It's pretty easy to convert basic non GUI programs that don't make a
lot of library use. So Modula-2, Pascal, C, VB6 and some Java  & C# is
easily manually converted.

If it uses GUI or a lot of libraries esp. OS APIs  and file operations
an automated C to Jal tool will be like a chocolate teapot.

I worked on the original C++ to C preprocessors. I think C is a
terrible language. It was NOT invented as a high level language but as
a portable abstract assembler to move UNIX software to different cpus
and HW.

So I have no interest in either automated tool C -> Jal (because of
library / API issue and bad C coding practices regarding type casting
and strings) nor JAL to C.
IMO the most worthwhile (only worthwhile) non PIC targets are the
CortexM0, CortexM1 and maybe CortexM3 ARMs. These should be done by
producing Thumb2 assembler, not an intermediate C. The larger or older
ARMs and ARMs for PDAs, Routers, Setboxes, Phones and MID/Netbooks are
best programmed with existing tools, libraries and languages for the
existing OSes.


The higher PICs than 18F4xxx type range are of zero interest,
especially the Microchip 32 bit which are MIPS licence core. ARM
cortex Mx series is the upgrade path. From $1 upwards. I should get my
entry level €5 Cortex M0  today in post :-)

For Cortex M0, M1, M3 I propose a JAL3 that will 100% compile JAL2
code but have extensions that take advantage of having indirection,
minimum 8K RAM, no paging on Flash Program space or RAM space and no
shared banks. "Handy" would be a good alias for Jal3 (Handy to make
JAL produce Thumb2 on ARMs)

Summary. Ignore C unless you have to take notice. My vote is zero
involvement of JAL with C.

My son is doing a C project on 16F877 using Microchip IDE and HiTech-
C. He is not a happy puppy despite all his desktop C and Java
experience on Windows and Linux. The ASM it is producing is rubbish
compared to JAL. He is plagued with bugs in #defines
C relies a lot on Macros and non-macro #defines. This is evil.
Larger projects rely on incomprehensible Make files. This is why good
C IDEs have their own project files too.

We want to avoid any more development of "pragma" in JAL and avoid "C"
contamination.  Even some long term official C libraries have evil
gotchas. If you think scanf is input persona of printf you are wrong.
If you think printf is good then you need therapy. Originally C++ and
Java had no printf. C programmers forced the inclusion.

Almost all bugs and security issues in Linux and Windows software is
due to C programming practices unchanged since the late 1970s. Much C+
+ code breaks the  real Strustrupp  intentions of C++ and is really C
with Class declarations replacing struct declarations.


-- 
You received this message because you are subscribed to the Google Groups 
"jallib" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/jallib?hl=en.

Reply via email to