 svenbarth
|
66fa732720
Merged revision(s) 47634, 47655 from trunk:
|
4 лет назад |
 marco
|
775567e8f7
# revisions: 44598,45635,45757,45764,45772
|
5 лет назад |
|
marco
|
fdb477df1e
# revisions: 43566,43567,43568,43586,43629,43823
|
5 лет назад |
 Jonas Maebe
|
f29598384b
* Objective-Pascal inferred result type and improved category method searching
|
5 лет назад |
 pierre
|
663c97ed23
Merge of revisions 41425 from trunk to fixes_3_2
|
6 лет назад |
|
marco
|
2b01261f36
--- Recording mergeinfo for merge of r39673 into '.':
|
6 лет назад |
|
marco
|
ee7090b118
--- Merging r39711 into '.':
|
6 лет назад |
|
marco
|
edf32cd5dc
The Important Ones:
|
6 лет назад |
 florian
|
68eb921d46
* broken compilation after r39347 fixed
|
7 лет назад |
|
svenbarth
|
cc153176f3
+ add a new symtable option sto_has_non_trivial_init that is true if the symtable contains a symbol with an
|
7 лет назад |
|
svenbarth
|
80ed66f349
* move calculation of sto_needs_init_final to a separate method do_init_final_check()
|
7 лет назад |
|
Jonas Maebe
|
cb7730a423
* fixed execution order of implicit finalization and class destructors: first
|
7 лет назад |
|
svenbarth
|
b322339758
+ extend tstoredsymtable with method register_children to register all its symbols and defs
|
7 лет назад |
|
svenbarth
|
83e4585b0f
* correctly handle unit identifiers inside specializations of generic routines
|
8 лет назад |
|
svenbarth
|
236a9b0aa5
* rework the checks for a helper's extended def to be a class so that it can be more easily extended by other objecttypes
|
8 лет назад |
 maciej-izak
|
751bde9792
Revert revision 36436 from trunk. Final fix for issue #31675.
|
8 лет назад |
|
maciej-izak
|
d3d29fa84c
Revert revisions 36133, 36128 from trunk.
|
8 лет назад |
|
maciej-izak
|
631d4a2d53
* Move global variable in_preproc_comp_expr into tscannerfile class.
|
8 лет назад |
|
maciej-izak
|
6d1ad52dc5
* for symbols used in preprocessor expressions, we don't want to increase references count (for smaller final binaries). Partial fix for problem presented in mantis 31675. Preprocessor functions like declared() in uses section will work only for symbols declared in implicytly included modules (for example like for heaptrc by option -gh, see example/test tw31675.pp)
|
8 лет назад |
|
florian
|
0897bd154a
+ align(<int64>,<int64>) and align(<qword>,<qword>), resolves an issue reported on the mailing list with large records
|
8 лет назад |
|
svenbarth
|
7dc4f16db3
* fix for Mantis #31431: allow specializations to allow private/protected variables that their generic could have accessed
|
8 лет назад |
|
maciej-izak
|
25db29d0a6
* symtable.pas:
|
8 лет назад |
|
maciej-izak
|
15ba9b54c6
Store informations about management operators in ppu files for records.
|
8 лет назад |
|
maciej-izak
|
48b149acd4
New tokens for new management operators: Initialize, Finalize, AddRef and Copy. Small code adjustment for parser (pdecobj.pas, pstatmnt.pas, psub.pas, ptype.pas) was required because few of old token has from now value which is outside range of set: _CONSTRUCTOR, _FINALIZATION.
|
8 лет назад |
|
svenbarth
|
969e19ba1a
* initialize module id of a symtable always to the id of the current module (at least if there is a current module, which is not the case for the initial macro symtable)
|
8 лет назад |
|
svenbarth
|
d9ea6aae4d
* fix for Mantis #30761: always return the symbol found in the helper instead of doing this dependant on the presence of the overload attribute; for this the collection of all suitable overloads is done in tcallcandidates instead.
|
8 лет назад |
|
svenbarth
|
d4e573c1e3
* fix for Mantis #27349: nested classes might also reside inside records
|
8 лет назад |
|
svenbarth
|
ce18b10c81
* fix for Mantis #30530: when the global/static symtable is reached defowner is Nil, so respect that in the loop condition
|
8 лет назад |
|
svenbarth
|
cb27889dea
* fix cycling by correctly handling the case that the global symtable of a program file is Nil
|
9 лет назад |
|
svenbarth
|
36e5d02e8e
* addsymref: do not directly check the top level symtable against the current module's global symtable as that would be false for a specialize symtable; check the module id instead
|
9 лет назад |
