info-kermit@ucbvax.ARPA (06/29/85)
From: Frank da Cruz <SY.FDC@CU20B.ARPA> Info-Kermit Digest Fri, 28 Jun 1985 Volume 2 : Number 40 Departments: C-KERMIT - Yet Another Unix Kermit Release (minor) MacKermit Observations MS-DOS KERMIT - Bug in MS-Kermit 2.28 and Tandy 2000 support MS-Kermit Reverse Wrap at Column 1 MISCELLANY - FTP Problems on CU20B ---------------------------------------------------------------------- Date: Fri 28 Jun 85 17:12:33-EDT From: Frank da Cruz <SY.FDC@CU20B> Subject: Yet Another Unix Kermit Release (minor) To: Info-Kermit@CU20B This is to announce C-Kermit 4C(055). No major changes, mainly just some enhancements to the dial command, support for a some additional systems and modems. Here, for those who understandably have a hard time keeping up with all this, is a list of changes since 4C(052) was announced on June 18: C-KERMIT FOR UNIX, CHANGES FROM 4C(054) TO 4C(055), 28 JUNE 85: ckudia.c (all changes by Dan Schullman, DEC): . Add support for US Robotics modem (untested) from Joe Orost at Berkeley. . Reorganize MDMINF data structure to accommodate US Robotics (some char fields had to become strings). . Allow interrupts (SIGINT, e.g. ^C) to cancel dialing in progress. . Ring bell when connection made successfully. . Close line on failures. . Allow stored numbers with DF100 and 200 modems. ckudia.c now supports the following modems: . Cermetek Info-Mate 212 A . DEC DF03-AC . DEC DF100 Series . DEC DF200 Series . General Data Comm 212A/ED . Hayes Smartmodem 1200 & compatibles . Penril . Racal Vadic . US Robotics 212A . Ventel Plus "unknown" modems and direct (modemless) connections. C-KERMIT FOR UNIX, CHANGES FROM 4C(053) TO 4C(054), 25 JUNE 85: ckuker.mak (makefile): . Add "make ft17" for Fortune 16:32 For:Pro 1.7. . Add "make uts24" for Amdahl UTS 2.4 . Add "make valid" for Valid Scaldstar CAD system . Add "make c70" for BBN C/70 IOS 2.4 ckcmai.c: . Add call to sysinit() ck[uvm]tio.c: . Add sysinit() function. For VMS, open console. For others, null for now. ckutio.c, ckufio.c: . Add support for Fortune 16:32, mostly like 4.1bsd. . Ditto for Amdahl UTS 2.4, mostly like V7. ckuus2.c: . Expand a couple tabs in hlp1 (-h help message) so things line up right. C-KERMIT FOR UNIX, CHANGES FROM 4C(052) TO 4C(053), 21 JUNE 85: ckcfn2.c: . Change dopar() to be of type CHAR. . Fix dopar() to calculate odd parity correctly. ckucon.c, ckuscr.c: . Add "extern CHAR dopar();" declarations. The files, as usual, are in KER:CK*.* on CU20B, available via anonymous FTP. There's no real need to get them unless one or more of the changes listed above is of use to you. ------------------------------ Date: Fri 28 Jun 85 10:32:33-EDT From: Frank da Cruz <SY.FDC@CU20B> Subject: MacKermit Observations To: manis%ubc.csnet@CSNET-RELAY In-Reply-To: Message from "Vincent Manis <manis%ubc.csnet@csnet-relay>" of Thu 27 Jun 85 22:49:46-EDT >Date: Thu, 27 Jun 85 19:42:15 pdt >From: Vincent Manis <manis%ubc.csnet@csnet-relay> >To: info-kermit@CU20B >Subject: MacKermit > >I've been using MacKermit for a week or so (0.8 (28)), and would like >to know whether the following are intended to be fixed: We're up to 0.8(32) now... >1) If I tell my local Unix that I'm a VT100, emacs has difficulty > refreshing the screen properly. Is the VT102 emulation imperfect, > or simply different enough from the VT100 that I should make a new > termcap entry? We use Mac Kermit with EMACS (both CCA and GNU) at 9600 baud on our VAX/Unix system and have not seen this problem. We tell it we're a VT100 since our EMACS's (or termcaps) don't yet have support for VT102. See if 0.8(32) exhibits the same behavior. >2) I'd like the ability to include an initial character string to send > automatically when I fire up a settings document. This would > provide some modem support (even if not terribly well). A reasonable idea; we'll add it to the list of things to do. A lot of other things have higher priority though, like screen save/print, saving lines off the top, cutting/pasting the screen, sizing the screen, allowing international character sets, etc. >3) Any possibility of mouse support (clicking the mouse somewhere away > from the current cursor position would send the requisite cursor > positioning string)? I doubt it. The program is already straining the limits of a 128K Mac. >4) Why doesn't KerKey have the ability to create a new settings > document? Why can't the settings menu from MacKermit be included > in KerKey as well? This would allow one to create a complete > settings file at one time. Simply a question of not having enough time to write the code. >5) I've had some sort of transient behaviour in which the first file > transfer in a session works ok, but subsequent ones just hang. > Firing up MacKermit again seems to cure the problem, but it's a > nuisance. (On the Unix end, this happens with both CKermit and the > old Unix Kermit, but not with enough regularity for me to pin it > down). You're not using parity, are you? If so, get 0.8(32), which fixes problems with parity. Otherise, it's hard to know what to say about this without more details about what machine, what Unix, what version of C-Kermit, what kind of connection, what baud rate, etc etc. This kind of behavior is always a possibility between any two Kermits and usually has more to do with the operating systems or communications equipment that Kermit itself. For instance, if you're running 4.2 bsd on a VAX with DMR's, you might be suffering from problems in Berkeley's DMR driver. Some other systems just get tired of allocating tty buffers after a while... We've used MacKermit to transfer very large batches of files at high baud rates without observing any consistent kind of failure. >6) The key in the upper right hand corner is labelled ``Backspace''. > By default it should generate a backspace. Matter of taste. You can always type a backspace using CTRL-H, but how do you type a DEL if there's no DEL key? Anyhow, that's what KerKey is for -- so people who don't like the defaults can change them. To be perfectly honest about the situation, Mac Kermit is "between maintainers" just now. The last two people who worked on it have both left, so I expect it will stay pretty much as it is for quite some time. - Frank ------------------------------ Date: Thu, 27 Jun 85 19:04:46 cdt From: goldberg@Uiuc.ARPA (Phil Goldberg) Subject: Bug in MS-Kermit 2.28 and Tandy 2000 support When you type 'C?' from the command prompt, instead of getting a 'Confirm with CR' message as in 2.27, you get a list of all commands from C-Z displayed. Since C is a valid command, isn't this a bug. I looked, but I couldn't figure it out. [Ed. - Hmmm... must have something to do with C being a synonym for CONNECT (necessary because of the addition of the CLEAR command). Added to the list of bugs, KER:MSKERM.BWR.] Steve Alexander has done mods to MSX/YIBM to get a Tandy 2000 module for v2.28. He swears up and down that this time he will send them in. Phil Goldberg goldberg@Uiuc [Ed. - Hope so...] ------------------------------ Date: Thu, 27 Jun 85 17:44:22 pdt From: gts%ucbpopuli.CC@Berkeley (Greg Small) To: info-kermit@cu20b Subject: MS-Kermit Reverse Wrap at Column 1 The backspace from column 1 to column 80 of the previous line was added for UNIX. For normal input echoing, UNIX assumes that the terminal handles all margin wraping. This includes both the normal forward wrap at the right margin and the less known reverse wrap at column 1. Many newer terminals do both. ANSI X3.64 enumerates some margin actions but does not specify any. Of course this only impacts those who enter and then wish to erase characters from lines longer than 80 characters. I had thought than no one would care, but our beta test release of MS-Kermit 2.27 turned up a number of users who complained that they could not erase characters accross the line wrap. I confess that I am not overly concerned with exact emulation of the limitations of the H19, our priority is delivering functionality. However we do try to avoid extensions which are likely to cause dificulties. Actually, we have abandoned pretending that a particular program emulates a real terminal. We now treat each emulator and version thereof as a seperate terminal type. Greg Small (415)642-5979 Microcomputer Communications gts@ucbpopuli.Berkeley.ARPA 214 Evans Hall CFO ucbvax!ucbpopuli!gts University of California SPGGTS@UCBCMSA.BITNET Berkeley, Ca 94720 ------------------------------ Date: Thu, 27 Jun 85 17:16:57 pdt From: rag@uw-june.arpa (David Ragozin) Subject: FTP problems on CU20B Today I tried to FTP some of the vms*.mar files on k2:. None of them would come. Each time I tried (with "TYPE ASCII") I received the error message: File not 7-bit - not text file. That sounds to me like a problem at your end, not mine. Could you have someone check it out? [Ed. - Oops! The bytesize was indeed 36, rather than 7. This is fixed now. Surprised nobody reported this before. In fact, all the following files had 36-bit bytesizes, which are now changed to 7: AP2KER.ASM AP2KER.TXT APPLEK.M65 CP4CPT.HEX CPMH8.HEX CPMPRO.HEX K10MIT.CCL K10V3.MEM MTSKERMIT.ASM MTSKERMIT.DOC MTSKERMIT.PAS PROV1.MEM RTED.TXT Sorry for the confusion; most of these files were created by TOPS-10 utilities that run on our DEC-20 system.] Also, is there some reason you know why I haven't been able to get any response from KERMSRV on BITNET for many days? [Ed. - KERMSRV has been undergoing reorganization the last few days. Should be up now.] ------------------------------ End of Info-Kermit Digest ************************* -------