GitLab

See pyb_usart_global_debug variable.

Also did some work on USB OTG, but nothing working yet.

You can now append a zipfile (containining uncomressed python sources)
to the micropython.hex file.

Use MEMZIP_DIR=directory when you call make, or set that in your
environment to include a different tree of source files.

Added sample /boot.py, /src/main.py, /test.py and /src/test.py files.

Added run command so that you can execute scripts from REPL (until import is implemented).

Added build directory to .gitignore

Fixes Issue #132.

* Fix usart_obj_tx_str to send the actual string.
* Fix issue #126

Creating of classes (types) and instances is much more like CPython now.
You can use "type('name', (), {...})" to create classes.

* Add a DEBUG flag to Makefile, enables: "-O0 -ggdb"

* Add Micro Python bindings to usart.c
* Export USART python object to namespace pyb

Use make V=1e make V=1 or set BUILD_VERBOSE in your environment to increase build verbosity.

This should fix issue #117

While updating the teensy Makfile, I noticed that the stm Makefile
was using calling arm-none-eabi-objcopy and size rather than
$(OBJCOPY) and $(SIZE), so I added these 2 for consistency.

* Fix rtc_init to use LSI for RTC clock source when LSE is not detected.

Updated mconfigport.h to enable GC

* Fix LED pin enum, first one should start at 1
* Fix LED initialization typo

* Fix LED pin enum, first one should start at 1
* Fix LED initialization typo

* Add a TARGET definition to Makefile (default PYBOARD).
* Add support for discovery LEDs in led module.
* Add support for discovery user switch in usersw
* Add EXTI interrupt handler for discovery user switch on (PA0).
* Parameterize led and usrsw modules pins and port.
* Issue #83

* Move user switch code from main.c into a separate module (usrsw)
* Add usrsw.c to Makefile

Now much more inline with how CPython does types.

Qstr's are now split into a linked-list of qstr pools.  This has 2
benefits: the first pool can be in ROM (huge benefit, since we no longer
use RAM for the core qstrs), and subsequent pools use m_new for the next
pool instead of m_renew (thus avoiding a huge single table for all the
qstrs).

Still would be better to use a hash table, but this scheme takes us part
of the way (eventually convert the pools to hash tables).

Also fixed bug with import.

Also improved the way the module code is referenced (not magic number 1
anymore).

Use the storage cache not only for writing but also for reading. This
avoids reading stale data and thus data loss.

So far, only start and stop integer indexes are supported. Step is
not supported, as well as objects of arbitrary types.