| |
2566 |
2566 |
the code generator usable for submodules as well |
| |
2567 |
2567 |
|
| |
2568 |
2568 |
- added flag ?-type ...? to "info lookup slots" |
| |
2569 |
2569 |
- made all useful converters external symbols |
| |
2570 |
2570 |
- added flag ?-type ...? to "info slots" |
| |
2571 |
2571 |
- delete accessor when slot is destroyed |
| |
2572 |
2572 |
- added pattern to "info slots" |
| |
2573 |
2573 |
- added to "info slot /attName/" |
| |
2574 |
2574 |
|
| |
2575 |
2575 |
- Fixed dispatch of defaultmethod for ensemble methods |
| |
2576 |
2576 |
- Added compile flag DISPATCH_ALWAYS_DEFINED_METHODS (deactivated). |
| |
2577 |
2577 |
So far, nx just uses dispatch on overloads or filters, but not on |
| |
2578 |
2578 |
defines (possible to call e.g. "destroy" from a script, but |
| |
2579 |
2579 |
internally the direct dispatch is used, as long there is no |
| |
2580 |
2580 |
overload). The compileflag would force to use the slower |
| |
2581 |
2581 |
dispatch always. |
| |
2582 |
2582 |
- Extended regression test |
| |
2583 |
2583 |
- Improve locality |
| |
2584 |
2584 |
|
| |
2585 |
2585 |
- Let "info slot" return the slot object |
| |
|
2586 |
- nx::mongo: Initial commit of the experimental mongoDB interface for nx |
| |
2586 |
2587 |
|
| |
2587 |
2588 |
TODO: |
| |
2588 |
2589 |
- how to delete attributes? |
| |
2589 |
2590 |
|
| |
2590 |
2591 |
- Higher binary compatibility for future versions: |
| |
2591 |
2592 |
* It is not nice to have the full Nsf_Param structure in nsf.h |
| |
2592 |
2593 |
(required for Nsf_methodDefinition in the c code generator) |
| |
2593 |
2594 |
* It is not nice to have the full ParseContext structure in nsfmongo |
| |
2594 |
2595 |
(required for the allocation of the parse context in the stubs) |
| |
2595 |
2596 |
Adding fields to these structures would kill alder binaries |
| |
2596 |
2597 |
- not all converters have already external symbols |
| |
2597 |
2598 |
|
| |
2598 |
2599 |
- when alloc|dealloc are loaded via require, we have |
| |
2599 |
2600 |
no redefined-protection on those. Since the script does not know, |
| |
2600 |
2601 |
on which class|object these are defined one cannot make assumption |
| |
2601 |
2602 |
on these. Problem? |
| |
2602 |
2603 |
- The structure of the script body requires currently that |
| |
2603 |
2604 |
"class|object ?-per-object?" is inserted at the 1nd pos. |
| |
2604 |
2605 |
This is not sufficient, if we would like to test in this |
| |
2605 |
2606 |
script, whether the first arg is e.g. a class. |