[hylafax-devel] Re: My RTN patch and many other changes/enhancements, third edition :-)

[Dmitry Bely <dbely@mail.ru>]

Steve Williams <steve@genie96.com> writes:

Sorry, I just missed you message, when answered to Tim...

> Alot of this is above my head, but a few things did jump out at me...
> 
> > 
> > 8. Default value for ModemResetCmds is now "ATZ". (ModemSoftResetCmd
> > parameter was never used :-)) (faxd/ModemConfig.c++, man/config.4f)
> > 
> 
> This is REALLY bad!!  ATZ sets modems to totally random settings, depending
> on how the last program did that did an AT&W had configured the modem.
> 
> I know FOR A FACT that one of our apps that dials out does an ATE0Q1&W ( turns
> off echo and result codes and saves it ).  When HylaFAX starts up, and does 
> an ATZ, things are going to break because it will NEVER see an OK come back.

If you do so, *you* are in trouble. ATZ not only restores user profile, but
is also (the only) universal method to reset the modem (almost any other
fax software do so). Also note, that it is a *default*, i.e. value that is
used then the parameter ModemResetCmds is *not* present in
config.modem. And explain me why the following is acceptable (from config
(5f)):

[---cut---]
Tag                     Type    Default Description
ModemSoftResetCmd       string  ATZ     command for doing a soft reset

[...]

ModemSoftResetCmd
The command to force a soft reset of the modem.
[---cut---]

?
Only because it's never used due to the Sam's (?) fault? :-)

> My personal opinion is that ATZ has ABSOLUTELY no place in ANY config file 
> because there is no predicting what exactly it will do.
> 
> > 
> > 10. Default value for ModemFlowControl is now XONXOFF -- I think, reasonable
> > default for the most modems. (faxd/ModemConfig.c++, man/config.4f)
> > 
> 
> Humm... what is the default now?

As I explained above, the value used then the parameter is *not* present
in your config.modem (man config). If it *is* present, nothing will change
for you.

> I would personally say that HARDWARE flow
> control would be the preferable Default flow control.

No. XONXOFF is supported by any modem in fax mode (an is set then the modem 
enters fax mode getting AT+FCLASS=2 command), but it is not true for
RTSCTS. The fact that the modem supports hardware flow control in data mode
does *not* mean, that it also supports it in data mode. (BTW, Harald, I
know you are reading this, what about XONXOFF in Class 1? Is it set after
AT+FCLASS=1?)

> That way, if people
> are using DATA mode for binary file transfers, there won't be suspicious 
> "hangs" when XOFF characters get sent to the modem.

Most modems *change* flow control scheme, then they enter fax mode.

Hope to hear from you soon,
Dmitry



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