Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | resolved conflicts with master | Araq | 2014-01-18 | 1 | -0/+1 |
|\ | |||||
| * | Added --noBabelPath override. | Dominik Picheta | 2013-12-25 | 1 | -0/+1 |
| | | |||||
* | | clean-up some obsolete code; close #602 | Zahary Karadjov | 2013-12-31 | 1 | -2/+0 |
| | | |||||
* | | case consistency: cs:partial bootstraps on windows | Araq | 2013-12-29 | 1 | -4/+4 |
| | | |||||
* | | case consistency: next steps | Araq | 2013-12-29 | 1 | -6/+6 |
| | | |||||
* | | case consistency part 4 | Araq | 2013-12-27 | 1 | -7/+7 |
| | | |||||
* | | case consistency part 1 | Araq | 2013-12-27 | 1 | -1/+1 |
|/ | |||||
* | Added jsondoc compiler switch | Erik O'Leary | 2013-12-12 | 1 | -0/+1 |
| | | | | Outputs top-level exported information in JSON | ||||
* | bugfix: package names should not contain '.' | Araq | 2013-10-07 | 1 | -2/+4 |
| | |||||
* | the compiler can now deal with multiple modules of the same name | Araq | 2013-09-26 | 1 | -4/+9 |
| | |||||
* | JS codegen: supports more builtins | Araq | 2013-09-24 | 1 | -10/+19 |
| | |||||
* | support for multiple modules of the same name; niminst supports 'platforms'; ↵ | Araq | 2013-09-24 | 1 | -1/+13 |
| | | | | minor bugfixes | ||||
* | -d:nocaas mode for easier bootstrapping on exotic OSes (Haiku) | Araq | 2013-09-01 | 1 | -0/+1 |
| | |||||
* | pass-through of static int generic params to arrays when late instantiation ↵ | Zahary Karadjov | 2013-08-23 | 1 | -1/+1 |
| | | | | is disabled | ||||
* | Experimental support for delayed instantiation of generics | Zahary Karadjov | 2013-08-23 | 1 | -0/+2 |
| | | | | | | | | | | | | | | | | | | | | This postpones the semantic pass over the generic's body until the generic is instantiated. There are several pros and cons for this method and the capabilities that it enables may still be possible in the old framework if we teach it a few new trick. Such an attempt will follow in the next commits. pros: 1) It allows macros to be expanded during generic instantiation that will provide the body of the generic. See ``tmacrogenerics``. 2) The instantiation code is dramatically simplified. Dealing with unknown types in the generic's body pre-pass requires a lot of hacky code and error silencing in semTypeNode. See ``tgenericshardcases``. cons: 1) There is a performance penalty of roughly 5% when bootstrapping. 2) Certain errors that used to be detected in the previous pre-pass won't be detected with the new scheme until instantiation. | ||||
* | new scoping rules for 'if' now active | Araq | 2013-06-04 | 1 | -1/+1 |
| | |||||
* | better error message for subtle nil errors | Araq | 2013-05-31 | 1 | -1/+1 |
| | |||||
* | Merge branch 'master' into newparser | Araq | 2013-05-07 | 1 | -0/+6 |
|\ | |||||
| * | support suggest after compile in caas mode | Zahary Karadjov | 2013-05-05 | 1 | -0/+6 |
| | | |||||
* | | Merge branch 'master' into newparser | Araq | 2013-05-04 | 1 | -2/+2 |
|\| | |||||
| * | nimrod dump can now produce a machine readable json report | Zahary Karadjov | 2013-05-01 | 1 | -2/+2 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | The data in the report includes necessary information for starting the compiler service and setting up the project paths in the IDE. the default verbosity of 1 is now set in the compiler code to fix an issue with verbosity being temporary set to 1 during config parsing even when it's explicitly overridden on the command-line. compiler/lexbase was temporary renamed to nimlexbase as a work-around for a codegen naming conflict with lib/pure/lexbase resulting in linking errors (further investigation needed). | ||||
* | | revert new scope for 'if' | Araq | 2013-05-02 | 1 | -1/+2 |
|/ | |||||
* | experimental support for outputting code snippets in error messages; ↵ | Zahary Karadjov | 2013-04-09 | 1 | -0/+6 |
| | | | | | | | implements #301 see #301 for description. currently, the feature is activated by setting the verbosity level to 2 or more | ||||
* | bugfix: fix linking when symbol files are used | Zahary Karadjov | 2013-04-08 | 1 | -0/+3 |
| | |||||
* | implemented --dynlibOverride option for static linking of 'dynlib' | Araq | 2013-04-08 | 1 | -0/+15 |
| | |||||
* | Removes executable bit for text files. | Grzegorz Adam Hankiewicz | 2013-03-16 | 1 | -0/+0 |
| | |||||
* | added isServing parameter | Simon Hafner | 2013-02-26 | 1 | -0/+1 |
| | |||||
* | Merge branch 'master' of github.com:Araq/Nimrod | Araq | 2013-02-16 | 1 | -1/+1 |
|\ | |||||
| * | EcmaScript => JS. Fixes #330 | Simon Hafner | 2013-02-15 | 1 | -1/+1 |
| | | | | | | | | No one calls it EcmaScript anymore. | ||||
* | | preparations for a generational GC | Araq | 2013-02-16 | 1 | -2/+2 |
|/ | |||||
* | first version of a simple mark&sweep GC; activate with --gc:markAndSweep | Araq | 2013-02-07 | 1 | -3/+2 |
| | |||||
* | cleaner GC switching | Araq | 2013-01-31 | 1 | -4/+8 |
| | |||||
* | the `mainmodule` option has effect only when no explicit module is given | Zahary Karadjov | 2013-01-28 | 1 | -1/+2 |
| | |||||
* | merged upstream master | Zahary Karadjov | 2013-01-27 | 1 | -8/+38 |
|\ | |||||
| * | implements #295 | Araq | 2013-01-08 | 1 | -1/+3 |
| | | |||||
| * | implemented --listFullPaths option for aporia | Araq | 2012-12-27 | 1 | -0/+1 |
| | | |||||
| * | first steps for FFI support at compile time | Araq | 2012-12-19 | 1 | -0/+12 |
| | | |||||
| * | lazy paths for Babel support | Araq | 2012-12-11 | 1 | -7/+22 |
| | | |||||
* | | adds an option to interleave the generated code with snippets from the ↵ | Zahary Karadjov | 2012-12-02 | 1 | -1/+4 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | original source Lines from the original source are outputted as comments next to line directives. Hopefully, this will make debugging codegen problems easier. Other changes: The frame setup code now uses a single-line C macro. My motivation was to reduce the noise in the generated output and make it easier to step over the boiler-plate code, but counter-intuitively this also improved the overall compilation speed a little bit so I applied the same treatment to line tracking too (this reduces the size of the generated files and the explanation is that probably the I/O overhead dominates the macro expansion costs). | ||||
* | | track the "owner" heap object in the ref write barrier | Zahary Karadjov | 2012-12-01 | 1 | -4/+4 |
| | | | | | | | | See the papers for reference counting with heap sliding views for details: | ||||
* | | CaaS in-memory caching | Zahary Karadjov | 2012-11-28 | 1 | -0/+3 |
| | | | | | | | | | | removed some redundant filepath params and variables and switched to canonical paths in most places | ||||
* | | caas is now drivable through stdin | Zahary Karadjov | 2012-11-28 | 1 | -1/+1 |
|/ | | | | | * added idetools --eval * streams.readLine recognises and applies the backspace character | ||||
* | doc2 improvements; small lexer bugfix: backslashes in comments | Araq | 2012-11-20 | 1 | -0/+2 |
| | |||||
* | next steps for exception tracking | Araq | 2012-11-03 | 1 | -1/+1 |
| | |||||
* | bugfix: exception tracking (still disabled) | Araq | 2012-11-01 | 1 | -0/+1 |
| | |||||
* | syntax compatibility between do blocks and stmt blocks | Zahary Karadjov | 2012-10-04 | 1 | -2/+4 |
| | | | | | | | | | | | | See the section `do notation` in the manual for more info. * nkMacroStmt has been removed Macro statements are now mapped to regular nkCall nodes. The support for additional clauses (such as else, except, of, etc) have been restored - they will now appear as additional arguments for the nkCall node (as nkElse, nkExcept, etc nodes) * fixed some regressions in the `is` operator and semCompiles | ||||
* | next steps towards term rewriting macros; simple examples work | Araq | 2012-08-30 | 1 | -2/+4 |
| | |||||
* | made compiler more robust for idetools; implemented idetools.usages | Araq | 2012-07-30 | 1 | -0/+1 |
| | |||||
* | implements #173 | Araq | 2012-07-28 | 1 | -1/+3 |
| | |||||
* | extracted documentation generator | Araq | 2012-05-09 | 1 | -2/+2 |
| |