v4.1.0

4.1.0 (2026-10-31) - The Halloween '26 edition 🎃

Features

• support the 'L' length modifier for floating point values (332524f)

Bug Fixes

• %g according to C11, '#' flag for floating point values (4fe92fd), closes #48 #55 #75 #111 #121 #122 #124
• avoid signed integer overflow, handle negative '*' precision (ecd5a1a), closes #118
• compiler warnings reported by GCC (83f01fb), closes #102
• exact decimal exponent for %e, zero, subnormals, 32-bit double (36f6079), closes #54 #94 #98 #119 #120
• honour the precision of integers with the '-' flag (b74de50), closes #49 #113
• NULL pointer for %s, %.0s string length, %c with '\0' (fb4c361), closes #63 #47
• rounding carry at a decimal tie, whole part on 16-bit platforms (abb4726), closes #51 #108
• rules of the '#' flag for integers (5bfef75), closes #50 #53 #99 #109 #110 #114

Refactoring

• fctprintf() delegates to vfctprintf() (a3de51c)

Build System

• add CMake build with CTest integration (38ec9dc), closes #106
• make: define CPPFILT for the demangled symbol list (d3f8d2e)

Documentation

• readme: update for the fixes, the test suite and CMake (f9f57d1)