Looks like my notes didn't save.
Basically python scripts and functions are limited to 1,000,000 lines of code. there is a PEP related to this.
In my profiling test I ran into this limit. Also importing takes longer when there are a ton of lines of code.
The basic idea behind this fix is to take the unrolled loops and convert them back into loops.
Should I do this when
- Within new object which walks the graph without executing it?
- creating the code?
- just before writing to file?
- just before importing?
My gut feeling is that the easier options are 2 and 3, but in the long term 1 would be the most maintainable since get_calc is quite bloated.
Looks like my notes didn't save.
Basically python scripts and functions are limited to 1,000,000 lines of code. there is a PEP related to this.
In my profiling test I ran into this limit. Also importing takes longer when there are a ton of lines of code.
The basic idea behind this fix is to take the unrolled loops and convert them back into loops.
Should I do this when
My gut feeling is that the easier options are 2 and 3, but in the long term 1 would be the most maintainable since get_calc is quite bloated.