HxD Hex Editor 2.5

What's new in 2.5

Hex editor: implement horizontal mouse wheel scrolling (including handling of Logitech driver bug)
Data inspector:
- New: added LEB128/ULEB128 types (variable width integer types used in WASM and dwarf debug information)
- New: introduce readonly type converters (to allow uniform error messages when attempting to change values, also for converter plugins)
- Enhancement: better error handling so that typical errors are caught and a more specific error message is given
- error message for plugin DLLs of wrong machine type (x86-32 vs. x86-64)
- Exporters:
- New: skip undefined or unreadable sections, such that Intel Hex and S-Record files can be created that have gaps
- New: source code exporter option to set the maximum text column or bytes per lines (= array elements per line)
- New: SRecord and IntelHex exporters can now have maximum of 250 and 255 bytes per line, respectively
- therefore there is also a new default block size/bytes per line setting for each of the exporters, which have a more common size of 32 (for both)
- allows user to select very large block sizes, yet default to a commonly used size
- Importers:
- New: Intel Hex and S-Record support undefined regions of arbitrary size, by creating undefined foldable regions, instead of creating zero filled blocks for them
- Enhancement: rewrite logic to allow for data records that are out of order (addresses aren’t strictly increasing), or overlapping, even when there are gaps of undefined data; previously this could lead to errors, or would only function without undefined data gaps
- Enhancement: Intel Hex and S-Record importers ignore leading and trailing whitespace in a line
- Checksums / digests:
- New: custom checksum
- can generate a checksum, with settable bitwidth for the checksum result, the checksum addends, and the endianness
- useful for various exotic formats, such as ROM files
- Search window:
Enhancement: added support for (U)Int24
use same code as datainspector to convert from string to integer
improved the naming of floating point and integer types (closer to data inspector)
searches for both signed and unsigned integers, choosing the right type depending on which fits best (negative numbers => negative type, positive numbers => unsigned type; this is ok, because the positive numbers in the signed type have the same encoding as in the unsigned type; this is a feature of two’s complement encoding of integers)
- Tool windows: …

Full release notes »