summary refs log tree commit diff stats
path: root/compiler/semmagic.nim
Commit message (Collapse)AuthorAgeFilesLines
* int128 on firstOrd, lastOrd and lengthOrd (#11701)Arne Döring2019-08-071-2/+4
| | | | * fixes #11847
* revert changesAndrii Riabushenko2019-07-251-2/+2
|
* fixes #11826Andrii Riabushenko2019-07-251-2/+2
|
* Offsetof fixes (#11690)Arne Döring2019-07-091-44/+3
| | | | | | | | * first fixes * more tests and fixes * code normalization
* fixes #11610 (#11659)Andreas Rumpf2019-07-051-0/+2
|
* fixes #7937 (#11244)Andreas Rumpf2019-05-141-3/+4
|
* added system.unown to make 'owned' sane to use in practice; later on we ↵Araq2019-04-291-0/+35
| | | | might change the type inference rules to make it more convenient
* more destructor based changes (#10885)Andreas Rumpf2019-03-231-0/+5
| | | | | | | | | | | * mark user defined destructors with sfOverriden to simplify the logic * refactoring in preparation to merge liftings and generic instantiations for destructors * ast: introduce nkHiddenTryStmt for destructor generation in order to be able to optimize better the code later on * renamed 'patterns' switch to 'trmacros' as it was totally misleading before * destructors: introduce tfCheckedForDestructor flag in preparation of strict =destroy checking * test for invalid/too late destructor introductions * liftdestructors: make code robust for nimsuggest * --newruntime works for hello world again * newruntime: code generation for closures
* fixes #10807 (#10814)cooldome2019-03-121-5/+5
| | | | | * fixes #10807 * use nkAddr instead of nkHiddenAddr
* 32 bit fixes (#10608)Arne Döring2019-02-131-3/+11
|
* Stop useless suggestion of unsafeAddr (#10598)LemonBoy2019-02-081-1/+6
| | | Fixes #10594
* Finalizer proc must be global (#10388)LemonBoy2019-01-221-0/+5
| | | Fixes #10376
* fix typetraits.`$` regression https://github.com/c-blake/cligen/issues/84 ↵Timothee Cour2018-12-301-2/+3
| | | | | | | (#10131) * fix typetraits.`$` regression https://github.com/c-blake/cligen/issues/84 * add test
* revives: Move typetraits.`$` to system. Fixes #5827 (#10071)Timothee Cour2018-12-301-1/+1
| | | | | | * Move typetraits.`$` to system. Fixes #5827. * revive PR; adjust code to make sure everything works and add tests * fix tests/concepts/tstackconcept.nim * address comments
* fixes #9868Araq2018-12-111-2/+0
|
* fixes #7365alaviss2018-11-261-1/+1
|
* Fixes multiple bugs with sink arguments (#9802)cooldome2018-11-261-1/+1
| | | | | | * fixes #9781 * fix spacing
* Of operator in vm fixes [backport] (#9717)cooldome2018-11-151-1/+7
| | | | | | * fixes #9701 * fixes #9702 * optimize of statement to bool
* fixes #9610Andreas Rumpf2018-11-031-17/+13
|
* fixes #9498, typeof is for everybodyAraq2018-10-251-3/+9
|
* implement sizeof and alignof operator (manually squashed #5664) (#9356)Timothee Cour2018-10-141-2/+55
|
* fixes #7827, bindSym enhancement (#8499)andri lim2018-08-021-1/+66
| | | | | | | | | | | | | | | | * bindSym power up, working prototype * update bindSym doc * add bindSym test * fix some typo * fix bindSym doc * get rid of specialops field from vm * add experimental: dynamicBindSym
* Do not consider enums with holes as ordinals (#8264)LemonBoy2018-07-121-1/+1
| | | | | | Make the compiler behave consistently with respect to what's written in the manual. Fixes #1239
* refactoring: remove idents.legacy global variable and pass the IdentCache ↵Andreas Rumpf2018-05-271-2/+2
| | | | around explicitly
* refactoring: fewer global variablesAndreas Rumpf2018-05-271-1/+1
|
* remove more global variables in the Nim compilerAndreas Rumpf2018-05-271-1/+1
|
* platform.nim doesn't use globals anymore; prepare msgs.nim for not using globalsAndreas Rumpf2018-05-181-5/+1
|
* preparations of making compiler/msgs.nim free of global variablesAndreas Rumpf2018-05-171-1/+1
|
* remove ast.emptyNode global; cleanup configuration.nimAraq2018-05-161-1/+1
|
* fixes testament compilationAraq2018-05-141-0/+4
|\
| * A motivating example for the new `bindSym` behavior.Zahary Karadjov2018-05-071-1/+1
| | | | | | | | | | | | | | | | The example is a proof-of-concept logging library, allowing you to define lexically-scoped environments where certain logging attributes are applied automatically to all logging statements. fixes tmacro1 (use of `bindSym` inside static blocks)
| * Better support for treating templates and macros as symbols.Zahary Karadjov2018-05-071-0/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This allows you to pass a template or a macro to another macro which can then inspect the implementation of the former template/macro using `getImpl`. Since templates can be freely redefined, this allows you to treat their symbols as compile-time variables that have lexical scope. A motivating PoC example for a logging library taking advantage of this will be provided in the next commit. Implementation details: * The name of a template or a macro will be consider a symbol if the template/macro requires parameters * For parameterless templates/macros, you can use `bindSym`, which was extended to also work outside of compile-time procs.
* | sem pass compiles againAndreas Rumpf2018-05-121-21/+21
| |
* | more modules compile againAndreas Rumpf2018-05-121-14/+14
|/
* introduce nkTupleConstr AST node for unary tuple construction; breaking changeAndreas Rumpf2018-04-131-1/+1
|
* Add column number to instantiation info (#7376)PMunch2018-04-121-0/+3
| | | | | | | | | | | | | | * Add column number to instantiation info Instantiation info left out column number for no good reason. This adds it in as the third element of the tuple. * Fix test that failed and added change to changelog An assertion test failed because it was declaring a type that was expected to be populated by instantiationInfo with the old signature. Also added the changes to the changelog as it is a breaking change.
* preparations for language extensions: 'sink' and 'lent' typesAndreas Rumpf2018-01-071-2/+2
|
* make tests green againAndreas Rumpf2017-11-181-1/+1
|
* wipAndreas Rumpf2017-11-021-1/+3
|
* remove old implementation of the roof operator; make tests green again; ↵Andreas Rumpf2017-10-291-31/+1
| | | | close #6292
* destructors: work in progressAraq2017-10-261-3/+8
|
* added typetraits.supportsCopyMemAndreas Rumpf2017-10-171-1/+6
|
* low/high/of are now overloadable operationsAndreas Rumpf2017-08-301-0/+36
|
* introduce a pre-processing pass for the concept bodiesZahary Karadjov2017-06-201-2/+2
| | | | | | | fixes #4982 fixes #3805 close #3414
* Allow tyOr,tyAnd and tyNot to be constructed in more contextsZahary Karadjov2017-04-281-16/+32
|
* use lower-case proc namesZahary Karadjov2017-03-241-4/+4
|
* new type traits: `GenericHead` and `StripGenericParams`Zahary Karadjov2017-03-241-2/+35
|
* implement the special treatment of explicit type params in conceptsZahary Karadjov2017-03-241-1/+1
|
* big compiler refactoring; avoid globals for multi method dispatcher generationAndreas Rumpf2017-02-221-1/+1
|
* recursive modules are only detected to improve error messagesAndreas Rumpf2016-11-241-1/+1
|