vieter-v
/
v
2
0
Fork 0
Code Issues Pull Requests Releases Activity

regex: fix for issue #7227, more tests (#7296)

pull/7300/head
penguindark 2020-12-13 02:04:53 +01:00 committed by GitHub
parent 6bd70bc88a
commit 4324be9bd8
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 80 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

77
vlib/regex/regex.v
View File

@ -256,6 +256,9 @@ mut:
// OR flag for the token
next_is_or bool // true if the next token is an OR
// last_dot flag
last_dot bool
}
[inline]
@ -1192,16 +1195,44 @@ fn (mut re RE) impl_compile(in_txt string) (int,int) {
// Post processing
//******************************************
//
// manage ist_dot_char
//
// count ist_dot_char to set the size of the state stack
mut pc1 := 0
mut tmp_count := 0
mut last_dot_pc := -1
for pc1 < pc {
if re.prog[pc1].ist == ist_dot_char {
tmp_count++
last_dot_pc = pc1
//println("Found dot_char pc:[${last_dot_pc}]")
}
pc1++
}
// if exist set the last dot_char token to manage the last .*
if last_dot_pc >= 0 {
re.prog[last_dot_pc].last_dot = true
mut last_dot_flag := true
mut tmp_pc := last_dot_pc + 1
for tmp_pc < pc {
if re.prog[tmp_pc].ist !in [rune(ist_prog_end),ist_group_end] {
last_dot_flag = false
break
}
tmp_pc++
}
re.prog[last_dot_pc].last_dot = last_dot_flag
//println("Our last dot flag pc: ${last_dot_pc} flag: ${last_dot_flag}")
}
//******************************************
// init the state stack
re.state_stack = []StateDotObj{len: tmp_count+1, init: StateDotObj{}}
@ -1325,6 +1356,12 @@ pub fn (re RE) get_code() string {
res.write("?")
}
}
// last dot char flag
if tk.last_dot == true {
res.write(" Last dot_char!")
}
res.write("\n")
if stop_flag {
break
@ -1502,7 +1539,7 @@ pub fn (mut re RE) match_base(in_txt byteptr, in_txt_len int ) (int,int) {
mut pc := -1 // program counter
mut state := StateObj{} // actual state
mut ist := rune(0) // actual instruction
mut l_ist :=rune(0) // last matched instruction
mut l_ist := rune(0) // last matched instruction
//mut group_stack := [-1].repeat(re.group_max)
//mut group_data := [-1].repeat(re.group_max)
@ -1581,6 +1618,9 @@ pub fn (mut re RE) match_base(in_txt byteptr, in_txt_len int ) (int,int) {
buf2.write("CHAR_CLASS_NEG[${re.get_char_class(pc)}]")
} else if ist == ist_dot_char {
buf2.write("DOT_CHAR")
if re.prog[pc].last_dot == true {
buf2.write(" Last dot_char!")
}
} else if ist == ist_group_start {
tmp_gi :=re.prog[pc].group_id
tmp_gr := re.prog[re.prog[pc].goto_pc].group_rep
@ -1638,6 +1678,11 @@ pub fn (mut re RE) match_base(in_txt byteptr, in_txt_len int ) (int,int) {
}
re.groups[g_index+1] = i
// manage last dot_char
if l_ist == ist_dot_char && re.prog[pc].last_dot == true {
re.groups[g_index+1]--
}
// continuous save, save until we have space
if re.group_csave_index > 0 {
// check if we have space to save the record
@ -1677,6 +1722,11 @@ pub fn (mut re RE) match_base(in_txt byteptr, in_txt_len int ) (int,int) {
return first_match,i
}
// manage last dot_char
if first_match >= 0 && l_ist == ist_dot_char && re.prog[pc].last_dot == true {
return first_match,i
}
//m_state = .end
//break
return no_match_found,0
@ -1798,6 +1848,7 @@ pub fn (mut re RE) match_base(in_txt byteptr, in_txt_len int ) (int,int) {
// check GROUP end
else if ist == ist_group_end {
// we are in matching streak
//println("Group END!! last ist: ${l_ist:08x}")
if state.match_index >= 0 {
// restore txt index stack and save the group data
@ -1816,9 +1867,14 @@ pub fn (mut re RE) match_base(in_txt byteptr, in_txt_len int ) (int,int) {
re.groups[g_index+1] = i
// if a group end with a dot, manage the not increased char index
/*
if i == re.groups[g_index] {
re.groups[g_index+1] = i+1
}
*/
if l_ist == ist_dot_char {
re.groups[g_index+1] = i+1
}
//println("GROUP ${re.prog[pc].group_id} END [${re.groups[g_index]}, ${re.groups[g_index+1]}]")
@ -1873,8 +1929,8 @@ pub fn (mut re RE) match_base(in_txt byteptr, in_txt_len int ) (int,int) {
state.match_index = i
re.prog[pc].rep++ // increase repetitions
//if re.prog[pc].rep >= re.prog[pc].rep_min && re.prog[pc].rep <= re.prog[pc].rep_max {
if re.prog[pc].rep >= 0 && re.prog[pc].rep <= re.prog[pc].rep_max {
if re.prog[pc].rep >= re.prog[pc].rep_min && re.prog[pc].rep <= re.prog[pc].rep_max {
//if re.prog[pc].rep >= 0 && re.prog[pc].rep <= re.prog[pc].rep_max {
//println("DOT CHAR save state : ${re.state_stack_index}")
// save the state
@ -1893,18 +1949,21 @@ pub fn (mut re RE) match_base(in_txt byteptr, in_txt_len int ) (int,int) {
}
if re.prog[pc].rep >= 1 && re.state_stack_index >= 0 {
//println("Save state char index.")
re.state_stack[re.state_stack_index].i = i + char_len
}
//i += char_len // next char
/*
// manage * and {0,} quantifier
if re.prog[pc].rep_max == max_quantifier {
//println("manage .*")
// manage last dot char
if re.prog[pc].last_dot == true
&& re.prog[pc].rep >= re.prog[pc].rep_min
&& re.prog[pc].rep <= re.prog[pc].rep_max
{
//println("We are the last dot_char in the query")
i += char_len
m_state = .ist_load
continue
}
*/
m_state = .ist_next
continue

14
vlib/regex/regex_test.v
View File

@ -112,6 +112,13 @@ match_test_suite = [
TestItem{"poth", r".(oth)|(eth)$",0,4},
TestItem{"poth", r"^.(oth)|(eth)$",0,4},
TestItem{"poth", r"^\w+$",0,4},
// test dot_char
TestItem{"8-11 l: qllllqllklhlvtl", r"^(\d+)-(\d+) ([a-z]): (.*)$",0,23},
TestItem{"accccb deer", r"^a(.*)b d(.+)r",0,11},
TestItem{"accccb deer", r"^a(.*)b d(.+)",0,11},
TestItem{"accccb deer", r"^(.*)$",0,11},
TestItem{"accccb deer", r"^a(.*)b d(.+)p",-1,0},
]
)
@ -167,7 +174,7 @@ struct TestItemCGroup {
q string
s int
e int
cg []int
cg []int // [number of items (3*# item), id_group_0, start_0, end_0, id_group_1, start1, start2,... ]
cgn map[string]int
}
const (
@ -181,7 +188,7 @@ cgroups_test_suite = [
TestItemCGroup{
"http://www.ciao.mondo/hello/pippo12_/pera.html",
r"(?P<format>https?)|(?P<format>ftps?)://(?P<token>[\w_]+.)+",0,46,
[2, 0, 0, 4, 1, 7, 10],
[8, 0, 0, 4, 1, 7, 12, 1, 11, 17, 1, 16, 23, 1, 22, 29, 1, 28, 38, 1, 37, 43, 1, 42, 46],
{'format':int(0),'token':1}
},
TestItemCGroup{
@ -247,7 +254,7 @@ fn test_regex(){
// check cgroups
if to.cgn.len > 0 {
if re.group_csave.len == 0 || re.group_csave[0] != to.cg[0] {
println("Capturing group len error! ${re.group_csave[0]}")
println("Capturing group len error! found: ${re.group_csave[0]} true ground: ${to.cg[0]}")
assert false
continue
}
@ -256,6 +263,7 @@ fn test_regex(){
mut ln := re.group_csave[0]*3
for ln > 0 {
if re.group_csave[ln] != to.cg[ln] {
println("Capturing group failed on $ln item!")
assert false
}
ln--