suzanne.soy/racket
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
38,543 Commits 17 Branches 54 Tags 278 MiB
a4eddbd63d
Commit Graph

38543 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Jay McCarthy
9b1c71a381 Catch all errors so credentials are tried 2017-10-27 16:20:11 -04:00
Robby Findler
259e3a6fe1 use the right identifier 
closes #1845
 2017-10-26 16:48:12 -05:00
Matthew Flatt
5f1b707881 unbreak no-futures/no-places build 2017-10-26 07:12:24 -06:00
Matthew Flatt
dfa5d48092 unbreak JIT and check limits 
Repairs a problem with ce9894c8bf, where a large "inlined" vector
allocation is not actually inlined, but other parts of the JIT
assume that it will behave as inlined --- which implies that the
runstack will be left unchanged after the call.

Closes #1868
 2017-10-25 17:48:31 -06:00
Matthew Flatt
55c1685526 JIT: inline vector and struct CAS primitives 2017-10-25 17:36:03 -06:00
Matthew Flatt
c52bd91c54 add unsafe-struct*-cas! 2017-10-25 17:36:03 -06:00
Daniel Mendler
bc26d29bf8 add vector-cas! and unsafe-vector*-cas! 2017-10-25 17:36:03 -06:00
Matthew Flatt
d8e2192145 Revert "Revert "JIT: fix inline allocation for large vectors"" 
This reverts commit f7b3d3f4b4.
 2017-10-25 17:34:54 -06:00
Matthew Flatt
f7b3d3f4b4 Revert "JIT: fix inline allocation for large vectors" 
This reverts commit ce9894c8bf.

I don't immediately see what has gone wrong, so reverting for now.
 2017-10-25 17:18:32 -06:00
Matthew Flatt
ce9894c8bf JIT: fix inline allocation for large vectors 
Cloaes #1800
 2017-10-25 09:55:41 -07:00
Matthew Flatt
22e8af32f3 Windows: avoid loop on reparse-point cycles 
In Windows 10 Fall Creators Update, a OneDrive mount looks to Racket like
a reparse point that refers to itself. Make Racket stop trying to resolve
the path further in that case.

Relevant to #1671
 2017-10-24 08:26:26 -06:00
Matthew Flatt
ae4e85a6a7 evt:struct with unsafe-poller: skip sleep if poll reports ready 
Intended to fix racket/gui#80
 2017-10-23 06:03:25 -07:00
Rommel Martinez
81d51789ad Fix typo 2017-10-22 01:48:21 -04:00
Matthew Flatt
e03ec08736 history notes for 6.11 2017-10-20 14:29:46 -06:00
Jay McCarthy
2b9b1b22f2 raco test --deps docs 2017-10-20 11:34:40 -04:00
Jay McCarthy
5662ed8e99 Adding --deps 2017-10-20 11:34:40 -04:00
Jay McCarthy
62b17db270 Adding docs for new proxy settings 2017-10-20 11:34:40 -04:00
Jay McCarthy
06e3205787 Fix PR1859 2017-10-20 11:34:40 -04:00
Matthew Flatt
e255c737b5 rktio: fix bug in 'text mode on Windows 
If a #\return lined up in just the wrong way with a file-read
request, then the #\return could get mishandled.
 2017-10-19 19:55:38 -06:00
Matthew Flatt
b6b9d0d409 repair unmarshling fix 
Fixes a problem with 78fdee6b6b
 2017-10-19 14:52:26 -06:00
Matthew Flatt
78fdee6b6b fix unmarshaling of module-namespace information 
Avoids errors that sometimes appear as "cannot redefine lifted.0".

Merge to 6.11.
 2017-10-19 07:45:51 -06:00
Andrew Kent
9e16d3f9c9 add #:result clause to for/fold forms 2017-10-18 07:15:15 -04:00
Andrew Kent
3119c5b732 better stx errs for/fold and for*/fold 2017-10-18 07:15:15 -04:00
Alexis King
1e38918aa9 Improve how splicing-syntax-parameterize interacts with module* forms 
Commit 00d438cfbe made an attempt at this,
but this commit does it in a much more careful way, based on manually
emulating how the macroexpander expands module* forms in order to allow
splicing-syntax-parameterize to apply even within #%module-begin forms
introduced by the expander.
 2017-10-17 14:57:09 -07:00
Matthew Flatt
dd585e7ff7 fix environment -> resolved path -> module path index 2017-10-17 08:23:02 -06:00
Leif Andersen
b940ee8440 tag-id -> tag-expr 
This is because the optional tag argument can be any expression,
as pointed out by Matthew.
 2017-10-16 16:31:38 -04:00
Leif Andersen
2a8bece8de Add #:tag to the docs for define-cpointer-type 2017-10-16 13:07:07 -04:00
Matthew Flatt
d64961cdba improve an error message from the expander 
Mostly, fix "compiled" versus "expanded" in an error message,
but this variant of the message copied from the new expander
implementation seems better overall.
 2017-10-16 10:00:15 -06:00
Matthew Flatt
2454fd6931 Fix hashing on lists containing self-references 
The hashing function was bounded, but with an accidental backtrack
so it could take time exponential in the intended bound.

Closes #1842
 2017-10-16 09:27:50 -06:00
Robby Findler
f24273d7ef bring down below 102 columns 2017-10-15 21:25:46 -05:00
Robby Findler
4354488453 add some more popular keys, based on TR's contract generation 2017-10-15 17:50:05 -05:00
Alexis King
00d438cfbe Make splicing-syntax-parameterize recur into module* without a mod-path 2017-10-15 15:53:59 -06:00
Matthew Flatt
0f425f2029 fix "subprocess.rktl" test to run in quiet mode 2017-10-12 17:45:04 -06:00
Matthew Flatt
402bf2173d subprocess: fix problems with adding subprocess to a group 2017-10-12 14:40:23 -06:00
Matthew Flatt
1cc55f30fe subprocess: support adding to an existing subprocess group 2017-10-12 12:37:21 -06:00
Matthew Flatt
9f5e9c8db3 fix documented contract for {absolute,relative,complete}-path? 
They accept any string, instead of just path strings.
 2017-10-12 12:37:20 -06:00
Matthew Flatt
afc5f919f1 ffi/unsafe/schedule: expose some scheduler internals 
These internals are needed for `racket/gui`, which currently
accesses them via C functions.
 2017-10-12 12:37:15 -06:00
Vincent St-Amour
fadf0d4a7d Post-release version for the v6.11 release 2017-10-12 13:15:45 -05:00
Justin Slepak
3935824922 Fix bug in hash-table key-value match pattern (#1532) 
Match pattern allowed `(hash-table)' to match on non-empty hashes
 2017-10-12 01:11:00 -04:00
Andrew Gwozdziewycz
0130662581 Update syntax/readerr documentation to match code. (#1838) 
* Update syntax/readerr documentation to match code.

This change simply updates the documentation to match the contract that's actually enforced on `raise-read-eof-error` and `raise-read-error`. 

Fixes #1837.

* Fix additional typo noticed by @bennn
 2017-10-10 15:24:08 -04:00
Robby Findler
404539c894 only do the in-tail-position contract elimination with chaperone (or flat) 
contracts

closes #1829
 2017-10-06 13:30:05 -05:00
Robby Findler
ce30687ec6 fix construction of test case failure message 2017-10-06 13:30:00 -05:00
Matthew Flatt
c7c8f6da7e fix expand corner to be consistent with eval 
Repairs a problem reported by Laurent Orseau, where running the
following program with errotrace enabled reports an error:

 #lang racket/base
 (require racket/class)
 (new object%)
 (define-namespace-anchor nsa)
 (define ns (namespace-anchor->namespace nsa))
 (eval '(new object%) ns)

I haven't been able to make a standalone test, yet, but I'll keep
trying.
 2017-10-05 10:41:19 -06:00
Matthew Butterick
43caae97b7 Doc typo 2017-10-04 21:14:12 -04:00
Daniel Feltey
3b56866fc1 Fix object/c contract-stronger? 2017-10-02 15:12:17 -05:00
Ben Greenman
c02d91d174 contract-stronger: swap order of cases 
Check that the input to the default `contract-stronger?` is
 not an impersonator before calling `equal?` on it
 2017-10-02 02:11:52 -04:00
Daniel Feltey
ed5911e531 make contract-stronger? return #f for impersonator contracts that don't implement stronger 2017-09-29 15:57:56 -05:00
Robby Findler
39dea70732 fix promise/c in the case of multiple values coming out of the promise 
closes #1821
 2017-09-29 07:11:48 -05:00
Leif Andersen
e5a14c4f9d Add make-serialize-info example. 2017-09-28 13:53:19 -04:00
Ben Greenman
56443e76a9 doc: remove extra '}' 2017-09-28 00:48:51 -04:00