| Summary: | Additional formats for Decoding Table | ||
|---|---|---|---|
| Product: | [Applications] okteta | Reporter: | Pavol Rusnak <stick> |
| Component: | general | Assignee: | Friedrich W. H. Kossebau <kossebau> |
| Status: | CONFIRMED --- | ||
| Severity: | wishlist | CC: | lllusion3418 |
| Priority: | NOR | ||
| Version First Reported In: | unspecified | ||
| Target Milestone: | --- | ||
| Platform: | openSUSE | ||
| OS: | Linux | ||
| Latest Commit: | Version Fixed/Implemented In: | ||
| Sentry Crash Report: | |||
|
Description
Pavol Rusnak
2009-02-12 11:01:00 UTC
Thanks for filing your wishes :) Good ideas, just would need some pointers to the format or, even better, some example code for decoding (and encoding, soon to come). With more entries the decoding table tool will need a config dialog to restrict the shown entries, best by user-definable sets, but that should be straightforward. It might make sense to create separate disassembly window and not integrate it with Decoding Table. For Assembler opcodes there are a few disassembler libraries: * libdisasm - http://bastard.sourceforge.net/libdisasm.html - last release January 2008 - seems to support only x86 * udis86 - http://udis86.sourceforge.net/ - last release June 2008 - supports both x86 and x86_64 * distorm64 - http://www.ragestorm.net/distorm/ - last release September 2008 - seems to have the most complete support (up to SSE4, VMX, ...) --------------------- Following datatypes could be later reused in Template editing mentioned in bug#184123. MS-DOS date & time (32-bit ) bits : meaning 0 - 4 : seconds/2 (0-29) 5 - 10 : minutes (0-59) 1 - 15 : hours (0-23) 16 - 20 : day (1-31) 21 - 24 : month (1-12) 25 - 31 : year-1980 Win32 FILETIME (64-bit) - integer number of 100 nanosecond intervals since 1.1.1601 (or milliseconds*10) OLE 2.0 date & time (64-bit) - real number (double precision) - integer part - number of days since 30.12.1899 - fraction part - time (.0 is 0:00, 0.25 is 6:00, 0.5 is 12:00, 0.75 is 18:00) * ANSI SQL date & time (64-bit) - two consecutive 32-bit integers - first: number of days since 17.11.1858 - number of 100 millisecond intervals since midnight * UNIX/C/FORTRAN date & time, secs (32-bit) - integer - number of seconds since 1.1.1970 * UNIX/C/FORTRAN date & time, mins (32-bit) - integer - number of seconds since 1.1.1970 * Macintosh HFS+ Date & Time (32-bit) - integer - number of seconds since 1.1.1904 * Java date & time (64-bit) - integer - number of milliseconds since 1.1.1970 Thanks for the pointers. Will not make it for KDE 4.3, but perhaps 4.4. Patches are welcome and will improve speed of implementation, as always :) For the disassembler, yes, that should be a separate tool/view. I hope to get to that one day, as I am interested to have that myself. It might also be nice to support integers other than {8,16,32,64}-bit.
One format I'm working with uses one byte to specify the "size" of the integer following this byte.
This can result in e.g 3-byte (24-bit) integers like this: 03 AB CD EF would be 0xEFCDAB.
I'm not asking for support for this specific binary format, but some way to decode these.
One way I'm thinking of would be to go by mouse/keyboard selection.
I.e. you would select AB CD EF in the file view and get the value in the decoding table either in a new row or by restricting the existing rows to the selection.
|