dots, and copying

pviton's picture

SWP 6.0.21 on MS Windows


Start from the Std Latex article and begin some display
math with Ctrl+d. Enter subscripted y's:

y = y_1,y_2,

at this point you want to enter "..." ie latex \dots.
Hit Ctrl+spacebar to open the MacroBox. Type in "dots".
Hit Return. The dots are inserted as expected.

You now want to complete the expression, with a comma
to be followed by y_n. Type the comma. What happens to me
is that the dots disappear, so I'm left with y1,y_2,, .

Note that this doesn't always happen when you use the
MacroBox: it doesn't when you use it to enter \alpha
and then type in a comma.

If you start over, and enter the \dots using the symbol
panel, then you can complete the expression as desired.
But it shouldn't be necessary to memorize cases where
the MacroBox works versus cases where it doesn't.


Suppose you've got the y_1,y_2, ..., y_n expression entered
succesfully. You now want to copy and paste it into a
new line for modification. So hit Return for a new line
in the display. Type z = . Now select and copy the
stuff on the rhs of the equal sign in line 1. Put the cursor
to the right of the equal sign on line 2. Press Ctrl+v to

What I get is the expression *reversed*, ie
y_n,...,y_2,y_1. See attached picture

After restarting the system I tried again.
The \dots behavior was unchanged; but this
time the copy gave y_1y_n,...,y_2, where
there was no comma between y_1 and y_n; and
note the trailing comma after y_2.

Other variants:

A. Line 1 is y_1y_2y_3 (product).
Line 2 with copy+paste is y_13yy_2

B. Line 1 is y_1,y_2,y_3 (ie comma-separated
but no \dots).
Line 2 with copy+paste is y_3,y_2,y_1
(ie backwards again), which suggests that
it's not the \dots that's at fault in the
original case.

C. Line 1 is : 12345. In this case copy+paste
works correctly, suggesting that the problem
may be with the subscripted variables, or perhaps
with math generally.

But the bottom line is that system fairly consistently
mangles simple copy operations, even if the form of
the mangling varies. As long as problems like this persist
I really don't see how SWP6 can be considered to be even a
minimally acceptable math editor for serious work.
I note that SWP55 has none of these problems.

copy-reverse.gif61.5 KB
jm's picture

I can reproduce the behavior

I can reproduce the behavior with dots using Mac OS, so it is obviously a general problem.

The issue of copy and paste reversing math expressions is, I think, the same problem that I reported in topic SWP 6.0.21 Math Bugs. It seems to be a quite general problem that is, I think, new to SWP 6.0.21 and needs to be fixed urgently for SWP to be at all useful for working with math expressions.