mirror of
https://github.com/golang/go
synced 2024-11-18 11:55:01 -07:00
runtime: speed up non-ASCII rune decoding
Copies utf8 constants and EncodeRune implementation from unicode/utf8. Adds a new decoderune implementation that is used by the compiler in code generated for ranging over strings. It does not handle ASCII runes since these are handled directly before calls to decoderune. The DecodeRuneInString implementation from unicode/utf8 is not used since it uses a lookup table that would increase the use of cpu caches. Adds more tests that check decoding of valid and invalid utf8 sequences. name old time/op new time/op delta RuneIterate/range2/ASCII-4 7.45ns ± 2% 7.45ns ± 1% ~ (p=0.634 n=16+16) RuneIterate/range2/Japanese-4 53.5ns ± 1% 49.2ns ± 2% -8.03% (p=0.000 n=20+20) RuneIterate/range2/MixedLength-4 46.3ns ± 1% 41.0ns ± 2% -11.57% (p=0.000 n=20+20) new: "".decoderune t=1 size=423 args=0x28 locals=0x0 old: "".charntorune t=1 size=666 args=0x28 locals=0x0 Change-Id: I1df1fdb385bb9ea5e5e71b8818ea2bf5ce62de52 Reviewed-on: https://go-review.googlesource.com/28490 Run-TryBot: Martin Möhrmann <martisch@uos.de> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
This commit is contained in:
Martin Möhrmann
Brad Fitzpatrick
parent
fe4307f060
commit
d295174030
@ -23,86 +23,86 @@ const runtimeimport = "" +
|
|||||||
"\x00\x01\x11\"\x00\x01 \x00\t!slicerunetostring\x00\x03\x17\x0f@\"\x00\x11|S\x00\x01 " +
|
"\x00\x01\x11\"\x00\x01 \x00\t!slicerunetostring\x00\x03\x17\x0f@\"\x00\x11|S\x00\x01 " +
|
||||||
"\x00\t!stringtoslicebyte\x00\x03\x17\x0f@\"\x00 \x00\x01\x11\"\x00\t'strin" +
|
"\x00\t!stringtoslicebyte\x00\x03\x17\x0f@\"\x00 \x00\x01\x11\"\x00\t'strin" +
|
||||||
"gtoslicebytetmp\x00\x01 \x00\x01\x11\"\x00\t!stringtosliceru" +
|
"gtoslicebytetmp\x00\x01 \x00\x01\x11\"\x00\t!stringtosliceru" +
|
||||||
"ne\x00\x03\x17\x0f@|S\x00 \x00\x01\x11|S\x00\t\x15charntorune\x00\x03 \x00\x02\x00\x04|S\r" +
|
"ne\x00\x03\x17\x0f@|S\x00 \x00\x01\x11|S\x00\t\x13decoderune\x00\x03 \x00\x02\x00\x04|S\rr" +
|
||||||
"retv·1\x00\x00\x02\rretk·2\x00\x00\t\x11slicecopy\x00\x06:\tto·2" +
|
"etv·1\x00\x00\x02\rretk·2\x00\x00\t\x11slicecopy\x00\x06:\tto·2\x00" +
|
||||||
"\x00\x00:\tfr·3\x00\x00\x16\vwid·4\x00\x1bunsafe-uintptr\x01\x02\x00\t\x1d" +
|
"\x00:\tfr·3\x00\x00\x16\vwid·4\x00\x1bunsafe-uintptr\x01\x02\x00\t\x1ds" +
|
||||||
"slicestringcopy\x00\x04:X\x00\x00:Z\x00\x00\x01\x02\x00\t\rconvI2E\x00\x02:" +
|
"licestringcopy\x00\x04:X\x00\x00:Z\x00\x00\x01\x02\x00\t\rconvI2E\x00\x02:\r" +
|
||||||
"\relem·2\x00\x00\x02:\vret·1\x00\x00\t\rconvI2I\x00\x04\x17\"\x06\x00\x00:\re" +
|
"elem·2\x00\x00\x02:\vret·1\x00\x00\t\rconvI2I\x00\x04\x17\"\x06\x00\x00:\rel" +
|
||||||
"lem·3\x00\x00\x02:f\x00\x00\t\rconvT2E\x00\x04\x17\"\x06\x00\x00\x17:j\x00\x00\x02:f\x00\x00\t" +
|
"em·3\x00\x00\x02:f\x00\x00\t\rconvT2E\x00\x04\x17\"\x06\x00\x00\x17:j\x00\x00\x02:f\x00\x00\t\r" +
|
||||||
"\rconvT2I\x00\x04\x17\"\vtab·2\x00\x00\x17:j\x00\x00\x02:f\x00\x00\t\x11assertE" +
|
"convT2I\x00\x04\x17\"\vtab·2\x00\x00\x17:j\x00\x00\x02:f\x00\x00\t\x11assertE2" +
|
||||||
"2E\x00\x06\x17\"\vtyp·1\x00\x00:\x0fiface·2\x00\x00\x17:\vret·3\x00\x00\x00\t" +
|
"E\x00\x06\x17\"\vtyp·1\x00\x00:\x0fiface·2\x00\x00\x17:\vret·3\x00\x00\x00\t\x13" +
|
||||||
"\x13assertE2E2\x00\x06\x17\"\x06\x00\x00:\x0fiface·3\x00\x00\x17:\vret·4\x00" +
|
"assertE2E2\x00\x06\x17\"\x06\x00\x00:\x0fiface·3\x00\x00\x17:\vret·4\x00\x00" +
|
||||||
"\x00\x01\x00\x00\t\x11assertE2I\x00\x06\x17\"t\x00\x00:v\x00\x00\x17:x\x00\x00\x00\t\x13assert" +
|
"\x01\x00\x00\t\x11assertE2I\x00\x06\x17\"t\x00\x00:v\x00\x00\x17:x\x00\x00\x00\t\x13assertE" +
|
||||||
"E2I2\x00\x06\x17\"\x06\x00\x00:||\x00\x00\x17:~\x00\x00\x01\x00\x00\t\x11assertE2T\x00\x06\x17\"t" +
|
"2I2\x00\x06\x17\"\x06\x00\x00:||\x00\x00\x17:~\x00\x00\x01\x00\x00\t\x11assertE2T\x00\x06\x17\"t\x00" +
|
||||||
"\x00\x00:v\x00\x00\x17:x\x00\x00\x00\t\x13assertE2T2\x00\x06\x17\"\x06\x00\x00:||\x00\x00\x17:~\x00" +
|
"\x00:v\x00\x00\x17:x\x00\x00\x00\t\x13assertE2T2\x00\x06\x17\"\x06\x00\x00:||\x00\x00\x17:~\x00\x00" +
|
||||||
"\x00\x01\x00\x00\t\x11assertI2E\x00\x06\x17\"t\x00\x00:v\x00\x00\x17:x\x00\x00\x00\t\x13assert" +
|
"\x01\x00\x00\t\x11assertI2E\x00\x06\x17\"t\x00\x00:v\x00\x00\x17:x\x00\x00\x00\t\x13assertI" +
|
||||||
"I2E2\x00\x06\x17\"\x06\x00\x00:||\x00\x00\x17:~\x00\x00\x01\x00\x00\t\x11assertI2I\x00\x06\x17\"t" +
|
"2E2\x00\x06\x17\"\x06\x00\x00:||\x00\x00\x17:~\x00\x00\x01\x00\x00\t\x11assertI2I\x00\x06\x17\"t\x00" +
|
||||||
"\x00\x00:v\x00\x00\x17:x\x00\x00\x00\t\x13assertI2I2\x00\x06\x17\"\x06\x00\x00:||\x00\x00\x17:~\x00" +
|
"\x00:v\x00\x00\x17:x\x00\x00\x00\t\x13assertI2I2\x00\x06\x17\"\x06\x00\x00:||\x00\x00\x17:~\x00\x00" +
|
||||||
"\x00\x01\x00\x00\t\x11assertI2T\x00\x06\x17\"t\x00\x00:v\x00\x00\x17:x\x00\x00\x00\t\x13assert" +
|
"\x01\x00\x00\t\x11assertI2T\x00\x06\x17\"t\x00\x00:v\x00\x00\x17:x\x00\x00\x00\t\x13assertI" +
|
||||||
"I2T2\x00\x06\x17\"\x06\x00\x00:||\x00\x00\x17:~\x00\x00\x01\x00\x00\t\x17panicdottype\x00\x06" +
|
"2T2\x00\x06\x17\"\x06\x00\x00:||\x00\x00\x17:~\x00\x00\x01\x00\x00\t\x17panicdottype\x00\x06\x17" +
|
||||||
"\x17\"\rhave·1\x00\x00\x17\"\rwant·2\x00\x00\x17\"||\x00\x00\x00\t\rifaceeq" +
|
"\"\rhave·1\x00\x00\x17\"\rwant·2\x00\x00\x17\"||\x00\x00\x00\t\rifaceeq\x00" +
|
||||||
"\x00\x04:\ti1·2\x00\x00:\ti2·3\x00\x00\x02\x00f\x00\x00\t\refaceeq\x00\x04:\x9c\x01\x00" +
|
"\x04:\ti1·2\x00\x00:\ti2·3\x00\x00\x02\x00f\x00\x00\t\refaceeq\x00\x04:\x9c\x01\x00\x00" +
|
||||||
"\x00:\x9e\x01\x00\x00\x02\x00f\x00\x00\t\rmakemap\x00\b\x17\"\x13mapType·2\x00\x00\n\rh" +
|
":\x9e\x01\x00\x00\x02\x00f\x00\x00\t\rmakemap\x00\b\x17\"\x13mapType·2\x00\x00\n\rhi" +
|
||||||
"int·3\x00\x00\x17:\x11mapbuf·4\x00\x00\x17:\x17bucketbuf·5\x00\x00\x02" +
|
"nt·3\x00\x00\x17:\x11mapbuf·4\x00\x00\x17:\x17bucketbuf·5\x00\x00\x02\x1d" +
|
||||||
"\x1d::\rhmap·1\x00\x00\t\x13mapaccess1\x00\x06\x17\"\xa4\x01\x00\x00\x1d::\rhma" +
|
"::\rhmap·1\x00\x00\t\x13mapaccess1\x00\x06\x17\"\xa4\x01\x00\x00\x1d::\rhmap" +
|
||||||
"p·3\x00\x00\x17:\vkey·4\x00\x00\x02\x17:\vval·1\x00\x00\t!mapaccess" +
|
"·3\x00\x00\x17:\vkey·4\x00\x00\x02\x17:\vval·1\x00\x00\t!mapaccess1" +
|
||||||
"1_fast32\x00\x06\x17\"\xa4\x01\x00\x00\x1d::\xb0\x01\x00\x00:\xb2\x01\x00\x00\x02\x17:\xb4\x01\x00\x00\t!map" +
|
"_fast32\x00\x06\x17\"\xa4\x01\x00\x00\x1d::\xb0\x01\x00\x00:\xb2\x01\x00\x00\x02\x17:\xb4\x01\x00\x00\t!mapa" +
|
||||||
"access1_fast64\x00\x06\x17\"\xa4\x01\x00\x00\x1d::\xb0\x01\x00\x00:\xb2\x01\x00\x00\x02\x17:\xb4\x01\x00" +
|
"ccess1_fast64\x00\x06\x17\"\xa4\x01\x00\x00\x1d::\xb0\x01\x00\x00:\xb2\x01\x00\x00\x02\x17:\xb4\x01\x00\x00" +
|
||||||
"\x00\t#mapaccess1_faststr\x00\x06\x17\"\xa4\x01\x00\x00\x1d::\xb0\x01\x00\x00:\xb2\x01\x00" +
|
"\t#mapaccess1_faststr\x00\x06\x17\"\xa4\x01\x00\x00\x1d::\xb0\x01\x00\x00:\xb2\x01\x00\x00" +
|
||||||
"\x00\x02\x17:\xb4\x01\x00\x00\t\x1bmapaccess1_fat\x00\b\x17\"\xa4\x01\x00\x00\x1d::\xb0\x01\x00\x00\x17" +
|
"\x02\x17:\xb4\x01\x00\x00\t\x1bmapaccess1_fat\x00\b\x17\"\xa4\x01\x00\x00\x1d::\xb0\x01\x00\x00\x17:" +
|
||||||
":\xb2\x01\x00\x00\x17\"\rzero·5\x00\x00\x02\x17:\xb4\x01\x00\x00\t\x13mapaccess2\x00\x06\x17\"" +
|
"\xb2\x01\x00\x00\x17\"\rzero·5\x00\x00\x02\x17:\xb4\x01\x00\x00\t\x13mapaccess2\x00\x06\x17\"\x13" +
|
||||||
"\x13mapType·3\x00\x00\x1d::\rhmap·4\x00\x00\x17:\vkey·5\x00\x00\x04\x17:" +
|
"mapType·3\x00\x00\x1d::\rhmap·4\x00\x00\x17:\vkey·5\x00\x00\x04\x17:\xb4" +
|
||||||
"\xb4\x01\x00\x00\x00\rpres·2\x00\x00\t!mapaccess2_fast32\x00\x06\x17\"\xc2\x01" +
|
"\x01\x00\x00\x00\rpres·2\x00\x00\t!mapaccess2_fast32\x00\x06\x17\"\xc2\x01\x00" +
|
||||||
"\x00\x00\x1d::\xc4\x01\x00\x00:\xc6\x01\x00\x00\x04\x17:\xb4\x01\x00\x00\x00\xc8\x01\x00\x00\t!mapaccess2_f" +
|
"\x00\x1d::\xc4\x01\x00\x00:\xc6\x01\x00\x00\x04\x17:\xb4\x01\x00\x00\x00\xc8\x01\x00\x00\t!mapaccess2_fa" +
|
||||||
"ast64\x00\x06\x17\"\xc2\x01\x00\x00\x1d::\xc4\x01\x00\x00:\xc6\x01\x00\x00\x04\x17:\xb4\x01\x00\x00\x00\xc8\x01\x00\x00\t#m" +
|
"st64\x00\x06\x17\"\xc2\x01\x00\x00\x1d::\xc4\x01\x00\x00:\xc6\x01\x00\x00\x04\x17:\xb4\x01\x00\x00\x00\xc8\x01\x00\x00\t#ma" +
|
||||||
"apaccess2_faststr\x00\x06\x17\"\xc2\x01\x00\x00\x1d::\xc4\x01\x00\x00:\xc6\x01\x00\x00\x04\x17:" +
|
"paccess2_faststr\x00\x06\x17\"\xc2\x01\x00\x00\x1d::\xc4\x01\x00\x00:\xc6\x01\x00\x00\x04\x17:\xb4" +
|
||||||
"\xb4\x01\x00\x00\x00\xc8\x01\x00\x00\t\x1bmapaccess2_fat\x00\b\x17\"\xc2\x01\x00\x00\x1d::\xc4\x01\x00\x00" +
|
"\x01\x00\x00\x00\xc8\x01\x00\x00\t\x1bmapaccess2_fat\x00\b\x17\"\xc2\x01\x00\x00\x1d::\xc4\x01\x00\x00\x17" +
|
||||||
"\x17:\xc6\x01\x00\x00\x17\"\rzero·6\x00\x00\x04\x17:\xb4\x01\x00\x00\x00\xc8\x01\x00\x00\t\x11mapassig" +
|
":\xc6\x01\x00\x00\x17\"\rzero·6\x00\x00\x04\x17:\xb4\x01\x00\x00\x00\xc8\x01\x00\x00\t\x11mapassign" +
|
||||||
"n\x00\x06\x17\"\xa4\x01\x00\x00\x1d::\xb0\x01\x00\x00\x17:\xb2\x01\x00\x00\x02\x17:\xb4\x01\x00\x00\t\x15mapiterin" +
|
"\x00\x06\x17\"\xa4\x01\x00\x00\x1d::\xb0\x01\x00\x00\x17:\xb2\x01\x00\x00\x02\x17:\xb4\x01\x00\x00\t\x15mapiterini" +
|
||||||
"it\x00\x06\x17\"\x13mapType·1\x00\x00\x1d::\rhmap·2\x00\x00\x17:\x0fhiter" +
|
"t\x00\x06\x17\"\x13mapType·1\x00\x00\x1d::\rhmap·2\x00\x00\x17:\x0fhiter\xc2" +
|
||||||
"·3\x00\x00\x00\t\x11mapdelete\x00\x06\x17\"\xd8\x01\x00\x00\x1d::\xda\x01\x00\x00\x17:\vkey·" +
|
"\xb73\x00\x00\x00\t\x11mapdelete\x00\x06\x17\"\xd8\x01\x00\x00\x1d::\xda\x01\x00\x00\x17:\vkey·3" +
|
||||||
"3\x00\x00\x00\t\x15mapiternext\x00\x02\x17:\x0fhiter·1\x00\x00\x00\t\x0fmakec" +
|
"\x00\x00\x00\t\x15mapiternext\x00\x02\x17:\x0fhiter·1\x00\x00\x00\t\x0fmakech" +
|
||||||
"han\x00\x04\x17\"\x15chanType·2\x00\x00\n\xa6\x01\x00\x00\x02\x1f\x06:\x0fhchan·1\x00" +
|
"an\x00\x04\x17\"\x15chanType·2\x00\x00\n\xa6\x01\x00\x00\x02\x1f\x06:\x0fhchan·1\x00\x00" +
|
||||||
"\x00\t\x11chanrecv1\x00\x06\x17\"\x15chanType·1\x00\x00\x1f\x02:\x0fhchan\xc2" +
|
"\t\x11chanrecv1\x00\x06\x17\"\x15chanType·1\x00\x00\x1f\x02:\x0fhchan·" +
|
||||||
"\xb72\x00\x00\x17:j\x00\x00\x00\t\x11chanrecv2\x00\x06\x17\"\xe8\x01\x00\x00\x1f\x02:\x0fhchan·" +
|
"2\x00\x00\x17:j\x00\x00\x00\t\x11chanrecv2\x00\x06\x17\"\xe8\x01\x00\x00\x1f\x02:\x0fhchan·3" +
|
||||||
"3\x00\x00\x17:\relem·4\x00\x00\x01\x00\x00\t\x11chansend1\x00\x06\x17\"\xee\x01\x00\x00\x1f\x04:" +
|
"\x00\x00\x17:\relem·4\x00\x00\x01\x00\x00\t\x11chansend1\x00\x06\x17\"\xee\x01\x00\x00\x1f\x04:\xf0" +
|
||||||
"\xf0\x01\x00\x00\x17:j\x00\x00\x00\t\x11closechan\x00\x02:\xea\x01\x00\x00\x00\a\x17writeBarr" +
|
"\x01\x00\x00\x17:j\x00\x00\x00\t\x11closechan\x00\x02:\xea\x01\x00\x00\x00\a\x17writeBarri" +
|
||||||
"ier\x00\x15\x06\renabled\x00\x00\x00\vneeded\x00\x00\x00\x05cgo\x00\x00\x00\t\x1dwrit" +
|
"er\x00\x15\x06\renabled\x00\x00\x00\vneeded\x00\x00\x00\x05cgo\x00\x00\x00\t\x1dwrite" +
|
||||||
"ebarrierptr\x00\x04\x17:\vdst·1\x00\x00:\vsrc·2\x00\x00\x00\t\x17typ" +
|
"barrierptr\x00\x04\x17:\vdst·1\x00\x00:\vsrc·2\x00\x00\x00\t\x17type" +
|
||||||
"edmemmove\x00\x06\x17\"t\x00\x00\x17:\vdst·2\x00\x00\x17:\vsrc·3\x00\x00\x00\t" +
|
"dmemmove\x00\x06\x17\"t\x00\x00\x17:\vdst·2\x00\x00\x17:\vsrc·3\x00\x00\x00\t\x1b" +
|
||||||
"\x1btypedslicecopy\x00\x06\x17\"\x06\x00\x00:\vdst·3\x00\x00:\vsrc·4" +
|
"typedslicecopy\x00\x06\x17\"\x06\x00\x00:\vdst·3\x00\x00:\vsrc·4\x00" +
|
||||||
"\x00\x00\x01\x02\x00\t\x17selectnbsend\x00\x06\x17\"\xe8\x01\x00\x00\x1f\x04:\xf4\x01\x00\x00\x17:\xf6\x01\x00\x00" +
|
"\x00\x01\x02\x00\t\x17selectnbsend\x00\x06\x17\"\xe8\x01\x00\x00\x1f\x04:\xf4\x01\x00\x00\x17:\xf6\x01\x00\x00\x01" +
|
||||||
"\x01\x00\x00\t\x17selectnbrecv\x00\x06\x17\"\xe8\x01\x00\x00\x17:j\x00\x00\x1f\x02:\x0fhchan\xc2" +
|
"\x00\x00\t\x17selectnbrecv\x00\x06\x17\"\xe8\x01\x00\x00\x17:j\x00\x00\x1f\x02:\x0fhchan·" +
|
||||||
"\xb74\x00\x00\x01\x00\x00\t\x19selectnbrecv2\x00\b\x17\"\xe8\x01\x00\x00\x17:j\x00\x00\x17\x00\x15re" +
|
"4\x00\x00\x01\x00\x00\t\x19selectnbrecv2\x00\b\x17\"\xe8\x01\x00\x00\x17:j\x00\x00\x17\x00\x15rec" +
|
||||||
"ceived·4\x00\x00\x1f\x02:\x0fhchan·5\x00\x00\x01\x00\x00\t\x11newselect\x00" +
|
"eived·4\x00\x00\x1f\x02:\x0fhchan·5\x00\x00\x01\x00\x00\t\x11newselect\x00\x06" +
|
||||||
"\x06\x17\"\vsel·1\x00\x00\n\x13selsize·2\x00\x00\b\rsize·3\x00\x00\x00\t\x13" +
|
"\x17\"\vsel·1\x00\x00\n\x13selsize·2\x00\x00\b\rsize·3\x00\x00\x00\t\x13s" +
|
||||||
"selectsend\x00\x06\x17\"\vsel·2\x00\x00\x1f\x04:\xf4\x01\x00\x00\x17:\xf6\x01\x00\x00\x02\x00\x15s" +
|
"electsend\x00\x06\x17\"\vsel·2\x00\x00\x1f\x04:\xf4\x01\x00\x00\x17:\xf6\x01\x00\x00\x02\x00\x15se" +
|
||||||
"elected·1\x00\x00\t\x13selectrecv\x00\x06\x17\"\xac\x02\x00\x00\x1f\x02:\xf4\x01\x00\x00\x17" +
|
"lected·1\x00\x00\t\x13selectrecv\x00\x06\x17\"\xac\x02\x00\x00\x1f\x02:\xf4\x01\x00\x00\x17:" +
|
||||||
":\xf6\x01\x00\x00\x02\x00\xae\x02\x00\x00\t\x15selectrecv2\x00\b\x17\"\xac\x02\x00\x00\x1f\x02:\xf4\x01\x00\x00\x17" +
|
"\xf6\x01\x00\x00\x02\x00\xae\x02\x00\x00\t\x15selectrecv2\x00\b\x17\"\xac\x02\x00\x00\x1f\x02:\xf4\x01\x00\x00\x17:" +
|
||||||
":\xf6\x01\x00\x00\x17\x00\x15received·5\x00\x00\x02\x00\xae\x02\x00\x00\t\x19selectdefau" +
|
"\xf6\x01\x00\x00\x17\x00\x15received·5\x00\x00\x02\x00\xae\x02\x00\x00\t\x19selectdefaul" +
|
||||||
"lt\x00\x02\x17\"\xac\x02\x00\x00\x02\x00\xae\x02\x00\x00\t\x0fselectgo\x00\x02\x17\"\xa4\x02\x00\x00\x00\t\tblo" +
|
"t\x00\x02\x17\"\xac\x02\x00\x00\x02\x00\xae\x02\x00\x00\t\x0fselectgo\x00\x02\x17\"\xa4\x02\x00\x00\x00\t\tbloc" +
|
||||||
"ck\x00\x00\x00\t\x11makeslice\x00\x06\x17\"\x06\x00\x00\x02\vlen·3\x00\x00\x02\vcap·" +
|
"k\x00\x00\x00\t\x11makeslice\x00\x06\x17\"\x06\x00\x00\x02\vlen·3\x00\x00\x02\vcap·4" +
|
||||||
"4\x00\x00\x02\x11:\vary·1\x00\x00\t\x15makeslice64\x00\x06\x17\"\x06\x00\x00\n\xbe\x02\x00\x00" +
|
"\x00\x00\x02\x11:\vary·1\x00\x00\t\x15makeslice64\x00\x06\x17\"\x06\x00\x00\n\xbe\x02\x00\x00\n" +
|
||||||
"\n\xc0\x02\x00\x00\x02\x11:\xc2\x02\x00\x00\t\x11growslice\x00\x06\x17\"\x06\x00\x00\x11:\vold·3\x00" +
|
"\xc0\x02\x00\x00\x02\x11:\xc2\x02\x00\x00\t\x11growslice\x00\x06\x17\"\x06\x00\x00\x11:\vold·3\x00\x00" +
|
||||||
"\x00\x02\xc0\x02\x00\x00\x02\x11:\xc2\x02\x00\x00\t\rmemmove\x00\x06\x17:\tto·1\x00\x00\x17:\vfrm" +
|
"\x02\xc0\x02\x00\x00\x02\x11:\xc2\x02\x00\x00\t\rmemmove\x00\x06\x17:\tto·1\x00\x00\x17:\vfrm\xc2" +
|
||||||
"·2\x00\x00\x16\x11length·3\x00^\x00\t\vmemclr\x00\x04\x17\"\vptr·1\x00\x00" +
|
"\xb72\x00\x00\x16\x11length·3\x00^\x00\t\vmemclr\x00\x04\x17\"\vptr·1\x00\x00\x16" +
|
||||||
"\x16\x11length·2\x00^\x00\t\x0fmemequal\x00\x06\x17:\ax·2\x00\x00\x17:\ay\xc2" +
|
"\x11length·2\x00^\x00\t\x0fmemequal\x00\x06\x17:\ax·2\x00\x00\x17:\ay·" +
|
||||||
"\xb73\x00\x00\x16\rsize·4\x00^\x01\x00\x00\t\x11memequal8\x00\x04\x17:\xda\x02\x00\x00\x17:\xdc" +
|
"3\x00\x00\x16\rsize·4\x00^\x01\x00\x00\t\x11memequal8\x00\x04\x17:\xda\x02\x00\x00\x17:\xdc\x02" +
|
||||||
"\x02\x00\x00\x01\x00\x00\t\x13memequal16\x00\x04\x17:\xda\x02\x00\x00\x17:\xdc\x02\x00\x00\x01\x00\x00\t\x13mem" +
|
"\x00\x00\x01\x00\x00\t\x13memequal16\x00\x04\x17:\xda\x02\x00\x00\x17:\xdc\x02\x00\x00\x01\x00\x00\t\x13meme" +
|
||||||
"equal32\x00\x04\x17:\xda\x02\x00\x00\x17:\xdc\x02\x00\x00\x01\x00\x00\t\x13memequal64\x00\x04\x17:" +
|
"qual32\x00\x04\x17:\xda\x02\x00\x00\x17:\xdc\x02\x00\x00\x01\x00\x00\t\x13memequal64\x00\x04\x17:\xda" +
|
||||||
"\xda\x02\x00\x00\x17:\xdc\x02\x00\x00\x01\x00\x00\t\x15memequal128\x00\x04\x17:\xda\x02\x00\x00\x17:\xdc\x02\x00\x00" +
|
"\x02\x00\x00\x17:\xdc\x02\x00\x00\x01\x00\x00\t\x15memequal128\x00\x04\x17:\xda\x02\x00\x00\x17:\xdc\x02\x00\x00\x01" +
|
||||||
"\x01\x00\x00\t\x0fint64div\x00\x03\n\x00\n\x00\x01\n\x00\t\x11uint64div\x00\x03\x14\x00\x14\x00\x01" +
|
"\x00\x00\t\x0fint64div\x00\x03\n\x00\n\x00\x01\n\x00\t\x11uint64div\x00\x03\x14\x00\x14\x00\x01\x14" +
|
||||||
"\x14\x00\t\x0fint64mod\x00\x03\n\x00\n\x00\x01\n\x00\t\x11uint64mod\x00\x03\x14\x00\x14\x00\x01\x14" +
|
"\x00\t\x0fint64mod\x00\x03\n\x00\n\x00\x01\n\x00\t\x11uint64mod\x00\x03\x14\x00\x14\x00\x01\x14\x00" +
|
||||||
"\x00\t\x1bfloat64toint64\x00\x01\x1a\x00\x01\n\x00\t\x1dfloat64touint6" +
|
"\t\x1bfloat64toint64\x00\x01\x1a\x00\x01\n\x00\t\x1dfloat64touint64" +
|
||||||
"4\x00\x01\x1a\x00\x01\x14\x00\t\x1dfloat64touint32\x00\x01\x1a\x00\x01\x12\x00\t\x1bint64t" +
|
"\x00\x01\x1a\x00\x01\x14\x00\t\x1dfloat64touint32\x00\x01\x1a\x00\x01\x12\x00\t\x1bint64to" +
|
||||||
"ofloat64\x00\x01\n\x00\x01\x1a\x00\t\x1duint64tofloat64\x00\x01\x14\x00\x01\x1a\x00\t" +
|
"float64\x00\x01\n\x00\x01\x1a\x00\t\x1duint64tofloat64\x00\x01\x14\x00\x01\x1a\x00\t\x1d" +
|
||||||
"\x1duint32tofloat64\x00\x01\x12\x00\x01\x1a\x00\t\x19complex128div\x00\x04" +
|
"uint32tofloat64\x00\x01\x12\x00\x01\x1a\x00\t\x19complex128div\x00\x04\x1e" +
|
||||||
"\x1e\vnum·2\x00\x00\x1e\vden·3\x00\x00\x02\x1e\vquo·1\x00\x00\t\x19racefun" +
|
"\vnum·2\x00\x00\x1e\vden·3\x00\x00\x02\x1e\vquo·1\x00\x00\t\x19racefunc" +
|
||||||
"center\x00\x01\x16^\x00\t\x17racefuncexit\x00\x00\x00\t\x0fraceread\x00\x01" +
|
"enter\x00\x01\x16^\x00\t\x17racefuncexit\x00\x00\x00\t\x0fraceread\x00\x01\x16" +
|
||||||
"\x16^\x00\t\x11racewrite\x00\x01\x16^\x00\t\x19racereadrange\x00\x04\x16\rad" +
|
"^\x00\t\x11racewrite\x00\x01\x16^\x00\t\x19racereadrange\x00\x04\x16\radd" +
|
||||||
"dr·1\x00^\x16\rsize·2\x00^\x00\t\x1bracewriterange\x00\x04\x16\x90\x03" +
|
"r·1\x00^\x16\rsize·2\x00^\x00\t\x1bracewriterange\x00\x04\x16\x90\x03\x00" +
|
||||||
"\x00^\x16\x92\x03\x00^\x00\t\x0fmsanread\x00\x04\x16\x90\x03\x00^\x16\x92\x03\x00^\x00\t\x11msanwri" +
|
"^\x16\x92\x03\x00^\x00\t\x0fmsanread\x00\x04\x16\x90\x03\x00^\x16\x92\x03\x00^\x00\t\x11msanwrit" +
|
||||||
"te\x00\x04\x16\x90\x03\x00^\x16\x92\x03\x00^\x00\v\xf6\x01\v\x00\x01\x00\n$$\n"
|
"e\x00\x04\x16\x90\x03\x00^\x16\x92\x03\x00^\x00\v\xf6\x01\v\x00\x01\x00\n$$\n"
|
||||||
|
|
||||||
const unsafeimport = "" +
|
const unsafeimport = "" +
|
||||||
"version 2\n\n\x00\x00\x01\vunsafe\x00\t\x0fOffsetof\x00\x01:\x00\x01\x16\x00\t" +
|
"version 2\n\n\x00\x00\x01\vunsafe\x00\t\x0fOffsetof\x00\x01:\x00\x01\x16\x00\t" +
|
||||||
|
|||||||
@ -53,7 +53,7 @@ func slicerunetostring(*[32]byte, []rune) string
|
|||||||
func stringtoslicebyte(*[32]byte, string) []byte
|
func stringtoslicebyte(*[32]byte, string) []byte
|
||||||
func stringtoslicebytetmp(string) []byte
|
func stringtoslicebytetmp(string) []byte
|
||||||
func stringtoslicerune(*[32]rune, string) []rune
|
func stringtoslicerune(*[32]rune, string) []rune
|
||||||
func charntorune(string, int) (retv rune, retk int)
|
func decoderune(string, int) (retv rune, retk int)
|
||||||
func slicecopy(to any, fr any, wid uintptr) int
|
func slicecopy(to any, fr any, wid uintptr) int
|
||||||
func slicestringcopy(to any, fr any) int
|
func slicestringcopy(to any, fr any) int
|
||||||
|
|
||||||
|
|||||||
@ -295,7 +295,7 @@ func walkrange(n *Node) {
|
|||||||
// if hv2 < utf8.RuneSelf {
|
// if hv2 < utf8.RuneSelf {
|
||||||
// hv1++
|
// hv1++
|
||||||
// } else {
|
// } else {
|
||||||
// hv2, hv1 = charntorune(ha, hv1)
|
// hv2, hv1 = decoderune(ha, hv1)
|
||||||
// }
|
// }
|
||||||
// v2 = hv2
|
// v2 = hv2
|
||||||
// // original body
|
// // original body
|
||||||
@ -334,9 +334,9 @@ func walkrange(n *Node) {
|
|||||||
eif := nod(OAS2, nil, nil)
|
eif := nod(OAS2, nil, nil)
|
||||||
nif.Rlist.Set1(eif)
|
nif.Rlist.Set1(eif)
|
||||||
|
|
||||||
// hv2, hv1 = charntorune(ha, hv1)
|
// hv2, hv1 = decoderune(ha, hv1)
|
||||||
eif.List.Set2(hv2, hv1)
|
eif.List.Set2(hv2, hv1)
|
||||||
fn := syslook("charntorune")
|
fn := syslook("decoderune")
|
||||||
eif.Rlist.Set1(mkcall1(fn, fn.Type.Results(), nil, ha, hv1))
|
eif.Rlist.Set1(mkcall1(fn, fn.Type.Results(), nil, ha, hv1))
|
||||||
|
|
||||||
body = append(body, nif)
|
body = append(body, nif)
|
||||||
|
|||||||
@ -1,204 +0,0 @@
|
|||||||
/*
|
|
||||||
* The authors of this software are Rob Pike and Ken Thompson.
|
|
||||||
* Copyright (c) 2002 by Lucent Technologies.
|
|
||||||
* Portions Copyright 2009 The Go Authors. All rights reserved.
|
|
||||||
* Permission to use, copy, modify, and distribute this software for any
|
|
||||||
* purpose without fee is hereby granted, provided that this entire notice
|
|
||||||
* is included in all copies of any software which is or includes a copy
|
|
||||||
* or modification of this software and in all copies of the supporting
|
|
||||||
* documentation for such software.
|
|
||||||
* THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
|
|
||||||
* WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR LUCENT TECHNOLOGIES MAKE ANY
|
|
||||||
* REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
|
|
||||||
* OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
|
|
||||||
*/
|
|
||||||
|
|
||||||
/*
|
|
||||||
* This code is copied, with slight editing due to type differences,
|
|
||||||
* from a subset of ../lib9/utf/rune.c [which no longer exists]
|
|
||||||
*/
|
|
||||||
|
|
||||||
package runtime
|
|
||||||
|
|
||||||
const (
|
|
||||||
bit1 = 7
|
|
||||||
bitx = 6
|
|
||||||
bit2 = 5
|
|
||||||
bit3 = 4
|
|
||||||
bit4 = 3
|
|
||||||
bit5 = 2
|
|
||||||
|
|
||||||
t1 = ((1 << (bit1 + 1)) - 1) ^ 0xFF /* 0000 0000 */
|
|
||||||
tx = ((1 << (bitx + 1)) - 1) ^ 0xFF /* 1000 0000 */
|
|
||||||
t2 = ((1 << (bit2 + 1)) - 1) ^ 0xFF /* 1100 0000 */
|
|
||||||
t3 = ((1 << (bit3 + 1)) - 1) ^ 0xFF /* 1110 0000 */
|
|
||||||
t4 = ((1 << (bit4 + 1)) - 1) ^ 0xFF /* 1111 0000 */
|
|
||||||
t5 = ((1 << (bit5 + 1)) - 1) ^ 0xFF /* 1111 1000 */
|
|
||||||
|
|
||||||
rune1 = (1 << (bit1 + 0*bitx)) - 1 /* 0000 0000 0111 1111 */
|
|
||||||
rune2 = (1 << (bit2 + 1*bitx)) - 1 /* 0000 0111 1111 1111 */
|
|
||||||
rune3 = (1 << (bit3 + 2*bitx)) - 1 /* 1111 1111 1111 1111 */
|
|
||||||
rune4 = (1 << (bit4 + 3*bitx)) - 1 /* 0001 1111 1111 1111 1111 1111 */
|
|
||||||
|
|
||||||
maskx = (1 << bitx) - 1 /* 0011 1111 */
|
|
||||||
testx = maskx ^ 0xFF /* 1100 0000 */
|
|
||||||
|
|
||||||
runeerror = 0xFFFD
|
|
||||||
runeself = 0x80
|
|
||||||
|
|
||||||
surrogateMin = 0xD800
|
|
||||||
surrogateMax = 0xDFFF
|
|
||||||
|
|
||||||
runemax = 0x10FFFF /* maximum rune value */
|
|
||||||
)
|
|
||||||
|
|
||||||
// charntorune returns the rune at the start of
|
|
||||||
// s[k:] and the index after the rune in s.
|
|
||||||
//
|
|
||||||
// If the string appears to be incomplete or decoding problems
|
|
||||||
// are encountered (runeerror, k + 1) is returned to ensure
|
|
||||||
// progress when charntorune is used to iterate over a string.
|
|
||||||
//
|
|
||||||
// Modified by Wei-Hwa Huang, Google Inc., on 2004-09-24
|
|
||||||
func charntorune(s string, k int) (rune, int) {
|
|
||||||
// When we're not allowed to read anything */
|
|
||||||
if len(s) <= k {
|
|
||||||
return runeerror, k + 1
|
|
||||||
}
|
|
||||||
|
|
||||||
s = s[k:]
|
|
||||||
|
|
||||||
// one character sequence (7-bit value)
|
|
||||||
// 00000-0007F => T1
|
|
||||||
c := s[0]
|
|
||||||
if c < tx {
|
|
||||||
return rune(c), k + 1
|
|
||||||
}
|
|
||||||
|
|
||||||
// If we can't read more than one character we must stop
|
|
||||||
if len(s) <= 1 {
|
|
||||||
return runeerror, k + 1
|
|
||||||
}
|
|
||||||
|
|
||||||
// two character sequence (11-bit value)
|
|
||||||
// 0080-07FF => t2 tx
|
|
||||||
c1 := s[1] ^ tx
|
|
||||||
if (c1 & testx) != 0 {
|
|
||||||
return runeerror, k + 1
|
|
||||||
}
|
|
||||||
if c < t3 {
|
|
||||||
if c < t2 {
|
|
||||||
return runeerror, k + 1
|
|
||||||
}
|
|
||||||
l := ((rune(c) << bitx) | rune(c1)) & rune2
|
|
||||||
if l <= rune1 {
|
|
||||||
return runeerror, k + 1
|
|
||||||
}
|
|
||||||
return l, k + 2
|
|
||||||
}
|
|
||||||
|
|
||||||
// If we can't read more than two characters we must stop
|
|
||||||
if len(s) <= 2 {
|
|
||||||
return runeerror, k + 1
|
|
||||||
}
|
|
||||||
|
|
||||||
// three character sequence (16-bit value)
|
|
||||||
// 0800-FFFF => t3 tx tx
|
|
||||||
c2 := s[2] ^ tx
|
|
||||||
if (c2 & testx) != 0 {
|
|
||||||
return runeerror, k + 1
|
|
||||||
}
|
|
||||||
if c < t4 {
|
|
||||||
l := ((((rune(c) << bitx) | rune(c1)) << bitx) | rune(c2)) & rune3
|
|
||||||
if l <= rune2 {
|
|
||||||
return runeerror, k + 1
|
|
||||||
}
|
|
||||||
if surrogateMin <= l && l <= surrogateMax {
|
|
||||||
return runeerror, k + 1
|
|
||||||
}
|
|
||||||
return l, k + 3
|
|
||||||
}
|
|
||||||
|
|
||||||
if len(s) <= 3 {
|
|
||||||
return runeerror, k + 1
|
|
||||||
}
|
|
||||||
|
|
||||||
// four character sequence (21-bit value)
|
|
||||||
// 10000-1FFFFF => t4 tx tx tx
|
|
||||||
c3 := s[3] ^ tx
|
|
||||||
if (c3 & testx) != 0 {
|
|
||||||
return runeerror, k + 1
|
|
||||||
}
|
|
||||||
if c < t5 {
|
|
||||||
l := ((((((rune(c) << bitx) | rune(c1)) << bitx) | rune(c2)) << bitx) | rune(c3)) & rune4
|
|
||||||
if l <= rune3 || l > runemax {
|
|
||||||
return runeerror, k + 1
|
|
||||||
}
|
|
||||||
return l, k + 4
|
|
||||||
}
|
|
||||||
|
|
||||||
// Support for 5-byte or longer UTF-8 would go here, but
|
|
||||||
// since we don't have that, we'll just return runeerror.
|
|
||||||
return runeerror, k + 1
|
|
||||||
}
|
|
||||||
|
|
||||||
// runetochar converts r to bytes and writes the result to str.
|
|
||||||
// returns the number of bytes generated.
|
|
||||||
func runetochar(str []byte, r rune) int {
|
|
||||||
/* runes are signed, so convert to unsigned for range check. */
|
|
||||||
c := uint32(r)
|
|
||||||
/*
|
|
||||||
* one character sequence
|
|
||||||
* 00000-0007F => 00-7F
|
|
||||||
*/
|
|
||||||
if c <= rune1 {
|
|
||||||
str[0] = byte(c)
|
|
||||||
return 1
|
|
||||||
}
|
|
||||||
/*
|
|
||||||
* two character sequence
|
|
||||||
* 0080-07FF => t2 tx
|
|
||||||
*/
|
|
||||||
if c <= rune2 {
|
|
||||||
_ = str[1]
|
|
||||||
str[0] = byte(t2 | (c >> (1 * bitx)))
|
|
||||||
str[1] = byte(tx | (c & maskx))
|
|
||||||
return 2
|
|
||||||
}
|
|
||||||
|
|
||||||
/*
|
|
||||||
* If the rune is out of range or a surrogate half, convert it to the error rune.
|
|
||||||
* Do this test here because the error rune encodes to three bytes.
|
|
||||||
* Doing it earlier would duplicate work, since an out of range
|
|
||||||
* rune wouldn't have fit in one or two bytes.
|
|
||||||
*/
|
|
||||||
if c > runemax {
|
|
||||||
c = runeerror
|
|
||||||
}
|
|
||||||
if surrogateMin <= c && c <= surrogateMax {
|
|
||||||
c = runeerror
|
|
||||||
}
|
|
||||||
|
|
||||||
/*
|
|
||||||
* three character sequence
|
|
||||||
* 0800-FFFF => t3 tx tx
|
|
||||||
*/
|
|
||||||
if c <= rune3 {
|
|
||||||
_ = str[2]
|
|
||||||
str[0] = byte(t3 | (c >> (2 * bitx)))
|
|
||||||
str[1] = byte(tx | ((c >> (1 * bitx)) & maskx))
|
|
||||||
str[2] = byte(tx | (c & maskx))
|
|
||||||
return 3
|
|
||||||
}
|
|
||||||
|
|
||||||
/*
|
|
||||||
* four character sequence (21-bit value)
|
|
||||||
* 10000-1FFFFF => t4 tx tx tx
|
|
||||||
*/
|
|
||||||
_ = str[3]
|
|
||||||
str[0] = byte(t4 | (c >> (3 * bitx)))
|
|
||||||
str[1] = byte(tx | ((c >> (2 * bitx)) & maskx))
|
|
||||||
str[2] = byte(tx | ((c >> (1 * bitx)) & maskx))
|
|
||||||
str[3] = byte(tx | (c & maskx))
|
|
||||||
return 4
|
|
||||||
}
|
|
||||||
@ -196,7 +196,7 @@ func slicerunetostring(buf *tmpBuf, a []rune) string {
|
|||||||
var dum [4]byte
|
var dum [4]byte
|
||||||
size1 := 0
|
size1 := 0
|
||||||
for _, r := range a {
|
for _, r := range a {
|
||||||
size1 += runetochar(dum[:], r)
|
size1 += encoderune(dum[:], r)
|
||||||
}
|
}
|
||||||
s, b := rawstringtmp(buf, size1+3)
|
s, b := rawstringtmp(buf, size1+3)
|
||||||
size2 := 0
|
size2 := 0
|
||||||
@ -205,7 +205,7 @@ func slicerunetostring(buf *tmpBuf, a []rune) string {
|
|||||||
if size2 >= size1 {
|
if size2 >= size1 {
|
||||||
break
|
break
|
||||||
}
|
}
|
||||||
size2 += runetochar(b[size2:], r)
|
size2 += encoderune(b[size2:], r)
|
||||||
}
|
}
|
||||||
return s[:size2]
|
return s[:size2]
|
||||||
}
|
}
|
||||||
@ -235,9 +235,9 @@ func intstring(buf *[4]byte, v int64) string {
|
|||||||
s, b = rawstring(4)
|
s, b = rawstring(4)
|
||||||
}
|
}
|
||||||
if int64(rune(v)) != v {
|
if int64(rune(v)) != v {
|
||||||
v = runeerror
|
v = runeError
|
||||||
}
|
}
|
||||||
n := runetochar(b, rune(v))
|
n := encoderune(b, rune(v))
|
||||||
return s[:n]
|
return s[:n]
|
||||||
}
|
}
|
||||||
|
|
||||||
@ -378,7 +378,7 @@ func gostringw(strw *uint16) string {
|
|||||||
str := (*[_MaxMem/2/2 - 1]uint16)(unsafe.Pointer(strw))
|
str := (*[_MaxMem/2/2 - 1]uint16)(unsafe.Pointer(strw))
|
||||||
n1 := 0
|
n1 := 0
|
||||||
for i := 0; str[i] != 0; i++ {
|
for i := 0; str[i] != 0; i++ {
|
||||||
n1 += runetochar(buf[:], rune(str[i]))
|
n1 += encoderune(buf[:], rune(str[i]))
|
||||||
}
|
}
|
||||||
s, b := rawstring(n1 + 4)
|
s, b := rawstring(n1 + 4)
|
||||||
n2 := 0
|
n2 := 0
|
||||||
@ -387,7 +387,7 @@ func gostringw(strw *uint16) string {
|
|||||||
if n2 >= n1 {
|
if n2 >= n1 {
|
||||||
break
|
break
|
||||||
}
|
}
|
||||||
n2 += runetochar(b[n2:], rune(str[i]))
|
n2 += encoderune(b[n2:], rune(str[i]))
|
||||||
}
|
}
|
||||||
b[n2] = 0 // for luck
|
b[n2] = 0 // for luck
|
||||||
return s[:n2]
|
return s[:n2]
|
||||||
|
|||||||
@ -92,6 +92,7 @@ func BenchmarkConcatStringAndBytes(b *testing.B) {
|
|||||||
var stringdata = []struct{ name, data string }{
|
var stringdata = []struct{ name, data string }{
|
||||||
{"ASCII", "01234567890"},
|
{"ASCII", "01234567890"},
|
||||||
{"Japanese", "日本語日本語日本語"},
|
{"Japanese", "日本語日本語日本語"},
|
||||||
|
{"MixedLength", "$Ѐࠀက퀀𐀀\U00040000\U0010FFFF"},
|
||||||
}
|
}
|
||||||
|
|
||||||
func BenchmarkRuneIterate(b *testing.B) {
|
func BenchmarkRuneIterate(b *testing.B) {
|
||||||
|
|||||||
123
src/runtime/utf8.go
Normal file
123
src/runtime/utf8.go
Normal file
@ -0,0 +1,123 @@
|
|||||||
|
// Copyright 2016 The Go Authors. All rights reserved.
|
||||||
|
// Use of this source code is governed by a BSD-style
|
||||||
|
// license that can be found in the LICENSE file.
|
||||||
|
|
||||||
|
package runtime
|
||||||
|
|
||||||
|
// Numbers fundamental to the encoding.
|
||||||
|
const (
|
||||||
|
runeError = '\uFFFD' // the "error" Rune or "Unicode replacement character"
|
||||||
|
runeSelf = 0x80 // characters below Runeself are represented as themselves in a single byte.
|
||||||
|
maxRune = '\U0010FFFF' // Maximum valid Unicode code point.
|
||||||
|
)
|
||||||
|
|
||||||
|
// Code points in the surrogate range are not valid for UTF-8.
|
||||||
|
const (
|
||||||
|
surrogateMin = 0xD800
|
||||||
|
surrogateMax = 0xDFFF
|
||||||
|
)
|
||||||
|
|
||||||
|
const (
|
||||||
|
t1 = 0x00 // 0000 0000
|
||||||
|
tx = 0x80 // 1000 0000
|
||||||
|
t2 = 0xC0 // 1100 0000
|
||||||
|
t3 = 0xE0 // 1110 0000
|
||||||
|
t4 = 0xF0 // 1111 0000
|
||||||
|
t5 = 0xF8 // 1111 1000
|
||||||
|
|
||||||
|
maskx = 0x3F // 0011 1111
|
||||||
|
mask2 = 0x1F // 0001 1111
|
||||||
|
mask3 = 0x0F // 0000 1111
|
||||||
|
mask4 = 0x07 // 0000 0111
|
||||||
|
|
||||||
|
rune1Max = 1<<7 - 1
|
||||||
|
rune2Max = 1<<11 - 1
|
||||||
|
rune3Max = 1<<16 - 1
|
||||||
|
|
||||||
|
// The default lowest and highest continuation byte.
|
||||||
|
locb = 0x80 // 1000 0000
|
||||||
|
hicb = 0xBF // 1011 1111
|
||||||
|
)
|
||||||
|
|
||||||
|
// decoderune returns the non-ASCII rune at the start of
|
||||||
|
// s[k:] and the index after the rune in s.
|
||||||
|
//
|
||||||
|
// decoderune assumes that caller has checked that
|
||||||
|
// the to be decoded rune is a non-ASCII rune.
|
||||||
|
//
|
||||||
|
// If the string appears to be incomplete or decoding problems
|
||||||
|
// are encountered (runeerror, k + 1) is returned to ensure
|
||||||
|
// progress when decoderune is used to iterate over a string.
|
||||||
|
func decoderune(s string, k int) (r rune, pos int) {
|
||||||
|
pos = k
|
||||||
|
|
||||||
|
if k >= len(s) {
|
||||||
|
return runeError, k + 1
|
||||||
|
}
|
||||||
|
|
||||||
|
s = s[k:]
|
||||||
|
|
||||||
|
switch {
|
||||||
|
case t2 <= s[0] && s[0] < t3:
|
||||||
|
// 0080-07FF two byte sequence
|
||||||
|
if len(s) > 1 && (locb <= s[1] && s[1] <= hicb) {
|
||||||
|
r = rune(s[0]&mask2)<<6 | rune(s[1]&maskx)
|
||||||
|
pos += 2
|
||||||
|
if rune1Max < r {
|
||||||
|
return
|
||||||
|
}
|
||||||
|
}
|
||||||
|
case t3 <= s[0] && s[0] < t4:
|
||||||
|
// 0800-FFFF three byte sequence
|
||||||
|
if len(s) > 2 && (locb <= s[1] && s[1] <= hicb) && (locb <= s[2] && s[2] <= hicb) {
|
||||||
|
r = rune(s[0]&mask3)<<12 | rune(s[1]&maskx)<<6 | rune(s[2]&maskx)
|
||||||
|
pos += 3
|
||||||
|
if rune2Max < r && !(surrogateMin <= r && r <= surrogateMax) {
|
||||||
|
return
|
||||||
|
}
|
||||||
|
}
|
||||||
|
case t4 <= s[0] && s[0] < t5:
|
||||||
|
// 10000-1FFFFF four byte sequence
|
||||||
|
if len(s) > 3 && (locb <= s[1] && s[1] <= hicb) && (locb <= s[2] && s[2] <= hicb) && (locb <= s[3] && s[3] <= hicb) {
|
||||||
|
r = rune(s[0]&mask4)<<18 | rune(s[1]&maskx)<<12 | rune(s[2]&maskx)<<6 | rune(s[3]&maskx)
|
||||||
|
pos += 4
|
||||||
|
if rune3Max < r && r <= maxRune {
|
||||||
|
return
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
return runeError, k + 1
|
||||||
|
}
|
||||||
|
|
||||||
|
// encoderune writes into p (which must be large enough) the UTF-8 encoding of the rune.
|
||||||
|
// It returns the number of bytes written.
|
||||||
|
func encoderune(p []byte, r rune) int {
|
||||||
|
// Negative values are erroneous. Making it unsigned addresses the problem.
|
||||||
|
switch i := uint32(r); {
|
||||||
|
case i <= rune1Max:
|
||||||
|
p[0] = byte(r)
|
||||||
|
return 1
|
||||||
|
case i <= rune2Max:
|
||||||
|
_ = p[1] // eliminate bounds checks
|
||||||
|
p[0] = t2 | byte(r>>6)
|
||||||
|
p[1] = tx | byte(r)&maskx
|
||||||
|
return 2
|
||||||
|
case i > maxRune, surrogateMin <= i && i <= surrogateMax:
|
||||||
|
r = runeError
|
||||||
|
fallthrough
|
||||||
|
case i <= rune3Max:
|
||||||
|
_ = p[2] // eliminate bounds checks
|
||||||
|
p[0] = t3 | byte(r>>12)
|
||||||
|
p[1] = tx | byte(r>>6)&maskx
|
||||||
|
p[2] = tx | byte(r)&maskx
|
||||||
|
return 3
|
||||||
|
default:
|
||||||
|
_ = p[3] // eliminate bounds checks
|
||||||
|
p[0] = t4 | byte(r>>18)
|
||||||
|
p[1] = tx | byte(r>>12)&maskx
|
||||||
|
p[2] = tx | byte(r>>6)&maskx
|
||||||
|
p[3] = tx | byte(r)&maskx
|
||||||
|
return 4
|
||||||
|
}
|
||||||
|
}
|
||||||
@ -54,14 +54,18 @@ var utf8map = []Utf8Map{
|
|||||||
{0x00ff, "\xc3\xbf"},
|
{0x00ff, "\xc3\xbf"},
|
||||||
{0x0100, "\xc4\x80"},
|
{0x0100, "\xc4\x80"},
|
||||||
{0x07ff, "\xdf\xbf"},
|
{0x07ff, "\xdf\xbf"},
|
||||||
|
{0x0400, "\xd0\x80"},
|
||||||
{0x0800, "\xe0\xa0\x80"},
|
{0x0800, "\xe0\xa0\x80"},
|
||||||
{0x0801, "\xe0\xa0\x81"},
|
{0x0801, "\xe0\xa0\x81"},
|
||||||
|
{0x1000, "\xe1\x80\x80"},
|
||||||
|
{0xd000, "\xed\x80\x80"},
|
||||||
{0xd7ff, "\xed\x9f\xbf"}, // last code point before surrogate half.
|
{0xd7ff, "\xed\x9f\xbf"}, // last code point before surrogate half.
|
||||||
{0xe000, "\xee\x80\x80"}, // first code point after surrogate half.
|
{0xe000, "\xee\x80\x80"}, // first code point after surrogate half.
|
||||||
{0xfffe, "\xef\xbf\xbe"},
|
{0xfffe, "\xef\xbf\xbe"},
|
||||||
{0xffff, "\xef\xbf\xbf"},
|
{0xffff, "\xef\xbf\xbf"},
|
||||||
{0x10000, "\xf0\x90\x80\x80"},
|
{0x10000, "\xf0\x90\x80\x80"},
|
||||||
{0x10001, "\xf0\x90\x80\x81"},
|
{0x10001, "\xf0\x90\x80\x81"},
|
||||||
|
{0x40000, "\xf1\x80\x80\x80"},
|
||||||
{0x10fffe, "\xf4\x8f\xbf\xbe"},
|
{0x10fffe, "\xf4\x8f\xbf\xbe"},
|
||||||
{0x10ffff, "\xf4\x8f\xbf\xbf"},
|
{0x10ffff, "\xf4\x8f\xbf\xbf"},
|
||||||
{0xFFFD, "\xef\xbf\xbd"},
|
{0xFFFD, "\xef\xbf\xbd"},
|
||||||
@ -228,6 +232,93 @@ func TestIntConversion(t *testing.T) {
|
|||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
|
var invalidSequenceTests = []string{
|
||||||
|
"\xed\xa0\x80\x80", // surrogate min
|
||||||
|
"\xed\xbf\xbf\x80", // surrogate max
|
||||||
|
|
||||||
|
// xx
|
||||||
|
"\x91\x80\x80\x80",
|
||||||
|
|
||||||
|
// s1
|
||||||
|
"\xC2\x7F\x80\x80",
|
||||||
|
"\xC2\xC0\x80\x80",
|
||||||
|
"\xDF\x7F\x80\x80",
|
||||||
|
"\xDF\xC0\x80\x80",
|
||||||
|
|
||||||
|
// s2
|
||||||
|
"\xE0\x9F\xBF\x80",
|
||||||
|
"\xE0\xA0\x7F\x80",
|
||||||
|
"\xE0\xBF\xC0\x80",
|
||||||
|
"\xE0\xC0\x80\x80",
|
||||||
|
|
||||||
|
// s3
|
||||||
|
"\xE1\x7F\xBF\x80",
|
||||||
|
"\xE1\x80\x7F\x80",
|
||||||
|
"\xE1\xBF\xC0\x80",
|
||||||
|
"\xE1\xC0\x80\x80",
|
||||||
|
|
||||||
|
//s4
|
||||||
|
"\xED\x7F\xBF\x80",
|
||||||
|
"\xED\x80\x7F\x80",
|
||||||
|
"\xED\x9F\xC0\x80",
|
||||||
|
"\xED\xA0\x80\x80",
|
||||||
|
|
||||||
|
// s5
|
||||||
|
"\xF0\x8F\xBF\xBF",
|
||||||
|
"\xF0\x90\x7F\xBF",
|
||||||
|
"\xF0\x90\x80\x7F",
|
||||||
|
"\xF0\xBF\xBF\xC0",
|
||||||
|
"\xF0\xBF\xC0\x80",
|
||||||
|
"\xF0\xC0\x80\x80",
|
||||||
|
|
||||||
|
// s6
|
||||||
|
"\xF1\x7F\xBF\xBF",
|
||||||
|
"\xF1\x80\x7F\xBF",
|
||||||
|
"\xF1\x80\x80\x7F",
|
||||||
|
"\xF1\xBF\xBF\xC0",
|
||||||
|
"\xF1\xBF\xC0\x80",
|
||||||
|
"\xF1\xC0\x80\x80",
|
||||||
|
|
||||||
|
// s7
|
||||||
|
"\xF4\x7F\xBF\xBF",
|
||||||
|
"\xF4\x80\x7F\xBF",
|
||||||
|
"\xF4\x80\x80\x7F",
|
||||||
|
"\xF4\x8F\xBF\xC0",
|
||||||
|
"\xF4\x8F\xC0\x80",
|
||||||
|
"\xF4\x90\x80\x80",
|
||||||
|
}
|
||||||
|
|
||||||
|
func runtimeDecodeRune(s string) rune {
|
||||||
|
for _, r := range s {
|
||||||
|
return r
|
||||||
|
}
|
||||||
|
return -1
|
||||||
|
}
|
||||||
|
|
||||||
|
func TestDecodeInvalidSequence(t *testing.T) {
|
||||||
|
for _, s := range invalidSequenceTests {
|
||||||
|
r1, _ := DecodeRune([]byte(s))
|
||||||
|
if want := RuneError; r1 != want {
|
||||||
|
t.Errorf("DecodeRune(%#x) = %#04x, want %#04x", s, r1, want)
|
||||||
|
return
|
||||||
|
}
|
||||||
|
r2, _ := DecodeRuneInString(s)
|
||||||
|
if want := RuneError; r2 != want {
|
||||||
|
t.Errorf("DecodeRuneInString(%q) = %#04x, want %#04x", s, r2, want)
|
||||||
|
return
|
||||||
|
}
|
||||||
|
if r1 != r2 {
|
||||||
|
t.Errorf("DecodeRune(%#x) = %#04x mismatch with DecodeRuneInString(%q) = %#04x", s, r1, s, r2)
|
||||||
|
return
|
||||||
|
}
|
||||||
|
r3 := runtimeDecodeRune(s)
|
||||||
|
if r2 != r3 {
|
||||||
|
t.Errorf("DecodeRuneInString(%q) = %#04x mismatch with runtime.decoderune(%q) = %#04x", s, r2, s, r3)
|
||||||
|
return
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
func testSequence(t *testing.T, s string) {
|
func testSequence(t *testing.T, s string) {
|
||||||
type info struct {
|
type info struct {
|
||||||
index int
|
index int
|
||||||
|
|||||||
Loading…
Reference in New Issue
Block a user