Source Code 1994 March/Source_Code_CD-ROM_Walnut_Creek_March_1994.iso/compsrcs/misc/volume32/ecu/part03

Text File | 1992-09-10 | 56.9 KB | 1,577 lines

Newsgroups: comp.sources.misc
From: [e-mail] (Warren Tucker)
Subject:  v32i038:  ecu - ECU Asynchronous Communications v3.20, Part03/40
Message-ID: <[e-mail]>
X-Md4-Signature: 5ce87794e0189803dfbd29f778611c49
Date: Fri, 11 Sep 1992 19:19:29 GMT
Approved: [e-mail]

Submitted-by: [e-mail] (Warren Tucker)
Posting-number: Volume 32, Issue 38
Archive-name: ecu/part03
Environment: SCO,XENIX,ISC,SUNOS,SYSVR4,HDB,Curses
Supersedes: ecu: Volume 21, Issue 53-89

---- Cut Here and feed the following to sh ----
#!/bin/sh
# this is ecu320.03 (part 3 of ecu320)
# do not concatenate these parts, unpack them in order with /bin/sh
# file OLD-HISTORY-1 continued
#
if test ! -r _shar_seq_.tmp; then
    echo 'Please unpack part 1 first!'
    exit 1
fi
(read Scheck
 if test "$Scheck" != 3; then
    echo Please unpack part "$Scheck" next!
    exit 1
 else
    exit 0
 fi
) < _shar_seq_.tmp || exit 1
if test ! -f _shar_wnt_.tmp; then
    echo 'x - still skipping OLD-HISTORY-1'
else
echo 'x - continuing file OLD-HISTORY-1'
sed 's/^X//' << 'SHAR_EOF' >> 'OLD-HISTORY-1' &&
X
X7.  If you are not using a 43-line screen, the list of commands
X    displayed by the interactive 'help' command is truncated.  I
X    apologize; this problem will be solved Real Soon Now.
X
XChanges since ecu 2.80 patchlevel 1
X
X1.  The built-in help system was reworked.  Interactive commands are
X    now divided into four categories: general, communications-related,
X    transfer-related and procedure-related.  The help command asks for
X    a category and displays the appropriate subset.
X
X2.  The built-in help file (ecuhelp.data) was revised.
X
XChanges since ecu 2.80 patchlevel 2
X
X1.  Patch 3 added a window mode selection to the sz file transfer menu;
X    also, the unacknowledged data quantity (called "window") was
X    added to the sz file transfer display.
X
XChanges since ecu 2.80 patchlevel 3
X
X1.  A problem with passing arguments to an initial procedure was
X    fixed.  Getopt() is now used, which requires that all switch
X    command arguments must precede the non-switch arguments.
X    This was not true in previous revisions of ecu.
X
X2.  A new (non-standard) escape sequence has been added to the
X    expresp procedure command.  \m### causes a ### millisecond delay.
X    Note this adds an ambiguity to the syntax of the send string:
X    you may not follow an \m delay with a literal digit to send.
X    For example, suppose you wish to delay 300 missileconds then
X    send "4A".  \m3004A will delay for 3004 (approximately) milliseconds,
X    then send an "A".  You must, instead, use \m300\034A to achieve
X    the desired result.
X
X    Also, a string of the form ~t[##] or ~m[##] may be prepended to
X    any expect vstring to change the timeout to ## seconds or milliseconds,
X    respectively.  '~t[8]gin:--~m[800]gin:' waits 8 seconds for
X    'gin:' and if not found, sends a carriage return and waits
X    an additional 800 milliseconds.
X
X3.  The method of choosing a shared memory segment key changed.
X    Previously, ftok() was used.  Now, the algorithm is
X        0xEC000000L | getpid();
X
X4.  The -D switch has been added to cause ecu to terminate
X    unconditionally upon completion of an initial procedure (-p).
X
XChanges since ecu 2.80 patchlevel 4
X
X1.  The malloc library is used, resulting in faster procedure
X    execution.
X
X2.  Restrictions on string operations limiting useful string length
X    to 256 characters have been lifted.  New restrictions limit
X    strings to 5120 characters in length :-).  This required some
X    major redesign and is responsible for the reissue of the entire
X    ECU program rather than merely patches.
X
X3.  A new procedure command, 'gosub' has been added, allowing
X    subroutines within a procedure file.
X
X    #---------
X    # example
X    #---------
X        echo 'main'; gosub SUB; echo 'back to main'
X        return           | return from proc
X
X    SUB
X        echo 'sub'
X        gosub SUB2
X        echo 'back to sub'
X        return           | return from subroutine
X
X
X    SUB2
X        echo 'sub2'
X        return           | return from subroutine
X
X4.  Other minor unreported bugs were fixed.
X
XChanges since ecu 3.00
X
X============= Patch 1:
X> Posting-number: Volume 16, Issue 60
X> Archive-name: ecu3/patch01
X>
X> This is patch 1 to ECU 3.  It provides speed improvements
X> and fixes for non-ANSI terminal support.  Also provided is an
X> experimental make file for GCC 1.37.1 on UNIX systems.
X
X============= Patch 2:
X> Posting-number: Volume 16, Issue 70
X> Archive-name: ecu3/patch2a
X>
X> Posting-number: Volume 16, Issue 71
X> Archive-name: ecu3/patch2b
X>
X> ECU 3.0 PATCH # 2 - Non-ansi/ISC
X>
X> 1.  Fix non-curses per-file transfer rate report
X>     (it was verrrrry optimistic :-)
X>
X> 2.  Add experimental support for ISC 2.2
X>
X> Note: if you are purely an SCO multiscreen user, you may wish to
X> save this patch and not apply it until further patches are issued.
X> By doing so, you will save an unnecessary compilation.
X> When another patch comes along, you must then apply this one so
X> that your source will be in sync.
X
X============= Patch 3:
X> Posting-number: Volume 16, Issue 87
X> Archive-name: ecu3/patch03
X>
X> Under XENIX,
X> 1. libmalloc.a and/or libc.a are broken such that calloc is
X>    resolved twice, once from each library.
X> 2. memmove() does not exist (in all or less recent versions?)
X>
X> To get ecu 3 to compile under XENIX, apply this patch with:
X>      patch < thismessage
X>
X> NOTE there is no test to see if patch 1 & 2 have been applied :-(
X> so make sure you have done so.  Also, I don't have access to XENIX
X> right now, so please forgive if this doesn't fly. The object is to
X> 1. supply -DDUFF to use Duff's device instead of missing memmove()
X> 2. remove #define MALLOC_3X and -lmalloc to avoid bug in libmalloc.a
X
X============= Patch 4:
X> Posting-number: Volume 16, Issue 88
X> Archive-name: ecu3/patch04
X>
X> ECU 3 PATCH 4 - miscellaneous, 1 serious
X>
X> 1. fix bug in non-ANSI key handling causing core dump fairly
X>    often (but not on my sys :-<)
X>
X> 2. duplicate definiton of ECULIBDIR in ecu.h fixed 3.  cosmetic
X>    fix in 'fi' interactive command prompt
X>
X> 4. comm line flusher lflush() was flushing console tty instead
X>
X> 5. illegal color name in 'color' procedure command caused
X>    hi_white on hi_white instead or error
X>
X> 6. prototypes for strchr/strrchr in stdio_lint.h caused problems
X>    on some systems
X>
X> 7. procedure sz was not reporting proper results if files skipped
X>
X> 8. miscellaneous cosmetic fixes and maintenance/experimental rearrangements
X
X============= Patch 5:
X> Posting-number: Volume 16, Issue 97
X> Archive-name: ecu3/patch05a
X>
X> 1. Fixed memory leak + bad free in feval_str().
X>
X> 2. Some common types of procedure string operations have been made faster
X> by redesign of the realloc_esd() function.
X>
X> 3. Octal integer constants were implemented; they may be denoted by
X> prefixing them with "0o".  Leading zero octal constants are not supported
X> due to ECUs mutiple use of the same integer constant parser.  I thought
X> most people would rather %itos("0200") return 200(10) than 128(10).
X>
X> 4. Many line drivers, particularly FAS, will wait a process closing
X> a tty line until all characters queued for output have been sent.
X> When closing a tty line that has been permanently XOFF'ed by a
X> dead remote system or a noisy XOFF, this can result in a hang.
X> Now, a SIGALRM/setjmp scheme is used to recover from this condition
X> and flush line output before reattempting the close.  Five
X> seconds are allowed for the draining to occur.
X>
X> 5. Multiscreen normal and reverse foreground and background colors
X> in effect at the beginning of ecu execution are restored when the
X> program terminates (thanks to staceyc@sco for illuminating the
X> undocumented GIO_IOCTL ioctl; I've wanted to be able to do this for
X> a long time).
X>
X> 6. [e-mail] reported a core dump immediately after
X> the one-line startup message under XENIX 286.  Others have had the
X> problem too.  I am puzzled -- it is happening in get_home_dir(),
X> which seems rather straightforward.  A patch to avoid the problem
X> is included: get the home directory from the environment variable
X> HOME if possible rather than invoke libpwent.a.
X>
X> 7. An embarassing nroff problem was reported by slerner@eingedi.
X> Patches are included for the nroff source.  Mr. Lerner also pointed
X> out an obsolete reference to samplproc.man.  The models subdirectory
X> contain many sample, even useful, procedures and samplproc.man
X> was discarded a long time ago.
X>
X> If you do not have nroff or the spare CPU cycles to make the
X> manual from nroff source, I'll email you one. Just mail me
X> at [e-mail] or emory!n4hgf!wht and put
X> "send ecu305 manual" in the subject line.  If I get enough
X> requests, I'll repost it to c.s.m as ecuman305.
X>
X> 8. A make bug in the trial ISC port was fixed in Make.ecu (thanks to
X> jmd@p1so).
X>
X> 9. During non-curses file transfer, the cumulative elapsed time
X> is reported once per complete line of 'RRRTRR..' status.
X>
X> 10. [e-mail] reported a bug in XENIX utmp handler
X> code.  This is a fix for most XENIX systems, but define OLD_WAY
X> in utmpstat.c if you have problems.
X>
X> 11. Sometimes upon adding a new phone entry, it would be marked
X> for redial upon creation.  This has been fixed by using calloc()
X> instead malloc().
X>
X> 12. We seem to be bumping some limits on 286 systems.  I added
X> more code to handle more core dumps reported by [e-mail]:
X> somehow newwin() managed to return zero (probably memory alloc fail).
X> The patch to ecuwinutil.c won't fix the problem, but will at least
X> report it and die gracefully, rather than splatter on the pavement
X> with a core dump.  Please write me with bug reports and crash info:
X> see below.
X>
X> .--------------------.
X> | NOTES and REQUESTS |
X> `--------------------'
X>
X> 1. I do not have any XENIX system to test with.  It is a tribute
X> to SCO orthogonality that 35,000+ line program works so well
X> across 2 CPUs and two operating systems of many minor releases
X> and many versions of development systems.  I continue to point
X> this out when I hear SCO-bashers do their thing.  However, since
X> s**t does happen, if you are reporting a problem that results in
X> a core dump, please run the following for me (I would APPRECIATE it):
X>
X>   % cd where ecu is
X>   % ./ecu
X>   ----- kaboom -----
X>   % ^Jstty sane^J if needed (probably)
X>   % adb ./ecu |& tee to.wht
X>   * $x
X>   * $c
X>   * $r
X>   * $q
X> If you are a really industrious one and have sdb and compiled with -g,
X>   % sdb ./ecu |& tee -a to.wht
X>   % t
X>   % q
X> Then
X>   % mail -s 'ecu 3.xx core dump' [e-mail] < to.wht
X>
X> 2. [e-mail] has made some excellent suggestions which
X> would enhance the line handling/choosing algorithms.  They will
X> require a change in the phone directory.  He suggests I interpret
X> old directory formats, while writing only new ones.  The changes
X> are far reaching and I haven't the time now to plug them in,
X> but kudos to him and I'll try to find A Round Tuit.
X
X============= Patch 6:
X> Posting-number: Volume 17, Issue 11
X> Archive-name: ecu3/patch06
X>
X> This patch (#6) fixes a bug in patch 5.  Things seem stable now.
X> I was going to wait for possible feedback from ISC porters, but
X> several folks have had this problem (no doubt since ECU won't work
X> without this patch).  This patch was also sent to comp.sources.bugs
X> on 13 February, but patchlevel.h was not updated.  You may get 1 hunk
X> rejected if you have already applied the patch.
X>
X> You need to remake/reinstall both ecu and ecuungetty.
X>
X> This man's hat sits on a quick-thinking head:
X> > Organization: DEMOS, Moscow, USSR
X> > Date: Wed, 13 Feb 91 02:15:23 +0300 (MSK)
X> > From: emory!hq.demos.su!ache (Andrew A. Chernov, canton Uri's citizen)
X> > Subject: ECU 3.05 utmpstat
X> >
X> > Hi, Warren, bad news.
X> > I just applied PATCH 5 for ECU, everything appears OK, but
X> > 'utmpstat' not work at all, I look in...
X> > IMHO it seems, like two strings are misplaced...
X> > (US_LOGIN and US_DIALOUT)
X> > I fix it, and ECU works correctly.
X> > Good Luck.
X
SHAR_EOF
echo 'File OLD-HISTORY-1 is complete' &&
chmod 0644 OLD-HISTORY-1 ||
echo 'restore of OLD-HISTORY-1 failed'
Wc_c="`wc -c < 'OLD-HISTORY-1'`"
test 21763 -eq "$Wc_c" ||
    echo 'OLD-HISTORY-1: original size 21763, current size' "$Wc_c"
rm -f _shar_wnt_.tmp
fi
# ============= OLD-HISTORY-2 ==============
if test -f 'OLD-HISTORY-2' -a X"$1" != X"-c"; then
    echo 'x - skipping OLD-HISTORY-2 (File already exists)'
    rm -f _shar_wnt_.tmp
else
> _shar_wnt_.tmp
echo 'x - extracting OLD-HISTORY-2 (Text)'
sed 's/^X//' << 'SHAR_EOF' > 'OLD-HISTORY-2' &&
X============= Changes for ECU 3.10 =======================================
X
XThis file gives hints about what has changed.  Rereading or skimming
Xthe manual is highly recommended.  My apologies if some changes did
Xnot get reflected in the manual.  Please let me know.
X
X1. Add nice interactive and procedure commands and %nice integer function.
X
X2. Add the %uid and %gid integer function.
X
X3. If ECU starts setuid, always reset the uid to the real uid.
X   Previously this only happened if ECU was setuid to root.
X
X4. Made compatible with Metro Link X11R4 xterm under SCO UNIX/386.
X   To use:
X   a. include -DMETROLINK_X11R4 in CFLAGS of ECU compilation.
X   b. edit the models/nonansikeys file patched by this patch
X      to comment out the SCO ODT xterm key definiton and
X      uncomment the Metro Link X11R4 definition.
X      Follow other instructions you see in the nonansikey file.
X
X5. Fixed miscellaneous typos and misspelled words in models/nonansikeys.
X
X6. Environment variables LINES and COLS override termcap li and co
X   values like terminfo curses does.  This gives you the correct
X   size screen on xterms, but the ecu curses stuff still sees
X   the termcap description un altered, giving you shorter (usually 24 line)
X   dialing and help menus.
X
X7. Support is provided in nonansikeys for Metro Link's MetroX newest
X   xterm (the one that uses pty pseudottys).  (Some of the key
X   bindings are differerent and some .xinitrc and xterm resources
X   are required; see 'models/nonansikeys'.)
X
X8. The -N switch for the procedure commands sx, sy, sz, rx, ry, and
X   rz forces the "no curses" mode for file transfer.  This mode
X   uses brief, tty-style reporting of file transfer progress.
X   The "no curses" mode is automatically invoked if the console
X   is not a pty or multiscreen and has a lower baud rate than the
X   attached communication line.
X
X9. I added a dialer for the "generic Hayes-style 2400" (dialgHA24) and
X   for the USR Courier 2400 (dialgUSR24).  My USR is at least
X   6 years old and I don't even know if they still make them or
X   if it is the same modem.
X
X10. A new config procedure greatly simplifies making and installing.
X
X11. I don't know if this will get it or not, but to try and help
X    our European friends to be able to use alt_[a-z].ep, the keycodes
X    required in the mapkeys file have changed from
X      128 through 153 (0x80 through 0x99)     to
X      229 through 249 (0xE0 through 0xF9).
X    Reread mapkeys/README.
X
X12. Termination of file transfers by SIGINT did not cause termination
X    of procedfure execution in pre-3.10 versions.  In this version, it does.
X
X    The appropriate manual sections have been updated to indicate:
X
X    Integer variable $i0 receives the exit status from the transfer.
X    Note: procedure execution IS terminated by a SIGINT to the spawned
X    file transfer process.  This is different than for versions prior
X    to 3.10.  If $i0 is set to -1, the file transfer program did not
X    begin.
X
X13. kbdtest3 is included to help figure out nonansikey configurations.
X    see models/nonansikeys.
X
X14. SCO UNIX MEMMOVE() AND GCC
X
Xmemmove() is a very, very fast non-overlapping memory copy, but
Xsince it does not save certain registers it uses, calling from
XGCC can be tricky.  GCC does a fabulous job of grinding all it
Xcan out of the available registers; MSC doesn't make very good
Xuse of registers, er, I mean keeps a lot of registers for scratch
Xuse.  I've never seen 386 MSC use dx except as a side effect of
Xmultiply.  I guess that's what somebody (Gates?) meant by "tuning"
Xone compiler to meet many needs being adequate. Enough philosophy,
Xthough.
X
XI am using of -fcall-save-{ax,bx,cx,dx} since in at least one
Xcase (ecufkey.c display_keyset()), rigorous optimization and
Xreally righteous register usage caused a call to strlen() to
Xscrew up since cx is not preserved by strlen.
X
Xstrlen:         push    edi
Xstrlen+0x1:     mov     edi,[esp+0x8]
Xstrlen+0x5:     xor     eax,eax               <-- goodbye ax
Xstrlen+0x7:     mov     ecx,0xffffffff        <-- goodbye cx
Xstrlen+0xc:     repne   scasb
Xstrlen+0xe:     inc     ecx
Xstrlen+0xf:     mov     eax,ecx
Xstrlen+0x11:    not     eax
Xstrlen+0x13:    pop     edi
Xstrlen+0x14:    ret
X
Xmemmove:        push    ebp
Xmemmove+0x1:    mov     ebp,esp
Xmemmove+0x3:    mov     edx,edi               <-- move rather than push
Xmemmove+0x5:    mov     ebx,esi               <-- move rather than push
Xmemmove+0x7:    mov     esi,[ebp+0xc]
Xmemmove+0xa:    mov     edi,[ebp+0x8]
Xmemmove+0xd:    mov     eax,edi               <-- goodbye ax
Xmemmove+0xf:    mov     ecx,[ebp+0x10]        <-- goodbye cx (OK w/MSC)
Xmemmove+0x12:   jcxz    memmove+0x43
Xmemmove+0x14:   cmp     edi,esi
Xmemmove+0x16:   jbe     memmove+0x2e
Xmemmove+0x18:   mov     eax,esi
Xmemmove+0x1a:   add     eax,ecx
Xmemmove+0x1c:   cmp     edi,eax
Xmemmove+0x1e:   jae     memmove+0x2e
Xmemmove+0x20:   mov     eax,edi
Xmemmove+0x22:   add     esi,ecx
Xmemmove+0x24:   add     edi,ecx
Xmemmove+0x26:   dec     esi
Xmemmove+0x27:   dec     edi
Xmemmove+0x28:   std
Xmemmove+0x29:   rep     movsb
Xmemmove+0x2b:   cld
Xmemmove+0x2c:   jmp     near memmove+0x43
Xmemmove+0x2e:   mov     eax,edi
Xmemmove+0x30:   test    Byte Ptr 0x1f:0x1,al
Xmemmove+0x36:   je      memmove+0x3a
Xmemmove+0x38:   movsb
Xmemmove+0x39:   dec     ecx
Xmemmove+0x3a:   shr     ecx,1
Xmemmove+0x3c:   rep     movsw
Xmemmove+0x3f:   adc     ecx,ecx
Xmemmove+0x41:   rep     movsb
Xmemmove+0x43:   mov     esi,ebx
Xmemmove+0x45:   mov     edi,edx
Xmemmove+0x47:   pop     ebp
Xmemmove+0x48:   ret
Xmemmove+0x49:   nop
Xmemmove+0x4a:   nop
Xmemmove+0x4b:   nop
X
X15.  A documented feature is that upon connecting to a remote
Xdialed through the directory, if a 'fkey' key set from ~/.ecu/keys
Xmatches the directory name, the key set is automatically loaded.
XThis feature has never worked (I never used it or tested it.)
XRecently I told someone they could use it and they reported it
Xdid not work.  It does now.
X
X16. The example file models/sz_update.ep was treating skipped files
Xas a failed transfer.  This was fixed.
X
X17. Interactive and procedure commands 'pushd' and 'popd' were added.
X
X18.  Adding pushd and popd documentation made ecu.man too long
Xgiving me "word overflow" and a core dump from my nroff.  So,
Xthere are now two documents, ecu.man and exits.man.  The latter
Xis very short and is just a kludge since the former is as long as
Xit can be.
X
X19. Added the procedure rlog command for procedure control of
Xreceiver logging.
X
X20. Added _initial.ep, _connect.ep and _hangup.ep. See "Special
XProcedures" in the manual.
X
X21. The use of #ifdef GCC140 is made to take advantage of a
Xassembly source generation bug fix in gcc 1.40.
X
X22. Running ecu with no pre-existing phone directory and specifying
Xa logical phone number (directory entry) to dial caused erratic
Xbehavior or core dump.  This has been fixed.  Thanks for the report
Xfrom [e-mail].  He did a good job of testing
Xan alpha version of 3.10.
X
X23. Pressing an invalid function key during interactive command
Xentry used to destroy the string being entered.  This has
Xbeen fixed.
X
X24. The ESCape vs. function key keyboard input discrimination
Xalgorithm has been improved.  Fewer, if any, function keys will now
Xmisinterpreted as ESCape keys followed by "junk."
XREADME.P1           ECU 3.10 Patch 1     Mon Aug 12 03:46:20 EDT 1991
X
XThis describes patch 1 to ECU 3.10, bringing the program to
Xversion 3.11.  Due to the manner in which the program builds its
Xversion strings, patchlevel.h should contain #define PATCHLEVEL
X11.
X
XTo use
X------
X
XTo apply the patch, unshar all of the parts.  This step will
Xcreate patch files and overwrite some files whose diffs were
Xlarger than the new files.
X
Xtype the command
X    sh PATCH1.APPLY
X
Xthen do the
X    Configure
Xsequence and re-make
X
XOnce you are satisfied the patch has been correctly applied,
Xyou may
X    rm PATCH1.*
Xand
X    make neat
Xto remove the usual crud and peanut hulls left on the floor after
Xa patch operation.
X
XThis patch fixes the following problems:
X----------------------------------------
X
X1.  ISC lock files are placed in /usr/spool/locks.  ECU now has a
Xconfigurable lock file directory whose name is automatically selected
Xdepending upon the machine type.
X
X1.  1.  Several problems with compiling the the makedirs program under
XXENIX were fixed.
X
X3.  The 'w' (wait) subcommand of the dialing directory screen was not
Xasking for both single and multiple entry list weait values, but only
Xthe first.
X
X4.  VT220 and perhaps other terminals insist upon generating the CSI
X(ESCape plus 0x80) character as a function prefix.  To handle the need
Xfor eight bit keryboard support for nonansikeys, I did the following:
X
X   a. fixed the ECU code to honor the full eight bits of keyboard
X      code sequences when processing function key reads.
X
X   b. removed other code that masks the parity bit of various data items.
X
X   c. added code the kbdtest3.c that includes an stty -a listing for
X      the keyboard at the time it is run.  I also specify in clearer
X      terms what parity was in use and that may be required to use
X      the keyboard.
X
X5.  Previously, you had to have a copy of nonansikeys in each user's
X.ecu directory who use non-ANSI consoles.  You may now put a single copy
Xin the ECU library directory (usually /usr/local/lib/ecu).  ECU searches
Xfirst in ~/.ecu then in the library directory.
X
X5a. The manual stated a number of restrictions regarding
Xthe codes which can be transmitted by a keyboard to be supported
Xby the non-ANSI console feature.  I missed updating that manual part
Xwhen I released 3.10.  ECU 3.10 merely requires the code sequence
Xto be free of nulls and less than 33 characters in length.
X
X6.  The models/nonansikeys file has been updated to include an entry for
Xthe latest Metro Link X11R4 server.  The new server has improved default
Xkey bindings and the previously dead KP_5 (unshifted Keypad 5) lives.
X
X7.  In previous revisions of ECU, when editing an existing string on a
Xscreen, the cursor was placed at the end of the line at the commencement
Xof editing.  This was untrue of 3.10 and this patch restores the
Xbehavior.
X
X8.  Baud rates below 300 now get two stop bits.  I doubt anyone is using
X110 baud but if you ever get the hankering, it will work now.
X
X9.  In the respond function of the expect-respond handler, \n sent CR
Xnot NL as documented.
X
X10.  The test for an environment HZ was incorrect causing the
Xsys/param.h value to be used always.
X
X11.  The 'nap' procedure command napped for the wrong amount of time if
Xthe -m option was used.
X
X12. Under some conditions, ecuungetty would find a lock file created by
Xits (parent) ECU and report that the line was in use by another dial out.
XThis has been around for a long time.
X
X13. The command history mechanism got botched up by some last minute
Xchanges to the tty line input editor.
X
X14. Some work (largely in the dark) has been done to handle
Xdifferences between ISC and SCO tty naming.  The new manual entry says
Xof the -l startup switch:
X
X  The argument to the switch is the base name of the tty (e.g., "tty1a" or
X  "acu0").  On SCO, since ttys are all named consistently according to the
X  "/dev/ttyxx" form, you may omit the "tty" (e.g., "1a" or "4g").
X
XThe new manual entry says of the tty name fierld in the dialing directory:
X
X  A specific line is specified by using the base name of the tty (e.g.,
X  "tty1a" or "acu0").  On SCO, since ttys are all named consistently
X  according to the "/dev/ttyxx" form, you may omit the "tty" (e.g., "1a"
X  or "4g").
X
XThis means than on ISC you must now fully specify the base name
Xof the tty.  On SCO, you now have the -option- of doing the same or
Xyou may continue to use the two character suffix.
X
XI believe that now:
X
X  on SCO, you may use any tty name whose base name begins with "tty".
X  on ISC, add tty names whose base name begins with "acu"
X
X15. Miscellaneous other (nonvisible or debugging) changes were made.
X
X:EDITS: summary:
X----------------
XConfigure                   change FASI_IN_USE to FASI
Xecu.c                       add memstat
Xecu.c                       jpm@logixwi fix: HZ getenv test wrong sense
Xecu.c                       soup up -l for ISC vs. SCO
Xecu.h                       configurable lock directory
XecuLCK.c                    SCO_TTY_NAMING considerations
XecuLCK.c                    configurable lock directory
XecuLCK.c                    race with ecuungetty over lock resolved
Xecucmd.h                    add memstat
Xecuicmd.c                   add memstat
Xecuicmd.c                   turn off memstat after frustrating evening
Xecuicmhelp.c                when editing string, set cursor to end
Xecuicmhist.c                new ttygets botched command history handler
Xeculine.c                   SCO_TTY_NAMING considerations
Xeculine.c                   baud rates below 300 get two stop bits
Xeculine.c                   remove unused externs
Xeculock.c                   US_WEGOTIT handling
Xeculock.c                   configurable lock directory
Xecuphone.c                  soup up tty name for ISC vs. SCO
Xecuphone.c                  w subcommand was not asking both questions
Xecuphone.c                  when editing string, set cursor to end
Xecusetup.c                  ISC tty names
Xecuungetty/ecuungetty.c     US_WEGOIT handling
Xecuungetty/ecuungetty.c     add debug log event code
Xecuwinutil.c                when editing string, set cursor to end
Xexpresp.c                   \n sent CR not NL
Xexpresp.c                   detect NULL expect string
Xexpresp.c                   nap min of hzmsec if \m
Xhdbintf.c                   US_WEGOTIT handling
Xkbdtest3.c                  add parity reporting
Xkbdtest3.c                  some terminals reinvent parity bit's use
Xlint_args.h                 afterlint-creation
Xlogevent.c                  use static logname
Xmakedirs.c                  need smart_fork for XENIX
Xmkdirs.c                    how did compile succeed without signal.h?
Xmkdirs.c                    no need for sys/wait.h + XENIX doesn't have it
Xmodels/nonansikeys          add new Metro Link server entry
Xnonansikey.c                allow any code as first in key sequence
Xnonansikey.c                look for nonansikeys in ECULIBDIR too
Xpcmd.c                      add nap -1 return and proctrace
Xpcmd.c                      nap -m test wrong sense ... old bug!
Xutmpstat.c                  US_WEGOTIT handling
Xutmpstatus.h                add US_WEGOTIT
Xz/Make.src                  neat used wrong rm args
Xzgcc                        thanks for the -p1 suggestion to [e-mail]
X
XREADME.P2           ECU 3.10 Patch 2        Tue Aug 13 16:29:11 EDT 1991
X
XThis describes patch 2 to ECU 3.10, bringing the program to
Xversion 3.12.
X
XThe majority of this patch deals with the way SCO and ISC misimplement
Xthe nap() system call.  A lingering problem with the interactive
Xcommand history handler is also fixed (pressing erroneous function keys
Xcaused the display to become garbaged).
X
XISC and SCO UNIX nap() misbehave.  This kludge doesn't return the
Xproper value (the actual time slept), but at least it does not make
Xa mockery of the manual page.  It says:
X
X     NAP(S)               UNIX System V            NAP(S)
X
X     Name
X      nap -    suspends execution for a short interval
X
X     Syntax
X      long nap(period)
X      long period;
X
X     Description
X      The current process is suspended from    execution for at least
X      the number of    milliseconds specified by period, or until a
X      signal is received.
X
X     Return Value
X      On successful    completion, a long integer indicating the
X      number of milliseconds actually slept    is returned. If    the
X      process received a signal while napping, the return value
X      will be -1, and errno    will be    set to EINTR.
X
X     See Also
X      sleep(S)
X
X     Notes
X      This function    is driven by the system    clock, which in    most
X      cases    has a granularity of tens of milliseconds.  This
X      function must    be linked with the linker option -lx.
X
XIt appears nap() under UNIX 3.2.x has departed virtually entirely from
Xthe manual page.  I'm beginning to look rather silly in several
Xmilleus since I keep telling people SCO UNIX is a viable upgrade from
XXENIX.  But process control people need some kind of timing capability
Xless than one second and we can't do it with nap or select.
X
Xnap(msec) is supposed to nap *at least* msec milliseconds.  However,
Xif msec is specified less than 1000/HZ + 1, it will not nap at all.
XThis was true for 3.2.0 and 3.2.1.
X
XIt is supposed to return the number of milliseconds it actually
Xslept.  Instead, it appears to "save up" the values and return them in
Xlots of 1000. This behavior is true for 3.2.2.
X
XAs it is nap() is nearly useless.  I believe select() suffers
Xfrom the same deficiency (< 1000 msec timeout becomes 1000 msec) but
XI haven't "proven" it yet.
X
XREADME.P3           ECU 3.10 Patch 3        Tue Aug 27 03:36:04 EDT 1991
X
XThis describes patch 3 to ECU 3.10, bringing the program to
Xversion 3.13.
X
XOverview
X--------
X
XThe patch fixes some problems with the previous release, adds new
Xfunctionality and adds support for SunOS 4.1 and ISC System V
XRelease 4.  I have tested it fairly well on a Sparc 1+ under
XX11R4 (MIT Sun server and xterm at PL18) and SunOS 4.1.  I did my
Xbest to integrate patches supplied by Lothar Hirschbiegel
X(aega!lh).  Lothar has ISC SVR4 and his changes worked for him,
Xbut I had to hack them since his patches were against the
X"official" 3.12 and I applied them to sources already half-ported
Xto the Sun.  I hope they go well, but write if they don't.  I want
Xto hear of any success you might have with other SVR4 systems, too.
X
XSystem V Release 4
X------------------
X
X> I cannot offer much information on this, since Lothar sent little more than
X> patches and I have no access to a system for testing.
X> This is what I have:
X> I've just finished porting ecu3.12 to SVR4. I have added my changes
X> as a diff file, maybe this is of interest for you.
X> The following points were added/changed:
X>
X> 1. I have included a configuration option for SVR4 in config.c
X> 2. ecu works fine with "ttymon" under SVR4. This means you don't need
X>    FAS or something like this - bidirectional access to the serial ports
X>    with the stock asy drivers is ok.
X> 3. Because of some strange(?) behaviour in SVR4 curses I had to use
X>    some terrible kludges to make it work clean. It's working as it is,
X>    but there is obvilously room for improvement...
X> 4. SVR4 has the same set of hardware handshake ioctl's as SCO does.
X>    It's named differently (RTSXOFF and CTSXON), but it seems to work exactly
X>    the same way. I have enabled the "RTS" option for SVR4 - hopefully this
X>    is a standard feature (termiox) in *all* SVR4 versions, not just in
X>    my ISC release... :-)
X> 5. I'm still working on ecusz/ecurz. Outgoing transfer is doing fine, but
X>    incoming transfer locks up the connection sometimes.
X>    As soon as I have finished this, I could send you the cdiffs too.
X
XSunOS
X-----
X
XI brought this up in relatively short order.  I may have gotten
Xwhat I paid for :-), but it seems stable.  The System V support
Xunder SunOS was pleasingly similar to the SVR3 environment ECU came
Xfrom.  This release works on my Sparc 1+ with SunOS 4.1.1 and X11R4.
XSome xterm VT100.Translations overrides are necessary to enable use of
Xcertain keys unrecognized by the default xterm configuration.
XDetails are provided in the updated models/nonansikeys.  If the
Xrecommended changes, you can do what I did to figure out how to
Xget this guy going:
X
X  1. run xev and press the various function keys to see the
Xnames you need to specify in the xterm translations.
X  2. edit the translations in nonansikeys example and put them in
Xyour .Xdefaults file
X  3. run kbdtest3 to get a kbdtest3.out file
X  4. put this file's output into your ~/.ecu/nonansikeys file
X(near the top, ahead of other xterm defs so it will be found first).
X  5. try it
X  6. when it works, send me the kbdtest3.out file along with the
XVT100.Translations overrides you casme up with and a description of
Xthe X environment you used.
X
XThere is also a new entry named 'sun' in models/nonansikeys for the
Xconsole.
X
XNote on 'nonansikeys'
X---------------------
X
XThis file was named when support was added to the original
XSCO-only version to support keyboard/screens other than the
Xnative multiscreen console.  Since the TERM name for multiscreens
Xis "ansi", it seemed reasonable to name the file nonansikeys.  It
Xnow should probably be called 'nonSCOmultiscreenskeys', but that
Xname is too long for System V :-).  Before too many more
Xrevisions, even SCO multiscreen key definitions will be read from
Xnonansikeys and maybe then will be a good time to rename the
Xfile.  Anyway, this file and a good termcap definition takes you
Xa long way to supporting ecu on a wide variety of System V{,-ish}
Xplatforms.
X
XFixes
X-----
X
X1.  The ANSI SD (scroll down) sequence was broken for consoles
Xother than SCO multiscreens.  It scrolled down n * COLS lines
Xinstead of n lines.
X
X2.  Editing the null phone directory entry (pressing 'e' in an
Xempty directory, editing and exiting) caused a core dump.  Now
Xyou are prevented from editing a non-existent entry.
X
X3.  Procedure commands with switch arguments longer than 8
Xcharacters caused stack corruption (unpredictable behavior
Xincluding core dump).  There is no call for switches this long,
Xbut if present, the program should not explode.  Now, switches
Xlonger thasn 8 characters (including the hyphen) are silently
Xtruncated.  Should commands ever be added which could conceivably
Xevoke switch arguments that might approach the limit, either the
Xlimit will be extended or error reporting will be added, or both,
Xas appropriate.
X
X4.  # comments to the right of nonansikey key definition lines
Xerroneously included the # in the definition.  This may not have
Xbeen a bug in 3.12.  I may have added it during 3.13 development,
Xbut it is gone now in any case.
X
X5.  Previously, if the console line baud rate was not at least 4
Xtimes the tty line rate (roughly), the "no curses" mode was
Xforced during file transfer and could not be overridden.  Since
Xvarious video driver implimentations express verious pseudo-"baud
Xrate" settings, this was more often a problem than a feature.  If
Xthe change causes you trouble, write me (or trying using Home pc
Xsz -N in lieu of Home pc sz, etc.).
X
X6.  Previously, an effort was made to editorialize on your choice
Xof a tty device name.  This is no longer done.  On SCO, you
Xshould use the lower case (direct) names, like tty1a, to get the
Xexpected lock file names, but trying to second guess all the
Xvendor and third-party names was an unnecessary and
Xcounter-productive stumpling block.  If you want to go for
X/dev/swap, go ahead (I don't expect the ioctls to work -- or your
Xsystem :-> ).
X
X7.  Andrew A. Chernov ([e-mail]) said:
X
X1) HDB dialers may return connect speed as return code (!= 0)
X [ and the CONNECT string built by hdbintf.c prints this code -- wht]
X [his patch used - wht]
X2) Using HDB Dialcodes file for phone numbers translation now
X   (\D,\T escape sequence)
X [his patch used - wht]
X
XIn other mail he wrote:
X
XI just found  little bugs:
X1) XENIX standard cc doesn't understand prototypes with names:
X"FILE *" is ok, but "FILE *fp" is bad. (patched sea/lint_args.h)
X [fixed - wht]
X2) 'makedirs' program NEED last '/' in path name to make directory.
XAll Make.src patched to add '/' at the end of dirs.
X [fixed n a different way; makedirs now works as advertised - wht]
X3) Some HDB dialers looks in argv[0] and change some parameters
X(f.e. speed, etc.), so name "ECUdial" is invalid, and need change to
Xdve->dialprog. WARNING: this patch required my previous patch
X(use 'token' instead of 'stripped_num')
X [his patch used - wht]
X
X8. Jan-Piet Mens <logixwi.uucp!jpm> reported problems
Xwith VT-220 support due to it's necessity of passing characters
Xwith the sign bit set.  As of now, characters are read by ecu and kept
X8-bit intact until they have been passed through the key mapping filter.
XKeys actually input to the program or sent over the line are
Xlimited to 7 bits for now (sorry to my international friends) since
Xcharacters with the sign-bit have internal significance to ECU.
XThis is a primary example of "American's Disease" (7-bit ASCII) and I'll get
Xit out of here as soon as I get over "Warren's Disease" (laziness).
X
XHe also reported:
X> In README.P1, you write that the "nonansikeys" file may be put in
X> ECULIBDIR.  This is incorrect: it is searched for in
X> ECULIBDIR/.ecu [Line 193 nonansikey.c]
X
XAcknowledgments
X---------------
X
XI've already mentioned Lothar's work in a matter-of-fact manner,
Xbut here is the place for extreme words of appreciation. SVR4 is
Xvery important and kudos to him for the leg up.
X
XRobert E.  Laughlin, Naval Ocean Systems Center (NOSC), [e-mail],
Xreported various successes and failures.  We had a good and continuing
Xdialog.  His interest, suggestions and moral support did more for me than my
Xsupport of ECU did for him.
X
XAndrew Chernov, of Relcom fame, made important suggestions and
Xfixes for XENIX support and for the product as a whole.  He made
Xsome important suggestions for an older revision as well, some of
Xwhich I put in and some not (Variously, I was too lazy, busy or
Xthe code had gone too far away from what he was patching from).
XHis work is laudable and important for many reasons far exceeding
Xhis support of this little program.  Keep him in your thoughts
Xthrough these strange and wonderous times.
X
XIt was good to hear from these folks for their comments asnd fixes:
X
X Andy Hall <relay.EU.net!ncduk!andyh>
X Dave Hammond <dsix2!daveh
X Greg Wilkins <sir-alan!gregw>
X Michael P. Deignan <anomaly.sbs.com!mpd>
X Michael Traub <rtf.bt.co.uk!traub>
X cmcl2.NYU.EDU!panix!tbetz (Tom Betz)
X Jan-Piet Mens <logixwi.uucp!jpm>
X mudos.ann-arbor.mi.us!mju (Marc Unangst)
X lll-winken.llnl.gov!abhg!tao!tempest!whitemp
X ucsd!jadpc.cts.com!jdeitch
X
XThis file is harder to write than all of the code changes.  Not
Xonly is it hard to remember what all bugs were reported by whom
Xand what changes were fixed by whom, I hope I haven't missed
Xcrediting anyone.  if I did, please forgive me.
X
XKeep dem cards and letters coming.
X
X
XREADME.P4           ECU 3.10 Patch 4       Wed Aug 28 09:43:44 EDT 1991
X
XThis describes patch 4 to ECU 3.10, bringing the program to
Xversion 3.14.
X
XThe majority of this patch deals with good work done by Paul
XSlootman ([e-mail]) on the SVR4 port.  From what he
Xencountered, I am very very disappointed in some of the more anal
XANSI C implementations out there.  Trigraphs suck and we all knew
Xthat, but having to cast a constant in a comparison with
Xstrlen???  (Excuse me -- that is a trigraph -- I must learn to
Xsay \?\?\?).
X
XOh well, there are a few other kinds of fixes in here too: ISC
Xshould be better off.  Some of the conditionally compiled code
Xthat was running for SCO only should have been in for ISC too.
X
XI thought I had fixed the bad search for nonansikeys in
XECULIBDIR/.ecu/ instead of ECULIBDIR/, but it is fixed now.
X
XTo use:
X       patch -p < PATCH4.01
X       Configure
X       make
X       su root (may not be necessary on Sun or SVR4 if you
X               have write access to the installed directories)
X       make install
X
XREADME.P6           ECU 3.10 Patch 6     Wed Sep  4 00:41:23 EDT 1991
X
XThis describes patch 5 to ECU 3.15, bringing the program to
Xversion 3.15.
X
XFiles in Patch 5:
XREADME.P5
XPATCH5.01
Xfasi/fas.h
Xckermit/ckutio-ecu.dif
Xmemmove/README
Xmemmove/memmove.s
Xmemmove/memmove286.asm
Xmemmove/memmove386.asm
Xtimetest/Makefile
Xtimetest/naptest.c
Xtimetest/naptest2.c
Xtimetest/naptest3.c
Xtimetest/seltest.c
X
XFirst, a note of apology to Sun users.  I used the -mm nroff macros
Xto write the manuals.  Sun provides -ms and -me, but not -mm.
XI'm looking for some PD replacements, but haven't succeeded yet.
XI'll be glad to mail an updated manual to anyone who asks.  I'm
Xalso looking for a friendly ftp site to hold updates, hence manuals.
X
X1. Improvements in the SVR4 port have been made, but now
XSVR4 almost certainly means an AT&T/Intel/ISC 386 port,
X                                              ^^^
Xsince we expect to find /usr/include/sys/at_ansi.h and
X/usr/include/sys/sys/kd.h files for console color capability.
X(Color is not yet supported on Sun -- only monochrome xterms
Xand the console without color manipulation).
X
XIf you have trouble with compiling ecutty.c for SVR4 (complaints
Xabout not finding either of these header files), temporarily remove the
XSVR4 condition for inclusion of them and for using the CONS_GET code
Xnear the bottom of the source.
X
X2.  Many improvements were made in the area of system independence.
XThe build procedure configures the location of HDB Devices,
XDialcodes and Dialers files among the various environments:
XSCO   /usr/lib/uucp
XISC   /usr/lib/uucp
XSun   /etc/uucp
XSVR4  /etc/uucp
X
X3. A new feature of Configure is the inclusion of local configuration
Xin the Makefiles it builds.  If you build a file named "config.local"
Xin the base source directory, it's contents will be placed ahead of
Xall other make statements in each Makefile.
X
XExample customization include:
XCFLAGS_EXTRA  = -Dwhatever
XGCC_CFLAGS_EXTRA = -fsoft-float  #just an example :-)
XCC_CGLAGS_EXTRA = -p
XSYMBOLIC = -g             # if you want sdb/dbx/gdb symbols in programs
XBINTIME = /bin/time       # if you want stats on each make step
XLDFLAGS_EXTRA  = -ljuicier-malloc
X
XJudicious use of this may help you avoid changing Make.src files
Xthus making patch complain about future patches, but
X
X4. A patch from Andrew Chernov that fell out of the last version
Xwas reapplied:  Many SCO-style dialer programs do not care
Xwhat is in argument 0 (the command name) when they are executed,
Xbut some do.  Before this patch, the argument passed was "ECUdial"
Xsince the gendial dialers *can* make use of it.  This usage
Xconflicted with some of Andrew's dialers which need the actual
Xname passed.  So, starting with this edition, if you invoke
Xa dialer program, its complete pathname will appear in argv[0].
XIf you want the old behavior, add a file to the base directory
Xcalled config.local and place in it the line:
X
X  CFLAGS_EXTRA = -DECUdial
X
X3. The usefulness of _rc.ep has been enhanced by adding the
X'setline' procedure command.    This command is only valid during the
Xexecution of _rc.ep.  It allows you to override the selection of
Xa default line as if you had used the -l switch.  In fact, using
Xthe commabnd overrides -l.
X
XUsage: setline <filename>
X
XThis command is an exact analog of the -l command line switch.
X<filename> is used on the setup screen (if it is presented)
Xas the default filename. If you do not edit this name, ECU will
Xattempt to open the line when you press End or ^D.
X
X<filename> has the format of a complete pathanme in /dev or simple
Xtty name.
X
XExample:
X  setline '/dev/tty1a'
X  setline 'tty1a'
X  setline 'cua0'
X
X5. Similarly, using the 'baud' and 'parity' procedure commands in
X_rc.ep serve as an exact analog of the -b, -e and -e command line switches.
XUsing 'baud' and 'parity' in rc_ep override the equivalent
Xcommand line switches.
X
X6.  On the Sun, STREAMS termios is used in lieu of termio.  The code
Xused for draining the line output queue ("wait for all transmitted
Xdata to be sent over the line") has been improved, making
Xexpect-respond, file transfer and other highly interactive operations
Xmore reliable.
X
X7.  The following files normally found in a personal ~/.ecu
Xdirectory may now have a systedm default in ECULIBDIR:
X
X   colors
X   keys
X   nonansikeys
X
X8. Unreported bugs in the format of the keys and nonansikeys files have
Xbben fixed.  Specifically, whitespace and comments are allowed after keyset
Xor terminal names.
X
X9. On ISC 2.2 systems, the console was left in standout or bold
Xmode at exit from ECU.  I believe this problem is now fixed.
X
X10. The GCC compilation flags for 386 versions have been updated to use
X-fcall-used-bx rather than -fcall-used-{ax,bx,cs,dx].
X
XIt turns out the only problem ECU had with strenuously optimized GCC
Xoptions was that SCO's memmove clobbers the bx register, which is
Xsupposed to be reserved for use as a register variable by all
Xcompilers, MSC or otherwise.
X
XIf you use the new memmove routines supplied with this patch, you may
Xomit this switch as well.  It is probably not worth fooling with any
Xof that unless you are just interested in the problem.
X
X11. I forgot to include the version 4F kermit diff file.  It's included
Xin here.
X
X12. In patch 2, I reported nap() in SCO UNIX systems to be faulty in varying
Xmanners depending on x in the 3.2.x version number.  ISC 2.2.1 nap also
Xmisbehaves in a similar manner.
X
XSCO very promptly supplied me with an /etc/conf/pack.d/kernel/os.o for
Xrelinking the kernel which fixes the problem somewhat, but not
Xcompletely.  Watch this space.  Some testing programs can be found in
Xtimetest/ as of this patch.
X
X13.  I've thought for some time SCO select might also be buggy.  I finally
Xwrote some test programs and proved it.  SCO select may appear to work
Xcorrectly for a few iterations, but rapidly degenerates into rounding
Xall timeout specifications to the nearest second.  Run timetest/seltest
Xfor a visual feel of what's going on.
X
XThe #define WORKING_SELECT is now supploied by the Configure procedure
Xfor systems that have a usable select.  See the updated README file
Xnote 3 and ecuutil.c Nap() for details.
X
X
XREADME.P5           ECU 3.10 Patch 5     Wed Sep  4 00:41:23 EDT 1991
X
XThis describes patch 6 to ECU 3.16, bringing the program to
Xversion 3.16.
X
XTwo quick fixes here:
XThe configuration procedure screws up when you cc on SVR4, Sun or ISC
X1. I forgot to put -DWORKING_SELECT in for cc on these machines.
X2. A '(' got changed to a '*' somewhere in the process.
X
XBad QA.  Sorry.
SHAR_EOF
chmod 0644 OLD-HISTORY-2 ||
echo 'restore of OLD-HISTORY-2 failed'
Wc_c="`wc -c < 'OLD-HISTORY-2'`"
test 33749 -eq "$Wc_c" ||
    echo 'OLD-HISTORY-2: original size 33749, current size' "$Wc_c"
rm -f _shar_wnt_.tmp
fi
# ============= ESOTERIC ==============
if test -f 'ESOTERIC' -a X"$1" != X"-c"; then
    echo 'x - skipping ESOTERIC (File already exists)'
    rm -f _shar_wnt_.tmp
else
> _shar_wnt_.tmp
echo 'x - extracting ESOTERIC (Text)'
sed 's/^X//' << 'SHAR_EOF' > 'ESOTERIC' &&
XThis interesting article appeared last summer:  much of the wisdom
Xapplys to the ECU procedure language :-)
X
X> From: [e-mail] (Guy Harris)
X> Newsgroups: comp.unix.questions
X> Subject: Re: Funny kill -9 behaviour
X> Message-ID: <[e-mail]>
X> Date: 18 Aug 90 21:27:07 GMT
X> References: <[e-mail]>
X> Organization: Auspex Systems, Santa Clara
X> Lines: 16
X>
X> >What I want to know is why in the hay does this happen?
X>
X> Because the C shell is flakier than a snowstorm.  To quote the BUGS
X> section of the SunOS 4.0.3 manual page (this quote is also in the S5R4
X> "csh" manual page):
X>
X>      Although robust enough for general use, adventures into  the
X>      esoteric  periphery  of  the  C  shell may reveal unexpected
X>      quirks.
X>
X> which translates as "the C shell is flakier than a snowstorm."
SHAR_EOF
chmod 0644 ESOTERIC ||
echo 'restore of ESOTERIC failed'
Wc_c="`wc -c < 'ESOTERIC'`"
test 835 -eq "$Wc_c" ||
    echo 'ESOTERIC: original size 835, current size' "$Wc_c"
rm -f _shar_wnt_.tmp
fi
# ============= Configure ==============
if test -f 'Configure' -a X"$1" != X"-c"; then
    echo 'x - skipping Configure (File already exists)'
    rm -f _shar_wnt_.tmp
else
> _shar_wnt_.tmp
echo 'x - extracting Configure (Text)'
sed 's/^X//' << 'SHAR_EOF' > 'Configure' &&
X:
X#!/bin/sh
X#+---------------------------------------------------------------------
X# ECU Configure
X#
X# PEDANTIC not recommended unless you want to do extensive
X# vendor-supplied header file hacking.
X#----------------------------------------------------------------------
X#+:EDITS:*/
X#:09-10-1992-13:58-wht@n4hgf-ECU release 3.20
X#:08-22-1992-15:37-wht@n4hgf-ECU release 3.20 BETA
X#:07-09-1992-18:09-wht@n4hgf-ecunumrev a bad idea
X#:07-07-1992-16:30-wht@gyro-add ecunumrev to config
X#:08-25-1991-05:11-wht@n4hgf2-added Sun n4hgf2 WHT automation
X#:08-06-1991-05:05-wht@n4hgf-change FASI_IN_USE to FASI
X#:07-12-1991-14:05-wht@n4hgf-ensure config.c compiles properly
X#:04-29-1991-02:39-wht@n4hgf-document/clean
X
X# --> add any of the following (99% of everybody will want to leave alone)
X# -DFASI     if you have installed and want to exploit the features
X#            of my FAS instrumentation extensions
X# -DMORSE if you have installed and want to use the n4hgf morse driver
X# -DPEDANTIC if you want to use -ansi -pedantic with gcc (not
X#            necessary or recommended unless you anticipate heavy
X#            code modification)
X# -DMETROLINK_X11R4_PTS if you have early 'pts-using' Metro Link
X#            xterms
XCFLAGS=
X
X# a bit of self indulgence for me (ignore it or be inspired)
XSYSTEM=
X[ -f /etc/systemid ] && SYSTEM=`cat /etc/systemid`
X[ "$SYSTEM" = "n4hgf" ] && CFLAGS='-DWHT'
X[ -f /etc/hostname.le0 ] && SYSTEM=`cat /etc/hostname.le0`
X[ "$SYSTEM" = "n4hgf2" ] && CFLAGS='-DWHT'
X
Xecho Please wait while I compile the config program.
Xrm -f ./config
Xcc $CFLAGS config.c -o config > /tmp/config.cc.log 2>&1
X[ -x ./config ] && rm -f /tmp/config.cc.log
X[ -x ./config ] || echo '/tmp/config.cc.log has compile errors'
X./config
X
X#end of Configure
SHAR_EOF
chmod 0755 Configure ||
echo 'restore of Configure failed'
Wc_c="`wc -c < 'Configure'`"
test 1752 -eq "$Wc_c" ||
    echo 'Configure: original size 1752, current size' "$Wc_c"
rm -f _shar_wnt_.tmp
fi
# ============= Make.src ==============
if test -f 'Make.src' -a X"$1" != X"-c"; then
    echo 'x - skipping Make.src (File already exists)'
    rm -f _shar_wnt_.tmp
else
> _shar_wnt_.tmp
echo 'x - extracting Make.src (Text)'
sed 's/^X//' << 'SHAR_EOF' > 'Make.src' &&
X#+-------------------------------------------------------------------
X# Make.src - master Makefile source for ECU and related programs
X# If THIS line is the third in the file you are reading, refer to
X# the files README and Configure.
X# [e-mail]
X#--------------------------------------------------------------------
X#+:EDITS:*/
X#:09-10-1992-13:58-wht@n4hgf-ECU release 3.20
X#:08-22-1992-15:37-wht@n4hgf-ECU release 3.20 BETA
X#:07-17-1992-18:23-wht@n4hgf-new module nap.c
X#:07-07-1992-20:04-wht@n4hgf-new installation mechanism
X#:05-17-1992-18:50-wht@n4hgf-add cpio
X#:02-10-1992-05:47-wht@n4hgf-mcs avail on some systems when non-COFF produced
X#:09-03-1991-23:47-wht@n4hgf2-add timetest directory
X#:09-02-1991-19:25-wht@n4hgf2-ckutio dif and fas.h left out of distribution
X#:09-01-1991-16:22-wht@n4hgf2-generalize HDB config location
X#:08-23-1991-13:35-wht@n4hgf-sun porting (3.13)
X#:07-25-1991-12:55-wht@n4hgf-ECU release 3.10
X#:04-21-1991-12:43-wht@n4hgf-convert to Make.src
X
XPROGRAM = ecu
XINSTALLABLE = $(LBIN)/$(PROGRAM)
X
XECUSHARNAME=ecu320
XMANSHARNAME=ecuman320
X
XLIBFUNCKEYMAP = $(ECULIBDIR)/funckeymap
X
XSHELL = /bin/sh
X
X.SUFFIXES:
X.SUFFIXES: .o .c .h .l
X.c.o:;    $(BINTIME) $(CC) -c $(CFLAGS) $*.c
X
X
XSRC    = \
X    bamboozle.c\
X    cmdtbl.c\
X    ecu.c\
X    ecuLCK.c\
X    ecuchdir.c\
X    ecudump.c\
X    ecufinsert.c\
X    ecufkey.c\
X    ecufork.c\
X    ecuDCE.c\
X    ecugrabbag.c\
X    eculine.c\
X    eculock.c\
X    ecunumrev.c\
X    ecuicmaux.c\
X    ecuicmhelp.c\
X    ecuicmhist.c\
X    ecuicmd.c\
X    ecupde.c\
X    ecuphdir.c\
X    ecuphrase.c\
X    ecurcvr.c\
X    ecuscrdump.c\
X    ecusetup.c\
X    ecushm.c\
X    ecusighdl.c\
X    ecutcap.c\
X    ecutime.c\
X    ecutty.c\
X    ecuuclc.c\
X    ecuusage.c\
X    ecuutil.c\
X    ecuwinutil.c\
X    ecuxfer.c\
X    esdutil.c\
X    fasiintf.c\
X    expresp.c\
X    feval.c\
X    gint.c\
X    gstr.c\
X    hdbintf.c\
X    logevent.c\
X    nap.c \
X    funckeymap.c\
X    pcmd.c\
X    pcmdif.c\
X    pcmdfile.c\
X    pcmdtty.c\
X    pcmdwhile.c\
X    pcmdxfer.c\
X    poutput.c\
X    pprintf.c\
X    proc.c\
X    proc_error.c\
X    procframe.c\
X    regexp.c\
X    ttynaming.c\
X    utmpstat.c\
X    var.c
X
XOBJ    = \
X    bamboozle.o\
X    cmdtbl.o\
X    ecu.o\
X    ecuLCK.o\
X    ecudump.o\
X    ecuchdir.o\
X    ecufinsert.o\
X    ecufkey.o\
X    ecufork.o\
X    ecuDCE.o\
X    eculine.o\
X    eculock.o\
X    ecunumrev.o\
X    ecuicmaux.o\
X    ecuicmhelp.o\
X    ecuicmhist.o\
X    ecuicmd.o\
X    ecugrabbag.o\
X    ecupde.o\
X    ecuphdir.o\
X    ecuphrase.o\
X    ecurcvr.o\
X    ecuscrdump.o\
X    ecusetup.o\
X    ecushm.o\
X    ecusighdl.o\
X    ecutcap.o\
X    ecutime.o\
X    ecutty.o\
X    ecuuclc.o\
X    ecuusage.o\
X    ecuutil.o\
X    ecuwinutil.o\
X    ecuxfer.o\
X    esdutil.o\
X    fasiintf.o\
X    expresp.o\
X    feval.o\
X    gint.o\
X    gstr.o\
X    hdbintf.o\
X    logevent.o\
X    lstat.o\
X    nap.o \
X    mkdirs.o\
X    funckeymap.o\
X    pcmd.o\
X    pcmdif.o\
X    pcmdfile.o\
X    pcmdtty.o\
X    pcmdwhile.o\
X    pcmdxfer.o\
X    poutput.o\
X    pprintf.o\
X    proc.o\
X    proc_error.o\
X    procframe.o\
X    regexp.o\
X    ttynaming.o\
X    utmpstat.o\
X    var.o
X
XSHARFLS = \
X    README \
X    README.* \
X    HISTORY \
X    OLD-HISTORY-* \
X    ESOTERIC \
X    Configure \
X    Make.src */Make.src \
X    *.[ch] \
X    bperr/bperr.c \
X    help/{*.[ch],ecuhelp.src} \
X    ecuungetty/*.[ch] \
X    z/*.[ch] \
X    sea/{*.[ch],*.{doc,imp}} \
X    gendial/{README,install_dialer,*.[ch]} \
X    fasi/{Makefile,Master,Node,PATCHLEVEL} \
X    fasi/{README,README.FASI,RELEASENOTES} \
X    fasi/{Space.c,System,digi-pc8.h,fas.c,fas.h,idfasi} \
X    ecufriend/*.[ch] \
X    mapkey/README mapkey/keys.usa.ecu \
X    xsel386/* \
X    memmove/{README,*.{s,asm}} \
X    timetest/{Makefile,*.c} \
X    doc/{README,Makefile,*.txt} models/* zgcc mkoldproto.l'
X
XMAKEDIRS_OBJ =\
X    makedirs.o\
X    mkdirs.o\
X    lstat.o
X
Xall: $(PROGRAM) .make.help .make.ug .make.z .make.s kbdtest3 makedirs
X
X$(PROGRAM): $(OBJ)
X    echo 'char *makedate="@(#)'ecu `date` $(SYSTEM)'";'  >makedate.c
X    $(CC) -c $(CFLAGS) makedate.c; rm makedate.c
X    rm -f $@; if [ -f $@ ]; then mv $@ $@~; fi
X    $(BINTIME) $(CC) -o $@ $(LDFLAGS) makedate.o $(OBJ) $(LIBS)
X    rm makedate.o
X
X$(OBJ): Makefile
X
X.make.help:
X    cd help; $(MAKE) -$(MAKEFLAGS)
X
X.make.z:
X    cd z; $(MAKE) -$(MAKEFLAGS)
X
X.make.s:
X    cd sea; $(MAKE) -$(MAKEFLAGS)
X
X.make.ug:
X    if [ $(USE_ECUUNGETTY) = yes ]; then \
X        cd ecuungetty; $(MAKE) -$(MAKEFLAGS); \
X    fi
X
Xbperr/bperr: bperr/bperr.c
X    cd bperr; $(MAKE) -$(MAKEFLAGS) bperr
X
Xproc_error.c: ecuerror.h
X    $(MAKE) -$(MAKEFLAGS) bperr/bperr
X    bperr/bperr
X
Xkbdtest3: kbdtest3.c
X    $(CC) -o $@ $(CFLAGS) kbdtest3.c
X
X$(OBJ): ecushm.h
X
Xecu.fls: Makefile
X    ls $(SRC) >ecu.fls
X
X# Not recommended
Xprotos: ecu.fls afterlint mkoldproto
X    echo ' ' > protos.h
X    csh ./zgcc ecu.fls protos.h .
X
X# Not recommended
Xallprotos:
X    $(MAKE) -$(MAKEFLAGS) protos
X    cd z; $(MAKE) -$(MAKEFLAGS) protos
X    cd sea; $(MAKE) -$(MAKEFLAGS) protos
X
Xmakedirs: $(MAKEDIRS_OBJ)
X    $(CC) -o $@ $(LDFLAGS) $(MAKEDIRS_OBJ) $(LIBS)
X
X# install ecu
Xiecu: makedirs ecu
X    @echo Installing ECU in $(LBIN) and $(ECULIBDIR) ... please wait
X    @if [ ! -d $(LBIN) ]; then \
X        echo '\012'Attempting to make $(LBIN); \
X        ./makedirs -m 755 $(LBIN); \
X        ls -ld $(LBIN); \
X    fi
X    @rm -f $(INSTALLABLE) $(INSTALLABLE)~
X    @if [ -f $(INSTALLABLE) ]; then \
X        echo '\012'Cannot remove $(INSTALLABLE) - moving to $(INSTALLABLE)~ \
X        mv $(INSTALLABLE) $(INSTALLABLE)~; \
X    fi
X    @cp $(PROGRAM) $(INSTALLABLE)
X    @strip $(INSTALLABLE)
X    @-if [ -x /usr/bin/mcs ]; then \
X        /usr/bin/mcs -d $(INSTALLABLE) || echo mcs present but not COFF? ; \
X    fi
X    @chown bin $(INSTALLABLE)
X    @chgrp bin $(INSTALLABLE)
X    @chmod 711 $(INSTALLABLE)
X    @echo '\012'Installed
X    @ls -l $(INSTALLABLE)
X    @if [ ! -d $(ECULIBDIR) ]; then \
X        echo '\012'Attempting to make $(ECULIBDIR); \
X        ./makedirs -m 755 $(ECULIBDIR); \
X        ls -ld $(ECULIBDIR); \
X    fi
X    @-if [ -f $(LIBFUNCKEYMAP) ]; then \
X        echo '\012'Saving your old $(LIBFUNCKEYMAP); \
X        for i in  2 1; \
X        do \
X            j=`expr $$i + 1`; \
X            mv $(LIBFUNCKEYMAP)-$$i $(LIBFUNCKEYMAP)-$$j 2>/dev/null; \
X        done; \
X        mv $(LIBFUNCKEYMAP) $(LIBFUNCKEYMAP)-1; \
X        ls -l $(LIBFUNCKEYMAP)*; \
X    fi
X    @cp models/funckeymap $(ECULIBDIR)
X    @chown bin $(LIBFUNCKEYMAP)
X    @chgrp bin $(LIBFUNCKEYMAP)
X    @chmod 644 $(LIBFUNCKEYMAP)
X    @echo '\012'Installed
X    @ls -l $(LIBFUNCKEYMAP)
X
X# install everything
Xinstall: all makedirs #run as root
X    @echo '\012'Starting full installation of ECU ... please wait
X    @make iecu
X    @if [ $(USE_ECUUNGETTY) = yes ]; then \
X        cd ecuungetty; $(MAKE) -$(MAKEFLAGS) install; \
X    fi
X    cd z; $(MAKE) -$(MAKEFLAGS) install
X    cd sea; $(MAKE) -$(MAKEFLAGS) install
SHAR_EOF
true || echo 'restore of Make.src failed'
fi
echo 'End of ecu320 part 3'
echo 'File Make.src is continued in part 4'
echo 4 > _shar_seq_.tmp
exit 0

exit 0 # Just in case...