October 2006 - Oberon - mailman3.inf.ethz.ch

Further experience with the CF card.
by Peter Easthope 31 Oct '06

31 Oct '06

Folk, After almost a year of routine use, the CF card refuses to mount. A backup is installed and little if any data is lost. The mount command and trap follow. The OFSBoot.BootSystem at the bottom of the trap seems odd. Is there any advice before I dash to reformat? Thanks, ... Peter E. FileSystem.Mount MY AosFS IDE1#3 ~ TRAP 7 Index out of range (PC Native 05.01.2003) OFSAosFiles.TraverseDir PC = 8627 a = 0000001CH dpg = 1624 i = 102 OFSAosFiles.TraverseDir PC = 8815 a = 0000001CH dpg = 29 i = 3 OFSAosFiles.DirInit PC = 8998 A = 37294, 74066, 115391, 145638, 156919, 177132, 201318, 236437 ... bad = FALSE files = 0 init = FALSE k = 386 vol = 0005D7A0H OFSAosFiles.NewFS PC = 20776 ch = 00000000X fs = 00068F40H i = 2 init = FALSE prefix = "MY" s = "" vol = 0005D7A0H Oberon.Call PC = 4860 Mod = 0003DFE0H P = OFSAosFiles.NewFS PC = 20239 i = 12 j = 17 name = "NewFS" new = FALSE par = 0038D120H res = 1 FileSystem.Mount PC = 1958 alias = "AosFS" cache = 00000000H cacheSize = 0 genfs = "OFSAosFiles.NewFS" genvol = "OFSDiskVolumes.New" hashSize = 0 i = 7 par = "IDE1#3 " prefix = "MY" res = 1 s = 00000014H Oberon.Call PC = 4860 Mod = 001C38E0H P = FileSystem.Mount PC = 930 i = 11 j = 16 name = "Mount" new = FALSE par = 0038D120H res = 1 TextFrames.CallCmd PC = 12027 F = 00354D40H cmd = "FileSystem.Mount" new = FALSE par = 0038D120H pos = 656 res = 1 TextFrames.Call PC = 12915 F = 00354D40H S = 00000048H h = 0 hint = "" i = 0 new = FALSE pos = 640 TextFrames.Edit PC = 20137 F = 00354D40H Keys = {1} M = 00000060H R = 0000004CH X = 90 Y = 571 beg = 0 buf = 00000000H ch = 00000000X end = 0 keysum = {1} pos = 640 text = 00000000H time = 0 TextFrames.Handle PC = 21222 F = 00354D40H F1 = 00000000H M = 00000020H ET.Handle PC = 10006 F = 00354D40H M = 0000000CH T = 003C37A0H handled = FALSE ml = FALSE tF = 00354D40H MenuViewers.Handle PC = 4365 M = 00000014H Main = 00354D40H Menu = 003B9440H V = 001CD400H V1 = 00000000H Oberon.Loop PC = 8578 M = 00000010H N = 0000000CH V = 001CD400H X = 90 Y = 571 ch = 00000020X keys = {1} oldshift = {} shift = {} System.Init PC = 20240 F = 003D1F00H S = 00000044H T = 003D1E80H Wt = 00000050H ok = FALSE OFSBoot.Call PC = 426 c = System.Init PC = 19716 cmd = "Init" i = 11 j = 4 m = 001636C0H protect = TRUE OFSBoot.BootSystem PC = 2780 s = "System.Init" OFSBoot.$$ PC = 13 @SELF = 0003F120H SELF = 0003F120H -- http://carnot.pathology.ubc.ca/

1 0

Refine *Mail*.Mod
by easlab＠absamail.co.za 29 Oct '06

29 Oct '06

edgarschwarz wrote: > I recently noticed that my (antique) Mail.Mod doesn't show the correct > date in Mail.Panel. > I traced and fixed the problem to MIME.Mod. See > "http://www.edgar-schwarz.de/repos/obedav/nopatch/MIME.Mod" > Other patches to NO (Meaning for NO in Bluebottle) are welcome. Does anybody else use an 'ETH' mailer ? AFAIK none of them does transmit-authentication. With the increase in spam this is surely now demanded by most ISPs ? Initially I was able to bluff transmit-authentication by doing: NetSystem.SetUser <args> TextMail.Directory before transmitting. Then my ISP required proper transmit-authentication. Ie. as per Micro$loth outsp00k -- the gold-standard !! Partly because I have to pay a dialup call for each test-cycle, I haven't managed to update the source code to do the required authentication yet. Under linux, the relevant rfc is easily confirmed, and my trace [I have to post this from linux because N-O fails] shows:----- # telnet smtp.absamail.co.za 25 Trying 196.35.40.10... Connected to smtp.absamail.co.za. Escape character is '^]'. 220 absamail.co.za Novonyx SMTP ready $Revision: 3.22.1.16 $ ehlo 250-absamail.co.za Pleased to meet you 250-HELP 250-EXPN 250-PIPELINING 250-8BITMIME 250-DSN 250-AUTH LOGIN <--- the Tx-authentication type available [per M$!] 250-AUTH=LOGIN 250 SIZE 3145728 AUTH LOGIN 334 VXNlcm5hbWU6 <--- Base64("Username:") Y3JnbHVyQGFic2FtYWlsLmNvLnph <--- Base64("crglur(a)absamail.co.za") 334 UGFzc3dvcmQ6 <--- Base64("Password:") NTMxMDQz <--- Base64(<my email password>) 235 Authentication successful! --------------- end of trace. A pop-side telnet trace is similarly simple as per rfc. BTW linux-oberon is convenient to do the Base64 encoding/decoding. N-O telnet is problematic. Telnet.Panel 's default "user@host" is IMO wrong, since valid arguments are: URL [port] ; which I can't relate to "user@host" ? The other problem, which I didn't solve, to automate the linux telnet session via a script either, is how to input, to the remote host, not by keyboard, but from a script. Under linux> telnet I can input the [eg.] Base64 string via pasting [gpm] text, followed by <return-key>. Analagously, N-O's V24.Panel [which I suspect Telnet.Panel is an incomplete copy attempt of] also sends to the remote host either what's keyed-in or pasted. AFAI can remember V24.Panel sends N-O line terminators, unless 'ET.OpenAscii' text is pasted. Importantly [AFAIK] telnet protocol expects CR/LF, which linux > telnet does. I wonder if N-O telnet does too ? Since the telnet access to pop and smtp servers is so simple and exactly as per rfc, I'm wondering how it is that my '90's V2.3.2 Mail.Panel would 'not connect'. I remember suspecting some subtle timing problem[s], which seemed to be confirmed when the DOSbased V2.3.6 did connect without any apparent change in the source code ? .. to be continued ....as a proposal HOW to fix ETH-mail... == Chris Glur.

4 4

Re: [Oberon] Refine *Mail*.Mod
by edgar＠edgarschwarz.de 29 Oct '06

29 Oct '06

easlab(a)absamail.co.za wrote: > Does anybody else use an 'ETH' mailer ? Guess we are the last ones :-) > AFAIK none of them does transmit-authentication. > With the increase in spam this is surely now demanded by most ISPs ? My ISP still has a POP3 mailer which seems to give an IP a limited transmit timeslice after an receive with authentication. When this should end ... .. I will have to look for a solution to my problem. Cheers, Edgar

1 0

Re: [Oberon] Refine *Mail*.Mod
by Peter Easthope 29 Oct '06

29 Oct '06

Chris wrote, > Does anybody else use an 'ETH' mailer ? Yes; I still use the Oberon Mail Panel with GulfNet. With cablelan.net there are problems. I have postponed investigating while cablelan upgrades server software. When the service is stable again, I'll see whether Oberon can use it. Regards, ... Peter

1 0

Re (2): [Oberon] supercall in AO
by shark＠gulfnet.sd64.bc.ca 28 Oct '06

28 Oct '06

Thomas, tf> ... Do^; (* re-uses the previous definition of Do in the "super object class" (A in this case) and is called a supercall *) ... OK, thanks. This structure of a module containing objects with associated methods appears to be very similar to the mathematical category containing objects and maps. So an AO module might be viewed as an instance of a category. Probably this aspect has been discussed in the literature. Anyone know of a specific publication? Thanks, ... Peter E. Desktops.OpenDoc http://carnot.pathology.ubc.ca/

1 0

Re: [Oberon] supercall in AO
by Thomas Frey 28 Oct '06

28 Oct '06

> Seems that an "overridden method" is one which > is redefined in an extended type. Correct? > > And what is a supercall? A call to the "previous definition" of a method. E.g. A = OBJECT PROCEDURE Do; BEGIN END Do; END A; B = OBJECT(A) PROCEDURE Do; (* overrides procedure Do *) BEGIN (* Do; (* would be a recursive call of Do *) *) Do^; (* re-uses the previous definition of Do in the "super object class" (A in this case) and is called a supercall *) END B; --Thomas

1 0

NS32000 Oberon Binaries
by bwc 27 Oct '06

27 Oct '06

I'm looking for the binaries for the ns32000 Oberon V4 system. Can anyone tell me how I can get a hold of the binary files? I'm especially interested in the compiler modules. Brantley Coile brantley(a)coraid.com

1 0

supercall in AO
by shark＠gulfnet.sd64.bc.ca 27 Oct '06

27 Oct '06

On page 12 of his dissertation, Pieter said "Supercalls in an overridden method are marked with the `^' character." Seems that an "overridden method" is one which is redefined in an extended type. Correct? And what is a supercall? Thanks, ... Peter E. Desktops.OpenDoc http://carnot.pathology.ubc.ca/

1 0

Re (2): [Oberon] Re: Oberon Digest, Vol 37, Issue 7
by edgar＠edgarschwarz.de 26 Oct '06

26 Oct '06

shark(a)gulfnet.sd64.bc.ca wrote: > Speaking as an unsophisticated user, the > Oberon TUI, with Gadget panels allowed in > tracks, remains the best interface I've ever > used. However posh the Aos GUI becomes, we > need to maintain the Oberon TUI. I also like the TUI. What about PET: File to user and Outline to system track ? Also would it be nice for PC.Compile to give error markers like Builder.Compile. Or is the TUI inherently thread unsafe so that it would be A LOT of work to do this ? Cheers, Edgar

1 0

Re: [Oberon] Re: Oberon Digest, Vol 37, Issue 7
by shark＠gulfnet.sd64.bc.ca 26 Oct '06

26 Oct '06

sr> ... editing programs is otherwise wonderfully easy in the modeless Oberon GUI. Speaking as an unsophisticated user, the Oberon TUI, with Gadget panels allowed in tracks, remains the best interface I've ever used. However posh the Aos GUI becomes, we need to maintain the Oberon TUI. Thanks again ETH, Peter E. Desktops.OpenDoc http://carnot.pathology.ubc.ca/

1 0