- Added encodeLatLonToSelectedMapcode as a convenience for languages
that use the C library, but have difficulties dealing with
multi-dimensional arrays (like Swift).
2.5.1
- Updated unit test to compile with plain C and added some test cases.
2.5.0
- Added support for getting territory names in English and local
alphabets.
- Added much improved unit test scripts to run gprof, valgrind,
the CLang address sanitize and compare the output of the new library
with and older version.
Changes in 2.4.0
- Added scripts for Korean (Choson'gul / Hangul), Burmese, Khmer,
Sinhalese, Thaana (Maldivan), Chinese (Zhuyin, Bopomofo), Tifinagh
(Berber), Tamil, Amharic, Telugu, Odia, Kannada, Gujarati.
- Renamed constants to more correct terms (Malay to Malayalam, Hindi to
Devanagari).
- Added getAlphabetsForTerritory(t), returning the most commonly used
alphabets for territory t.
- Improved some characters for Arabic and Devanagari.
- Fixed Bengali to also support Assamese.
- Improved readability of implementation.
- Added conditional define option for unittest: NO_POSIX_THREADS. Add
-DNO_POSIX_THREADS to your compiler command-line to not use
pthreads.h, for example, on MSVC platforms.
- Added conditional define option library: NO_FAST_ENCODE. Add
-DNO_FAST_ENCODE to your compiler command-line to not use the fast
encoding (default is to use fast encoding). Using fast encoding speeds
up de encoding by a factor of 2. For normal use, there is no reason not
to use fast encoding.
- Added parseMapcodeString to get individual mapcode elements, like the
territory code, the 'proper' mapcode (without the territory and
precision extension) and the precision extension parsed from a (user)
input string.
- Additional bug fixes.
Based on a PR by Aleksej Lebedev.
A mapcode represents a location. Every location on Earth can be
represented by a mapcode. Mapcodes were designed to be short,
easy to recognise, remember and communicate. They are precise
to a few meters, which is good enough for every-day use.
