iTU4l Changelog

ReadStringFromDevice(): $string to '' initialized (Thanks to Pierre Bernard)

Skip empty 0xFFFFFFFF records instead of recognize eof because there appear data after such empty records.

Improved gpx export: waypoints and tracklog are separate parts of the gpx file (Thanks to Piotr Gapiński)

Altitude value separatorin in dcsv export changed from , to ; (thanks to Holger Kirsch)

Whitelisted /dev/ttyUSB* and /dev/*acm* instead of any blacklisting in device name autodetection.

Time zone handling fixed by Denis N. Antonioli: kml and GPX is now GMT, csv, sql and print is still local time.

At least one user reported, that there is the same firmware problem like one year ago with tracks from Dec 31th. They show a wrong year (2034). It should be the same with the Windows software. At least after you erase all track data from your device the new recorded tracks should be ok again.

Kml Waypoints bug fixed by Denis N. Antonioli.

Improved waypoints by Denis N. Antonioli:
  1. If the gps device wakes up from sleep (Tag 99), the existing Placemark will be closed and a new one is started. As a consequence, the Placemark name is now the timestamp of the 1st gps record.
  2. A marker is made only for tag with a value of 254.
  3. Placemarks reference the line style, instead of re-defining it every time (cosmetic).
Support for Tag-based waypoints in kml export added and fixed several typos, thanks to Denis N. Antonioli.

Removed baud rate setting from the stty call because it is not longer supported by stty of actual linux distributions (and seamingly it was an NOP all the time). The tool should now run without problems with current linux kernels like 2.6.27-9 from Ubuntu 8.10 (Intrepid).
Added Postgres SQL export for, thanks to B.W.H. van Beest.

Tom Porcher reported, that these tools work under Windows with Cygwin. There is a non critical error message from stty.

The kernel 2.6.24-19 shipped with Ubuntu 8.04 (Hardy Heron) freezes on access to the USB device using the usbserial driver. Updated build with kernelcheck works. Ubuntu 8.10 (Intrepid) Alpha6 with its kernel works too. So I will not spend more time with in Hardy Heron. You still need the usb tricks.

Added skip of CheckForStty() if option -m is given.

Added option -m for skipping of the stty-based ConfigureSerialDevice(). Thanks to Glenn Hoffman, who reported, that this helps to get the script running on MacOS X Leopard.

To connect the device to an Asus Eee-PC, please read this: Thanks to Peter Gutmann.

Improved autodetect:
Support added for DPL700 firmware 4.1, where the name reported by the device itself changed from "WP GPS+BT" to "WondeProud Tech. Co. BT-CD110" (Thanks to Piotr Gapinski and Yuri Pirola).

Dirk Stelloh reported problems with the autodetect mechanism under Ubuntu 8.04. I cannot fix this at the moment, because I have not yet installed this Ubuntu version. Please specify your device with the option -d if you come into this trouble. Check syslog for you device name after you plugged in the device. In most cases -d /dev/ttyUSB0 will do the job.

Justin Clements reported, that readed data is scrambled under Fedora as guest OS running in the Parallels Virtual Machine under MacOS X as host OS. For me this points to the MacOS X lowlevel USB driver too (see 16.03.2008)

Steffen Schulz reported, that runs fine on an Intel based MacBook Pro running Linux instead of MacOS X after he changed the baudrate from115200 to 9600. That means, the problem seems to be the MacOS X (=BSD) lowlevel USB driver.

Since today a complete rewrite is online. Now writing logger settings is also supported. Use option -w with care and on your own risk! Be warned, you can destroy your device with this option ! You are still reading ? Ok. No damage is reported at all until now. Type iTU4l.plwithout any options or arguments and you get the usage information. You need the wav-files from the original CD in ~/wav or in your current directory for writing settings, because the wav-files have to be written always together with te settings.

I give up making the script MacOS compatible, because this task is not realy productive without having access to such a computer. See 16.03.2008 too.

The system freezes reported under Kernel 2.6.23 (Fedora) and 2.6.24 (Kubuntu Hardy Alpha) seem to come from the generic usb-serial driver (see here). The usb driver developes suggest to use the cdc_acm driver instead usbserial vendor=0x03EB product=0x6126. I cannot test it because my own 2.6.15 system does not know this driver. However in the thread there is also a patch for the generc usb-serial driver which shoulf fix the system freeze, as Dmitry Burstein reported (Thanks!). I hope that thos patch will come in the official kernel soon.

Dmitry Burstein reported, that freezes his Fedora 8. Because there is no problem with other Knoppix on the same hardware it points to a problem with the Fedora AT91xx usbserial driver in my eyes.

Added tag support for gpx, added short arguments for option -i, added memory dump read from STDIN, removed <course>0.000000</course> in gpx because it is optional

It seems to be hopefull, to get the scripts running under MacOS X. It is reported, that the USB device is detected automatically but the data readed are scrambled at all. I cannot test this because I have no access to MacOS X. Please report any additional expierences under MacOS X.

Since Version from today data from sr-files which overlap the 31.12.2007 are readed correct before and after the 31.12.2007 (the day on which the loggers firmware had problems according to this thread). Thanks you to Sven Haberer, who points me to the firmware because I searched in the for a bug, which was not there. If you still have problems erase all data in your logger: -e -d <your device>

There are several improvments in now:

23.06.2009, iTU4l⟨ΑΤ⟩