mirror/otto
1
0
Fork 0
mirror of https://github.com/robertkrimen/otto synced 2025-09-28 18:45:22 +08:00
Code Issues Releases Wiki Activity

More (complex) RegExp => RE2 transforming and tests

Browse Source 
The unescaping of that which does not need escaping
This commit is contained in:
Robert Krimen 2012-10-20 18:06:04 -07:00
parent bf38778ba2
commit 68943ba1e4
5 changed files with 88 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
.test262/Makefile
View File

@ -45,3 +45,4 @@ try-tmp: tmp
look: .fail
cat $<
@echo `readlink $<`

3
.test262/cull
View File

@ -67,4 +67,7 @@ rm -f ch07/7.6/7.6.1/7.6.1-8-8.js
rm -f ch07/7.6/7.6.1/7.6.1-8-9.js
# Skip, no backreferencing in re2
rm -f ch07/7.8/7.8.5/S7.8.5_A1.4_T1.js
# We WANT to squawk on \1, \2, \3, ..., \9
# TODO Come up with replacement test?
rm -f ch07/7.8/7.8.5/S7.8.5_A1.4_T2.js
popd

2
Makefile
View File

@ -23,10 +23,10 @@ TEST := -v --run String_fromCharCode
TEST := -v --run Lexer\|Parse
TEST := -v --run Lexer
TEST := -v --run String_
TEST := -v --run RegExp
TEST := -v --run ParseSuccess
TEST := -v --run Parse
TEST := -v --run ParseFailure
TEST := -v --run RegExp
TEST := .
test: test-i

15
otto_test.go
View File

@ -57,6 +57,7 @@ func TestTransformRegExp(t *testing.T) {
Terst(t)
Is(transformRegExp(`\\|'|\r|\n|\t|\u2028|\u2029`), `\\|'|\r|\n|\t|\x{2028}|\x{2029}`)
Is(transformRegExp(`\x`), `x`)
}
func TestIsValidRegExp(t *testing.T) {
@ -1567,6 +1568,20 @@ func TestRegExp(t *testing.T) {
test(`/\a/.source`, "\\a")
test(`/\;/.source`, "\\;")
test(`/\ /.source`, "\\ ")
// Start sanity check...
test("eval(\"/abc/\").source", "abc")
test("eval(\"/\u0023/\").source", "#")
test("eval(\"/\u0058/\").source", "X")
test("eval(\"/\\\u0023/\").source == \"\\\u0023\"", "true")
test("'0x' + '0058'", "0x0058")
test("'\\\\' + '0x' + '0058'", "\\0x0058")
// ...stop sanity check
test(`abc = '\\' + String.fromCharCode('0x' + '0058'); eval('/' + abc + '/').source`, "\\X")
test(`abc = '\\' + String.fromCharCode('0x0058'); eval('/' + abc + '/').source == "\\\u0058"`, "true")
test(`abc = '\\' + String.fromCharCode('0x0023'); eval('/' + abc + '/').source == "\\\u0023"`, "true")
test(`abc = '\\' + String.fromCharCode('0x0078'); eval('/' + abc + '/').source == "\\\u0078"`, "true")
}
func TestNewFunction(t *testing.T) {

71
type_regexp.go
View File

@ -3,6 +3,7 @@ package otto
import (
"fmt"
"regexp"
"strings"
)
type _regExpObject struct {
@ -98,11 +99,75 @@ func execResultToArray(runtime *_runtime, target string, result []int) *_object
return runtime.newArray(valueArray)
}
/*var transformRegExp_matchSlashU *regexp.Regexp = regexp.MustCompile(`\\u([:xdigit:]{1-4})`)*/
var transformRegExp_matchSlashU *regexp.Regexp = regexp.MustCompile(`\\u([[:xdigit:]]{1,4})`)
// 0031,0032,0033,0034,0035,0036,0037,0038,0039 // 1 - 9
// 0043,0045,0046,0047,0048,0049,004A,004B,004C,004D,004E,004F
// 0050,0052,0054,0055,0056,0058,0059,005A
// 0063,0065,0067,0068,0069,006A,006B,006C,006D,006F
// 0070,0071,0075,0078,0079
// 0080,0081,0082,0083,0084,0085,0086,0087,0088,0089,008A,008B,008C,008D,008E,008F
// 0090,0091,0092,0093,0094,0095,0096,0097,0098,0099,009A,009B,009C,009D,009E,009F
// 00A0,00A1,00A2,00A3,00A4,00A5,00A6,00A7,00A8,00A9,00AA,00AB,00AC,00AD,00AE,00AF
// 00B0,00B1,00B2,00B3,00B4,00B5,00B6,00B7,00B8,00B9,00BA,00BB,00BC,00BD,00BE,00BF
// 00C0,00C1,00C2,00C3,00C4,00C5,00C6,00C7,00C8,00C9,00CA,00CB,00CC,00CD,00CE,00CF
// ...
// c = 63* c[A-Z]
// p = 70
// u = 75* u[:xdigit:]{4}
// x = 78* x[:xdigit:]{2}
//\x{0031}-\x{0039}
var transformRegExp_matchSlashU = regexp.MustCompile(`\\u([[:xdigit:]]{1,4})`)
var transformRegExp_unescape = regexp.MustCompile(strings.NewReplacer("\n", "", "\t", "", " ", "").Replace(`
(?:
\\(
[
\x{0043}\x{0045}-\x{004F}
\x{0050}\x{0052}\x{0054}-\x{0056}\x{0058}-\x{005A}
\x{0065}\x{0067}-\x{006D}\x{006F}
\x{0070}\x{0071}\x{0079}
\x{0080}-\x{FFFF}
]
)()
) |
(?:
\\(c)([^A-Z])
) |
(?:
\\(u)([^[:xdigit:]])
) |
(?:
\\(u)([:xdigit:][^[:xdigit:]])
) |
(?:
\\(u)([:xdigit:][:xdigit:][^[:xdigit:]])
) |
(?:
\\(u)([:xdigit:][:xdigit:][:xdigit:][^[:xdigit:]])
) |
(?:
\\(x)([^[:xdigit:]])
) |
(?:
\\(x)([:xdigit:][^[:xdigit:]])
)
`))
var transformRegExp_unescapeDollar = regexp.MustCompile(strings.NewReplacer("\n", "", "\t", "", " ", "").Replace(`
(?:
\\([cux])$
)
`))
// TODO Go "regexp" bug? Can't do: (?:)|(?:$)
func transformRegExp(ecmaRegExp string) (goRegExp string) {
return transformRegExp_matchSlashU.ReplaceAllString(ecmaRegExp, `\x{$1}`)
tmp := []byte(ecmaRegExp)
tmp = transformRegExp_unescape.ReplaceAll(tmp, []byte(`$1$2`))
tmp = transformRegExp_unescapeDollar.ReplaceAll(tmp, []byte(`$1`))
tmp = transformRegExp_matchSlashU.ReplaceAll(tmp, []byte(`\x{$1}`))
return string(tmp)
}
func isValidRegExp(ecmaRegExp string) bool {