perm filename DIRECT.DOC[SYS,HE] blob
sn#341273 filedate 1978-03-18 generic text, type C, neo UTF8
COMMENT ⊗ VALID 00004 PAGES
C REC PAGE DESCRIPTION
C00001 00001
C00002 00002 INFORMATION ON DISK AREA SYS,HE
C00006 00003
C00010 00004
C00013 ENDMK
C⊗;
INFORMATION ON DISK AREA SYS,HE
EDGE and CURVE are documented quite thoroughly on file EDGCUR.DOC
Note that EDGE must be loader either with DDT, RAID, or
dummy file DD if its command decoder is to function properly.
DO DOHE will compile, load, and save HE after it has been changed.
G. Grapes segmenter worked, after a fashion, when I stopped playing
with it. It needs to have the main matching loop rewritten to
correct a design flaw if it is to be seriously used. I suspect it
will never be run again.
The utility package contains message procedure drivers for many jobs
run under the hand/eye monitor. Many of these jobs no longer are
maintained and this package will not run, but contains some examples
of drivers which could be of use for those which are still used. A
better set of drivers for EDGE and CURVE can be found in
COLECT.SAI[CAL,HE].
The CAMERA program and associated files must be rewritten for the new
table and calibration format. A start at this exists on CAL,HE.
I know nothing about the SHY files, nor do I know of any
documentation for them.
See Taylor, Finkel, or Shimano for information on these file (and to
purge the excess)
The RPO files never did work well and no documentation exists, unless
it is on file COMPLX.
I know nothing about the WAP files. Binford may know something about
them.
See Arthur about the colour files. The stereo files listed here are
for a format no longer used.
The PDP-11 Monitor and Stereo files are documented on INTERF.DOC and
STEREO.DOC. Arthur can run them and the code is fairly well
commented, especially the 11 code.
WHLTST, DACTST, SERTST, and TBLTST are test program which may no
longer work. There is no documentation unless they tell you
something when you start them.
POTCAL was used to do some calibration of the pan/tilt pots. It may
no longer work. It is not documented.
APPLY is the Huckel operator manipulator. It is documented in
APPLY.DOC. It contains the only code ever written to get double
resolution TV pictures.
FLIST, documented on FLIST.DOC, lists files, given the page numbers
in the test. I wrote it for listing pieces of the HELIB
documentation. It has a few bugs.
PREAMB is the second segment definitions for programs running under
HE. the TXT file is the original. Running PREMAK generates the SAI
version with no comments.
MACROS is a set of HE macros which can be used to load many of the
hand/eye programs in parallel after they have had to be recompiled
because some hacker has changed SAIL or the runtime routines.
COMP @GLORPG is reputed to compile all hand/eye programs needing it.
COMP @SAIRPG does the same for all (most?) other SAIL programs on the
area.
DPYSUB.HDR is the version of the header for the QUAM display routines
used by programs running under HE. It modifies a few procedures so
the piece-of-glass numbers allocated are unique over all jobs running
under HE instead of just for routines within one job.
SAITRG probably is no longer used since the trig functions are in
SAIL, but I am scared to remove it just in case.
PICTUR and SPICT are the TV picture input and manipulating programs.
PICTUR has its dump on 1,3. SPICT is the program for inputting
STEREO pictures in the current format.
GCREF is a handy program to create a cross reference listings of a
set of programs loaded together from their individual cross reference
listings. Unfortunately it is not documented. But its command
string is of the form
DEST.EXT←/sCREF1,CREF2, ... , CREFn
where CREFi are the cross reference listings. They must be made with
the switches (to CREF) set so that the macro and instruction listings
are not made. They will handle most text found in program listing
part of the output, but run much faster if it was suppressed also.
The switches can occur most anywhere in the command line, are
designated by /s, where s can be
L to output only local variables (those appearing in only on
file
G to output only global variables (if neither L or G is speci-
fied, you get both.
S to output short lines so the output will fit in binders.
Otherwise it is full lines.
The output looks similar to that from CREF except that it is
formatted somewhat better and each variable is followed by the file
name(s) in which it is found before the block names, if any. Note
that while the program lists in one block all files using a given
variable name, this does not necessarily mean they are the same
variable in each file, since the program does not check the
INTERNAL/EXTERNAL declarations. Assembly language freaks might like
the magic 6-state, table driven input parser in this program.