Rick et al,
  Yes, I should have given a little more detail. One problem exists when
using z\" interpretively:

The following shows that no zero byte is put in the scratch buffer:

z\" Hello" zcount 5 + dump
  411429 48 65 6C 6C 6F 20 65 6E 64 69 6E 67 20 70 72 69 Hello ending
pri
  411439 6E 74 65 64 20 70 61 67 65 00 00 00 00 00 nted page.....
ok

This one shows that \" is the escape sequence for the embedded ASCII $22
character. \' does not work, as stated in the manual on page 84 in table
14. This is a simple misprint in the book, or the glyph doesn't look
like a double quote mark:

z\" \'Hello\'" zcount 5 + dump
  4114A9 5C 27 48 65 6C 6C 6F 5C 27 69 6E 67 20 70 72 69 \'Hello\'ing
pri
  4114B9 6E 74 65 64 20 64 6F 63 75 6D 65 6E 74 6E 73 00 nted
documentns.
  4114C9 00 00 00 00 ....
ok
z\" \"Hello\"" zcount 5 + dump
  411529 22 48 65 6C 6C 6F 22 65 74 20 64 65 66 61 75 6C "Hello"et
defaul
  411539 74 20 70 72 69 6E 74 65 72 49 46 54 42 41 52 2E t
printerIFTBAR.
  411549 42 4D 50 00 00 00 00 00 BMP.....
ok

These words DO work as expected when compiled into colon definitions. I
have not yet done an exhaustive check on all the words, nor the U\"
words , but offhand it seems that the the z" word works correctly
interpretively, but the z\" word does not.

> Would you please give an example of how you are using the
> string compiler, and a memory dump of the result?
> *********** REPLY SEPARATOR ***********
>
> On 7/19/2000 at 10:02 PM Bob Dickow wrote:
>
> >My SwiftForth version 2.00.2 26 Oct 1999 seems to fail to compile the
> >ending 0 byte on words such as Z\". Also, these words don't always
work
> >quite as expected in some other ways for me too. Bug? Or just my own

Bob
Received on Thu Jul 20 2000 - 11:03:11 PDT