summaryrefslogtreecommitdiff
path: root/modules/regex
diff options
context:
space:
mode:
Diffstat (limited to 'modules/regex')
-rw-r--r--modules/regex/doc_classes/RegEx.xml13
-rw-r--r--modules/regex/regex.cpp229
2 files changed, 72 insertions, 170 deletions
diff --git a/modules/regex/doc_classes/RegEx.xml b/modules/regex/doc_classes/RegEx.xml
index c00fa96b2e..312275842a 100644
--- a/modules/regex/doc_classes/RegEx.xml
+++ b/modules/regex/doc_classes/RegEx.xml
@@ -11,7 +11,7 @@
regex.compile("\\w-(\\d+)")
[/codeblock]
The search pattern must be escaped first for GDScript before it is escaped for the expression. For example, [code]compile("\\d+")[/code] would be read by RegEx as [code]\d+[/code]. Similarly, [code]compile("\"(?:\\\\.|[^\"])*\"")[/code] would be read as [code]"(?:\\.|[^"])*"[/code].
- Using [method search] you can find the pattern within the given text. If a pattern is found, [RegExMatch] is returned and you can retrieve details of the results using functions such as [method RegExMatch.get_string] and [method RegExMatch.get_start].
+ Using [method search], you can find the pattern within the given text. If a pattern is found, [RegExMatch] is returned and you can retrieve details of the results using methods such as [method RegExMatch.get_string] and [method RegExMatch.get_start].
[codeblock]
var regex = RegEx.new()
regex.compile("\\w-(\\d+)")
@@ -19,7 +19,7 @@
if result:
print(result.get_string()) # Would print n-0123
[/codeblock]
- The results of capturing groups [code]()[/code] can be retrieved by passing the group number to the various functions in [RegExMatch]. Group 0 is the default and will always refer to the entire pattern. In the above example, calling [code]result.get_string(1)[/code] would give you [code]0123[/code].
+ The results of capturing groups [code]()[/code] can be retrieved by passing the group number to the various methods in [RegExMatch]. Group 0 is the default and will always refer to the entire pattern. In the above example, calling [code]result.get_string(1)[/code] would give you [code]0123[/code].
This version of RegEx also supports named capturing groups, and the names can be used to retrieve the results. If two or more groups have the same name, the name would only refer to the first one with a match.
[codeblock]
var regex = RegEx.new()
@@ -34,6 +34,15 @@
print(result.get_string("digit"))
# Would print 01 03 0 3f 42
[/codeblock]
+ [b]Example of splitting a string using a RegEx:[/b]
+ [codeblock]
+ var regex = RegEx.new()
+ regex.compile("\\S+") # Negated whitespace character class.
+ var results = []
+ for match in regex.search_all("One Two \n\tThree"):
+ results.push_back(match.get_string())
+ # The `results` array now contains "One", "Two", "Three".
+ [/codeblock]
[b]Note:[/b] Godot's regex implementation is based on the [url=https://www.pcre.org/]PCRE2[/url] library. You can view the full pattern reference [url=https://www.pcre.org/current/doc/html/pcre2pattern.html]here[/url].
[b]Tip:[/b] You can use [url=https://regexr.com/]Regexr[/url] to test regular expressions online.
</description>
diff --git a/modules/regex/regex.cpp b/modules/regex/regex.cpp
index 50ca01067b..c10a276eae 100644
--- a/modules/regex/regex.cpp
+++ b/modules/regex/regex.cpp
@@ -156,26 +156,13 @@ void RegExMatch::_bind_methods() {
}
void RegEx::_pattern_info(uint32_t what, void *where) const {
- if (sizeof(CharType) == 2) {
- pcre2_pattern_info_16((pcre2_code_16 *)code, what, where);
-
- } else {
- pcre2_pattern_info_32((pcre2_code_32 *)code, what, where);
- }
+ pcre2_pattern_info_32((pcre2_code_32 *)code, what, where);
}
void RegEx::clear() {
- if (sizeof(CharType) == 2) {
- if (code) {
- pcre2_code_free_16((pcre2_code_16 *)code);
- code = nullptr;
- }
-
- } else {
- if (code) {
- pcre2_code_free_32((pcre2_code_32 *)code);
- code = nullptr;
- }
+ if (code) {
+ pcre2_code_free_32((pcre2_code_32 *)code);
+ code = nullptr;
}
}
@@ -187,39 +174,20 @@ Error RegEx::compile(const String &p_pattern) {
PCRE2_SIZE offset;
uint32_t flags = PCRE2_DUPNAMES;
- if (sizeof(CharType) == 2) {
- pcre2_general_context_16 *gctx = (pcre2_general_context_16 *)general_ctx;
- pcre2_compile_context_16 *cctx = pcre2_compile_context_create_16(gctx);
- PCRE2_SPTR16 p = (PCRE2_SPTR16)pattern.c_str();
-
- code = pcre2_compile_16(p, pattern.length(), flags, &err, &offset, cctx);
+ pcre2_general_context_32 *gctx = (pcre2_general_context_32 *)general_ctx;
+ pcre2_compile_context_32 *cctx = pcre2_compile_context_create_32(gctx);
+ PCRE2_SPTR32 p = (PCRE2_SPTR32)pattern.get_data();
- pcre2_compile_context_free_16(cctx);
+ code = pcre2_compile_32(p, pattern.length(), flags, &err, &offset, cctx);
- if (!code) {
- PCRE2_UCHAR16 buf[256];
- pcre2_get_error_message_16(err, buf, 256);
- String message = String::num(offset) + ": " + String((const CharType *)buf);
- ERR_PRINT(message.utf8());
- return FAILED;
- }
+ pcre2_compile_context_free_32(cctx);
- } else {
- pcre2_general_context_32 *gctx = (pcre2_general_context_32 *)general_ctx;
- pcre2_compile_context_32 *cctx = pcre2_compile_context_create_32(gctx);
- PCRE2_SPTR32 p = (PCRE2_SPTR32)pattern.c_str();
-
- code = pcre2_compile_32(p, pattern.length(), flags, &err, &offset, cctx);
-
- pcre2_compile_context_free_32(cctx);
-
- if (!code) {
- PCRE2_UCHAR32 buf[256];
- pcre2_get_error_message_32(err, buf, 256);
- String message = String::num(offset) + ": " + String((const CharType *)buf);
- ERR_PRINT(message.utf8());
- return FAILED;
- }
+ if (!code) {
+ PCRE2_UCHAR32 buf[256];
+ pcre2_get_error_message_32(err, buf, 256);
+ String message = String::num(offset) + ": " + String((const char32_t *)buf);
+ ERR_PRINT(message.utf8());
+ return FAILED;
}
return OK;
}
@@ -234,69 +202,39 @@ Ref<RegExMatch> RegEx::search(const String &p_subject, int p_offset, int p_end)
length = p_end;
}
- if (sizeof(CharType) == 2) {
- pcre2_code_16 *c = (pcre2_code_16 *)code;
- pcre2_general_context_16 *gctx = (pcre2_general_context_16 *)general_ctx;
- pcre2_match_context_16 *mctx = pcre2_match_context_create_16(gctx);
- PCRE2_SPTR16 s = (PCRE2_SPTR16)p_subject.c_str();
-
- pcre2_match_data_16 *match = pcre2_match_data_create_from_pattern_16(c, gctx);
-
- int res = pcre2_match_16(c, s, length, p_offset, 0, match, mctx);
+ pcre2_code_32 *c = (pcre2_code_32 *)code;
+ pcre2_general_context_32 *gctx = (pcre2_general_context_32 *)general_ctx;
+ pcre2_match_context_32 *mctx = pcre2_match_context_create_32(gctx);
+ PCRE2_SPTR32 s = (PCRE2_SPTR32)p_subject.get_data();
- if (res < 0) {
- pcre2_match_data_free_16(match);
- return nullptr;
- }
-
- uint32_t size = pcre2_get_ovector_count_16(match);
- PCRE2_SIZE *ovector = pcre2_get_ovector_pointer_16(match);
-
- result->data.resize(size);
-
- for (uint32_t i = 0; i < size; i++) {
- result->data.write[i].start = ovector[i * 2];
- result->data.write[i].end = ovector[i * 2 + 1];
- }
-
- pcre2_match_data_free_16(match);
- pcre2_match_context_free_16(mctx);
-
- } else {
- pcre2_code_32 *c = (pcre2_code_32 *)code;
- pcre2_general_context_32 *gctx = (pcre2_general_context_32 *)general_ctx;
- pcre2_match_context_32 *mctx = pcre2_match_context_create_32(gctx);
- PCRE2_SPTR32 s = (PCRE2_SPTR32)p_subject.c_str();
+ pcre2_match_data_32 *match = pcre2_match_data_create_from_pattern_32(c, gctx);
- pcre2_match_data_32 *match = pcre2_match_data_create_from_pattern_32(c, gctx);
+ int res = pcre2_match_32(c, s, length, p_offset, 0, match, mctx);
- int res = pcre2_match_32(c, s, length, p_offset, 0, match, mctx);
-
- if (res < 0) {
- pcre2_match_data_free_32(match);
- pcre2_match_context_free_32(mctx);
+ if (res < 0) {
+ pcre2_match_data_free_32(match);
+ pcre2_match_context_free_32(mctx);
- return nullptr;
- }
+ return nullptr;
+ }
- uint32_t size = pcre2_get_ovector_count_32(match);
- PCRE2_SIZE *ovector = pcre2_get_ovector_pointer_32(match);
+ uint32_t size = pcre2_get_ovector_count_32(match);
+ PCRE2_SIZE *ovector = pcre2_get_ovector_pointer_32(match);
- result->data.resize(size);
+ result->data.resize(size);
- for (uint32_t i = 0; i < size; i++) {
- result->data.write[i].start = ovector[i * 2];
- result->data.write[i].end = ovector[i * 2 + 1];
- }
-
- pcre2_match_data_free_32(match);
- pcre2_match_context_free_32(mctx);
+ for (uint32_t i = 0; i < size; i++) {
+ result->data.write[i].start = ovector[i * 2];
+ result->data.write[i].end = ovector[i * 2 + 1];
}
+ pcre2_match_data_free_32(match);
+ pcre2_match_context_free_32(mctx);
+
result->subject = p_subject;
uint32_t count;
- const CharType *table;
+ const char32_t *table;
uint32_t entry_size;
_pattern_info(PCRE2_INFO_NAMECOUNT, &count);
@@ -304,7 +242,7 @@ Ref<RegExMatch> RegEx::search(const String &p_subject, int p_offset, int p_end)
_pattern_info(PCRE2_INFO_NAMEENTRYSIZE, &entry_size);
for (uint32_t i = 0; i < count; i++) {
- CharType id = table[i * entry_size];
+ char32_t id = table[i * entry_size];
if (result->data[id].start == -1) {
continue;
}
@@ -344,7 +282,7 @@ String RegEx::sub(const String &p_subject, const String &p_replacement, bool p_a
const int safety_zone = 1;
PCRE2_SIZE olength = p_subject.length() + 1; // space for output string and one terminating \0 character
- Vector<CharType> output;
+ Vector<char32_t> output;
output.resize(olength + safety_zone);
uint32_t flags = PCRE2_SUBSTITUTE_OVERFLOW_LENGTH;
@@ -357,55 +295,28 @@ String RegEx::sub(const String &p_subject, const String &p_replacement, bool p_a
length = p_end;
}
- if (sizeof(CharType) == 2) {
- pcre2_code_16 *c = (pcre2_code_16 *)code;
- pcre2_general_context_16 *gctx = (pcre2_general_context_16 *)general_ctx;
- pcre2_match_context_16 *mctx = pcre2_match_context_create_16(gctx);
- PCRE2_SPTR16 s = (PCRE2_SPTR16)p_subject.c_str();
- PCRE2_SPTR16 r = (PCRE2_SPTR16)p_replacement.c_str();
- PCRE2_UCHAR16 *o = (PCRE2_UCHAR16 *)output.ptrw();
+ pcre2_code_32 *c = (pcre2_code_32 *)code;
+ pcre2_general_context_32 *gctx = (pcre2_general_context_32 *)general_ctx;
+ pcre2_match_context_32 *mctx = pcre2_match_context_create_32(gctx);
+ PCRE2_SPTR32 s = (PCRE2_SPTR32)p_subject.get_data();
+ PCRE2_SPTR32 r = (PCRE2_SPTR32)p_replacement.get_data();
+ PCRE2_UCHAR32 *o = (PCRE2_UCHAR32 *)output.ptrw();
- pcre2_match_data_16 *match = pcre2_match_data_create_from_pattern_16(c, gctx);
-
- int res = pcre2_substitute_16(c, s, length, p_offset, flags, match, mctx, r, p_replacement.length(), o, &olength);
-
- if (res == PCRE2_ERROR_NOMEMORY) {
- output.resize(olength + safety_zone);
- o = (PCRE2_UCHAR16 *)output.ptrw();
- res = pcre2_substitute_16(c, s, length, p_offset, flags, match, mctx, r, p_replacement.length(), o, &olength);
- }
+ pcre2_match_data_32 *match = pcre2_match_data_create_from_pattern_32(c, gctx);
- pcre2_match_data_free_16(match);
- pcre2_match_context_free_16(mctx);
+ int res = pcre2_substitute_32(c, s, length, p_offset, flags, match, mctx, r, p_replacement.length(), o, &olength);
- if (res < 0) {
- return String();
- }
-
- } else {
- pcre2_code_32 *c = (pcre2_code_32 *)code;
- pcre2_general_context_32 *gctx = (pcre2_general_context_32 *)general_ctx;
- pcre2_match_context_32 *mctx = pcre2_match_context_create_32(gctx);
- PCRE2_SPTR32 s = (PCRE2_SPTR32)p_subject.c_str();
- PCRE2_SPTR32 r = (PCRE2_SPTR32)p_replacement.c_str();
- PCRE2_UCHAR32 *o = (PCRE2_UCHAR32 *)output.ptrw();
-
- pcre2_match_data_32 *match = pcre2_match_data_create_from_pattern_32(c, gctx);
-
- int res = pcre2_substitute_32(c, s, length, p_offset, flags, match, mctx, r, p_replacement.length(), o, &olength);
-
- if (res == PCRE2_ERROR_NOMEMORY) {
- output.resize(olength + safety_zone);
- o = (PCRE2_UCHAR32 *)output.ptrw();
- res = pcre2_substitute_32(c, s, length, p_offset, flags, match, mctx, r, p_replacement.length(), o, &olength);
- }
+ if (res == PCRE2_ERROR_NOMEMORY) {
+ output.resize(olength + safety_zone);
+ o = (PCRE2_UCHAR32 *)output.ptrw();
+ res = pcre2_substitute_32(c, s, length, p_offset, flags, match, mctx, r, p_replacement.length(), o, &olength);
+ }
- pcre2_match_data_free_32(match);
- pcre2_match_context_free_32(mctx);
+ pcre2_match_data_free_32(match);
+ pcre2_match_context_free_32(mctx);
- if (res < 0) {
- return String();
- }
+ if (res < 0) {
+ return String();
}
return String(output.ptr(), olength);
@@ -435,7 +346,7 @@ Array RegEx::get_names() const {
ERR_FAIL_COND_V(!is_valid(), result);
uint32_t count;
- const CharType *table;
+ const char32_t *table;
uint32_t entry_size;
_pattern_info(PCRE2_INFO_NAMECOUNT, &count);
@@ -453,39 +364,21 @@ Array RegEx::get_names() const {
}
RegEx::RegEx() {
- if (sizeof(CharType) == 2) {
- general_ctx = pcre2_general_context_create_16(&_regex_malloc, &_regex_free, nullptr);
-
- } else {
- general_ctx = pcre2_general_context_create_32(&_regex_malloc, &_regex_free, nullptr);
- }
+ general_ctx = pcre2_general_context_create_32(&_regex_malloc, &_regex_free, nullptr);
code = nullptr;
}
RegEx::RegEx(const String &p_pattern) {
- if (sizeof(CharType) == 2) {
- general_ctx = pcre2_general_context_create_16(&_regex_malloc, &_regex_free, nullptr);
-
- } else {
- general_ctx = pcre2_general_context_create_32(&_regex_malloc, &_regex_free, nullptr);
- }
+ general_ctx = pcre2_general_context_create_32(&_regex_malloc, &_regex_free, nullptr);
code = nullptr;
compile(p_pattern);
}
RegEx::~RegEx() {
- if (sizeof(CharType) == 2) {
- if (code) {
- pcre2_code_free_16((pcre2_code_16 *)code);
- }
- pcre2_general_context_free_16((pcre2_general_context_16 *)general_ctx);
-
- } else {
- if (code) {
- pcre2_code_free_32((pcre2_code_32 *)code);
- }
- pcre2_general_context_free_32((pcre2_general_context_32 *)general_ctx);
+ if (code) {
+ pcre2_code_free_32((pcre2_code_32 *)code);
}
+ pcre2_general_context_free_32((pcre2_general_context_32 *)general_ctx);
}
void RegEx::_bind_methods() {
generated by cgit v1.2.3 (git 2.25.1) at 2025-04-08 10:58:57 +0000