Re: Proposed General Release (Dev: 2020/05/01 21:21)

Rob Giuliano

I updated the WIKI on this subject.
  There are 2 sections now -> Pre WINE 2.8 (same info as before: WINE Registry) and Post WINE 2.8 (updated for USB devices: reassign existing Ports). 

Unfortunately, the explanation is a bit wordy.  If you are familar with Linux and WINE, please take a look and provide any feedback.

I have 3 TNCs that connect through USB devices, and I have successfully changed TNCs with APRSIS32 matching the COM port for each.  In my case, each USB interface had a unique Vendor and Product code. 

My one comment about the updated WINE versions is:
"The good news is that you no longer have to mess with the Registry (which can cause other problems).  Older WINE versions had a pretty simple registry, but the newer versions are much more Windows like."

Robert Giuliano

On Saturday, May 2, 2020, 12:57:42 PM EDT, Rob Giuliano via <kb8rco@...> wrote:

I would argue their purpose was to ensure anything Linux 'sees' is mapped to WINE.
Why does Linux have
    tty - okay I know this one!
    tty0 - tty63
    ttyS0 - ttyS32
when only a few devices are actually present?
Isn't that what things like udev are for?

I will be doing a little investigating into possibly deleting some of these nonexistent devices and see if that cuts down what WINE sees.

Robert Giuliano

On Saturday, May 2, 2020, 12:25:44 PM EDT, Greg D <ko6th.greg@...> wrote:

Hi Rob,

It sounds like the Wine team is trying to be annoyance-for-annoyance compatible with Windows :(. 

I'm on Wine 1.8.6 on an old OpenSuSE LEAP 42.1 system, and the upgrade to the latest went without a hitch.  It's an iGate that I run 24x7, and all is working fine.  Thanks for the "heads-up" for the future when I eventually upgrade the system.

Greg  KO6TH

Rob Giuliano via wrote:
I'll try a few more things and update the wiki - soon.

Robert Giuliano

On Saturday, May 2, 2020, 10:36:15 AM EDT, Rob Giuliano via <kb8rco@...> wrote:

As near as I can tell, there is something in the WINE setup that automatically (and "unchangeably") links every ttyS## device in /dev to a com## in the .wine folder.  Linux (at least my XUbuntu) has ttyS0 - ttyS32.  Each USB serial device is added after that. 

Good news is that APRSIS32 "sees" all 33 (in my case) serial ports.  The bad news of course is that there is no guarantee of the USB order if you have more than 1 USB serial device.

Through investigation, I learned that udev rules can reallocate any of these com# device to a symlink device link other than an ttyS#, BUT WINE changes them back.  However, udev rule CAN override an existing device (with no hardware), that worked for me.  My desktop has 1 physical serial port on ttyS0 (com1), so I can reallocate any ttyS# with # > 0.  Just remember that COM# ports start with 1 and ttyS# devices start with 0 (zero), so my override of ttyS3 became com4.  A section of my udev rules file in /etc/udev/rules.d called 65-tnc.rules:

# SUBSYSTEM=="tty", ATTRS{idVendor}=="0403", ATTRS{idProduct}=="6001", SYMLINK+="tncx"
SUBSYSTEM=="tty", ATTRS{idVendor}=="0403", ATTRS{idProduct}=="6001", SYMLINK+="ttyS3"

I have a line for each TNC device I connect, including some for Bluetooth connections, but I haven't tested those with WINE 5.0 yet.

Robert Giuliano

On Saturday, May 2, 2020, 8:58:33 AM EDT, Michael Coslo via <mjc_n3li@...> wrote:

Hi Rob,

WINES’s handling of com ports flummoxed me. What used to be a simple symbolic link became something else indeed  ver 2.8 and later. The people at Codeweavers tried to help, but they had instructions for Mac, which were different enough that it didn’t work for me.

Needing my station up and running, I ended up removing Mint and reinstalling Windows 8.1, which is just too cruel! 8^)

When I’m not pressed for time, I’ll reinstall mint and try again.

But you might have the secret sauce for getting new versions of WINE to do the link?

Otherwise I might try again with an old version of WINE.

- 73 Mike N3LI -

> On May 2, 2020, at 12:56 AM, Rob Giuliano via <> wrote:
> Just upgraded my setup.  This includes latest XUbuntu (20.04) with WINE 5.0.
> Some features are new to me with the updated WINE version.
> APRSIS32 ran the normal update after clicking HELP.
> Other than connecting to my TNC (WINE changed COM port stuff), NO ISSUES here!
> I'll let it run now.
> Robert Giuliano
> On Friday, May 1, 2020, 10:02:06 PM EDT, Lynn Deffenbaugh <kj4erj@...> wrote:
> I just pushed the subject version to development.  Please let me know if
> you successfully upgrade to this and if it initially looks ok.
> The only change since the last Development release is a spelling
> correction, changing the name for APRSISMO (check KJ4ERJ-12), and
> setting the new release timestamp.
> If this checks out, I'll push it to release sometime over the weekend.
> Lynn (D) - KJ4ERJ - Author of APRSISCE for Windows Mobile and Win32
> Here are the proposed release notes.  They will be posted in full to the
> Wiki.
> Here's the major features in version 2020/05/01 21:21
> Updated ToCall and Mic-E definitions from:
> as of 17 Apr 2020 (updated
> 2020/04/27)
> through 4 Jun 2019 (Updated
> 2020/04/27)
> No longer suppress gating messages just because a station was heard
> INSIDE a 3rd party packet on RF.  See notes about KO6TH from 2018/09/03
> 11:36
> Support "Common Talker IDs" from
> See notes from
> 2017/07/17 15:05
> Support Always-On-Top in right click menu in both main and MultiTrack
> windows (not on CE)
> Support proprietary 4 digit precision extension (!DAO!xx) in Configure /
> Beacon / Precision / 4 Digits
> Draw a small heading arrow near symbols that beacon speed and course
> Support a Syslog-type port for APRISMO diagnostic capture (Development
> mode only)
> Added Screen / Labels / Nicknames / Delete All (n)
> Also, if you try to define a new nickname that uses the same label as an
> existing nickname, the Accept button will disable until the Label is unique.
> There's a new Configure / Aliases / Accumulate feature to "learn" local
> digipeater aliases for actual hop counting.
> (Note: these are things like WIDE, RELAY as well as the State and
> Regional aliases (SSn-N))
> See:
> Show both Non-alias used hop count and Simplistic used hop count in the
> Digi/IGate column of the packet scroller.
> Filtered gating of packets from -IS to RF, but only in Development
> mode.  See notes from 2012/08/31 08:02 below.
> There's a bunch of updates to the View / Platform submenu population and
> counting sprinkled throughout.
> There's a new Configure / Companions option described in 2012/09/04
> 21:26, 2012/09/06 02:03, 02:51, 13:35
> Support MiniDump creation on a detected crash.  This may create
> APRSIS32.DMP files in your execution directory that will assist me in
> determining the cause of crashes.
> If APRSIS32 generates a .DMP/DMZ file, it will automatically restart.
> Eventually it'll even auto-detect the .DMPs and offer to send them to me.
> Support port type of "Dummy" (connection type doesn't matter) to be a
> Transmit-Only Dummy Load for testing -IS to RF IGating in an RF-less
> environment.
> Count "Oth" packets for any packet that doesn't fit a defined type.
> These are the packets that a type (t/) filter can't pick up from a
> filtered feed.  These packet counts are display on the port status
> windows (right click in APRS-IS OK pane).
> There's a new set of "Busy Stations" windows available on the APRS-IS OK
> right click menu.  Read about them in 2012/09/27 07:39 below.
> Support Screen / Brightness / Bright/Dim which changes the background of
> the transparent maps from White to Black. Discovered on APRSISDR.  (Adam
> The NMEA port now supports a direct TCP/IP connection to GPSd. Simply
> configure the NMEA port to use TCP/IP and point it to your GPSd
> instance.  If you've never heard of GPSd, don't worry about it.  If
> you've got a GPSd already running, then I don't need to tell you more,
> right?  (Steve G6UIM)
> RF Ports now have a "NoGate ME" checkbox to disable gating your own
> generated packets to the -IS if copied on a digipeat.
> Added zoom number in center of slider (breaks if/when I do fractional
> zooms).  (Greg KB3KBR, Fred N7FMH)
> The zoom number turns green (it might be hard to read) when you're
> zoomed in close enough to "Move ME".  (Greg KB3KBR)
> Ignore single character !shriek!s (!x! being put out by N5ETC-2)
> Properly parse local time (/) format specifier (Jerry KF5AOK)
> New support for JS8Call and FT8 log-scraping objects.  This is similar
> to the JT65 scraper that already existed.  Simply create new object of
> the appropriate type and point it to your ALL.TXT log file for the
> corresponding program.  Object will be created for any detected station
> for which a grid square is available. These are objects local to your
> instance only.
> Support for "Hashes".  These are similar to !shriek!s, but are delimited
> by a single preceding #, as in #JS8 or #EVENT.  More details on this to
> follow at some later date.
> Note that only 32 unique Hashes are currently supported.  This limit
> will be removed before this support is finalized.
> KISS (aka AX.25) callsigns are now limited to A..Z and 0..9. Invalid
> characters are simply removed from the transmitted callsign, so beware!

Join to automatically receive all group messages.