Skip to content
Projects
Groups
Snippets
Help
Loading...
Help
Support
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
G
go
Project overview
Project overview
Details
Activity
Releases
Repository
Repository
Files
Commits
Branches
Tags
Contributors
Graph
Compare
Issues
0
Issues
0
List
Boards
Labels
Milestones
Merge Requests
0
Merge Requests
0
Analytics
Analytics
Repository
Value Stream
Wiki
Wiki
Snippets
Snippets
Members
Members
Collapse sidebar
Close sidebar
Activity
Graph
Create a new issue
Commits
Issue Boards
Open sidebar
Kirill Smelkov
go
Commits
1f4d54ea
Commit
1f4d54ea
authored
Oct 28, 2010
by
Rob Pike
Browse files
Options
Browse Files
Download
Email Patches
Plain Diff
regexp: eliminate vector in favor of append.
R=rsc CC=golang-dev
https://golang.org/cl/2795041
parent
fafb116c
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
25 additions
and
29 deletions
+25
-29
src/pkg/regexp/regexp.go
src/pkg/regexp/regexp.go
+25
-29
No files found.
src/pkg/regexp/regexp.go
View file @
1f4d54ea
...
@@ -60,7 +60,6 @@ package regexp
...
@@ -60,7 +60,6 @@ package regexp
import
(
import
(
"bytes"
"bytes"
"container/vector"
"io"
"io"
"os"
"os"
"strings"
"strings"
...
@@ -117,7 +116,7 @@ type Regexp struct {
...
@@ -117,7 +116,7 @@ type Regexp struct {
expr
string
// the original expression
expr
string
// the original expression
prefix
string
// initial plain text string
prefix
string
// initial plain text string
prefixBytes
[]
byte
// initial plain text bytes
prefixBytes
[]
byte
// initial plain text bytes
inst
*
vector
.
Vecto
r
inst
[]
inst
r
start
instr
// first instruction of machine
start
instr
// first instruction of machine
prefixStart
instr
// where to start if there is a prefix
prefixStart
instr
// where to start if there is a prefix
nbra
int
// number of brackets in expression, for subexpressions
nbra
int
// number of brackets in expression, for subexpressions
...
@@ -190,8 +189,8 @@ func newChar(char int) *_Char {
...
@@ -190,8 +189,8 @@ func newChar(char int) *_Char {
type
_CharClass
struct
{
type
_CharClass
struct
{
common
common
negate
bool
// is character class negated? ([^a-z])
negate
bool
// is character class negated? ([^a-z])
//
vector
of int, stored pairwise: [a-z] is (a,z); x is (x,x):
//
slice
of int, stored pairwise: [a-z] is (a,z); x is (x,x):
ranges
*
vector
.
IntVector
ranges
[]
int
cmin
,
cmax
int
cmin
,
cmax
int
}
}
...
@@ -202,9 +201,9 @@ func (cclass *_CharClass) print() {
...
@@ -202,9 +201,9 @@ func (cclass *_CharClass) print() {
if
cclass
.
negate
{
if
cclass
.
negate
{
print
(
" (negated)"
)
print
(
" (negated)"
)
}
}
for
i
:=
0
;
i
<
cclass
.
ranges
.
Len
(
);
i
+=
2
{
for
i
:=
0
;
i
<
len
(
cclass
.
ranges
);
i
+=
2
{
l
:=
cclass
.
ranges
.
At
(
i
)
l
:=
cclass
.
ranges
[
i
]
r
:=
cclass
.
ranges
.
At
(
i
+
1
)
r
:=
cclass
.
ranges
[
i
+
1
]
if
l
==
r
{
if
l
==
r
{
print
(
" ["
,
string
(
l
),
"]"
)
print
(
" ["
,
string
(
l
),
"]"
)
}
else
{
}
else
{
...
@@ -215,8 +214,7 @@ func (cclass *_CharClass) print() {
...
@@ -215,8 +214,7 @@ func (cclass *_CharClass) print() {
func
(
cclass
*
_CharClass
)
addRange
(
a
,
b
int
)
{
func
(
cclass
*
_CharClass
)
addRange
(
a
,
b
int
)
{
// range is a through b inclusive
// range is a through b inclusive
cclass
.
ranges
.
Push
(
a
)
cclass
.
ranges
=
append
(
cclass
.
ranges
,
a
,
b
)
cclass
.
ranges
.
Push
(
b
)
if
a
<
cclass
.
cmin
{
if
a
<
cclass
.
cmin
{
cclass
.
cmin
=
a
cclass
.
cmin
=
a
}
}
...
@@ -229,7 +227,7 @@ func (cclass *_CharClass) matches(c int) bool {
...
@@ -229,7 +227,7 @@ func (cclass *_CharClass) matches(c int) bool {
if
c
<
cclass
.
cmin
||
c
>
cclass
.
cmax
{
if
c
<
cclass
.
cmin
||
c
>
cclass
.
cmax
{
return
cclass
.
negate
return
cclass
.
negate
}
}
ranges
:=
[]
int
(
*
cclass
.
ranges
)
ranges
:=
cclass
.
ranges
for
i
:=
0
;
i
<
len
(
ranges
);
i
=
i
+
2
{
for
i
:=
0
;
i
<
len
(
ranges
);
i
=
i
+
2
{
if
ranges
[
i
]
<=
c
&&
c
<=
ranges
[
i
+
1
]
{
if
ranges
[
i
]
<=
c
&&
c
<=
ranges
[
i
+
1
]
{
return
!
cclass
.
negate
return
!
cclass
.
negate
...
@@ -240,7 +238,7 @@ func (cclass *_CharClass) matches(c int) bool {
...
@@ -240,7 +238,7 @@ func (cclass *_CharClass) matches(c int) bool {
func
newCharClass
()
*
_CharClass
{
func
newCharClass
()
*
_CharClass
{
c
:=
new
(
_CharClass
)
c
:=
new
(
_CharClass
)
c
.
ranges
=
new
(
vector
.
IntVector
)
c
.
ranges
=
make
([]
int
,
0
,
4
)
c
.
cmin
=
0x10FFFF
+
1
// MaxRune + 1
c
.
cmin
=
0x10FFFF
+
1
// MaxRune + 1
c
.
cmax
=
-
1
c
.
cmax
=
-
1
return
c
return
c
...
@@ -298,8 +296,8 @@ func (nop *_Nop) kind() int { return _NOP }
...
@@ -298,8 +296,8 @@ func (nop *_Nop) kind() int { return _NOP }
func
(
nop
*
_Nop
)
print
()
{
print
(
"nop"
)
}
func
(
nop
*
_Nop
)
print
()
{
print
(
"nop"
)
}
func
(
re
*
Regexp
)
add
(
i
instr
)
instr
{
func
(
re
*
Regexp
)
add
(
i
instr
)
instr
{
i
.
setIndex
(
re
.
inst
.
Len
(
))
i
.
setIndex
(
len
(
re
.
inst
))
re
.
inst
.
Push
(
i
)
re
.
inst
=
append
(
re
.
inst
,
i
)
return
i
return
i
}
}
...
@@ -380,15 +378,15 @@ func (p *parser) charClass() instr {
...
@@ -380,15 +378,15 @@ func (p *parser) charClass() instr {
p
.
error
(
ErrBadRange
)
p
.
error
(
ErrBadRange
)
}
}
// Is it [^\n]?
// Is it [^\n]?
if
cc
.
negate
&&
cc
.
ranges
.
Len
(
)
==
2
&&
if
cc
.
negate
&&
len
(
cc
.
ranges
)
==
2
&&
cc
.
ranges
.
At
(
0
)
==
'\n'
&&
cc
.
ranges
.
At
(
1
)
==
'\n'
{
cc
.
ranges
[
0
]
==
'\n'
&&
cc
.
ranges
[
1
]
==
'\n'
{
nl
:=
new
(
_NotNl
)
nl
:=
new
(
_NotNl
)
p
.
re
.
add
(
nl
)
p
.
re
.
add
(
nl
)
return
nl
return
nl
}
}
// Special common case: "[a]" -> "a"
// Special common case: "[a]" -> "a"
if
!
cc
.
negate
&&
cc
.
ranges
.
Len
()
==
2
&&
cc
.
ranges
.
At
(
0
)
==
cc
.
ranges
.
At
(
1
)
{
if
!
cc
.
negate
&&
len
(
cc
.
ranges
)
==
2
&&
cc
.
ranges
[
0
]
==
cc
.
ranges
[
1
]
{
c
:=
newChar
(
cc
.
ranges
.
At
(
0
)
)
c
:=
newChar
(
cc
.
ranges
[
0
]
)
p
.
re
.
add
(
c
)
p
.
re
.
add
(
c
)
return
c
return
c
}
}
...
@@ -606,8 +604,7 @@ func unNop(i instr) instr {
...
@@ -606,8 +604,7 @@ func unNop(i instr) instr {
}
}
func
(
re
*
Regexp
)
eliminateNops
()
{
func
(
re
*
Regexp
)
eliminateNops
()
{
for
i
:=
0
;
i
<
re
.
inst
.
Len
();
i
++
{
for
_
,
inst
:=
range
re
.
inst
{
inst
:=
re
.
inst
.
At
(
i
)
.
(
instr
)
if
inst
.
kind
()
==
_END
{
if
inst
.
kind
()
==
_END
{
continue
continue
}
}
...
@@ -621,8 +618,7 @@ func (re *Regexp) eliminateNops() {
...
@@ -621,8 +618,7 @@ func (re *Regexp) eliminateNops() {
func
(
re
*
Regexp
)
dump
()
{
func
(
re
*
Regexp
)
dump
()
{
print
(
"prefix <"
,
re
.
prefix
,
">
\n
"
)
print
(
"prefix <"
,
re
.
prefix
,
">
\n
"
)
for
i
:=
0
;
i
<
re
.
inst
.
Len
();
i
++
{
for
_
,
inst
:=
range
re
.
inst
{
inst
:=
re
.
inst
.
At
(
i
)
.
(
instr
)
print
(
inst
.
index
(),
": "
)
print
(
inst
.
index
(),
": "
)
inst
.
print
()
inst
.
print
()
if
inst
.
kind
()
!=
_END
{
if
inst
.
kind
()
!=
_END
{
...
@@ -664,17 +660,17 @@ func (re *Regexp) setPrefix() {
...
@@ -664,17 +660,17 @@ func (re *Regexp) setPrefix() {
var
b
[]
byte
var
b
[]
byte
var
utf
=
make
([]
byte
,
utf8
.
UTFMax
)
var
utf
=
make
([]
byte
,
utf8
.
UTFMax
)
// First instruction is start; skip that.
// First instruction is start; skip that.
i
:=
re
.
inst
.
At
(
0
)
.
(
instr
)
.
next
()
.
index
()
i
:=
re
.
inst
[
0
]
.
next
()
.
index
()
Loop
:
Loop
:
for
i
<
re
.
inst
.
Len
(
)
{
for
i
<
len
(
re
.
inst
)
{
inst
:=
re
.
inst
.
At
(
i
)
.
(
instr
)
inst
:=
re
.
inst
[
i
]
// stop if this is not a char
// stop if this is not a char
if
inst
.
kind
()
!=
_CHAR
{
if
inst
.
kind
()
!=
_CHAR
{
break
break
}
}
// stop if this char can be followed by a match for an empty string,
// stop if this char can be followed by a match for an empty string,
// which includes closures, ^, and $.
// which includes closures, ^, and $.
switch
re
.
inst
.
At
(
inst
.
next
()
.
index
())
.
(
instr
)
.
kind
()
{
switch
re
.
inst
[
inst
.
next
()
.
index
()]
.
kind
()
{
case
_BOT
,
_EOT
,
_ALT
:
case
_BOT
,
_EOT
,
_ALT
:
break
Loop
break
Loop
}
}
...
@@ -683,7 +679,7 @@ Loop:
...
@@ -683,7 +679,7 @@ Loop:
i
=
inst
.
next
()
.
index
()
i
=
inst
.
next
()
.
index
()
}
}
// point prefixStart instruction to first non-CHAR after prefix
// point prefixStart instruction to first non-CHAR after prefix
re
.
prefixStart
=
re
.
inst
.
At
(
i
)
.
(
instr
)
re
.
prefixStart
=
re
.
inst
[
i
]
re
.
prefixBytes
=
b
re
.
prefixBytes
=
b
re
.
prefix
=
string
(
b
)
re
.
prefix
=
string
(
b
)
}
}
...
@@ -700,7 +696,7 @@ func Compile(str string) (regexp *Regexp, error os.Error) {
...
@@ -700,7 +696,7 @@ func Compile(str string) (regexp *Regexp, error os.Error) {
}
}
}()
}()
regexp
.
expr
=
str
regexp
.
expr
=
str
regexp
.
inst
=
new
(
vector
.
Vector
)
regexp
.
inst
=
make
([]
instr
,
0
,
10
)
regexp
.
doParse
()
regexp
.
doParse
()
return
return
}
}
...
@@ -830,8 +826,8 @@ func (a *matchArena) addState(s []state, inst instr, prefixed bool, match *match
...
@@ -830,8 +826,8 @@ func (a *matchArena) addState(s []state, inst instr, prefixed bool, match *match
// If bytes == nil, scan str.
// If bytes == nil, scan str.
func
(
re
*
Regexp
)
doExecute
(
str
string
,
bytestr
[]
byte
,
pos
int
)
[]
int
{
func
(
re
*
Regexp
)
doExecute
(
str
string
,
bytestr
[]
byte
,
pos
int
)
[]
int
{
var
s
[
2
][]
state
var
s
[
2
][]
state
s
[
0
]
=
make
([]
state
,
10
)[
0
:
0
]
s
[
0
]
=
make
([]
state
,
0
,
10
)
s
[
1
]
=
make
([]
state
,
10
)[
0
:
0
]
s
[
1
]
=
make
([]
state
,
0
,
10
)
in
,
out
:=
0
,
1
in
,
out
:=
0
,
1
var
final
state
var
final
state
found
:=
false
found
:=
false
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment