10-binkd\binkd.exe : fatal error LNK1169: one or more multiply defined symbols found

    2020/05/26 20:48:08 1.1a-111 git
    binkd.c,2.127,2.128
    Print "Facilities" without space at the end of line. Make sense for further "bi nkd.exe -vv>file_id.diz" processing. (#21)

    2020/05/25 14:12:23 git
    configure,2.57,2.58 configure.in,2.57,2.58
    Fix check perl multiplicity in configure

    2020/05/25 08:48:41 1.1a-110 git
    readcfg.c,2.116,2.117
    Revert 4a528544 (perl multiplicity)

    2020/05/17 23:32:09 1.1a-109 git
    binkd.c,2.126,2.127 perlhooks.c,2.95,2.96 perlhooks.h,2.18,2.19 readcfg.c,2.115 ,2.116
    Call PERL_SYS_INIT3() only once

    2020/05/17 20:52:52 1.1a-108 git
    exitproc.c,2.48,2.49
    tab to spaces

    2020/05/17 20:52:14 1.1a-107 git
    readcfg.c,2.114,2.115
    Require PERL_MULTIPLICITY for run multiple perl interpreters

    2020/05/17 20:52:07 1.1a-106 git
    perlhooks.c,2.94,2.95
    Set PL_perl_destruct_level to 1 for correct restart perl interpreter Signed-off-by: Pavel Gulchuk <gul@gul.kiev.ua>

    2020/05/17 20:11:12 1.1a-105 git
    perlhooks.c,2.93,2.94
    Fix init perlsem

    2020/04/30 21:30:12 git
    Makefile,1.1,1.2
    [DOS] fix making directory tree for DOS build

    2020/04/29 09:03:10 1.1a-104 git
    ftnnode.c,2.51,2.52 ftnq.c,2.44,2.45 protocol.c,2.237,2.238
    Silence warnings: remove superfluous nullity checks
    struct `FTN_NODE` defined in `btypes.h` contains
    the member `pwd`, which is a character array.
    Several places in the code, this array was checked
    for nullity; however, this was always done in a place
    where the validity of the referring `FTN_NODE` struct
    pointer had already been established, and as `pwd` is
    an array (and not a generic pointer) it can never be
    NULL. These superflous checks caused warnings when
    compiling under `clang`.

    This change removes those checks to silence these warnings.

    Additionally, at those call sites often the code called
    `strcmp` to compare `p->pwd` against some other string,
    invariably inside of a conditional. Modify the
    conditionals to always explicity compare the value
    returned from `strcmp` against e.g. 0.

    Signed-off-by: Dan Cross <cross@fat-dragon.org>

    2020/04/25 15:53:52 git
    Makefile,2.28,2.29 Makefile.emo,1.11,1.12 Makefile.klibc,1.6,1.7
    [OS/2] change .EXE naming scheme and library definitions

    2020/04/25 15:53:47 1.1a-103 git
    ftnq.c,2.43,2.44 run.c,2.15,2.16 sys.h,2.47,2.48
    compile Binkd for DOS with djgpp 2.05 and watt32 library

    2020/04/25 15:53:17 git
    Makefile,NONE,1.1
    compile Binkd for DOS with djgpp 2.05 and watt32 library

    2020/04/25 15:53:10 1.1a-102 git
    client.c,2.110,2.111
    Fix build for OS/2 and Windows 9x on ancient compilers
    Original patch by Dan Cross,
    additional testing and improvements by Max Vasilyev
    Neither Win9x nor OS/2 define `struct sockaddr_storage`.
    To work around this, we resort to casts to make the
    `ss_family` member of that struct look like the `sa_family`
    member in a `struct sockaddr`. According to POSIX, all
    of these structs should be defined in such a way that
    this will work without surprises.

    2020/04/05 20:33:09 git
    README.md,2.2,2.3
    Update README.md
    Fixed incorrect instructions and beautified the steps into a tested and verifie d cookbook style Howto.

    2020/01/30 10:21:49 1.1a-101 git
    client.c,2.109,2.110
    Fix an out-of-bounds error on sockaddrs.
    The `invalidAddresses` vector in `client.c` is used
    to hold invalid addresses `binkd` should not use.
    However, the array was of type `struct sockaddr`,
    which is not large enough to hold all of the
    protocol-specific data of the `struct sockaddr_*`
    structures. As a result, `binkd` would access
    out-of-bounds memory when examining elements of
    the array.

    Fixed by redefining `invalidAddresses` to be an
    array of type `struct sockaddr_storage`, which is
    guaranteed by POSIX to be large enough to hold all
    data associated with a socket address for any
    protocol family.

    Signed-off-by: Dan Cross <cross@fat-dragon.org>

    2020/01/30 10:21:13 1.1a-100 git
    client.c,2.108,2.109 ftnnode.c,2.50,2.51 iphdr.h,2.28,2.29 protocol.c,2.236,2.2 37 readcfg.c,2.113,2.114 readcfg.h,2.44,2.45
    Fix use-after-free bug in get_host_and_port.
    Don't use pointer assignment in this function,
    but rather, copy into a fixed-length buffer.

    Fixes #15

    Signed-off-by: Dan Cross <cross@fat-dragon.org>

    Ttyl :-),
    Al

    ... Always remember you're unique - just like everyone else.
    --- SBBSecho 3.14-Linux
    * Origin: The Rusty MailBox - Penticton, BC Canada (1:153/757.2)
  • From Carlos Navarro@2:341/66.3463 to Bj÷rn Felten on Sat May 22 08:47:34 2021
    21 May 2021 22:39, you wrote to me:

    I've been using this version for almost seven years now without any problems:

    R:\binkd> binkd -v
    Binkd 1.1a-65 (Sep 21 2014 18:59:33/Win32)

    Stable enough for me... 8-)

    Did you start using Binkd with version 1.1a? Or did you upgrade from 1.0.x or older?

    Carlos

    --- GoldED+/W32-MSVC 1.1.5-b20180707
    * Origin: .:. (2:341/66.3463)
  • From Carlos Navarro@2:341/66.3463 to Daniel Path on Sat May 22 08:49:22 2021
    21 May 2021 22:29, you wrote to me:

    What's new in 1.1a since 1.0?

    i've put the changelog for you here from the debian distribution:

    https://tentacle.studio64.hu/~danielp/binkd-changelog.gz

    Got it. Thank you, but think I will not read a 2800+line text file of commits since version 0.9.6 :-)
    I just wanted to know some new feature...

    Carlos

    --- GoldED+/W32-MSVC 1.1.5-b20180707
    * Origin: .:. (2:341/66.3463)
  • From Carlos Navarro@2:341/66.3463 to Alan Ianson on Sat May 22 08:50:54 2021
    21 May 2021 18:40, you wrote to me:

    I am using BinkD 1.0.4 because AFAIK it is the latest stable
    version. However, most nodes I connect to are using 1.1a-112.

    I was using that for a long time too.. but I am not sure the release version is getting any love anymore so I moved to the git 1.1a
    version.

    Ok but... did you remember if there was anything in 1.1a that you missed in 1.0.4 when you upgraded, or was it just a love affair? :-)

    Isn't that an alpha version? Is it recommended to use that one
    instead of stable?

    I find it good and stable.

    I assume it is, as I see many (most?) Binkd nodes are using 1.1a.

    What's new in 1.1a since 1.0?

    Here's the newest that I see in the HISTORY file..

    [snip]

    OMG... I just want to know which new features it has, not another list of commits... ':-)

    Carlos

    --- GoldED+/W32-MSVC 1.1.5-b20180707
    * Origin: .:. (2:341/66.3463)
  • From Oli@2:280/464.47 to Carlos Navarro on Sat May 22 08:12:22 2021
    Carlos wrote (2021-05-21):

    Hi,

    I am using BinkD 1.0.4 because AFAIK it is the latest stable version. However, most nodes I connect to are using 1.1a-112.

    Isn't that an alpha version? Is it recommended to use that one instead of stable?

    Is it? I really don't know. There was a time were versions like 2.3g were not uncommon. I guess 1.1a-x was meant as a snapshot from the current branch and the official release would become 1.1.0.

    There is not much developing going on. It's mostly a bug fix or other minor changes every couple of month. In that sense it's pretty stable. I think you also could call it a rolling release.

    What's new in 1.1a since 1.0?

    No need to wait for a "stable" 1.1 release, it's stable enough. Just make sure you get at least 1.1a-101, 1.1a-99 and below had some nasty bugs.

    ---
    * Origin: . (2:280/464.47)
  • From Alan Ianson@1:153/757 to Carlos Navarro on Sat May 22 00:05:50 2021
    Hello Carlos,

    OMG... I just want to know which new features it has, not another list
    of commits... ':-)

    Aside from the few commits I listed there was also some changes around IPv6 that Michiel had asked for, but I don't quite remember what they were now.

    It's mostly code fixes or beautification more than new features.

    I found 1.0.4 good and stable but it doesn't include the above.

    Ttyl :-),
    Al

    --- GoldED+/LNX 1.1.5-b20180707
    * Origin: The Rusty MailBox - Penticton, BC Canada (1:153/757)
  • From Michiel van der Vlist@2:280/5555 to Alan Ianson on Sat May 22 09:37:32 2021
    Hello Alan,

    On Saturday May 22 2021 00:05, you wrote to Carlos Navarro:

    OMG... I just want to know which new features it has, not another
    list of commits... ':-)

    Aside from the few commits I listed there was also some changes around IPv6 that Michiel had asked for, but I don't quite remember what they
    were now.

    The -64 and -46 options come to mind.


    Cheers, Michiel
    --- GoldED+/W32-MSVC 1.1.5-b20170303
    * Origin: http://www.vlist.eu (2:280/5555)
  • From Carlos Navarro@2:341/66.3463 to Oli on Sat May 22 11:09:04 2021
    22 May 2021 08:12, you wrote to me:

    I am using BinkD 1.0.4 because AFAIK it is the latest stable
    version. However, most nodes I connect to are using 1.1a-112.

    Isn't that an alpha version? Is it recommended to use that one
    instead of stable?

    Is it? I really don't know.

    That's what the FidoNews' software listing says. :-)

    There was a time were versions like 2.3g
    were not uncommon. I guess 1.1a-x was meant as a snapshot from the
    current branch and the official release would become 1.1.0.

    There is not much developing going on. It's mostly a bug fix or other minor changes every couple of month. In that sense it's pretty stable.
    I think you also could call it a rolling release.

    Ok.

    What's new in 1.1a since 1.0?

    No need to wait for a "stable" 1.1 release, it's stable enough. Just
    make sure you get at least 1.1a-101, 1.1a-99 and below had some nasty bugs.

    Thanks. I would upgrade to the one I see others are using (1.1a-111 or 1.1a-112)

    Carlos

    --- GoldED+/W32-MSVC 1.1.5-b20180707
    * Origin: .:. (2:341/66.3463)
  • From Carlos Navarro@2:341/66.3463 to Alan Ianson on Sat May 22 11:16:50 2021
    22 May 2021 00:05, you wrote to me:

    OMG... I just want to know which new features it has, not another
    list of commits... ':-)

    Aside from the few commits I listed there was also some changes around IPv6 that Michiel had asked for, but I don't quite remember what they
    were now.

    It's mostly code fixes or beautification more than new features.

    I found 1.0.4 good and stable but it doesn't include the above.

    Now that's something. Thanks for the info.

    Carlos

    --- GoldED+/W32-MSVC 1.1.5-b20180707
    * Origin: .:. (2:341/66.3463)
  • From Phil Taylor@1:275/201.100 to Carlos Navarro on Fri Jun 4 03:29:04 2021
    22 May 2021 00:05, you wrote to me:

    OMG... I just want to know which new features it has, not another
    list of commits... ':-)

    Aside from the few commits I listed there was also some changes around IPv6 that Michiel had asked for, but I don't quite remember what they were now.

    It's mostly code fixes or beautification more than new features.

    I found 1.0.4 good and stable but it doesn't include the above.

    Now that's something. Thanks for the info.

    Carlos

    For which operating system?
    --- SBBSecho 3.14-Linux
    * Origin: sbbs.dynu.net 2025 (1:275/201.100)