bitpim-devel Mailing List for BitPim (Page 265)

SourceForge Headquarters 225 Broadway Suite 1600 San Diego, CA 92101 +1 (858) 422-6466

Okay-it tells me it will be on COM4-so i select it, but it says  access is 
denied.

> I am trying all the ports....I can't get it to let me

On Windows 98/ME:

   Control Panel -> System -> Device Manager

   Scroll down to Ports (COM and LPT)

   Look for one labelled USB to Serial Port

On Windows 2000/XP:

   Control Panel -> System -> Hardware -> Device Manager

   Scroll down to Ports (COM & LPT)

   Look for one labelled Prolific USB-to-Serial Comm Port

I use the Radio Shack cable which uses the prolific drivers.  Other vendors
cables will have different names, but they should be similar.

Roger

If you are on could we email without this post so you can talk me through it? 
I appreciate everything

I am trying all the ports....I can't get it to let me

> wxPyAssertionError: C++ assertion "wxAssertFailure" failed in 
> e:\projects\wx\src\common\image.cpp(720): invalid image

Delete My Documents\bitpim\wallpaper

This error occurs because XP helpfully dumps some extra files
in the directory without me asking (specifically thumbs.db).
Bitpim then falls over trying to read it.

Fixed in 0.5

Roger

> File "serial\serialwin32.pyo", line 186, in setBaudrate
> error: (87, 'SetCommState', 'The parameter is incorrect.')

This is because whatever you told bitpim is a com port isn't a
com port.  Change it in Edit -> Settings.  You need to restart
bitpim for it to pick up the new setting.

Roger

I deleted and reinstalled the program and it gives me a warning no type 
selected then says the same thing   almost    

An unexpected exception has occurred.
Please report the following information to the developers

Traceback (most recent call last):
  File "<string>", line 20, in ?
  File "gui.pyo", line 238, in run
  File "gui.pyo", line 222, in __init__
  File "gui.pyo", line 373, in __init__
  File "gui.pyo", line 527, in OnPopulateEverythingFromDisk
  File "guiwidgets.pyo", line 807, in populate
  File "wxPython\image.pyo", line 252, in GetWidth
wxPyAssertionError: C++ assertion "wxAssertFailure" failed in 
e:\projects\wx\src\common\image.cpp(720): invalid image

Im getting this error: 

An unexpected exception has occurred.
Please report the following information to the developers

Traceback (most recent call last):
File "gui.pyo", line 180, in run
File "gui.pyo", line 121, in __call__
File "gui.pyo", line 736, in getdata
File "com_lgvx4400.pyo", line 230, in getwallpapers
File "com_lgvx4400.pyo", line 505, in setmode
File "com_lgvx4400.pyo", line 528, in _setmodebrew
File "com_lgvx4400.pyo", line 540, in _setmodelgdmgo
File "commport.pyo", line 40, in setbaudrate
File "serial\serialwin32.pyo", line 186, in setBaudrate
error: (87, 'SetCommState', 'The parameter is incorrect.')

Will someone please help Me ASAP I am trying so hard to get this to work!! 
sorry if this has been answered before or something! Thank you in advance!

I don't consider the Bitpim user interface to be that good.  In particular the
menu structure could do with some thinking through.  There is also something
to be said for using MDI instead of the notebook.

If you would like to contribute to bitpim, this is one area you can help.
See what other programs do, and try and figure out what makes sense to
bitpim users.  (Remember the average user is not a coldly logical
programmer :-)

Produce some screenshots or other mockup of your ideas, and after a small
amount of discussion I'll implement whatever is best.

Roger

> Yep.  wxpython 2.2 is the version available in Debian testing

You can't even get that version anymore from the main download
page on the wxPython site!

Roger

> > BTW, I grabbed the CVS snapshot and had to fetch the latest wxpython
> > (the version that's part of Debian testing doesn't support all the
> > features you're using).  Just an FYI for the readme.
> 
> I'll probably do a seperate developer readme/pages at some point.
> wxpython 2.3 and 2.4 work for me.  Are Debian REALLY behind on
> wxpython?

Yep.  wxpython 2.2 is the version available in Debian testing
(unstable has 2.3 and 2.4 in it).  Dunno if the holdup is some
dependency or a slow maintainer...

For the archive, if others need to upgrade wxpython getting and
installing the source wasn't difficult.  It even includes the
wxWindows distro and is generally nicely put together:
	http://prdownloads.sourceforge.net/wxpython/wxPythonSrc-2.4.0.4.tar.gz

> > Also, a couple of the button graphics seem to be missing from the
> > CVS repository.  (They just show up as "?" onscreen.)
> 
> That is correct.

Ah, a todo item then...

> I use Redhat 8.0 and the cable shows up as /dev/usb/ttyUSB0.  RH doesn't
> use devfs so it is really hard to tell what is actually connected.  I
> was thinking of parsinng the dmesg stuff to figure out something pretty
> to display to the user!

On my box its probably going to show up as /dev/ttyUSB1 (no
intervening 'usb').  Amusingly it will probably show up in a different
place depending on if my USB Visor is plugged in or not....

I should probably wait until I actually get a cable before making
pronouncements, though.  :)

-Pat

----- ----- ---- ---  ---  --   -    -      -         -               -
Pat Tullmann                                       tul...@cs...
                     "Forty-Two." -- Deep Thought

> In the map file (brewImageIndex.map) the files are layed out like so:
> 
> 03 00 03 00 xx xx xx xx xx xx xx .... 00 02 00 xx xx xx ... 00 01 00 xx 
> xx xx xx

The integer fields are two bytes in lsb order.

The first field is the total number of entries.

Each entry is a serial number and then the filename in a fixed length
field.

There is space for 30 entries in the mapfile.  Unused entries 
have a serial of 0xffff (or -1 if you prefer) with the name blanked
out.

When you delete an entry from the phone, it deletes the file on
the filesystem, and marks the entry with the 0xffff serial.  It
also decrements the count (first field of file).  This means that
you can have "holes" in the index.

Just like you, I initially thought that ordering would make a
difference ....

Roger

Whoops, found the archives and noticed I was way behind the curve... :)

Sorry about that.

Mike

On Tuesday, March 11, 2003, at 08:51 PM, Mike wrote:

>
> Hello all,
>
> I just wanted to drop in and see if I could offer some help with this 
> image index problem your having... I've only started with bitpim 
> today, happily modifying my vx4400, and tried to delete an image off 
> the phone.  After looking into it a bit more, I've come to a solution 
> for the index to get back in sync...
>
> In the map file (brewImageIndex.map) the files are layed out like so:
>
> 03 00 03 00 xx xx xx xx xx xx xx .... 00 02 00 xx xx xx ... 00 01 00 
> xx xx xx xx
>
> The first byte being the number of images stored, the third byte 
> starting the index... what I found was that all of the image files 
> that I had a number in front of them... such as 03, 02, 01 for the 
> three images that I had... the problem is that it is supposed to count 
> from 0, not 1, so I was missing a 00 index... I changed the 03 in byte 
> 3 to be 00 and all my indexes on the phone worked fine (even though 
> they're not in line with any file order that I can think of)... the 
> images now match up with the names and they stay that way...
>
> I hope this helps your dilemma, if there's any questions just hollar.
>
> Mike
>
>
>
>
>
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by:Crypto Challenge is now open! Get 
> cracking and register here for some mind boggling fun and the chance 
> of winning an Apple iPod:
> http://ads.sourceforge.net/cgi-bin/redirect.pl?thaw0031en
> _______________________________________________
> Bitpim-devel mailing list
> Bit...@li...
> https://lists.sourceforge.net/lists/listinfo/bitpim-devel
>
>
>
http://www.parrotlaw.us -- The Professional Law Student's forums

Hello all,

I just wanted to drop in and see if I could offer some help with this 
image index problem your having... I've only started with bitpim today, 
happily modifying my vx4400, and tried to delete an image off the 
phone.  After looking into it a bit more, I've come to a solution for 
the index to get back in sync...

In the map file (brewImageIndex.map) the files are layed out like so:

03 00 03 00 xx xx xx xx xx xx xx .... 00 02 00 xx xx xx ... 00 01 00 xx 
xx xx xx

The first byte being the number of images stored, the third byte 
starting the index... what I found was that all of the image files that 
I had a number in front of them... such as 03, 02, 01 for the three 
images that I had... the problem is that it is supposed to count from 
0, not 1, so I was missing a 00 index... I changed the 03 in byte 3 to 
be 00 and all my indexes on the phone worked fine (even though they're 
not in line with any file order that I can think of)... the images now 
match up with the names and they stay that way...

I hope this helps your dilemma, if there's any questions just hollar.

Mike

> Roger wrote:
> > So far I have tried the following:
> > 
> >   - Sorting index entries numerically and/or alphapbetically
> >   - Not deleting files I am overwriting
> >   - Ensuring entries are contiguous
> >   - Forcing starting at zero and working up
> >   - Starting at one and working up

I have now also tried:

  - the index is actually how many characters of the name to display
  - the index needs to be in the same order that the files are on the
    filesystem

I think I am just going to add a menu entry 'Reindex' that goes and
rebuilds the index on the phone.  (My developer prototype before
bitpim had this 'feature').

> Can you leave some sort of "null" entry in the deleted spot?  If other
> parts of the phone refer to images by entry number, that seems the
> only correct choice...

The problem is that the phone is the one doing the deletion!  It does
enter a null entry.  It then also gets confused and displays bad things
when selecting the wallpapers.  To be honest this is more of a cosmetic
issue than anything else.

If you want to see what I am talking about, do the following steps:

  - Use filesystem view to look at dloadindex/brewImageIndex.map
    (the record format is pretty obvious)
  - Use bitpim to download a few wallpapers
  - Check out the index again to see what bitpim wrote
  - Go into the phone menus and select wallpapers, and delete one
    or two
  - Note how when you are browsing wallpaper it may show ones that
    you have already deleted.  Also on the selection screen 
    (Menu -> 8 -> 2 -> 3) move up and down.  See how it displays
    junk at the bottom
  - Use filesystem view to see what it has done to the index again

I am pretty much convinced that this is a plain old bug in the
phone.

> BTW, I grabbed the CVS snapshot and had to fetch the latest wxpython
> (the version that's part of Debian testing doesn't support all the
> features you're using).  Just an FYI for the readme.

I'll probably do a seperate developer readme/pages at some point.
wxpython 2.3 and 2.4 work for me.  Are Debian REALLY behind on wxpython?

> Also, a couple of the button graphics seem to be missing from the CVS
> repository.  (They just show up as "?" onscreen.)

That is correct.  Look in the terminal where you start bitpim.  It has
lovely reminders of the files that need to be created.  (The code falls
back on the question mark when it can't find the desired graphic).
And just to make life a little more fun, on Windows they need to be
16x16 and on Linux it is 32x32.  I think I will just have them all
as 32x32 on disk and then scale them when reading them in.

> I wasn't going to bother buying a data cable since I don't have Win
> XP/2K...

bitpim works fine under Windows 98 as well.

> but bitpim lets me play with it under Linux!  Sweet!  Got to
> get to Radio Shack...  Thanks!  :)

I aim to please :-)  Once you have it working, and if you have a little
time to spare, I would like some code that detects which ports are
available on Linux.

I use Redhat 8.0 and the cable shows up as /dev/usb/ttyUSB0.  RH doesn't
use devfs so it is really hard to tell what is actually connected.  I
was thinking of parsinng the dmesg stuff to figure out something pretty
to display to the user!

Roger

Roger wrote:
> So far I have tried the following:
> 
>   - Sorting index entries numerically and/or alphapbetically
>   - Not deleting files I am overwriting
>   - Ensuring entries are contiguous
>   - Forcing starting at zero and working up
>   - Starting at one and working up

Can you leave some sort of "null" entry in the deleted spot?  If other
parts of the phone refer to images by entry number, that seems the
only correct choice...

BTW, I grabbed the CVS snapshot and had to fetch the latest wxpython
(the version that's part of Debian testing doesn't support all the
features you're using).  Just an FYI for the readme.

Also, a couple of the button graphics seem to be missing from the CVS
repository.  (They just show up as "?" onscreen.)

I wasn't going to bother buying a data cable since I don't have Win
XP/2K... but bitpim lets me play with it under Linux!  Sweet!  Got to
get to Radio Shack...  Thanks!  :)

-Pat

----- ----- ---- ---  ---  --   -    -      -         -               -
Pat Tullmann                                     tul...@fl...
      That which does not kill you just didn't try hard enough.

I have been investigating an issue with the phone.  Quite simply, 
try deleting a wallpaper on the phone.  The other entries will
do wierd things (display truncated names), display the filename
of the file, but just white etc.

Can anyone figure out what the contents of the index file should
be so that the phone deletes things correctly (it always deletes
the image file fine, and appears to update the index correctly
but then behaves badly).

So far I have tried the following:

  - Sorting index entries numerically and/or alphapbetically
  - Not deleting files I am overwriting
  - Ensuring entries are contiguous
  - Forcing starting at zero and working up
  - Starting at one and working up

I've double checked that the index contents are identical to 
what mapedit generates.

I'm coming to the conclusion that this is a bug in the phone.

Roger

> Please let me know if I can help out.

One thing that would be useful is if someone can figure out what on
earth is going on with the image index.  In particular the phone
doesn't seem to delete images well.  Bitpim 0.4 writes out the
indices in hashtable (random) order.  My dev code sorts the order.
This didn't make any difference.

mapedit just uses consecutive numbers starting at 0.  I try to 
preserve the numbers since they can be referenced by phonebook
entries, so I can have missing numbers in the sequence.

It may just be that index numbers aren't index numbers, or that the
phone software is incomplete (won't be the first time).

> Did you completely reverse engineer the serial protocol, or did you have
> some documentation to work from? 

I used a comm port sniffer and worked it out from there.  I am not
aware of any documentation.

Roger

Roger,

Please let me know if I can help out.

Did you completely reverse engineer the serial protocol, or did you have
some documentation to work from? 

-Jon

>       I've check out the scheduler file format and so far it looks
>       very random... why cant those idiots use comma-delimited files
>       for storage or something :)

Because it isn't a PC.  It has very limited CPU and RAM (both need power).
They use fixed length records for everything.  This is very easy to
read/write as C structs.

>       Ill take a look at phone book etc over the spring break. I am
>       very interested in creating an application that either
>       synchronizes with outlook calendar or emulates it.

I'm going to be adding that functionality to bitpim anyway.  Initially
it will be direct editing in bitpim itself, and then will expand to
include Outlook and Palm Desktop on Windows and Mozilla on Linux.

Roger

> There is a specific phonebook protocol.  (And boy is it a piece of junk.)
> For everything else you access the filesystem of the phone.  You can use

BTW there is also a protocol viewer tab in bitpim.  That way you can actually
see what is sent/received.

The phone is very finnicky with the protocols.  If you get one slight
thing wrong, it will either ignore you, pretend that it actually 
succeeded, or my favourite which is to generate an erroneous error.

Roger

(You are actually the third member :-)

>   Roger, would you mind posting source code and/or file format that
>   the phone uses? I am very interested in developing a package to
>   synchronize phone book and/or scheduler with PC
>   [either in c++ or Delphi]

Bitpim already does the synchronizing so you would be duplicating it!

You can get the sourcecode from http://sourceforge.net/cvs/?group_id=75211

There is a specific phonebook protocol.  (And boy is it a piece of junk.)
For everything else you access the filesystem of the phone.  You can use
the filesystem browser in bitpim to see stuff (double clicking on a file
gives you a hex dump).

To see the specific protocol information, look in the file
com_lgvx4400.py

Roger

Hello bitpim-devel,

  Roger, would you mind posting source code and/or file format that
  the phone uses? I am very interested in developing a package to
  synchronize phone book and/or scheduler with PC
  [either in c++ or Delphi]

-- 
Best regards,
 Vic                          mailto:drg...@op...

Bitpim 0.3 is now released.  There is now a vague resemblence to a home page
at bitpim.sf.net.  I will be announcing the existence of bitpim to a wider
audience with this release.

The changelog:

 - Cope with ringtones directory not existing on phone 
 - Cope with wallpapers directory not existing on phone 
 - Can now write out your phonebook. 
 - Can now crudely edit your phonebook 
 - Changed wallpaper size to be 120x98 instead of 120x100 so that it comes out 
   correctly centered vertically 
 - The ondisk format of the phonebook has changed. If you used 0.2, please delete
   any stored phonebook (defaults to My Documents\bitpim\phonebook\index.idx) 
   otherwise bitpim may refuse to start. 

Roger

This is a first post to make SourceForge create the list archives.

Roger

2003	Jan	Feb	Mar (53)	Apr (48)	May (14)	Jun (3)	Jul (21)	Aug (11)	Sep (77)	Oct (67)	Nov (28)	Dec (163)
2004	Jan (112)	Feb (143)	Mar (114)	Apr (138)	May (173)	Jun (119)	Jul (119)	Aug (117)	Sep (187)	Oct (170)	Nov (254)	Dec (193)
2005	Jan (336)	Feb (284)	Mar (189)	Apr (100)	May (89)	Jun (52)	Jul (85)	Aug (138)	Sep (181)	Oct (137)	Nov (104)	Dec (98)
2006	Jan (76)	Feb (106)	Mar (224)	Apr (270)	May (103)	Jun (144)	Jul (77)	Aug (38)	Sep (37)	Oct (20)	Nov (14)	Dec (73)
2007	Jan (130)	Feb (68)	Mar (78)	Apr (60)	May (45)	Jun (63)	Jul (84)	Aug (45)	Sep (40)	Oct (12)	Nov (71)	Dec (56)
2008	Jan (44)	Feb (20)	Mar (25)	Apr (17)	May (33)	Jun (60)	Jul (97)	Aug (38)	Sep (10)	Oct (20)	Nov (13)	Dec (19)
2009	Jan (7)	Feb (5)	Mar (23)	Apr (10)	May (6)	Jun (5)	Jul (17)	Aug (7)	Sep (14)	Oct (27)	Nov (13)	Dec (12)
2010	Jan (37)	Feb (9)	Mar (13)	Apr (12)	May (8)	Jun (3)	Jul (1)	Aug (9)	Sep (3)	Oct (1)	Nov (1)	Dec (4)
2011	Jan (1)	Feb	Mar (1)	Apr (4)	May	Jun (2)	Jul (8)	Aug	Sep	Oct	Nov	Dec
2012	Jan (2)	Feb	Mar	Apr	May (5)	Jun	Jul	Aug	Sep (5)	Oct	Nov (3)	Dec
2015	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov (2)	Dec
2017	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep (1)	Oct	Nov	Dec
2021	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug (2)	Sep (2)	Oct	Nov	Dec

bitpim-devel Mailing List for BitPim (Page 265)

bitpim-devel — For bitpim developers