rminnich@udel.EDU (Ron Minnich) (05/04/88)
1) WHy do i want to do this?
Cause all the io, in lattice and manx, is in terms of file handles.
I just don't LIKE the console.device nonsense. Every program i
have seen does it a little differently. This seems silly.
Which brings me to the thing i don't want to ask:
Anybody out there disassembled newcli yet?
After all, newcli, or maybe the con: driver, does what i want
to do! OR maybe there is a library function to do it?
Like i say, i would rather not know, but, lacking any REASONABLE
alternative ...
--
ron (rminnich@udel.edu)cmcmanis%pepper@Sun.COM (Chuck McManis) (05/04/88)
In article <2370@louie.udel.EDU> rminnich@udel.EDU (Ron Minnich) writes: >1) WHy do i want to do this? > Cause all the io, in lattice and manx, is in terms of file handles. > I just don't LIKE the console.device nonsense. >ron (rminnich@udel.edu) Ron has been sending these out for a bit and while I thought the answer was fairly clear I suspect it is not. The 'simple' answer is to put a window pointer into the startup message of the console device when starting it, and then jamming that value into a file handle at the appropriate place. Basically, there are three things we are dealing with here : CON: - can be opened and accessed with read(3), and write(3), as well as the stdio library. Limited to opening windows on the WORKBENCH screen. console.device - can be opened with the AmigaDOS Open(2) call and accessed with SendIO(2) and DoIO(2), etc. Normally limited to the WORKBENCH but some tricks can be used (described in AmigaMail) to make it open on a different screen. Window - Can be opened with OpenWindow() and written to using the Text() call. This is what vt100 does, the extremely patient person could possibly turn the VT100 program into a library routine for opening 'terminal' windows. The problem with the solution above is the 'jamming' part. There is no officially sanctioned way to create a filehandle out of an open file descriptor. About the only think I can think of would be to open a CON: window, using the ACTION_INFO packet get a pointer to the ConUnit structure, put in the parameters from your opened console.device, and then send it an esc-c to get it to re-sync. Not pretty and not easy either. (Nor is it likely to work in future versions of the OS). My real suggestion is to put some wrappers around the calls to the console device to make them more usable, or rethink the problem. --Chuck McManis uucp: {anywhere}!sun!cmcmanis BIX: cmcmanis ARPAnet: cmcmanis@sun.com These opinions are my own and no one elses, but you knew that didn't you.
rminnich@udel.EDU (Ron Minnich) (05/05/88)
In article <51947@sun.uucp> cmcmanis@sun.UUCP (Chuck McManis) writes: >In article <2370@louie.udel.EDU> rminnich@udel.EDU (Ron Minnich) writes: >>1) WHy do i want to do this? >> Cause all the io, in lattice and manx, is in terms of file handles. >Ron has been sending these out for a bit and while I thought the answer >was fairly clear I suspect it is not. Yup, that is my problem. There are three or four ways to get a half-assed solution, most of them very unsatisfying and version-dependent. I have looked at the Process, ConUnit, etc. structures and folks, this is a REAL mess. So far no one has proposed a general solution that uses standard ADOS/Rom Kernel stuff- it all involves some pretty bizarre trickery. The worst part is that it seems to me you ought to WANT to do this!File Handles are the standard io mechanism for both lattice and manx now. Ergo, i think, you want file handles to windows. I am looking at a solution that involves using ConMan; i will let you know what i find. ron -- ron (rminnich@udel.edu)
peter@sugar.UUCP (Peter da Silva) (05/05/88)
In article <51947@sun.uucp>, cmcmanis%pepper@Sun.COM (Chuck McManis) writes: > console.device - can be opened with the AmigaDOS Open(2) call and accessed ^^^^^^^ ??????? Don't you mean OpenDevice? By the way, the code fragment on page 279 of the AW version of RKM:L&D is of interest: OpenConsole(writerequest,readrequest,window) struct IOStdReq *writerequest; struct IOStdReq *readrequest; struct Window *window; { int error; writerequest->io_Data = (APTR)window; writerequest->io_Length = sizeof(*window); error = OpenDevice("console.device", 0, writerequest, 0); readrequest->io_Device = writerequest->io_Device; readrequest->io_Unit = writerequest->io_Unit; return error; } -- -- Peter da Silva `-_-' ...!hoptoad!academ!uhnix1!sugar!peter -- "Have you hugged your U wolf today?" ...!bellcore!tness1!sugar!peter -- Disclaimer: These aren't mere opinions, these are *values*.
ewhac@well.UUCP (Leo 'Bols Ewhac' Schwab) (05/06/88)
In article <51947@sun.uucp> cmcmanis@sun.UUCP (Chuck McManis) writes: >console.device - can be opened with the AmigaDOS Open(2) call and accessed > with SendIO(2) and DoIO(2), etc. Normally limited to the > WORKBENCH but some tricks can be used (described in AmigaMail) > to make it open on a different screen. Say what, Chuck? CON: is the DOS domain goodie, and console.device is the Exec domain goodie. CON: is opened with Open() and console.device is opened with OpenDevice(). The console.device can be attached to any window on any screen. This is explained quite well in the AutoDocs. Or am I missing some underlying Deep Meaning you were trying to impart? _-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_ Leo L. Schwab -- The Guy in The Cape ihnp4!pacbell -\ \_ -_ Recumbent Bikes: dual ---> !{well,unicom}!ewhac O----^o The Only Way To Fly. hplabs / (pronounced "AE-wack") "Work FOR? I don't work FOR anybody! I'm just having fun." -- The Doctor
cmcmanis%pepper@Sun.COM (Chuck McManis) (05/07/88)
In article <5887@well.UUCP> ewhac@well.UUCP (Leo 'Bols Ewhac' Schwab) writes: ->In article <51947@sun.uucp> FumbleFingers writes: ->>console.device - can be opened with the AmigaDOS Open(2) call and accessed ->> with SendIO(2) and DoIO(2), etc. Normally limited to the ->> WORKBENCH but some tricks can be used (described in AmigaMail) ->> to make it open on a different screen. -> -> Say what, Chuck? I would like to say it was all planned and that I was just checking to see if anyone was reading, unfortunately this particular passage required some "name expansion" on the part of the reader. Yes, I meant OpenDevice() and not Open(). The DoIO and SendIO should have given it away :-) --Chuck McManis uucp: {anywhere}!sun!cmcmanis BIX: cmcmanis ARPAnet: cmcmanis@sun.com These opinions are my own and no one elses, but you knew that didn't you.