Skip to content

Releases: mcci-catena/arduino-lmic

Maintenance release

10 Feb 20:25
87b6398
Compare
Choose a tag to compare
  • v2.3.2 is a patch release. It incorporates two pull requests.

    • #204 eliminates a warning if using a custom pinmap.
    • #206 updates CI testing to Arduino IDE v1.8.8.

Maintenance release

29 Nov 04:43
23e6d5e
Compare
Choose a tag to compare

This release has the fix for #199, which prevents https://github.com/mcci-catena/arduino-lorawan from building with v2.3.0 (when 2.3.0 is installed over the internet from the IDE).

Support TCXO, add pinmaps to library, refactor

26 Nov 09:22
2f6cabe
Compare
Choose a tag to compare

This is a substantial release. It's believed to be backward compatible; please report any issues.

The release introduces two important changes.

  1. The pinmap is extended with an additional field pConfig, pointing to a C++ class instance. This instance, if provided, has extra methods for dealing with TCXO control and other fine details of operating the radio. It also gives a natural way for us to extend the behavior of the HAL.

  2. Pinmaps can be pre-configured into the library, so that users don't have to do this in every sketch.

Accompanying this is a fairly large refactoring of inner header files. We now have top-level header file <arduino_lmic_hal_configuration.h>, which provides much the same info as the original <hal/hal.h>, without bringing most of the LMIC internal definitions into scope. We also changed the SPI API based on a suggestion from @manuelbl, making the HAL more friendly to structured BSPs (and also making the SPI API potentially faster).

Bug fixes: added a new API (radio_irq_handler_v2()), which allows the caller to provide the timestamp of the interrupt. This allows for more accurate timing, because the knowledge of interrupt overhead can be moved to a platform-specific layer (#148). Fixed compile issues on ESP32 (#140 and #153). We added ESP32 and 32u4 as targets in CI testing. We switched CI testing to Arduino IDE 1.8.7. Fixed issue #161 selecting the Japan version of as923 using CFG_as923jp (selecting via CFG_as923 and LMIC_COUNTRY_CODE=LMIC_COUNTRY_CODE_JP worked). Fixed #38 -- now any call to hal_init() will put the NSS line in the idle (high/inactive) state. As a side effect, RXTX is initialized, and RESET code changed to set value before transitioning state. Likely no net effect, but certainly more correct.

Add feather ABP example

04 Oct 03:29
32846d1
Compare
Choose a tag to compare

This release primarily adds feather ABP example. Also added: more CI testing, doc improvements, #131 for lorawan.io compatibility.

Patch release

24 Sep 04:19
30c766d
Compare
Choose a tag to compare

V2.2.1 corrects the value of ARDUINO_LMIC_VERSION (#123), allows ttn-otaa-feather-us915 example to compile for the Feather 32u4 LoRa (#116), and addresses documentation issues (#122, #120).

Stable code release with numerous fixes and enhancements

19 Sep 17:26
2e76a81
Compare
Choose a tag to compare

Biggest enhancement from the point of view of usability: this release adds a Feather example that sends sensor data, along with some helper routines for floating-point encoding.

Biggest technical enhancement: official Listen-Before-Talk support (e.g., for Japan).

Lots of bug fixes and documentation adjustments (we hope they're improvements, but not up to us to judge).

No APIs were deprecated or removed. Multiples were added.

Documentation and example improvements

24 Mar 21:34
44ae64d
Compare
Choose a tag to compare

This release was primarily made to fix #56. While at it, we updated README.md, and did some light updates of the abp examples.

Fix radio driver power control bugs

07 Feb 04:22
Compare
Choose a tag to compare

Fixes #50 and #47, making the radio control for the SX1276 match the datasheet.

Fix issue #43 (asserts for US/AU after LinkAdrRequest)

02 Jan 19:36
Compare
Choose a tag to compare

This is a bug patch release, correcting issue #43.

Fix issue #39

15 Dec 16:39
Compare
Choose a tag to compare

V2.1.2 has a fix for issue #39 (adding a prototype for LMIC_DEBUG_PRINTF if needed). Fully compatible, and in theory just a bug fix, so we just incremented the patch level.