[Date Prev][Date Next][Subject Prev][Subject Next][ Date Index][ Subject Index]

Re: email-style italics: MARKUP.PM

Robert, re

Skipping for the nonce the issue of Email addresses, in your
sample paragraph, I really don't understand WHY there are
unbalanced tags. E.g. what is the purpose of unbalanced "I
don't think it's _usable in any later versions"? If that's an
error, then it's an error. I don't see how the parent program
that created this usage of underscore could possibly
discriminate between a tag that means Bold On/Off, and a tag
that means something else... Or are we to assume that these
tags are always inserted manually, and recipients are always
assumed to discriminate -- to translate them appropriately as
Bold in their heads when that's clearly what they mean, and
otherwise to interpret them as something else?
This discussion began with Bill's question concerning converting
text-based email italics tags (_) to XyWrite or Quark markup, if I
recall correctly. So my assumption is that these tags would usually be
inserted manually and would be subject to error, and that the emails
would not nec. be written by someone who cared or knew that someone else
might want to convert text-based email style markup to some other sort
of markup. So I did a couple tests to see what might cause problems,
creating various possible errors (an admittedly non-real-world test).


Actually, you _can_ stop, fix, and resume from the current

location in the file: the program operates from the *current
location* to End_Of_File (not from Top_Of_File to EOF).


Then this pretty much take care of what I was writing about; obviously I
hadn't tried doing that.

(The following comment is sent with the caveat that you should read it
in light of my other comments below about my conversion needs:) if the
user could reload the last set of input tags without rekeying the tags,
after having stopped the program to make a correction, that might speed
use. Also, if the user had the option to force the program to start,
having been warned that there are mismatched tags, instead of having the
program abort, you might be able to correct the problems as you work
through the file.

The way MARKUP.PM is designed, if you're verifying each change,
it treats them as balanced pairs, not as individual tags, so
when you elect NOT to make the change, it skips both members of
the pair and moves on to the first tag in the next-following
pair. That halves the number of "change" PRompts you have to
deal with (no small matter when there are, say, 518 underscores
as in one of my test files).


Understood; I noticed that when I used the program.
Obviously we could stop on each
tag, and then, if skipped, treat the next following tag as the
first member of a new pair. And the user could be forced to
verify each beginning/ending tag -- and be shown the exact tag
that will be put if they proceed with the change. That would
solve the problem with tags that appear in Email addresses or
other similar situations. It's adds to the user's burden, but
ensures accuracy. Is that what you want? Or do you prefer the
"convert this pair, correct this pair, or skip this _ and move
to next _" formula? And dump any check for balance or
even-numberedness, except a simple check for basic existence?
Here's where I should shut up. This was Bill Troop's issue, not mine; I
played with the program to see what it would do, and was interested in
how it does and might work, based on my experience, but I'm not the
likely end-user here.

 I
don't really care -- although you have to remember that when
you're staring at the Begin tag, it isn't always obvious or even
visible where the End tag is (for example, in a long
"quotation").

Understood, and occasionally experienced with frustration.
Remember, I'm trying to get away from the notion
that * and _ are the only tags involved -- far too parochial.
You should be able to change "quotation" to “-” or

whatever...


Also understood. But my comments focused on issues related to _ and *
because that was how this all began.


Paul