Active Topics
-
The N900 is now 15 years old! (9)
to Nokia N900 by Arno_11 - 1 day, 1 hr ago -
Sailfish OS for the Motorola Moto G7 Power (XT1955-5) - (ocean) (4)
to SailfishOS by Maemish - 6 days, 13 hrs ago - more...
|
2010-01-17
, 18:25
|
|
Posts: 61 |
Thanked: 36 times |
Joined on Feb 2006
@ Harpenden
|
#52
|
OK, next plan of attack is to use the Bluetooth Packet Logger in the XCode utilities to see if the N900 is spitting it out like that, or if it is getting munged within the Mac. If the N900 is doing that and the spec says no LF or CR in those places then we have a bona fide bug in Maemo.
|
|
2010-01-17
, 19:12
|
|
Posts: 356 |
Thanked: 172 times |
Joined on Jan 2010
@ Canada
|
#53
|
Originally Posted by GaryHT627
Hear Hear! Many thanks to both of you for sorting through this so diligently
I really appreciate all the effort on this and I am thrilled that I can do at least a basic sync on my contacts
It's fascinating reading -- I only wish I had even a tiny bit of programming skill with which to help, but sadly I don't.
So guinea pig it is!
I noticed a couple of small quirks with Contacts myself:
• AIM user names didn't sync from Address Book to the N900
• The 'Me' card in Address Book didn't sync to the N900's 'My information' data, and I can't seem to find any way of linking the two after the fact
Last edited by Bingley Joe; 2010-01-17 at 19:15.
|
|
2010-01-17
, 19:27
|
|
Posts: 33 |
Thanked: 10 times |
Joined on Jan 2010
|
#54
|
Originally Posted by scudderfish
Well I don't know about you Scud but the packet logger doesn't seem to offer any useful information (that I can understand!).
OK, next plan of attack is to use the Bluetooth Packet Logger in the XCode utilities to see if the N900 is spitting it out like that, or if it is getting munged within the Mac. If the N900 is doing that and the spec says no LF or CR in those places then we have a bona fide bug in Maemo.
It's frustrating that it just seems to come down to those CR's.
Any luck with the packet logger?
|
|
2010-01-17
, 19:47
|
|
Posts: 356 |
Thanked: 172 times |
Joined on Jan 2010
@ Canada
|
#55
|
Ok, this is weird
Up to now I haven't really had any issues outside of those I've already mentioned (all minor to say the least).
Just now, however, after doing some editing of a couple of contacts on the N900, I tried to do a sync to see how the changes would be reflected in Address Book..
The answer -- it wanted to create duplicate entries for EVERYONE
Up to now I haven't really had any issues outside of those I've already mentioned (all minor to say the least).
Just now, however, after doing some editing of a couple of contacts on the N900, I tried to do a sync to see how the changes would be reflected in Address Book..
The answer -- it wanted to create duplicate entries for EVERYONE
|
|
2010-01-17
, 20:15
|
|
Posts: 61 |
Thanked: 36 times |
Joined on Feb 2006
@ Harpenden
|
#56
|
Originally Posted by 3quid
OK, it is definitely the N900 that is spitting out the extra CRs. In the attachment is a bluetooth packet capture of a sync where I had created an alarmed event. If you examine the packet at 19:32:18.762, there is definitely 0D/0A sequences in the string 17T194000..;PT3M..;..END before the semi-colons. Now we need to read the specs to ensure this is definitely disallowed and then take a trip to bugzilla.
Well I don't know about you Scud but the packet logger doesn't seem to offer any useful information (that I can understand!).
It's frustrating that it just seems to come down to those CR's.
Any luck with the packet logger?
|
|
2010-01-17
, 20:37
|
|
Posts: 61 |
Thanked: 36 times |
Joined on Feb 2006
@ Harpenden
|
#57
|
OK if I can read my BNF correctly (haven't done that since Uni many moons ago)
it looks like the extra CRLF we've been worried about are actually allowed by the spec and it is Apple that is incorrectly parsing it.
Code:
CR = <ASCII CR, carriage return> ; ( 15, 13.)
LF = <ASCII LF, linefeed> ; ( 12, 10.)
CRLF = CR LF
SPACE = <ASCII SP, space> ; ( 40, 32.)
HTAB = <ASCII HT, horizontal-tab> ; ( 11, 9.)
All literal property names are valid as upper, lower, or mixed case.
ws = 1*(SPACE / HTAB)
; "whitespace," one or more spaces or tabs
entprop = [ws] simprop
[params] ":" value CRLF
/ [ws] "AALARM"
[params] ":" aalarmparts CRLF
aalarmparts = 0*3(strnosemi ";") strnosemi
; runTime, snoozeTime, repeatCount, audioContent
strnosemi = *(*nonsemi ("\;" / "\" CRLF)) *nonsemi
; To include a semicolon in this string, it must be escaped
; with a "\" character.
nonsemi = <any non-control ASCII except ";">
|
|
2010-01-17
, 20:54
|
|
Posts: 33 |
Thanked: 10 times |
Joined on Jan 2010
|
#58
|
Hmm ok. I guess I will have to look up some more info on the iCal/vCal parser! Hopefully it will be something that we can remedy
|
|
2010-01-17
, 22:00
|
|
Posts: 33 |
Thanked: 10 times |
Joined on Jan 2010
|
#59
|
There is also a possible issue with the CRLF just in front of the SUMMARY value
Code:
SUMMARY ENCODING=QUOTED-PRINTABLE;CHARSET=UTF-8:= TestEntry
|
|
2010-01-18
, 06:37
|
|
Posts: 1 |
Thanked: 0 times |
Joined on Jan 2010
|
#60
|
Originally Posted by scudderfish
I haven't done BNF in a while myself either, but I think you have misread the grammar.
it looks like the extra CRLF we've been worried about are actually allowed by the spec and it is Apple that is incorrectly parsing it.
Code:
strnosemi = *(*nonsemi ("\;" / "\" CRLF)) *nonsemi
It looks like someone needs to open a bug with Maemo on this.
and the summary line also had a line break after the = sign
its not all down to forum software. i will email you the log when i get back in an hour or so (or recreate it)
i am pretty sure that the formatting of the aalarm line is the main issue for working syncs. i just dont know if its the ical parser or what is sent by the n900