Commit Graph

4152 Commits (752c895d2364e7331501876cca62da74d6456c61)

Author SHA1 Message Date
Björn Lindqvist 46105b62f6 VM: change callstack>array primitive so it doesn't reverse anymore
This fixes one part of the #452 bug.
2016-10-13 01:34:55 +02:00
Björn Lindqvist 32825b9f5f bootstrap.image.primitives: array>quotation can't be foldable 2016-10-09 00:47:37 +02:00
Björn Lindqvist 88b2ed482c math.parser: (format-float) can only be flushable 2016-10-09 00:47:37 +02:00
Björn Lindqvist bbd71ae975 math.parser: (format-float) doesn't need to zero-terminate returned byte-array 2016-10-09 00:47:37 +02:00
John Benediktsson 8f6ac5d457 Revert "hash-sets,hashtables: make it so the array backing the hash is non-empty"
This reverts commit 6f10f06c27.
2016-10-06 11:18:41 -07:00
John Benediktsson db46b16d10 Revert "hash-sets,hashtables: improved key@ word"
This reverts commit 70c7f9e029.
2016-10-06 11:18:35 -07:00
Björn Lindqvist 6f10f06c27 hash-sets,hashtables: make it so the array backing the hash is non-empty
Commit 70c7f9e029 made it so the code
assumes the array is not empty. But it can be empty if the hashtable is
created using "hashtable new" and then it can crash because it reads
uninitialized memory. Setting the initial of the array slot to
a valid hash-array should fix that.
2016-10-05 19:19:39 +02:00
Björn Lindqvist a6df4168af bootstrap.layouts: sync note in comments 2016-09-25 20:10:16 +02:00
Björn Lindqvist 70c7f9e029 hash-sets,hashtables: improved key@ word
Don't need to check the length of the backing array because it can be
assumed to be > 0. That should make hash lookups a little faster.
2016-09-25 20:07:08 +02:00
Alexander Iljin 3cfc4c69ff checksums-docs: add link to checksums.process-docs 2016-09-19 20:18:06 -07:00
Björn Lindqvist 3bbf421043 io.files: the paths are absolute so cwd prepend-path isn't needed 2016-09-14 14:02:11 +02:00
Björn Lindqvist 20711a0fd3 Adding escaping to CHAR: " -> CHAR: \"
It is not necessary to escape the " character, but it doesn't hurt and
is necessary for syntax-highlighting (FUEL) to work correctly.
2016-09-04 06:40:43 +02:00
Björn Lindqvist b82ea14af0 sets,compiler.cfg.stacks.global: putting the refine word in sets 2016-08-30 02:34:37 +02:00
John Benediktsson acb4dec55d Revert "io.pathnames: fix separator when appending paths in Windows"
This reverts commit b477c43293.
2016-08-23 09:54:16 -07:00
masweeney b477c43293 io.pathnames: fix separator when appending paths in Windows 2016-08-23 08:53:38 -07:00
Björn Lindqvist 67f38302d8 system: looks like the name>> accessor can be used here
So then the cpu>string and os>string words aren't needed.
2016-08-11 17:32:02 +02:00
John Benediktsson fa5441f70a Revert "assocs-docs: mark assoc-partition and partition as related words"
This reverts commit 892e41c1ce.
2016-08-10 19:42:41 -07:00
Alexander Iljin 892e41c1ce assocs-docs: mark assoc-partition and partition as related words 2016-08-10 19:42:20 -07:00
Björn Lindqvist ddc5ece757 alien.*: frontend varargs support! #1677
alien-invoke gets an extra parameter indicating if the call uses varargs
or not. In 99.9% of the cases, the parameter should be f, but is t if
varargs are indicated. E.g

: do-printf ( fmt d -- st ) int f "printf" { c-string double } t alien-invoke ;
2016-08-10 21:34:37 +02:00
Björn Lindqvist dee9cf0a70 alien: replacing the alien-*-error family of errors with callsite-not-compiled
All the errors did the same thing anyway. Also adding a summary method
to it.
2016-08-04 11:51:59 +02:00
John Benediktsson 851398898b checksums.crc32: verify checksum-state is working. 2016-08-02 13:16:30 -07:00
John Benediktsson 55df8437df checksums: fix stack effect of with-checksum-state. 2016-08-02 10:58:48 -07:00
John Benediktsson 1b54f0a434 checksums: make checksum-state implement dispose. 2016-08-01 15:46:30 -07:00
John Benediktsson 83a7592a86 checksums: making "checksum-state" a universal concept.
I'm not super happy with the design yet, partly because OpenSSL doesn't
subclass ``checksum-state`` so all ``initialize-checksum-state`` will not
return a ``checksum-state`` instance.  That could maybe be changed by making
``checksum-state`` a mixin, or perhaps some other way.
2016-07-30 09:31:10 -07:00
Alexander Iljin acd22d1218 combinators-docs: fix wording 2016-07-30 09:25:30 -07:00
Alexander Iljin 1cec145e46 math: fix stack effect declaration for the neg? generic 2016-07-30 09:25:30 -07:00
Alexander Iljin 1aded9abdd docs: replace $description with $class-description for class words 2016-07-30 09:25:30 -07:00
Alexander Iljin 2fd5654473 Add EOL at EOF for all authors.txt files for consistency
In some instances replaced CR/LF with LF.
2016-07-30 09:25:30 -07:00
Alexander Iljin 4e3dab7237 math-docs: fix a copy-paste error 2016-07-30 09:25:30 -07:00
John Benediktsson ccbde89968 vocabs.loader: one vocab-exists? word. 2016-07-16 16:21:45 -07:00
John Benediktsson 8d077a96ad using [ ... ] [ drop ] if* instead of dup [ ... ] [ 2drop ] if. 2016-07-11 21:01:03 -07:00
John Benediktsson bd630d1464 math.integers: use must-fail instead of ignore-errors. 2016-07-10 20:28:28 -07:00
John Benediktsson 5abccb1391 io.backend: remove unused normalize-directory word. 2016-06-30 14:11:24 -07:00
John Benediktsson 180bb32e86 checksums: remove doc for stream-checksum. 2016-06-30 12:34:36 -07:00
John Benediktsson 2255d6d876 checksums: some cleanup.
- define checksum-bytes and checksum-stream in terms of each other
- delete stream-checksum now that it's no longer needed
- move checksum-state stuff into checksums.common
- add a block-checksum that uses the checksum-state stuff
- change checksums.openssl to not use the checksum-state
2016-06-30 12:30:58 -07:00
John Benediktsson e146309a0c io.encodings.binary: move to core. 2016-06-30 12:30:58 -07:00
Alexander Iljin 66b9770f77 checksums: make add-checksum-bytes a generic word 2016-06-29 14:17:57 -07:00
Alexander Iljin c054cb7a78 checksums-docs: add note on using checksums.stream 2016-06-29 14:17:57 -07:00
Alexander Iljin 0c56a7cec1 checksums-docs: fix heading in checksum-file 2016-06-29 14:17:57 -07:00
Björn Lindqvist ee80cb0752 ui.backend.gtk: use a global variable to store the icon data
It's easier than the previous code that replaced the get-icon-data word
during deployment. Also GTK support for icon.ico icons.
2016-06-17 16:39:11 +02:00
Björn Lindqvist 6984bae7ca VM: rename of JIT_PUSH_IMMEDIATE to JIT_PUSH_LITERAL
The code template pushes both immediates and references so the name was
wrong.
2016-06-07 14:54:23 +02:00
Björn Lindqvist 91f335b424 style thing: indenting tuple slots 2016-05-31 03:37:23 +02:00
Doug Coleman 21461a1b75 core: selector-for and collector-for are the wrong naming convention--they should be selector-as and collector-as because they take exemplars.
The exemplar is the starting length of the output sequence in collector-as, so add another collector-for-as that takes a seq that we are making a collector for.
2016-05-26 14:36:11 -07:00
Alexander Iljin bdd000c7a2 syntax-docs: fix the code snippet in the ERROR: article 2016-05-25 15:56:05 +03:00
Alexander Iljin 0b3310c326 words-docs: fix wording 2016-05-25 13:23:28 +03:00
Alexander Iljin 06ec320eca words-docs: add "recursive" to the list of word properties
Put "flushable" in quotes for consistency.
2016-05-25 13:23:01 +03:00
Alexander Iljin b8476d7081 io.streams-docs: fix some copy-paste errors 2016-05-25 11:38:28 +03:00
Alexander Iljin 1e0bf40b7f docs: minor fixes and amendments. 2016-05-22 07:53:21 -07:00
Doug Coleman 839a5a22ec effects: make :type print as :type instead of ``_: type`` 2016-05-14 17:03:09 -07:00
Doug Coleman 3fa9ad91f6 effects: Add support for :type as standalone, unnamed types. ( :float -- ) pprints as ( _: float -- ), and ( :( a -- b ) -- ) works, etc. :union{ foo bar } pprints as object but could be improved if the effects parser uses the prettyprinter somehow.
Move ?execute-parsing to parser to avoid circularity.
2016-05-12 16:27:50 -07:00