d62a6c4b09
Minor core cleanup
2008-11-18 22:08:02 -06:00
a9a28a3231
Trying to blindly fix Win64 unit tests
2008-11-17 20:28:52 -06:00
f29300c6ba
Better error message when vocab top level forms leave crap on the stack
2008-11-16 19:47:52 -06:00
251f9213c3
Fix typo
2008-11-16 19:19:53 -06:00
fbe29ceca8
format-table should not be private since ui.gadgets.grids uses it
2008-11-16 15:31:17 -06:00
e6218fdc71
Move words from compiler.errors.private to compiler.errors
2008-11-16 13:46:21 -06:00
acc1dc23fa
More usages of $quotation
2008-11-16 09:08:38 -06:00
9bf63b1613
New $quotation markup element
2008-11-16 09:03:30 -06:00
93e9e34175
Add $maybe markup element
2008-11-16 06:02:13 -06:00
99fd539b01
Rename io.streams.string:null to null-encoding
2008-11-16 05:59:38 -06:00
ff95802c46
Change tags a bit
2008-11-14 20:17:27 -06:00
dffbb120a3
SLOT: now defines the accessor words instead of just deferring them
2008-11-14 00:39:28 -06:00
4858a2ea74
Simplify inline? now that generic words can't be inline anymore
2008-11-13 09:33:18 -06:00
ef6206d4bb
Try to optimize generic dispatch to speed up + on fixnums, nth on arrays for example
2008-11-13 03:51:04 -06:00
9f52ee50b0
Another speedup to >alist: create vector with right size up-front. 10 second bootstrap time reduction
2008-11-12 00:10:50 -06:00
eed1dd3d2b
Fix byte vector prettyprinting
2008-11-11 23:38:19 -06:00
57cf756c0e
Use eq? instead of number= in a few places
2008-11-11 23:04:15 -06:00
ed3a81a503
Faster M: hashtable >alist performs less dispatch and allocates less junk
2008-11-11 23:03:50 -06:00
0edfc83d50
document file-systems
2008-11-11 15:41:27 -06:00
78ab09d8bb
move file-systems word to core
2008-11-11 14:04:05 -06:00
a32acdcff0
Tweak grouping code for better method inlining
2008-11-11 12:11:13 -06:00
1d630b6cf2
Minor documentation fixes
2008-11-11 10:51:15 -06:00
e49ee006d1
Fix loop combinator docs
2008-11-11 08:50:48 -06:00
33a082c361
Fix reference to obsolete G: word
2008-11-10 02:43:16 -06:00
57242951b9
Better error reporting in FROM: and RENAME: words
2008-11-08 19:33:15 -06:00
78eeaddcf2
Add winnt-x86.64 boot image name
2008-11-07 20:34:26 -06:00
cc7ab11881
Add more unit tests
2008-11-07 20:34:04 -06:00
63a9975a0e
Support inline, foldable, flushable on methods; add declarations in a couple of places for ricing purposes
2008-11-06 13:13:37 -06:00
d2b2cda596
Sort methods by tag, this speeds up >fixnum in the common case where the input is a fixnum
2008-11-06 10:44:06 -06:00
1c1333fbe9
Compile not and >boolean as branchless intrinsics by having the CFG builder detect certain code patterns
2008-11-06 09:09:21 -06:00
a95bb533b5
Remove more redundant branches from tuple type predicates and generic words with methods on tuple classes
2008-11-06 09:08:17 -06:00
9160e667a7
Fix tuple>array and tuple reshaping
2008-11-06 01:10:56 -06:00
bfd119e3b5
I/O ricing: various hints added so that UTF8 and ASCII fastpaths compile with less dispatch. 25% improvement on reverse-complement
2008-11-06 00:02:44 -06:00
aac256324f
lo-tag-dispatch-engine now sorts tags, this ensures that >fixnum compiles more efficiently
2008-11-06 00:02:10 -06:00
cc879fa9b7
Tuple layouts are now arrays, instead of built-in types. The superclass
...
array is now part of the tuple layout object itself, and class hashcodes
are stored alongside class words there. This removes 2 indirections when
reading a superclass, and 3 when reading a superclass hashcode.
2008-11-05 22:20:29 -06:00
4e98751ce0
Remove a conditional branch from all tuple dispatches, since we don't have to check if the class height is at least 1; and remove memory accesses from tuple dispatch where all tuples are height 1
2008-11-05 20:23:38 -06:00
4fd1767768
Old fix for classes-intersect? no-method bug was incorrect; we were
...
ignoring anonymous classes in compiled-generic-crossref. Also,
forget* now calls reset-word so that references to predicates of
forgotten classes don't break the compiler with a similar error.
2008-11-05 18:32:02 -06:00
fb64c1cb45
Fix class<=; a predicate class derived from a union was not reported as being contained in the union
2008-11-04 04:59:54 -06:00
cc94894441
Fix notorious classes-intersect? bug, or at least one manifestation thereof. Turns out that we may temporarily end up with forgotten classes in the compiled-generic-crossref table. This is not a problem, since subsequently the words that reference forgotten classes will presumably be redefined and recompiled, but it does mean that (compiled-generic-usage) does need to handle this case
2008-11-04 03:38:44 -06:00
9b492b55e0
Fix minor leak: class-caches were not reset if a class was forgotten
2008-11-04 02:17:37 -06:00
7134fad54c
Add more unit tests highlighting the problem: predicate -vs- union comparison is broken
2008-11-04 02:17:22 -06:00
b4d3473d5e
Tuple parsing didn't handle line breaks properly; if scan returns f, must throw unexpected-eof
2008-11-03 21:01:50 -06:00
dc85ed3d56
Fix help lint: links to obsolete article
2008-11-03 08:08:28 -06:00
cdb5c30bd3
Only build a generic word once, even if a compilation unit defines several methods on it, by adding generics to a set and building them at the end. 25% improvement on bootstrap time
2008-11-03 03:51:28 -06:00
44a2499c17
fixnum-shift-fast and fixnum-mod subprimitives
2008-11-03 01:18:35 -06:00
998637109a
Merge branch 'master' into new_codegen
2008-11-02 03:10:40 -06:00
673fda2281
Fix bogus error message
2008-11-02 03:10:27 -06:00
b5c3c19ebe
Merge branch 'master' of git://factorcode.org/git/factor into new_codegen
2008-11-01 16:38:04 -05:00
7fc44aa444
Add failing tests
2008-10-31 21:07:57 -05:00
2fb17458e3
Fix duplication
2008-10-28 16:22:54 -07:00
Slava Pestov
Doug Coleman
unknown