If the source and replacement strings were the same length, the code was performing undefined pointer arithmetic involving a NULL pointer. I'm not aware of any observable effects of this in practice, but it's potentially problematic. It's detected by ubsan, for example when running `make check-python-test-suite`: DOH/string.c:839:4: runtime error: applying non-zero offset to non-null pointer 0x602000001558 produced null pointer SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior DOH/string.c:839:4 in |
||
|---|---|---|
| .. | ||
| CParse | ||
| DOH | ||
| Doxygen | ||
| Include | ||
| Modules | ||
| Preprocessor | ||
| Swig | ||
| Makefile.am | ||
| README | ||
SWIG Source directory
Source/DOH - A core set of basic datatypes including
strings, lists, hashes, and files. Used
extensively by the rest of SWIG.
Source/Swig - Swig core. Type-system, utility functions.
Source/Preprocessor - SWIG C Preprocessor
Source/CParse - SWIG C Parser (still messy)
Source/Modules - Language modules.
Source/Include - Include files.
Historic directories which may be in CVS, but have been removed:
Source/Modules1.1 - Old SWIG-1.1 modules. Empty.
Source/LParse - Experimental parser. Officially dead
as CParse is more capable.
Source/SWIG1.1 - Old SWIG1.1 core. Completely empty now.