Re: [hylafax-users] Howto enforce Hylafax to correctly encode incoming faxes?

[Dmitry Bely <dbely@mail.ru>]

"Bernd Proissl" <lists@proissl.de> writes:

> I cannot force Hylafax to correctly encode incoming faxes.
> I follow the instructions of Dmitry:
> 
> http://www.hylafax.org/archive/2000-08/msg00081.html
> 
> which says add
> 
> Class2CQQueryCmd:       "!(0),(0)"
> RecvDataFormat:         "2-D MMR"
> 
> to your config file but this does not change anything here.

Very strange.

> The log is:

Can you show your modem init log from /var/log/daemon.log?
ServerTracing=0xFFF is desireable.

[...]

>     ---- Transcript of session follows ----
> Sep 25 12:29:26.46: [28093]: SESSION BEGIN 00000033 xxx
> Sep 25 12:29:26.46: [28093]: <-- [4:ATA\r]
> Sep 25 12:29:38.46: [28093]: --> [4:+FCO]
> Sep 25 12:29:38.46: [28093]: ANSWER: FAX CONNECTION
> Sep 25 12:29:38.46: [28093]: STATE CHANGE: ANSWERING -> RECEIVING
> Sep 25 12:29:38.46: [28093]: MODEM input buffering enabled
> Sep 25 12:29:38.47: [28093]: RECV FAX: begin
> Sep 25 12:29:41.16: [28093]: --> [4:+FCO]
> Sep 25 12:29:43.21: [28093]: --> [27:+FTI:"xxx    "]
> Sep 25 12:29:43.21: [28093]: REMOTE TSI "xxx"
> Sep 25 12:29:43.21: [28093]: --> [20:+FCS:0,5,0,2,0,0,0,0]
> Sep 25 12:29:43.21: [28093]: REMOTE wants 14400 bit/s
> Sep 25 12:29:43.21: [28093]: REMOTE wants page width 1728 pixels in 215 mm
> Sep 25 12:29:43.22: [28093]: REMOTE wants unlimited page length
> Sep 25 12:29:43.22: [28093]: REMOTE wants 3.85 line/mm
> Sep 25 12:29:43.22: [28093]: REMOTE wants 1-D MR
> Sep 25 12:29:43.22: [28093]: --> [2:OK]
> Sep 25 12:29:43.22: [28093]: <-- [7:AT+FDR\r]
> Sep 25 12:29:48.15: [28093]: --> [7:CONNECT]
> Sep 25 12:29:48.15: [28093]: RECV: begin page
> Sep 25 12:29:48.15: [28093]: RECV: send trigger 022
> Sep 25 12:29:48.15: [28093]: <-- data [1]
> Sep 25 12:30:06.24: [28093]: RECV/CQ: Bad 1D pixel count, row 1146, got 0,
> expected 1728
> Sep 25 12:30:06.24: [28093]: RECV/CQ: Bad 1D pixel count, row 1147, got 0,
> expected 1728
> Sep 25 12:30:06.24: [28093]: RECV/CQ: Bad 1D pixel count, row 1148, got 0,
> expected 1728
> Sep 25 12:30:06.24: [28093]: RECV/CQ: Bad 1D pixel count, row 1149, got 0,
> expected 1728
> Sep 25 12:30:06.24: [28093]: RECV/CQ: Bad 1D pixel count, row 1150, got 0,
> expected 1728

Hylafax had not catched RTC here; that's known bug, already fixed in my
local sources (not commited yet because I am still trying to clean up some
Class1 receive protocol issues).

> Sep 25 12:30:06.25: [28093]: RECV/CQ: Invalid BlackTable code word, row
> 1151, x 241
> Sep 25 12:30:06.25: [28093]: RECV/CQ: Bad 1D pixel count, row 1151, got 241,
> expected 1728
> Sep 25 12:30:06.36: [28093]: RECV: 32573 bytes of data, 1152 total lines
> Sep 25 12:30:06.37: [28093]: --> [17:+FPS:1,1152,0,0,0]
> Sep 25 12:30:07.74: [28093]: --> [6:+FET:0]

[...]

> TIFF Directory at offset 0x7f46
>   Subfile Type: multi-page document (2 = 0x2)
>   Image Width: 1728 Image Length: 1152
>   Resolution: 204, 98 pixels/inch
>   Bits/Sample: 1
>   Compression Scheme: CCITT Group 3

Yes, reencoding does not work -- should be CCITT Group 4 here.

>   Photometric Interpretation: min-is-white
>   FillOrder: lsb-to-msb
>   Date & Time: "2000:09:25 12:29:48"
>   Host Computer: "linux1"
>   Software: "HylaFAX (tm) Version 4.1beta44"
>   Image Description: "xx"
>   Make: "RING\nU.S. ROBOTICS 56K VOICE"

Oh, USR Sportster in Class 2.0 mode. Are you really looking for trouble? :-)

Hope to hear from you soon,
Dmitry




____________________ HylaFAX(tm) Users Mailing List _______________________
 To unsub: mail -s unsubscribe hylafax-users-request@hylafax.org < /dev/null