wader/fq
1
1
Fork 0
mirror of https://github.com/wader/fq.git synced 2024-12-24 05:41:38 +03:00
Code Issues Projects Releases Wiki Activity

interp: find buffer should always use ByteRuneReader

Browse Source 
Fixes grep and bgrep with buffer
This commit is contained in:
Mattias Wadman 2021-10-18 01:34:05 +02:00
parent d6d94842b5
commit 16d1f45de1
2 changed files with 16 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
pkg/interp/funcs.go
View File

@ -841,7 +841,7 @@ func (i *Interp) find(c interface{}, a []interface{}) gojq.Iter {
}
var re string
var flags string
var byteRunes bool
switch a0 := a[0].(type) {
case string:
@ -855,10 +855,11 @@ func (i *Interp) find(c interface{}, a []interface{}) gojq.Iter {
for _, b := range reBuf {
reRs = append(reRs, rune(b))
}
flags = "b"
byteRunes = true
re = string(reRs)
}
var flags string
if len(a) > 1 {
flags, ok = a[1].(string)
if !ok {
@ -866,6 +867,10 @@ func (i *Interp) find(c interface{}, a []interface{}) gojq.Iter {
}
}
if strings.Contains(flags, "b") {
byteRunes = true
}
// TODO: err to string
// TODO: extract to regexpextra? "all" FindReaderSubmatchIndex that can iter?
sre, err := gojqextra.CompileRegexp(re, "gimb", flags)
@ -885,7 +890,7 @@ func (i *Interp) find(c interface{}, a []interface{}) gojq.Iter {
// raw bytes regexp matching is a bit tricky, what we do is to read each byte as a codepoint (ByteRuneReader)
// and then we can use UTF-8 encoded codepoint to match a raw byte. So for example \u00ff (encoded as 0xc3 0xbf)
// will match the byte \0xff
if strings.Contains(flags, "b") {
if byteRunes {
// byte mode, read each byte as a rune
rr = ioextra.ByteRuneReader{RS: bb}
} else {

8
pkg/interp/testdata/grep.fqtest vendored
View File

@ -6,6 +6,10 @@ mp3> grep(44100, "ID", "^ID3$", "^ID.?$", "Info", "magic", "\u00ff", [0x49, 0x44
0xe0| 50 | P |.frames[1].header.sample_rate: 44100
|00 01 02 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f|0123456789abcdef|
0x1b0| 52 | R |.frames[2].header.sample_rate: 44100
|00 01 02 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f|0123456789abcdef|
0x1c0| 11 4b 36 4a 08 83 58 c9| .K6J..X.|.frames[2].data: 114b364a088358c920d4295298c8c8f9...
0x1d0|20 d4 29 52 98 c8 c8 f9 13 80 40 24 bc 91 23 42| .)R......@$..#B|
* |until 0x279.7 (178) | |
|00 01 02 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f|0123456789abcdef|
0x0|49 44 33 |ID3 |.headers[0].magic: "ID3" (Correct)
|00 01 02 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f|0123456789abcdef|
@ -35,6 +39,10 @@ mp3> fgrep(44100, "ID", "^ID3$", "^ID.?$", "Info", "magic", "\u00ff", [0x49, 0x4
|00 01 02 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f|0123456789abcdef|
0x0|49 44 33 |ID3 |.headers[0].magic: "ID3" (Correct)
mp3> bgrep(44100, "ID", "^ID3$", "^ID.?$", "Info", "magic", "\u00ff", [0x49, 0x44])
|00 01 02 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f|0123456789abcdef|
0x1c0| 11 4b 36 4a 08 83 58 c9| .K6J..X.|.frames[2].data: 114b364a088358c920d4295298c8c8f9...
0x1d0|20 d4 29 52 98 c8 c8 f9 13 80 40 24 bc 91 23 42| .)R......@$..#B|
* |until 0x279.7 (178) | |
|00 01 02 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f|0123456789abcdef|
0x0|49 44 33 |ID3 |.headers[0].magic: "ID3" (Correct)
|00 01 02 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f|0123456789abcdef|