History

Damien George ad4656b861 all: Rename BYTES_PER_WORD to MP_BYTES_PER_OBJ_WORD. The "word" referred to by BYTES_PER_WORD is actually the size of mp_obj_t which is not always the same as the size of a pointer on the target architecture. So rename this config value to better reflect what it measures, and also prefix it with MP_. For uses of BYTES_PER_WORD in setting the stack limit this has been changed to sizeof(void *), because the stack usually grows with machine-word sized values (eg an nlr_buf_t has many machine words in it). Signed-off-by: Damien George <damien@micropython.org>		2021-02-04 22:46:42 +11:00
..
Makefile	examples/embedding: Example for embedding MicroPython in an app.	2016-07-22 22:13:03 +03:00
Makefile.upylib	examples/embedding: Remove obsolete fatfs files from build.	2019-11-11 11:37:32 +11:00
README.md	examples/embedding: Add code markup and fix typo in README.md.	2018-06-18 12:29:22 +10:00
hello-embed.c	all: Rename BYTES_PER_WORD to MP_BYTES_PER_OBJ_WORD.	2021-02-04 22:46:42 +11:00
mpconfigport.h	examples/embedding: Replace symlink of mpconfigport.h with real file.	2019-10-29 22:53:34 +11:00
mpconfigport_minimal.h	all: Fix auto-enable of MICROPY_GCREGS_SETJMP to select GC behaviour.	2020-04-30 16:49:42 +10:00

README.md

Example of embedding MicroPython in a standalone C application

This directory contains a (very simple!) example of how to embed a MicroPython in an existing C application.

A C application is represented by the file hello-embed.c. It executes a simple Python statement which prints to the standard output.

Building the example

Building the example is as simple as running:

make

It's worth to trace what's happening behind the scenes though:

As a first step, a MicroPython library is built. This is handled by a separate makefile, Makefile.upylib. It is more or less complex, but the good news is that you won't need to change anything in it, just use it as is, the main Makefile shows how. What may require editing though is a MicroPython configuration file. MicroPython is highly configurable, so you would need to build a library suiting your application well, while not bloating its size. Check the options in the file mpconfigport.h. Included is a copy of the "minimal" Unix port, which should be a good start for minimal embedding. For the list of all available options, see py/mpconfig.h.
Once the MicroPython library is built, your application is compiled and linked it. The main Makefile is very simple and shows that the changes you would need to do to your application's Makefile (or other build configuration) are also simple:

a) You would need to use C99 standard (you're using this 15+ years old standard already, not a 25+ years old one, right?).

b) You need to provide a path to MicroPython's top-level dir, for includes.

c) You need to include -DNO_QSTR compile-time flag.

d) Otherwise, just link with the MicroPython library produced in step 1.

Out of tree build

This example is set up to work out of the box, being part of the MicroPython tree. Your application of course will be outside of its tree, but the only thing you need to do is to pass MPTOP variable pointing to MicroPython directory to both Makefiles (in this example, the main Makefile automatically passes it to Makefile.upylib; in your own Makefile, don't forget to use a suitable value).

A practical way to embed MicroPython in your application is to include it as a git submodule. Suppose you included it as libs/micropython. Then in your main Makefile you would have something like:

MPTOP = libs/micropython

my_app: $(MY_OBJS) -lmicropython

-lmicropython:
	$(MAKE) -f $(MPTOP)/examples/embedding/Makefile.upylib MPTOP=$(MPTOP)