Well you are going to have to grasp the nettle sooner or later. I think that most of the things that are going to be changed have been changed by now and most of the issues are understood so there is not match gain in waiting now.
Some libraries still #include WProgram.h, and have to be updated. This is the only issue I found so far (but TBH I've not tried all of my previous sketches / libraries with 1.0).
well ok so far
the main sketch that I was nervous about didn't need too much surgery
main change was from using SoftwareSerial as BYTE is no longer supported
made the suggested change to Serial.write() and it - er well it failed badly!
fortunately I recognised the symptoms as I had seen them before
so I moved all the strings by wrapping F() around them
phew all working again!
couple of observations:
a) it would be nice to view some sort of map so I could anticipate running out of <>
b) the sketch is about 1500 bytes bigger than under IDE 21
I'll persist with 1.0 for now!
thanks for the encouragement
mmcp42:
b) the sketch is about 1500 bytes bigger than under IDE 21
As I recall, this is mostly due to the changes in Serial. You said you moved your strings with F(), were they in PROGMEM before? That could also account for the increase if they were not.
Yeah actually after thinking about it for a minute, what I said doesn't really make sense. Whether you store the strings in PROGMEM or not won't change the size of the sketch. They're going into flash one way or another...
If you believe there is a per-string overhead in the latter, please let me know. Mikal and I spent a few hours ensuring the F-macro is as efficient as possible. If there is a per-string overhead then we made a mistake that should be corrected.