9
edits
Fixed-width strings: Difference between revisions
→Sample instance data
No edit summary |
|||
| Line 74: | Line 74: | ||
We need to discuss whether strings should be 0-terminated. Processing a string buffer with C/C++ routines would be much easier if it did. Also, in debug mode, the display of a string is easier. | We need to discuss whether strings should be 0-terminated. Processing a string buffer with C/C++ routines would be much easier if it did. Also, in debug mode, the display of a string is easier. | ||
''SRJ: IMHO we'd be better off not requiring null termination; this will give us more flexibility if in the future we want to have strings point at existing arbitrary data structures that contain string data. As far as C interop, this will only be easier with 8-bit strings, and when combined with the inherent lack of safety in the typical C string routines, I think we're better off avoiding ease of interop, lest someone decide that strcat() is a good way to add to a Tamarin string... debug modes can of course always provide wrappers that allow access to zero-terminated strings (TT has such a thing now).'' | |||
The listing uses bit fields, but I think that we should use masks and shifts to ensure the correct width and alignment of data, along with inline methods to access the fields. | The listing uses bit fields, but I think that we should use masks and shifts to ensure the correct width and alignment of data, along with inline methods to access the fields. | ||