Commit Graph

15860 Commits

Author SHA1 Message Date
pgkos
296dfae8af Fixes splitfile (#11918) [bugfix] 2019-08-15 15:22:14 +02:00
Dominik Picheta
b095203f28 Fixes regression introduced by #11904. (#11948) 2019-08-15 06:05:38 +02:00
Timothee Cour
630e18a6eb refs #11947; FieldError fixed (#11950) 2019-08-15 06:04:38 +02:00
Arne Döring
a25d43ca6b fix warnings in system.nim (#11944) 2019-08-14 17:48:11 +02:00
Arne Döring
65ed2d7fe8 fixes #11942 (#11945) 2019-08-14 17:41:44 +02:00
Clyybber
e6cb7e34ef Make name mangling reversable unambiguously (#11939) [bugfix] 2019-08-14 08:11:50 +02:00
cooldome
5ef62fcb25 newruntime for vcc (#11934) [bugfix] 2019-08-14 08:10:59 +02:00
narimiran
1874b3e786 [nobackport] add a changelog entry for the previous commit 2019-08-13 07:38:07 +02:00
Fabian Keller
83397a69fc Consistent behavior of exec; Improved docs around exec, staticExec, gorgeEx, etc. (#10967)
* improved docs around exec, staticExec, gorgeEx, etc.
* incorporate review comment; made behavior of exec consistent
2019-08-13 07:41:08 +02:00
Andreas Rumpf
433613e267 fixes #11844 (#11935) 2019-08-13 06:37:37 +02:00
Araq
322ce1872f fixes #10689 2019-08-12 14:06:48 +02:00
Araq
289b5e9ef9 fixes #11254 2019-08-12 13:58:35 +02:00
sschwarzer
c9c28bf85e Update documentation on xmltree.items/mitems (#11930)
* Update documentation on `xmltree.items`/`mitems`

So far the documentation on `items` and `mitems` wasn't explicit about whether the iteration recurses down the node's children or not. I assumed recursion, which was wrong.

* Improve wording in comment

Use the more common and shorter word "direct".
2019-08-12 11:12:22 +02:00
Andreas Rumpf
da64c8762f destructors: spec reflects reality, =sink is here to stay 2019-08-11 21:55:47 +02:00
Andreas Rumpf
212ae2f125 fixes #11891 2019-08-10 15:41:24 +02:00
Araq
0e4a8bfb28 fixes 'unused module' feature for 32 bit builds of Nim for good [nobackport] 2019-08-09 16:38:10 +02:00
Araq
da37177ac2 make tests green again on 32bit systems 2019-08-09 10:12:32 +02:00
Andreas Rumpf
3b8f721460 merged #11813 manually [bugfix] 2019-08-08 23:09:27 +02:00
Zachary Carter
082c440f13 Fixes effect error in pure/logging.nim (#11898)
* Fix effect error in pure/logging.nim

Currently calling `logging.log` will produce - `Error: can have an unlisted effect: RootEffect`. Add `RootEffect` to list of tags.

* the only tag necessary is RootEffect since all other Effects are descendants of RootEffect
2019-08-08 22:48:25 +02:00
Andreas Rumpf
81ddc67785 [refactoring] compiler: simplified markUsed 2019-08-08 22:43:58 +02:00
Andreas Rumpf
04708742e7 fixes #11809 2019-08-08 21:41:05 +02:00
Andreas Rumpf
e3c0bf48c5 tiny style change [refactoring] 2019-08-08 18:30:10 +02:00
Andreas Rumpf
0d5bd7e8b5 removed unused imports [refactoring] 2019-08-08 17:03:19 +02:00
Arne Döring
44e7a7b6c2 Lock semchecked ast for macros (#11883) [bugfix]
* reject to modify type checked AST
* add flag to back out
* Introduce legacy feature set.
2019-08-08 16:57:06 +02:00
Timothee Cour
fda51b6ca2 fix crash on sigmatch (#11913) 2019-08-08 14:41:23 +02:00
Dmitry Matveyev
756fabb979 Fix typo (#11910) 2019-08-08 13:17:53 +02:00
Arne Döring
48968f3c84 delete expensive call (#11909) [refactoring] 2019-08-08 09:16:07 +02:00
konradmb
addd7b5e20 Fix issue #10726 - HTTP response without Content-Length is not accessible (#11904)
* Add patch by @xenogenesi

* Async test for HTTP/1.1 without Content-Length

* Apply suggestions from code review

Co-Authored-By: Dominik Picheta <dominikpicheta@googlemail.com>
2019-08-08 08:41:56 +02:00
Andreas Rumpf
c8cffaf420 Incremental compilation (IC): Improvements (#11881)
* IC: C codegen is aware of IC
* manual: minor change to make VSCode's RST plugin render it properly
* IC: minor refactoring
* testament: code refactorings
* rodutils: removed dead code
* IC: always build the compiler with the IC feature
* IC: C codegen improvements
* IC: implement the undocumented -d:nimMustCache option for testing purposes
* IC: added first basic tests
* IC: extensive testing of the deserialization feature
* testament: refactoring; better IC tests
* IC: removes 'nimMustCache' flag; readonly does the same
* testament: minor refactoring
* update Nimble version
* testament: removed dead code and imports; IC: added simple test
* IC: progress
2019-08-08 08:41:05 +02:00
Andreas Rumpf
c0d240b8cd fixes #11807 (#11900)
* fixes #11807
* make tests green again
2019-08-07 22:40:58 +02:00
amzak
f34ae81971 Fixed handling of empty strings for encodings conversions (#11897) 2019-08-07 19:56:42 +02:00
awr1
9321b33cd2 Typeclass/Variant clarification in Manual (#11901)
* [Docs] clarified what type classes do in manual

* [Docs] fixed 'dynamism', added to object variant section
2019-08-07 16:03:32 +02:00
Arne Döring
afbcd1b330 int128 on firstOrd, lastOrd and lengthOrd (#11701)
* fixes #11847
2019-08-07 15:53:16 +02:00
Andy Davidoff
8407a57499 note set[T].len alias in future changelog (#11893)
per comments in #11885
2019-08-06 21:07:26 +02:00
amzak
3d7d5cf121 Fixed utf8<->utf16 conversions on windows (#11888) 2019-08-06 03:30:47 -04:00
Arne Döring
ce148e71ef disallow static in return type (#9686) [nobackport] 2019-08-06 01:26:53 +02:00
Andreas Rumpf
42e83ac24c fixes #11817 (#11889) 2019-08-05 16:48:26 +02:00
Andy Davidoff
d208d7a995 set[T].len is an alias for set[T].card (#11885) [feature] 2019-08-05 14:27:45 +02:00
Araq
61c66abbfe newruntime: make system.delete and friends work with owned ref (bugfix); reported on the forum 2019-08-05 00:14:30 +02:00
Araq
7024bb800c fixes #11844 2019-08-04 23:54:29 +02:00
John Paul Adrian Glaubitz
65d1140d11 Add build support for Linux/ia64 (#11880) [feature]
* compiler: Add ia64 as target architecture on Linux
* build.sh: Enable CPU detection for ia64
2019-08-04 23:16:30 +02:00
Arne Döring
b73d097d36 fixes #9829 (#11849) 2019-08-04 19:31:55 +02:00
Chris Heller
343ace8194 Skip db_postgres.setRow if getRow returns 0 rows (#11859)
Fixes nim-lang/Nim#11821
2019-08-04 19:28:24 +02:00
Leorize
51d998390c tgetaddrinfo: disable ICMP tests for Haiku
They're not supported by Haiku's getaddrinfo()
2019-08-04 12:15:17 +01:00
Leorize
df32c45757 tgetaddrinfo: use sizeof() instead of hardcoding
OS like Haiku have a bigger sockaddr_in struct for legacy(?) reasons.
Using sizeof() should account for the different struct sizes.
2019-08-04 12:15:17 +01:00
Oscar Nihlgård
bcfb540e57 Allow typeof(nil) as generic parameter (#11869) 2019-08-03 10:16:07 +02:00
Oscar Nihlgård
50e921bb94 VM exception fixes (#11868) 2019-08-02 22:59:04 +01:00
Jjp137
a906b3952b random: minor documentation updates (#11831)
Add docs for the sample procs that take in sets, and clean up a
few other minor things.
2019-08-02 02:42:18 -04:00
Oscar Nihlgård
829f719670 Fix VM conversion to var type [bugfix] (#11866) 2019-08-01 17:44:12 +02:00
Arne Döring
803406d07c fix #11854 (#11857) 2019-07-30 17:24:42 +02:00