CP/M/CPM_CDROM.iso/cpminfo/cpm-cc05.art

Text File | 1994-07-13 | 17KB | 306 lines

==============================================================================
[   THE KAY*FOG RBBS  |  CPM-CC05.ART  |  posted 01/18/86  |  303 lines 17k  ]

          The CP/M Connection                   Originally published in
                  by                               Computer Currents
             Ted Silveira                           2550 9th Street
  (copyright and all rights reserved)             Berkeley, CA  94710


                                 May 7, 1985
                              WORDSTAR UTILITIES

     If the most common thing people do with computers is shuffle disks,
then the second-most common thing they do is shuffle text--word-processing
and all its related activities.  In the CP/M public domain, there are
enough text-handling programs to fill several columns, but since WordStar
is word-processing to most CP/M users, let's start with a batch of
utilities made especially for WordStar.


MAGE31, RESQ12

     It was almost midnight, and I had a deadline.  Outside it was windy
and raining.  The work was going so slowly and I was so tired that I
stupidly hadn't saved anything for at least an hour.  Suddenly the lights
dimmed, only briefly but long enough.  When I hit the next key, nothing
happened--the screen still showed my text, intact, but the computer was
frozen, locked up, out to lunch, total amnesia.  The results:  two minutes
of frantic pounding on the keyboard, five minutes of stunned silence, a
resigned push of the reset button, and an hour spent redoing my lost work.

     Even if you've never been struck by this disaster, don't push your
luck.  Anyone who works with WordStar needs either MAGE31 or RESQ12, two
programs that recover text after disk errors, locked-up computers, or
almost any disaster short of a complete power failure.  A few months after
my midnight disaster, I was hit by the same problem, but this time, I had
MAGE31.  Total time lost:  2 minutes.

     The secret is that when you have to restart your computer by hitting
^C or even the reset button, your computer doesn't erase the part of your
text that's in memory.  It just forgets the text is there.  So MAGE31 and
RESQ12 go to the section of memory where your text should be and save what
they find into a disk file.  You can then combine that file with the
incomplete text file (filetype $$$) you'll find on your disk or with your
latest complete version.

     MAGE31 is slightly easier to use than RESQ12 but also slightly harder
to set up because it must be adapted to the different versions of WordStar
(3.3, 3.0, 2.26).  If you find a copy of MAGE31.COM that's set for your
WordStar, you're in business.  Otherwise, you must get the assembly
language file (MAGE31.ASM) and adapt it to your WordStar by changing an
address in the file and reassembling it.  The ASM file includes directions.
Once it's set, MAGE31 is quick, simple, and foolproof.

     RESQ12 does the same job in a slightly different way.  When you run
RESQ12, it asks for a string (a combination of characters or words) from
your lost text.  It searches upward in memory until it finds that string,
moves back downward to what it thinks is the start of your text, and then
saves everything from that point to the end of your text.  In practice, it
works quickly and pretty accurately (assuming the string really is in your
text).  Unlike MAGE31, RESQ12 comes ready-to-run and also works with
programs other than WordStar.

     If you use WordStar, make sure you have one or the other of these
programs.


PAIRX

     WordStar nightmare #2:  You start printing a long file.  You watch the
first half-page come out, just to make sure everything's OK, then wander
off to do something else.  You return half an hour later to discover your
printer has been underlining every word from page 2 on--you left out a
single ^S needed to stop the underlining.

     PAIRX, by Eric Gans, is a simple, fast solution to that problem.  It
checks WordStar files to make sure that the commands for underlining (^S),
boldface (^B), double-strike (^D), strike-out (^X), superscript (^T), and
subscript (^V) all come in matched pairs.  If it finds any of these
characters unmatched (turned on without being turned off), it gives you the
page and line number of the isolated character.  As it comes, PAIRX expects
to find a match within one line, but the documentation explains how to
increase this range, if you want to.

     PAIRX isn't as essential as MAGE31 or RESQ12, but it saves time,
frustration, and tractor-feed paper.


FILTER PROGRAMS

     WordStar document files are not just ordinary files--they have many
special characters in them to represent the soft carriage returns, soft
hyphens, soft spaces, and print control characters that make WordStar's
formatting so versatile.  These special characters can cause problems if
you transfer a WordStar file to another word processor, an unprepared
typesetter, or many bulletin boards and information services.  So there are
many programs available to "harden" WordStar's soft format by filtering out
the special characters, resulting in a file (often called a standard ASCII
file) acceptable in even the most conservative circles.  And because there
are so many WordStar users, there are also filter programs to "soften"
standard ASCII files, converting them to WordStar document files so they
can be reformatted with all WordStar's tricks.

     HRDSFT, by Kenneth Toy, is a lovely program.  It's small (2K) and
fast, and it can either harden or soften a file.  It does the standard job
of hardening, converting all soft characters to normal ones and removing
all control characters except carriage returns, line feeds, and tabs.

     It also does a good job of softening a hard file, converting all hard
carriage returns to soft ones unless followed by a space, a tab, a period
(dot commands), or another carriage return.  This method works pretty well
on single-spaced text, though some things, like columns of numbers at the
left margin can trip it up (that's true of all softeners).  HRDSFT has two
weak points, which it shares with similar programs:  it doesn't soften
double-spaced text properly, and it doesn't harden soft hyphens but simply
removes them.  HRDSFT also can't handle files larger than will fit in
memory, about 48K or so (but I rarely have files that long).

     ENSOFT and UNSOFT are a matched set of filters; ENSOFT softens and
UNSOFT hardens.  Both work well but have problems with the same things
HRDSFT does:  double-spaced text, soft hyphens, and occasional odd formats.
They are bigger (about 8K each) than HRDSFT but can handle files of any
length.

     FILT6, by Irv Hoff, is a hardener only; it turns any text file into a
standard ASCII file by cleaning out all special characters and all control
characters except carriage returns, line feeds, and tabs.  It works with
any word processor's files but has special WordStar options, such as
retaining or removing dot commands.  You can also have it replace spaces
with tabs wherever possible (which saves space in long files) or replace
tabs with spaces (some programs choke on tabs).  FILT6 also has special
options for handling assembly language files, which make it handy for
programmers.

     I keep HRDSFT on my word processing disks, because it's smaller and
faster than ENSOFT/UNSOFT, and FILT6 on my programming and communications
disks, because of it extra options.


REPCS, CHGCHR

     These two programs solve one of WordStar's nagging problems--how to
search and replace ^S.  WordStar's find and find/replace functions can
search for almost any character, including control characters.
Unfortunately, they can't search for the ^S WordStar uses for underlining
because the search command uses ^S as its match-any-character wildcard.  So
you can't, for example, use the search command to find every word in a file
that you've underlined.

     REPCS (sometimes called REPLACE), by Maryanne Weston and Donald Hay,
solves that problem.  It will search a WordStar file and replace every ^S
with the character @.  You can then use WordStar's search command to find
the @ and either delete it or change it to something else.  That's all
REPCS does, though you can use DDT to patch the search and replacement
characters to new values.

     The more versatile CHGCHR, by Al de la Torre, will search a file for
any character and replace it with any other, including non-printing control
characters like ^S.  You specify the search character and replacement
character on the command line when you run the program, and CHGCHR does the
rest.  CHGCHR is not limited to working on WordStar files.

     You probably won't need these programs often, but I've known people
driven to distraction trying to find every ^S in a long WordStar file
without them.


FTNT14

     Unlike Perfect Writer and some other word processors, WordStar can't
do footnotes on its own.  There is, however, a public domain footnoting
program, FTNT14 by Eric Meyer.  FTNT14 has so many features and subtleties
I can't cover them all, but here's a quick rundown.

     FTNT14 will extract notes from a WordStar file and print them as
either footnotes or endnotes.  It will number them automatically (even
starting over on each page, if you want), format them in various ways,
carry long footnotes over to the bottom of the next page, and allow you to
use separate print commands (like condensed printing) in the notes.

     FTNT14 will also handle internal page references for up to 52
locations within your text.  You can refer to a chapter, table, or other
location from any point in the text and have FTNT14 insert the correct page
reference when it finally formats your text.

     FTNT14 also does what Meyer calls "figure block" formatting.  This
command allows you to specify a certain number of lines of text (a table,
perhaps, or an example) as a figure block.  When FTNT14 encounters this
command, it will print the figure block at the top of the following page,
but unlike WordStar's .CP command, FTNT14 will fill in the rest of the
space on the current page with text from after the block.  In other words,
there won't be any gaping whole at the bottom of the page the figure block
came from.  That's neat.

     FTNT14 works in two passes, one to extract the notes and one to create
a formatted file for final printing.  It obeys all WordStar control
characters and most dot commands.  It makes minor changes in the behavior
of a few dot commands (to accommodate the notes) and won't let you use dot
commands that affect vertical page format:  top and bottom margins, heading
and footing margins, line height.

     Formatting footnotes, especially in a long file, is a touchy, complex
business, but FTNT14 does its job well.  It's a class program.


INDEX PROGRAMS

     Along with footnotes, go indexes.  There are three public domain
indexing programs available for WordStar, and though none is as classy as
FTNT14, all can save you some drudgery.

     WINDEX12, by Eric Gans, is the newest and slickest of the three.  It
works on WordStar document files only, can index both words and phrases,
and correctly handles words divided by soft hyphens.  It can index up to
254 keywords, handles files of any length and references up to 9999 pages,
and, like the other programs, alphabetizes the index.

     Unlike the following two programs, WINDEX12 doesn't require you to
mark the words to be indexed in your text.  Instead, you build a separate
file containing the keywords (up to 254 of them) you want indexed, or you
enter the keywords from your keyboard at indexing time.  That does make
your job easier, but it also means you may miss some phrases you waned to
index, unless you search through the file on your own.  It also means that
you can't make WINDEX12 ignore certain occurrences of a word.

     INDEX101, by Tom Jennings, has you mark the words and phrases you want
indexed by using the unused WordStar control characters ^PK and ^PP.
(Actually, ^PP is used if you're trying out WordStar's undocumented and
very buggy proportional spacing, in which case you'll have to remove the
^PP characters after you index.)

     INDEX101 adds the index to the end of your WordStar file, unlike the
other two programs, which write the index to a separate file.  It also
erases the old index each time you run the program.

     GENINDEX also has you mark the words to be indexed in your text, using
^PQ for major references (printed boldface) and ^PW for minor references
(printed normally).  These control characters are two of WordStar's user-
definable print control characters; you'll have conflicts if you're already
using these (to control a dot matrix printer, for example).

     GENINDEX writes its index into a separate file and updates (rather
than erases) any old index it finds.  However, GENINDEX doesn't work on
normal WordStar document files, only on files that have been printed-to-
disk using the "disk file output" option of WordStar's print command.

     I rarely index anything, so my experience is limited, but WINDEX12 is
the easiest program to use.  It's also the program to use if you want to
index every occurrence of a keyword.  INDEX101 gives you more control if
you don't want to index every occurrence of a keyword and probably also if
you're going to index many phrases.  None of the three programs puts out an
index that's really ready to print; you'll want to do some cleaning up.


[FTNT14]

     Unlike Perfect Writer and some other word processors, WordStar can't
do footnotes on its own.  There is, however, a public domain footnoting
program, FTNT14 by Eric Meyer.  FTNT14 has so many features and subtleties
I can't cover them all, but here's a quick rundown.

     FTNT14 will extract notes from a WordStar file and print them as
either footnotes or endnotes.  It will number them automatically (even
starting over on each page, if you want), format them in various ways,
carry long footnotes over to the bottom of the next page, and allow you to
use separate print commands (like condensed printing) in the notes.

     FTNT14 will also handle internal page references for up to 52
locations within your text.  You can refer to a chapter, table, or other
location from any point in the text and have FTNT14 insert the correct page
reference when it finally formats your text.

     In addition, FTNT14 does what Meyer calls "figure block" formatting,
which allows you to specify a certain number of lines of text (a table,
perhaps, or an example) as a block.  This command causes FTNT14 to print
the block at the top of the following page, but unlike WordStar's .CP
command, FTNT14's figure block command fills in the rest of the space on
the current page with text from after the block.  In other words, there
won't be any gaping hole at the bottom of the page the figure block came
from.  That's neat.

     FTNT14 works in two passes, one to extract the notes and one to create
a formatted file for final printing.  It obeys all WordStar control
characters and most dot commands.  It makes minor changes in the behavior
of a few dot commands (to accommodate the notes) and won't let you use dot
commands that affect vertical page format:  top and bottom margins, heading
and footing margins, line height.

     Formatting footnotes, especially in a long file, is a touchy, complex
business, but FTNT14 does its job well.  It's a class program.


     All these programs are available in the usual channels, except
WINDEX12, which is so new that it's only available on RCP/Ms at the moment.
Next issue, I'll cover some text utilities for use with any word processor.

------------------------------------------------------------------------------
      Ted Silveira is a freelance writer and contributing editor to several
   computer-oriented publications.  He appreciates suggestions or feedback
   and can be reached through the KAY*FOG RBBS [phone] and CompuServe
   (72135,1447) or by mail to 2756 Mattison Lane, Santa Cruz, CA 95065.

-------------------------  End of CPM-CC05.ART Text  -------------------------