fsave with csglobals

[jtaseff]

Hi there,

We're testing out using fsave/fload instead of our implementation of something similar. Couple of questions.

 

Is there a way to deal with csglobal variables? I tried putting this in the .tpl file:

&1 csvar1

&2 csvar1

 

And in the .cfg file, I only get this, which won't work to reload to the correct coordinate systems:

Q1024 = 10

Q1024 = 20

 

 

 

 

Additionally, is there any way to have it work with variable names? For example, if I always have it fsave a global by name, myvar1 gets stored as P1111. But if I update the project to add a new global before that, myvar1 is now in P1112, and the next fload will put the saved value into the wrong place in P1111.

[steve.milici]

Use the Non-Saved Setup Data Structure Elements Coord[x].Q.

[jtaseff]

Is there a way to do it by name when accessing Coord[x].Q, or a way to programatically find the Q-number of a named variable? Would much rather be able to save by name if the project changes around, instead of tracking Q numbers around by hand

[steve.milici]

Not at this time. We will have this in the 4.x.x.x version of the IDE.

[jtaseff]

For anyone else trying this, I did find a way around the csglobal problem:

 

In the .tpl file, I can have it run this to output the currently addressed CS:

&1
Ldata.coord
csvar1
csvar2
&2
Ldata.coord
csvar1
csvar2

 

This produces the following in the .cfg file on fsave, which seems to load things in properly on fload.

Ldata.coord=1
Q1024=10
Q1025=20
Ldata.coord=2
Q1024=30
Q1025=40

[steve.milici]

Yes – this is the “valid” correct mechanism to use.

 

I was testing this yesterday as a possibility but wanted to verify with the firmware programmes that the “tpl” file will be processed in the order the command are presented (they did verify this).