Commit Graph

11 Commits

Author SHA1 Message Date
fc774fd0ff Refactor, separate docs from headers, add ru docs
Doxygen kind'a sucks and I need multilanguage documentation, so I did
that. Also, separated massive Math.h file into smaller files.
2025-06-21 20:12:15 +03:00
dd15b42b44 Add Unicode tests, fix whitespace, fix docs
Added Unicode tests (and fixed few bugs), changed whitespace script to
proccess markdown files, added new guide to showcase how to work with
Unicode.
2025-04-06 14:11:38 +03:00
6aee5a83aa Refactor string functions, add unicode support
Refactored existing functions. Added Unicode support and UTF-8, UTF-16,
and UTF-32 encoding/decoding.
2025-04-05 13:56:00 +03:00
f9ebeabb18 Refactor strings (avoid using dynamically allocated memory)
Refactored strings functions to make them use supplied string
instead of allocating new ones.

Additionally, split String/Int.c into several files to avoid define
mess that was present before.
2025-04-05 13:56:00 +03:00
9025c3d945 Fix bug related to char being signed or unsigned on different platforms 2025-04-05 13:56:00 +03:00
4b2e3da567 Add int tests, fix bugs, add docs
Added tests for string to/from integers and added documentation.
While adding tests caught few bugs and shortcomings (now everything
works as expected)
2025-04-05 13:56:00 +03:00
b7fc93a490 Refactor string to/from double conversion, fix bugs, use bigger ints
Finally, fixed bugs in bigint implementation as well as switched to
32/64 bit integer implementation (gained 2x speed up in some cases).

Additionally, I decided to split Float.c into Float.c and BInt.h for
better readability.
2025-04-05 13:56:00 +03:00
b943135d71 Fix whitespace, add asserts, add string to double function, fix bugs
Finally added StringToDouble function (should work for the majority of
the cases). Additionally fixed bug in StringFromDouble related to
incorrect rounding and added asserts (should add more asserts in the
following commits).

Also implemented some optimizations from Burger and Dybvig paper.
2025-04-05 13:56:00 +03:00
82bea0ebf8 Refactor bigints, add int and float conv functions
Added functions to convert from/to ints/floats. Floats are converted
according to basic Steele&White algorithm (without speedup).
2025-04-05 13:56:00 +03:00
8009498331 Fix bug in setting precomputed power-of-10 2025-04-05 13:56:00 +03:00
fc3d4e22d5 Add initial bigint implementation for the fmt library 2025-04-05 13:56:00 +03:00