|
From: Daniel J S. <dan...@ie...> - 2003-10-23 21:59:55
|
Here is an observation in x11.trm that technically isn't a bug, but there is a small issue with consistency. I'm working on a short patch right now that highlights this issue. I'm debating a work around but it might be nice for the list to think whether a small alteration should be made for consistency, in which case I wouldn't have to do a work around. If one starts up gnuplot and issues a set term x11 3 the X11_ipc to gplt_x11 is not valid yet. That means that anything in the X11_options() routine that checks if X11_ipc is valid will not send information to gplt_x11 at that point in time. There currently is some code there, but it has no real effect because what would have been sent over is something saved internal to x11.trm code and gets resent. (You can probably guess I'd like to put something there that won't be saved necessarily. But don't let that sway you, I'm arguing consistency here.) Once something is plotted, say plot sin(x) X11_ipc is up and running and all things are hunky dory. So, the consistency issue... The ramification of the above is that set term x11 3 will cause the x11 window to appear on the desktop, _unless no plots have been created yet_. In other words, from a fresh invocation set term x11 3 creates no window. But plot sin(x) set term x11 3 creates _two_ windows. Any comments on this? Thanks, Dan |