Re: Change Invisibles: migrating from Xy 3.55 to 4.017

[Robert Holmgren holmgren@xxxxxxxx]

** Reply to note from xywrite@xxxxxxxx Sat, 11 Mar 2000 11:04:04 -0500

Leslie:

> : Let us say you want to search for the italicized Jack in the example
> below:
>
> All work and no play makes Jack a dull boy.
>
> If you write your pgm using the ctrl < and > characters (here
> represented by a single angle bracket):
>
> BX se /Jack/Q2
>
> All is well.
>
> But if you are in the habit (as I am, for the sake of less visual
> confusion when in normal mode) of using the ascii characters 174/175
> (here represented by a double angle bracket):
>
> BX se /≪MDUL≫Jack≪MDNM≫/Q2
>
> You get a "not found" message.

In the first instance, you use 1-byte 174/175 guillemets. They work.
In the second instance, you are putting 3-byte 174/175 into your program.
It's a tenet of XyWrite that if you move 3-byte characters from text to
the command line (as happens when you use BC ... XC in a program written in the text window),
they are transformed from 3-byte to 1-byte chars -- thus it works; whereas if you use BX ...Q2,
they remain 3-byte chars and therefore it doesn't work. This is simply a
matter of correct coding, not of any flaw in BX ...Q2.

At this moment, I can recall only one instance where BC ...XC is necessary, namely:
CD commands (Change_Directory) sometimes need to be issued on the
CMline, especially when a DIRectory is displayed in the text window. But otherwise BX ...Q2 is
absolutely flawless, and in every respect superior and preferable (faster, less disruptive,
capable of executing commands much longer than the CMline could accommodate) to BC ...XC. I'd
wager that every time BX "fails", there's actually a coding error.



---------
Robert Holmgren
holmgren@xxxxxxxx
---------