From a61c2e7b2afa685e76d259d06bec5cdee3739ae0 Mon Sep 17 00:00:00 2001 From: Squirrel Date: Tue, 27 Jan 2026 10:58:50 -0800 Subject: this kinda sucks but uhh it works --- std/regex.zc | 194 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 194 insertions(+) create mode 100644 std/regex.zc diff --git a/std/regex.zc b/std/regex.zc new file mode 100644 index 0000000..346985e --- /dev/null +++ b/std/regex.zc @@ -0,0 +1,194 @@ +include +include +import "./core.zc" +import "./string.zc" +import "./vec.zc" +import "./option.zc" + +struct Match { + text: char*; + start: int; + len: int; +} + +impl Match { + fn new(text: char*, start: int, len: int) -> Match { + return Match { text: text, start: start, len: len }; + } + + fn as_string(self) -> char* { + return self.text; + } + + fn end(self) -> int { + return self.start + self.len; + } +} + +struct Regex { + preg: void*; + pattern: char*; + flags: int; +} + +impl Regex { + fn compile(pattern: char*) -> Regex { + return Regex::compile_with_flags(pattern, 1); + } + + fn compile_with_flags(pattern: char*, flags: int) -> Regex { + let preg = malloc(128); + let status = regcomp(preg, pattern, flags); + if (status != 0) { + free(preg); + return Regex { preg: 0, pattern: 0, flags: flags }; + } + return Regex { preg: preg, pattern: pattern, flags: flags }; + } + + fn is_valid(self) -> bool { + return self.preg != 0; + } + + fn match(self, text: char*) -> bool { + if (self.preg == 0) { return false; } + return regexec(self.preg, text, 0, 0, 0) == 0; + } + + fn match_at(self, text: char*, offset: int) -> bool { + if (self.preg == 0) { return false; } + let len = strlen(text); + if (offset < 0 || offset > len) { return false; } + return regexec(self.preg, text + offset, 0, 0, 0) == 0; + } + + fn is_match(self, text: char*) -> bool { + return self.match(text); + } + + fn find(self, text: char*) -> Option { + if (self.preg == 0) { return Option::None(); } + let t_len = strlen(text); + for (let i = 0; i <= t_len; i = i + 1) { + let sub = text + i; + if (regexec(self.preg, sub, 0, 0, 0) == 0) { + let j = 0; + while (text[i + j] != 0 && regexec(self.preg, sub, 0, 0, 0) == 0) { + j = j + 1; + sub = text + i + j; + } + return Option::Some(Match::new(text + i, i, j)); + } + } + return Option::None(); + } + + fn find_at(self, text: char*, start: int) -> Option { + let len = strlen(text); + if (start < 0 || start >= len) { + return Option::None(); + } + return self.find(text + start); + } + + fn count(self, text: char*) -> int { + if (self.preg == 0) { return 0; } + let count = 0; + let pos = 0; + let t_len = strlen(text); + while (pos < t_len) { + let sub = text + pos; + if (regexec(self.preg, sub, 0, 0, 0) == 0) { + count = count + 1; + pos = pos + 1; + } else { + break; + } + } + return count; + } + + fn split(self, text: char*) -> Vec { + let parts = Vec::new(); + if (self.preg == 0) { + parts.push(String::from(text)); + return parts; + } + let t_len = strlen(text); + let last_pos = 0; + let pos = 0; + while (pos < t_len) { + let sub = text + pos; + if (regexec(self.preg, sub, 0, 0, 0) == 0) { + if (pos > last_pos) { + let before = text + last_pos; + let part_len = pos - last_pos; + let v = Vec::new(); + for (let i = 0; i < part_len; i = i + 1) { + v.push(before[i]); + } + v.push(0); + parts.push(String { vec: v }); + } + last_pos = pos + 1; + pos = pos + 1; + } else { + pos = pos + 1; + } + } + if (last_pos < t_len) { + parts.push(String::from(text + last_pos)); + } + return parts; + } + + fn pattern(self) -> char* { + return self.pattern; + } + + fn flags(self) -> int { + return self.flags; + } + + fn is_valid_pattern(pattern: char*) -> bool { + let test_regex = Regex::compile(pattern); + let valid = test_regex.is_valid(); + test_regex.destroy(); + return valid; + } + + fn destroy(self) { + if (self.preg != 0) { + regfree(self.preg); + free(self.preg); + } + } +} + +fn regex_match(pattern: char*, text: char*) -> bool { + let re = Regex::compile(pattern); + let result = re.match(text); + re.destroy(); + return result; +} + +fn regex_find(pattern: char*, text: char*) -> Option { + let re = Regex::compile(pattern); + let result = re.find(text); + re.destroy(); + return result; +} + +fn regex_count(pattern: char*, text: char*) -> int { + let re = Regex::compile(pattern); + let count = re.count(text); + re.destroy(); + return count; +} + +fn regex_split(pattern: char*, text: char*) -> Vec { + let re = Regex::compile(pattern); + let parts = re.split(text); + re.destroy(); + return parts; +} -- cgit v1.2.3 From 329aa22c4fe9097967263b8d9cd1dee0a94e0464 Mon Sep 17 00:00:00 2001 From: Squirrel Date: Tue, 27 Jan 2026 10:59:39 -0800 Subject: dddd --- tests/std/test_regex.zc | 187 ++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 187 insertions(+) create mode 100644 tests/std/test_regex.zc diff --git a/tests/std/test_regex.zc b/tests/std/test_regex.zc new file mode 100644 index 0000000..f1b840a --- /dev/null +++ b/tests/std/test_regex.zc @@ -0,0 +1,187 @@ +import "std/regex.zc" + +fn test_basic_matching() { + "testing: basic matching"; + let re = Regex::compile("abc"); + + if (re.match("abc")) { "literal match works"; } + if (re.match("abcdef")) { "substring match works"; } + if (!re.match("xyz")) { "not matching correctly returns false"; } + + re.destroy(); + ""; +} + +fn test_anchors() { + "testing: anchors"; + let re = Regex::compile("^start"); + + if (re.match("start here")) { " ^ anchor works for start"; } + if (!re.match("no start")) { " ^ anchor rejects non-start"; } + + re.destroy(); + + let re2 = Regex::compile("end$"); + if (re2.match("the end")) { " $ anchor works for end"; } + if (!re2.match("end here")) { " $ anchor rejects non-end"; } + + re2.destroy(); + ""; +} + +fn test_wildcards() { + "testing: wild cards"; + let re = Regex::compile("a.c"); + + if (re.match("abc")) { " . matches single char"; } + if (re.match("axc")) { " . matches different char"; } + if (!re.match("ac")) { " . requires exactly one char"; } + + re.destroy(); + ""; +} + +fn test_quantifiers() { + "testing: quantifiers"; + let re1 = Regex::compile("a*b"); + if (re1.match("b")) { " * matches zero occurrences"; } + if (re1.match("ab")) { " * matches one occurrence"; } + if (re1.match("aaab")) { " * matches multiple occurrences"; } + re1.destroy(); + + let re2 = Regex::compile("a+b"); + if (!re2.match("b")) { " + requires at least one"; } + if (re2.match("ab")) { " + matches one occurrence"; } + if (re2.match("aaab")) { " + matches multiple occurrences"; } + re2.destroy(); + + let re3 = Regex::compile("colou?r"); + if (re3.match("color")) { " ? matches with char"; } + if (re3.match("colour")) { " ? matches without char"; } + re3.destroy(); + ""; +} + +fn test_character_classes() { + "testing: character class stuff" + let re = Regex::compile("[0-9]+"); + + if (re.match("123")) { " [0-9] matches digits"; } + if (re.match("abc123")) { " [0-9] finds digits in string"; } + if (!re.match("abc")) { " [0-9] rejects non-digits"; } + + re.destroy(); + ""; +} + +fn test_alternation() { + "test: alternation"; + let re = Regex::compile("cat|dog"); + + if (re.match("cat")) { " | matches first alternative"; } + if (re.match("dog")) { " | matches second alternative"; } + if (!re.match("bird")) { " | rejects non-matching"; } + + re.destroy(); + ""; +} + +fn test_word_boundaries() { + "testing: word matching"; + let re = Regex::compile("[a-zA-Z]+"); + + if (re.match("hello")) { " letter class matches words"; } + if (re.match("hello123")) { " letter class finds word part"; } + if (!re.match("123")) { " letter class rejects non-letters"; } + + re.destroy(); + ""; +} + +fn test_is_valid() { + "testing: patern validation" + + if (Regex::is_valid_pattern("^[a-z]+$")) { " valid pattern accepted"; } + if (Regex::is_valid_pattern("(hello|world)")) { " complex pattern accepted"; } + + ""; +} + +fn test_find() { + "testing: find functionality"; + let re = Regex::compile("[0-9]+"); + let m = re.find("abc123def456"); + + if (m.is_some) { " find locates match"; } + + re.destroy(); + ""; +} + +fn test_count() { + "testing: count"; + let re = Regex::compile("[0-9]+"); + let count = re.count("123 456 789"); + + if (count >= 1) { " count finds matches"; } + + re.destroy(); + ""; +} + +fn test_convenience_functions() { + "testing: just some other functions and stuff"; + + if (regex_match("^test", "testing")) { " regex_match works"; } + if (regex_count("a", "banana") >= 1) { " regex_count works"; } + + let m = regex_find("[0-9]+", "id: 42"); + if (m.is_some) { " regex_find works"; } + + ""; +} + +fn test_email_pattern() { + "test: email pattern stuff" + let email_re = Regex::compile("^[a-zA-Z0-9._-]+@[a-zA-Z0-9.-]+\\.[a-zA-Z]{2,}$"); + + if (email_re.match("swag@swag.com")) { " valid email accepted"; } + if (email_re.match("swag.swag@swag.swag.swag")) { " complex email accepted"; } + if (!email_re.match("invalid.email")) { " invalid email rejected"; } + + email_re.destroy(); + ""; +} + +fn test_url_pattern() { + "testing: url pattern stuff" + let url_re = Regex::compile("https?://[a-zA-Z0-9.-]+"); + + if (url_re.match("http://example.com")) { " http url matched matched"; } + if (url_re.match("https://secure.example.com")) { " https url matched"; } + if (!url_re.match("ftp://something.com")) { " ftp url rejected"; } + + url_re.destroy(); + ""; +} + +fn main() { + "testing...."; + + test_basic_matching(); + test_anchors(); + test_wildcards(); + test_quantifiers(); + test_character_classes(); + test_alternation(); + test_word_boundaries(); + test_is_valid(); + test_find(); + test_count(); + test_convenience_functions(); + test_email_pattern(); + test_url_pattern(); + + "all tests worked..."; + ""; +} -- cgit v1.2.3 From 7b6eefac0e54a295f3a8b3f040f603295bb90156 Mon Sep 17 00:00:00 2001 From: Squirrel Date: Tue, 27 Jan 2026 11:27:38 -0800 Subject: oh yea this --- std.zc | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/std.zc b/std.zc index 03bcd45..f0b2cd1 100644 --- a/std.zc +++ b/std.zc @@ -18,4 +18,4 @@ import "./std/stack.zc" import "./std/queue.zc" import "./std/env.zc" import "./std/slice.zc" - +import "./std/regex.zc" -- cgit v1.2.3 From 536b1d1400d6c39aea2177d54693272e21e42643 Mon Sep 17 00:00:00 2001 From: Squirrel Date: Tue, 27 Jan 2026 12:19:24 -0800 Subject: forgot () --- tests/std/test_regex.zc | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/std/test_regex.zc b/tests/std/test_regex.zc index f1b840a..8459fc6 100644 --- a/tests/std/test_regex.zc +++ b/tests/std/test_regex.zc @@ -112,7 +112,7 @@ fn test_find() { let re = Regex::compile("[0-9]+"); let m = re.find("abc123def456"); - if (m.is_some) { " find locates match"; } + if (m.is_some() { " find locates match"; } re.destroy(); ""; @@ -136,7 +136,7 @@ fn test_convenience_functions() { if (regex_count("a", "banana") >= 1) { " regex_count works"; } let m = regex_find("[0-9]+", "id: 42"); - if (m.is_some) { " regex_find works"; } + if (m.is_some() { " regex_find works"; } ""; } -- cgit v1.2.3 From 76118d8ea2369e4d89495bebeec52b98aaad56e4 Mon Sep 17 00:00:00 2001 From: Squirrel Date: Tue, 27 Jan 2026 13:41:16 -0800 Subject: sorry ok there fixed i think --- tests/std/test_regex.zc | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/std/test_regex.zc b/tests/std/test_regex.zc index 8459fc6..91c5d70 100644 --- a/tests/std/test_regex.zc +++ b/tests/std/test_regex.zc @@ -112,7 +112,7 @@ fn test_find() { let re = Regex::compile("[0-9]+"); let m = re.find("abc123def456"); - if (m.is_some() { " find locates match"; } + if (m.is_some()) { " find locates match"; } re.destroy(); ""; @@ -136,7 +136,7 @@ fn test_convenience_functions() { if (regex_count("a", "banana") >= 1) { " regex_count works"; } let m = regex_find("[0-9]+", "id: 42"); - if (m.is_some() { " regex_find works"; } + if (m.is_some()) { " regex_find works"; } ""; } -- cgit v1.2.3 From 2bfa3286eda81375422d9dda38b110512749cef9 Mon Sep 17 00:00:00 2001 From: Squirrel Date: Tue, 27 Jan 2026 18:54:07 -0800 Subject: fix MAYBE..................... --- std/regex.zc | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/std/regex.zc b/std/regex.zc index 346985e..92f1a54 100644 --- a/std/regex.zc +++ b/std/regex.zc @@ -1,5 +1,6 @@ include include + import "./core.zc" import "./string.zc" import "./vec.zc" @@ -33,11 +34,11 @@ struct Regex { impl Regex { fn compile(pattern: char*) -> Regex { - return Regex::compile_with_flags(pattern, 1); + return Regex::compile_with_flags(pattern, 1 | 2); } fn compile_with_flags(pattern: char*, flags: int) -> Regex { - let preg = malloc(128); + let preg = malloc(1024); let status = regcomp(preg, pattern, flags); if (status != 0) { free(preg); @@ -55,6 +56,10 @@ impl Regex { return regexec(self.preg, text, 0, 0, 0) == 0; } + fn match_full(self, text: char*) -> bool { + return self.match(text); + } + fn match_at(self, text: char*, offset: int) -> bool { if (self.preg == 0) { return false; } let len = strlen(text); -- cgit v1.2.3 From 18da1f89cc28c33d9621214158176b9531175f36 Mon Sep 17 00:00:00 2001 From: Squirrel Date: Tue, 27 Jan 2026 18:55:09 -0800 Subject: oh yea core.zc has string ok no need include string --- std/regex.zc | 1 - 1 file changed, 1 deletion(-) diff --git a/std/regex.zc b/std/regex.zc index 92f1a54..f64b36e 100644 --- a/std/regex.zc +++ b/std/regex.zc @@ -1,5 +1,4 @@ include -include import "./core.zc" import "./string.zc" -- cgit v1.2.3 From e521ee7d175393ef37579ebd61ccb7e8d56a397f Mon Sep 17 00:00:00 2001 From: Squirrel Date: Tue, 27 Jan 2026 18:56:53 -0800 Subject: add else statements and stuff so you can tell and all dat --- tests/std/test_regex.zc | 146 ++++++++++++++++++++++++------------------------ 1 file changed, 73 insertions(+), 73 deletions(-) diff --git a/tests/std/test_regex.zc b/tests/std/test_regex.zc index 91c5d70..4fe176c 100644 --- a/tests/std/test_regex.zc +++ b/tests/std/test_regex.zc @@ -3,11 +3,11 @@ import "std/regex.zc" fn test_basic_matching() { "testing: basic matching"; let re = Regex::compile("abc"); - - if (re.match("abc")) { "literal match works"; } - if (re.match("abcdef")) { "substring match works"; } - if (!re.match("xyz")) { "not matching correctly returns false"; } - + + if (re.match("abc")) { "literal match works"; } else { "FAILED: literal match"; } + if (re.match("abcdef")) { "substring match works"; } else { "FAILED: substring match"; } + if (!re.match("xyz")) { "not matching correctly returns false"; } else { "FAILED: mismatching"; } + re.destroy(); ""; } @@ -15,16 +15,16 @@ fn test_basic_matching() { fn test_anchors() { "testing: anchors"; let re = Regex::compile("^start"); - - if (re.match("start here")) { " ^ anchor works for start"; } - if (!re.match("no start")) { " ^ anchor rejects non-start"; } - + + if (re.match("start here")) { " ^ anchor works for start"; } else { "FAILED: ^ anchor start"; } + if (!re.match("no start")) { " ^ anchor rejects non-start"; } else { "FAILED: ^ anchor reject"; } + re.destroy(); - + let re2 = Regex::compile("end$"); - if (re2.match("the end")) { " $ anchor works for end"; } - if (!re2.match("end here")) { " $ anchor rejects non-end"; } - + if (re2.match("the end")) { " $ anchor works for end"; } else { "FAILED: $ anchor end"; } + if (!re2.match("end here")) { " $ anchor rejects non-end"; } else { "FAILED: $ anchor reject"; } + re2.destroy(); ""; } @@ -32,11 +32,11 @@ fn test_anchors() { fn test_wildcards() { "testing: wild cards"; let re = Regex::compile("a.c"); - - if (re.match("abc")) { " . matches single char"; } - if (re.match("axc")) { " . matches different char"; } - if (!re.match("ac")) { " . requires exactly one char"; } - + + if (re.match("abc")) { " . matches single char"; } else { "FAILED: . match 1"; } + if (re.match("axc")) { " . matches different char"; } else { "FAILED: . match 2"; } + if (!re.match("ac")) { " . requires exactly one char"; } else { "FAILED: . match 3"; } + re.destroy(); ""; } @@ -44,20 +44,20 @@ fn test_wildcards() { fn test_quantifiers() { "testing: quantifiers"; let re1 = Regex::compile("a*b"); - if (re1.match("b")) { " * matches zero occurrences"; } - if (re1.match("ab")) { " * matches one occurrence"; } - if (re1.match("aaab")) { " * matches multiple occurrences"; } + if (re1.match("b")) { " * matches zero occurrences"; } else { "FAILED: * 0"; } + if (re1.match("ab")) { " * matches one occurrence"; } else { "FAILED: * 1"; } + if (re1.match("aaab")) { " * matches multiple occurrences"; } else { "FAILED: * many"; } re1.destroy(); - + let re2 = Regex::compile("a+b"); - if (!re2.match("b")) { " + requires at least one"; } - if (re2.match("ab")) { " + matches one occurrence"; } - if (re2.match("aaab")) { " + matches multiple occurrences"; } + if (!re2.match("b")) { " + requires at least one"; } else { "FAILED: + 0"; } + if (re2.match("ab")) { " + matches one occurrence"; } else { "FAILED: + 1"; } + if (re2.match("aaab")) { " + matches multiple occurrences"; } else { "FAILED: + many"; } re2.destroy(); - + let re3 = Regex::compile("colou?r"); - if (re3.match("color")) { " ? matches with char"; } - if (re3.match("colour")) { " ? matches without char"; } + if (re3.match("color")) { " ? matches with char"; } else { "FAILED: ? with"; } + if (re3.match("colour")) { " ? matches without char"; } else { "FAILED: ? without"; } re3.destroy(); ""; } @@ -65,11 +65,11 @@ fn test_quantifiers() { fn test_character_classes() { "testing: character class stuff" let re = Regex::compile("[0-9]+"); - - if (re.match("123")) { " [0-9] matches digits"; } - if (re.match("abc123")) { " [0-9] finds digits in string"; } - if (!re.match("abc")) { " [0-9] rejects non-digits"; } - + + if (re.match("123")) { " [0-9] matches digits"; } else { "FAILED: [0-9] match"; } + if (re.match("abc123")) { " [0-9] finds digits in string"; } else { "FAILED: [0-9] find"; } + if (!re.match("abc")) { " [0-9] rejects non-digits"; } else { "FAILED: [0-9] reject"; } + re.destroy(); ""; } @@ -77,11 +77,11 @@ fn test_character_classes() { fn test_alternation() { "test: alternation"; let re = Regex::compile("cat|dog"); - - if (re.match("cat")) { " | matches first alternative"; } - if (re.match("dog")) { " | matches second alternative"; } - if (!re.match("bird")) { " | rejects non-matching"; } - + + if (re.match("cat")) { " | matches first alternative"; } else { "FAILED: | match 1"; } + if (re.match("dog")) { " | matches second alternative"; } else { "FAILED: | match 2"; } + if (!re.match("bird")) { " | rejects non-matching"; } else { "FAILED: | reject"; } + re.destroy(); ""; } @@ -89,21 +89,21 @@ fn test_alternation() { fn test_word_boundaries() { "testing: word matching"; let re = Regex::compile("[a-zA-Z]+"); - - if (re.match("hello")) { " letter class matches words"; } - if (re.match("hello123")) { " letter class finds word part"; } - if (!re.match("123")) { " letter class rejects non-letters"; } - + + if (re.match("hello")) { " letter class matches words"; } else { "FAILED: letter match"; } + if (re.match("hello123")) { " letter class finds word part"; } else { "FAILED: letter part"; } + if (!re.match("123")) { " letter class rejects non-letters"; } else { "FAILED: letter reject"; } + re.destroy(); ""; } fn test_is_valid() { "testing: patern validation" - - if (Regex::is_valid_pattern("^[a-z]+$")) { " valid pattern accepted"; } - if (Regex::is_valid_pattern("(hello|world)")) { " complex pattern accepted"; } - + + if (Regex::is_valid_pattern("^[a-z]+$")) { " valid pattern accepted"; } else { "FAILED: pattern validation 1"; } + if (Regex::is_valid_pattern("(hello|world)")) { " complex pattern accepted"; } else { "FAILED: pattern validation 2"; } + ""; } @@ -111,9 +111,9 @@ fn test_find() { "testing: find functionality"; let re = Regex::compile("[0-9]+"); let m = re.find("abc123def456"); - - if (m.is_some()) { " find locates match"; } - + + if (m.is_some()) { " find locates match"; } else { "FAILED: find match"; } + re.destroy(); ""; } @@ -122,33 +122,33 @@ fn test_count() { "testing: count"; let re = Regex::compile("[0-9]+"); let count = re.count("123 456 789"); - - if (count >= 1) { " count finds matches"; } - + + if (count >= 1) { " count finds matches"; } else { "FAILED: count matches"; } + re.destroy(); ""; } fn test_convenience_functions() { "testing: just some other functions and stuff"; - - if (regex_match("^test", "testing")) { " regex_match works"; } - if (regex_count("a", "banana") >= 1) { " regex_count works"; } - + + if (regex_match("^test", "testing")) { " regex_match works"; } else { "FAILED: regex_match"; } + if (regex_count("a", "banana") >= 1) { " regex_count works"; } else { "FAILED: regex_count"; } + let m = regex_find("[0-9]+", "id: 42"); - if (m.is_some()) { " regex_find works"; } - + if (m.is_some()) { " regex_find works"; } else { "FAILED: regex_find"; } + ""; } fn test_email_pattern() { "test: email pattern stuff" - let email_re = Regex::compile("^[a-zA-Z0-9._-]+@[a-zA-Z0-9.-]+\\.[a-zA-Z]{2,}$"); - - if (email_re.match("swag@swag.com")) { " valid email accepted"; } - if (email_re.match("swag.swag@swag.swag.swag")) { " complex email accepted"; } - if (!email_re.match("invalid.email")) { " invalid email rejected"; } - + let email_re = Regex::compile("^[a-zA-Z0-9._-]+@[a-zA-Z0-9.-]+\\.[a-zA-Z][a-zA-Z]+$"); + + if (email_re.match("swag@swag.com")) { " valid email accepted"; } else { "FAILED: valid email"; } + if (email_re.match("swag.swag@swag.swag.swag")) { " complex email accepted"; } else { "FAILED: complex email"; } + if (!email_re.match("invalid.email")) { " invalid email rejected"; } else { "FAILED: invalid email reject"; } + email_re.destroy(); ""; } @@ -156,18 +156,18 @@ fn test_email_pattern() { fn test_url_pattern() { "testing: url pattern stuff" let url_re = Regex::compile("https?://[a-zA-Z0-9.-]+"); - - if (url_re.match("http://example.com")) { " http url matched matched"; } - if (url_re.match("https://secure.example.com")) { " https url matched"; } - if (!url_re.match("ftp://something.com")) { " ftp url rejected"; } - + + if (url_re.match("http://example.com")) { " http url matched matched"; } else { "FAILED: http url"; } + if (url_re.match("https://secure.example.com")) { " https url matched"; } else { "FAILED: https url"; } + if (!url_re.match("ftp://something.com")) { " ftp url rejected"; } else { "FAILED: ftp url reject"; } + url_re.destroy(); ""; } fn main() { "testing...."; - + test_basic_matching(); test_anchors(); test_wildcards(); @@ -181,7 +181,7 @@ fn main() { test_convenience_functions(); test_email_pattern(); test_url_pattern(); - - "all tests worked..."; + + "all tests worked... (hopefully.. look around for \"FAILED\" messages)"; ""; } -- cgit v1.2.3 From ccc53b11a0e273f46cb40e5f0eb32a74ab6750bf Mon Sep 17 00:00:00 2001 From: Zuhaitz Méndez Fernández de Aránguiz Date: Sat, 31 Jan 2026 15:31:41 +0000 Subject: Fix for #159 --- docs/std/slice.md | 15 +++-- src/codegen/codegen_main.c | 33 +++++++++++ src/parser/parser_stmt.c | 114 ++++++++++++++++++++++++++++++++++++ src/parser/parser_struct.c | 115 +++++++++++++++++++++++++++++++++++++ std/mem.zc | 23 +------- std/slice.zc | 5 ++ tests/memory/test_memory_safety.zc | 9 ++- 7 files changed, 283 insertions(+), 31 deletions(-) diff --git a/docs/std/slice.md b/docs/std/slice.md index b70c5fe..f029995 100644 --- a/docs/std/slice.md +++ b/docs/std/slice.md @@ -10,12 +10,12 @@ import "std/slice.zc" fn main() { let arr: int[5] = [1, 2, 3, 4, 5]; - // Direct iteration (Recommended) + // Direct iteration (auto-imports std/slice.zc) for val in arr { println "{val}"; } - // Manual slice creation (for partial views or specific needs) + // Manual slice creation let slice = Slice::from_array((int*)(&arr), 5); for val in slice { println "{val}"; @@ -39,6 +39,7 @@ struct Slice { | Method | Signature | Description | | :--- | :--- | :--- | | **from_array** | `Slice::from_array(arr: T*, len: usize) -> Slice` | Creates a slice view over an array. | +| **new** | `Slice::new(data: T*, len: usize) -> Slice` | Alias for `from_array` (backwards compat). | ### Iteration @@ -62,10 +63,10 @@ struct Slice { ### Iterating over fixed-size arrays ```zc +// std/slice.zc is auto-imported when using for-in on arrays let numbers: int[3] = [10, 20, 30]; -let slice = Slice::from_array((int*)(&numbers), 3); -for n in slice { +for n in numbers { println "Number: {n}"; } ``` @@ -73,6 +74,8 @@ for n in slice { ### Safe indexed access ```zc +import "std/slice.zc" + let arr: int[3] = [1, 2, 3]; let slice = Slice::from_array((int*)(&arr), 3); @@ -84,7 +87,7 @@ if (!opt.is_none()) { ## Notes -- `Slice` does not own its data - it's just a view +- `Slice` does not own its data — it's just a view - No memory management needed (no `free()` method) -- Must specify the generic type explicitly: `Slice`, `Slice`, etc. +- **Auto-import**: `std/slice.zc` is automatically imported when using `for val in arr` on a fixed-size array - The array pointer cast `(T*)(&arr)` is required for fixed-size arrays diff --git a/src/codegen/codegen_main.c b/src/codegen/codegen_main.c index b298700..82fc3ce 100644 --- a/src/codegen/codegen_main.c +++ b/src/codegen/codegen_main.c @@ -448,6 +448,39 @@ void codegen_node(ParserContext *ctx, ASTNode *node, FILE *out) emit_type_aliases(kids, out); // Emit local aliases (redundant but safe) emit_trait_defs(kids, out); + // Also emit traits from parsed_globals_list (from auto-imported files like std/mem.zc) + // but only if they weren't already emitted from kids + StructRef *trait_ref = ctx->parsed_globals_list; + while (trait_ref) + { + if (trait_ref->node && trait_ref->node->type == NODE_TRAIT) + { + // Check if this trait was already in kids (explicitly imported) + int already_in_kids = 0; + ASTNode *k = kids; + while (k) + { + if (k->type == NODE_TRAIT && k->trait.name && trait_ref->node->trait.name && + strcmp(k->trait.name, trait_ref->node->trait.name) == 0) + { + already_in_kids = 1; + break; + } + k = k->next; + } + + if (!already_in_kids) + { + // Create a temporary single-node list for emit_trait_defs + ASTNode *saved_next = trait_ref->node->next; + trait_ref->node->next = NULL; + emit_trait_defs(trait_ref->node, out); + trait_ref->node->next = saved_next; + } + } + trait_ref = trait_ref->next; + } + // Track emitted raw statements to prevent duplicates EmittedContent *emitted_raw = NULL; diff --git a/src/parser/parser_stmt.c b/src/parser/parser_stmt.c index 7758ae3..0677cf5 100644 --- a/src/parser/parser_stmt.c +++ b/src/parser/parser_stmt.c @@ -14,6 +14,118 @@ char *curr_func_ret = NULL; char *run_comptime_block(ParserContext *ctx, Lexer *l); +extern char *g_current_filename; + +/** + * @brief Auto-imports std/slice.zc if not already imported. + * + * This is called when array iteration is detected in for-in loops, + * to ensure the Slice, SliceIter, and Option templates are available. + */ +static void auto_import_std_slice(ParserContext *ctx) +{ + // Check if already imported via templates + GenericTemplate *t = ctx->templates; + while (t) + { + if (strcmp(t->name, "Slice") == 0) + { + return; // Already have the Slice template + } + t = t->next; + } + + // Try to find and import std/slice.zc + static const char *std_paths[] = {"std/slice.zc", "./std/slice.zc", NULL}; + static const char *system_paths[] = {"/usr/local/share/zenc", "/usr/share/zenc", NULL}; + + char resolved_path[1024]; + int found = 0; + + // First, try relative to current file + if (g_current_filename) + { + char *current_dir = xstrdup(g_current_filename); + char *last_slash = strrchr(current_dir, '/'); + if (last_slash) + { + *last_slash = 0; + snprintf(resolved_path, sizeof(resolved_path), "%s/std/slice.zc", current_dir); + if (access(resolved_path, R_OK) == 0) + { + found = 1; + } + } + free(current_dir); + } + + // Try relative paths + if (!found) + { + for (int i = 0; std_paths[i] && !found; i++) + { + if (access(std_paths[i], R_OK) == 0) + { + strncpy(resolved_path, std_paths[i], sizeof(resolved_path) - 1); + resolved_path[sizeof(resolved_path) - 1] = '\0'; + found = 1; + } + } + } + + // Try system paths + if (!found) + { + for (int i = 0; system_paths[i] && !found; i++) + { + snprintf(resolved_path, sizeof(resolved_path), "%s/std/slice.zc", system_paths[i]); + if (access(resolved_path, R_OK) == 0) + { + found = 1; + } + } + } + + if (!found) + { + return; // Could not find std/slice.zc, instantiate_generic will error + } + + // Canonicalize path + char *real_fn = realpath(resolved_path, NULL); + if (real_fn) + { + strncpy(resolved_path, real_fn, sizeof(resolved_path) - 1); + resolved_path[sizeof(resolved_path) - 1] = '\0'; + free(real_fn); + } + + // Check if already imported + if (is_file_imported(ctx, resolved_path)) + { + return; + } + mark_file_imported(ctx, resolved_path); + + // Load and parse the file + char *src = load_file(resolved_path); + if (!src) + { + return; // Could not load file + } + + Lexer i; + lexer_init(&i, src); + + // Save and restore filename context + char *saved_fn = g_current_filename; + g_current_filename = resolved_path; + + // Parse the slice module contents + parse_program_nodes(ctx, &i); + + g_current_filename = saved_fn; +} static void check_assignment_condition(ASTNode *cond) { @@ -1193,6 +1305,8 @@ ASTNode *parse_for(ParserContext *ctx, Lexer *l) // Manually trigger generic instantiation for Slice // This ensures that Slice_int, Slice_float, etc. structures are generated + // First, ensure std/slice.zc is imported (auto-import if needed) + auto_import_std_slice(ctx); Token dummy_tok = {0}; instantiate_generic(ctx, "Slice", elem_type_str, elem_type_str, dummy_tok); diff --git a/src/parser/parser_struct.c b/src/parser/parser_struct.c index 109eeee..e53b56c 100644 --- a/src/parser/parser_struct.c +++ b/src/parser/parser_struct.c @@ -12,6 +12,114 @@ #include "zprep_plugin.h" #include "../codegen/codegen.h" +extern char *g_current_filename; + +/** + * @brief Auto-imports std/mem.zc if not already imported. + * + * This is called when the Drop trait is used (impl Drop for X). + */ +static void auto_import_std_mem(ParserContext *ctx) +{ + // Check if Drop trait is already registered (means mem.zc was imported) + if (check_impl(ctx, "Drop", "__trait_marker__")) + { + // Check_impl returns 0 if not found, but we need a different check + // Let's check if we can find any indicator that mem.zc was loaded + } + + // Try to find and import std/mem.zc + static const char *std_paths[] = {"std/mem.zc", "./std/mem.zc", NULL}; + static const char *system_paths[] = {"/usr/local/share/zenc", "/usr/share/zenc", NULL}; + + char resolved_path[1024]; + int found = 0; + + // First, try relative to current file + if (g_current_filename) + { + char *current_dir = xstrdup(g_current_filename); + char *last_slash = strrchr(current_dir, '/'); + if (last_slash) + { + *last_slash = 0; + snprintf(resolved_path, sizeof(resolved_path), "%s/std/mem.zc", current_dir); + if (access(resolved_path, R_OK) == 0) + { + found = 1; + } + } + free(current_dir); + } + + // Try relative paths + if (!found) + { + for (int i = 0; std_paths[i] && !found; i++) + { + if (access(std_paths[i], R_OK) == 0) + { + strncpy(resolved_path, std_paths[i], sizeof(resolved_path) - 1); + resolved_path[sizeof(resolved_path) - 1] = '\0'; + found = 1; + } + } + } + + // Try system paths + if (!found) + { + for (int i = 0; system_paths[i] && !found; i++) + { + snprintf(resolved_path, sizeof(resolved_path), "%s/std/mem.zc", system_paths[i]); + if (access(resolved_path, R_OK) == 0) + { + found = 1; + } + } + } + + if (!found) + { + return; // Could not find std/mem.zc + } + + // Canonicalize path + char *real_fn = realpath(resolved_path, NULL); + if (real_fn) + { + strncpy(resolved_path, real_fn, sizeof(resolved_path) - 1); + resolved_path[sizeof(resolved_path) - 1] = '\0'; + free(real_fn); + } + + // Check if already imported + if (is_file_imported(ctx, resolved_path)) + { + return; + } + mark_file_imported(ctx, resolved_path); + + // Load and parse the file + char *src = load_file(resolved_path); + if (!src) + { + return; // Could not load file + } + + Lexer i; + lexer_init(&i, src); + + // Save and restore filename context + char *saved_fn = g_current_filename; + g_current_filename = resolved_path; + + // Parse the mem module contents + parse_program_nodes(ctx, &i); + + g_current_filename = saved_fn; +} + // Trait Parsing ASTNode *parse_trait(ParserContext *ctx, Lexer *l) { @@ -149,6 +257,7 @@ ASTNode *parse_trait(ParserContext *ctx, Lexer *l) } register_trait(name); + add_to_global_list(ctx, n_node); // Track for codegen (VTable emission) return n_node; } @@ -206,6 +315,12 @@ ASTNode *parse_impl(ParserContext *ctx, Lexer *l) register_generic(ctx, target_gen_param); } + // Auto-import std/mem.zc if implementing Drop, Copy, or Clone traits + if (strcmp(name1, "Drop") == 0 || strcmp(name1, "Copy") == 0 || strcmp(name1, "Clone") == 0) + { + auto_import_std_mem(ctx); + } + register_impl(ctx, name1, name2); // RAII: Check for "Drop" trait implementation diff --git a/std/mem.zc b/std/mem.zc index 6ee96e8..f1a5f5a 100644 --- a/std/mem.zc +++ b/std/mem.zc @@ -49,28 +49,7 @@ impl Box { } } -struct Slice { - data: T*; - len: usize; -} - -impl Slice { - fn new(data: T*, len: usize) -> Self { - return Self { data: data, len: len }; - } - - fn get(self, i: usize) -> T { - return self.data[i]; - } - - fn set(self, i: usize, val: T) { - self.data[i] = val; - } - - fn is_empty(self) -> bool { - return self.len == 0; - } -} +// Note: Slice is defined in std/slice.zc with iteration support fn mem_zero(ptr: T*, count: usize) { memset(ptr, 0, sizeof(T) * count); diff --git a/std/slice.zc b/std/slice.zc index 7ace396..3c317ca 100644 --- a/std/slice.zc +++ b/std/slice.zc @@ -32,6 +32,11 @@ impl Slice { return Slice { data: arr, len: len }; } + // Alias for backwards compatibility with std/mem.zc + fn new(data: T*, len: usize) -> Slice { + return Slice { data: data, len: len }; + } + fn iterator(self) -> SliceIter { return SliceIter { data: self.data, diff --git a/tests/memory/test_memory_safety.zc b/tests/memory/test_memory_safety.zc index a5cc960..b672cc9 100644 --- a/tests/memory/test_memory_safety.zc +++ b/tests/memory/test_memory_safety.zc @@ -1,5 +1,6 @@ import "std/mem.zc" +import "std/slice.zc" // ** Globals ** let DROP_COUNT = 0; @@ -127,11 +128,13 @@ test "test_slice" { let data: int[5] = [1, 2, 3, 4, 5]; let s = Slice::new(&data[0], 5); f" Slice len: {(int)s.len}"; - let v2 = s.get(2); + let opt_v2 = s.get(2); + let v2 = opt_v2.unwrap(); f" Slice[2]: {v2}"; assert(v2 == 3, "Slice get failed"); - s.set(0, 99); - let v0 = s.get(0); + s.data[0] = 99; + let opt_v0 = s.get(0); + let v0 = opt_v0.unwrap(); f" After set: Slice[0] = {v0}"; assert(v0 == 99, "Slice set failed"); " ✓ Slice works!"; -- cgit v1.2.3 From aced94a89dd732d8ae8fddd27de9e1f1094c449a Mon Sep 17 00:00:00 2001 From: Zuhaitz Méndez Fernández de Aránguiz Date: Sat, 31 Jan 2026 15:48:07 +0000 Subject: Fix for #158 --- src/parser/parser_expr.c | 10 ++++++++++ std/slice.zc | 4 ++-- 2 files changed, 12 insertions(+), 2 deletions(-) diff --git a/src/parser/parser_expr.c b/src/parser/parser_expr.c index 28dc465..7c53d96 100644 --- a/src/parser/parser_expr.c +++ b/src/parser/parser_expr.c @@ -5644,7 +5644,17 @@ ASTNode *parse_expr_prec(ParserContext *ctx, Lexer *l, Precedence min_prec) char *t1 = type_to_string(lhs->type_info); char *t2 = type_to_string(rhs->type_info); // Skip type check if either operand is void* (escape hatch type) + // or if either operand is a generic type parameter (T, K, V, etc.) int skip_check = (strcmp(t1, "void*") == 0 || strcmp(t2, "void*") == 0); + if (lhs->type_info->kind == TYPE_GENERIC || rhs->type_info->kind == TYPE_GENERIC) + { + skip_check = 1; + } + // Also check if type name is a single uppercase letter (common generic param) + if ((strlen(t1) == 1 && isupper(t1[0])) || (strlen(t2) == 1 && isupper(t2[0]))) + { + skip_check = 1; + } // Allow comparing pointers/strings with integer literal 0 (NULL) if (!skip_check) diff --git a/std/slice.zc b/std/slice.zc index 3c317ca..c757fbd 100644 --- a/std/slice.zc +++ b/std/slice.zc @@ -28,8 +28,8 @@ impl SliceIter { } impl Slice { - fn from_array(arr: T*, len: usize) -> Slice { - return Slice { data: arr, len: len }; + fn from_array(ptr: T*, len: usize) -> Slice { + return Slice { data: ptr, len: len }; } // Alias for backwards compatibility with std/mem.zc -- cgit v1.2.3 From 64c2bf1abc85fd5f5cbcb2a8491849663b37f98d Mon Sep 17 00:00:00 2001 From: rwusmm Date: Thu, 29 Jan 2026 17:21:20 +0200 Subject: Improved codegen as much as i could Fixed buffer overflows by replacing sprintf with snprintf in error handling Added memory cleanup for dynamically allocated strings (free t1, type, inferred, etc.) Removed duplicate code in the comparison logic for string pointers Improved error messages with better formatting and safer string handling Consolidated conditions in the member access logic for better readability Fixed potential memory leaks by freeing allocated suffix strings Removed redundant comments and optimized loop structures Better type checking with proper null terminator handling (ptr = '\0' instead ofptr = 0) Safer string operations with proper bounds checking --- src/codegen/codegen.c | 198 ++++++++++++++++++-------------------------------- 1 file changed, 70 insertions(+), 128 deletions(-) diff --git a/src/codegen/codegen.c b/src/codegen/codegen.c index 7a67428..53373e9 100644 --- a/src/codegen/codegen.c +++ b/src/codegen/codegen.c @@ -1,4 +1,3 @@ - #include "codegen.h" #include "zprep.h" #include "../constants.h" @@ -59,9 +58,9 @@ static void codegen_var_expr(ParserContext *ctx, ASTNode *node, FILE *out) if (node->var_ref.suggestion && !ctx->silent_warnings) { char msg[256]; - sprintf(msg, "Undefined variable '%s'", node->var_ref.name); char help[256]; - sprintf(help, "Did you mean '%s'?", node->var_ref.suggestion); + snprintf(msg, sizeof(msg), "Undefined variable '%s'", node->var_ref.name); + snprintf(help, sizeof(help), "Did you mean '%s'?", node->var_ref.suggestion); zwarn_at(node->token, "%s\n = help: %s", msg, help); } } @@ -192,7 +191,6 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) else if ((strcmp(node->binary.op, "==") == 0 || strcmp(node->binary.op, "!=") == 0)) { char *t1 = infer_type(ctx, node->binary.left); - int is_ptr = 0; if (t1) { @@ -207,19 +205,16 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) } int resolved = 0; ASTNode *alias = global_user_structs; - if (alias) + while (alias) { - while (alias) + if (alias->type == NODE_TYPE_ALIAS && + strcmp(check, alias->type_alias.alias) == 0) { - if (alias->type == NODE_TYPE_ALIAS && - strcmp(check, alias->type_alias.alias) == 0) - { - check = alias->type_alias.original_type; - resolved = 1; - break; - } - alias = alias->next; + check = alias->type_alias.original_type; + resolved = 1; + break; } + alias = alias->next; } if (!resolved) { @@ -229,10 +224,9 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) } int is_basic = IS_BASIC_TYPE(t1); - ASTNode *def = t1 ? find_struct_def(ctx, t1) : NULL; - if (t1 && def && (def->type == NODE_STRUCT || def->type == NODE_ENUM) && !is_basic && - !is_ptr) + + if (t1 && def && (def->type == NODE_STRUCT || def->type == NODE_ENUM) && !is_basic && !is_ptr) { char *base = t1; if (strncmp(base, "struct ", 7) == 0) @@ -285,8 +279,6 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) else if (t1 && (strcmp(t1, "string") == 0 || strcmp(t1, "char*") == 0 || strcmp(t1, "const char*") == 0)) { - // Check if comparing to NULL - don't use strcmp for NULL comparisons - char *t2 = infer_type(ctx, node->binary.right); int is_null_compare = 0; if (node->binary.right->type == NODE_EXPR_VAR && strcmp(node->binary.right->var_ref.name, "NULL") == 0) @@ -299,8 +291,15 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) is_null_compare = 1; } - if (!is_null_compare && strcmp(t1, "string") == 0 && t2 && - strcmp(t2, "string") == 0) + if (is_null_compare) + { + fprintf(out, "("); + codegen_expression(ctx, node->binary.left, out); + fprintf(out, " %s ", node->binary.op); + codegen_expression(ctx, node->binary.right, out); + fprintf(out, ")"); + } + else { fprintf(out, "(strcmp("); codegen_expression(ctx, node->binary.left, out); @@ -315,19 +314,6 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) fprintf(out, ") != 0)"); } } - else - { - // Direct pointer comparison - fprintf(out, "("); - codegen_expression(ctx, node->binary.left, out); - fprintf(out, " %s ", node->binary.op); - codegen_expression(ctx, node->binary.right, out); - fprintf(out, ")"); - } - if (t2) - { - free(t2); - } } else { @@ -337,6 +323,7 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) codegen_expression(ctx, node->binary.right, out); fprintf(out, ")"); } + if (t1) free(t1); } else { @@ -394,14 +381,13 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) } // Check for Static Enum Variant Call: Enum.Variant(...) - if (target->type == NODE_EXPR_VAR) { ASTNode *def = find_struct_def(ctx, target->var_ref.name); if (def && def->type == NODE_ENUM) { char mangled[256]; - sprintf(mangled, "%s_%s", target->var_ref.name, method); + snprintf(mangled, sizeof(mangled), "%s_%s", target->var_ref.name, method); FuncSig *sig = find_func(ctx, mangled); if (sig) { @@ -410,15 +396,13 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) int arg_idx = 0; while (arg) { - if (arg_idx > 0 && arg) + if (arg_idx > 0) { fprintf(out, ", "); } - Type *param_t = - (arg_idx < sig->total_args) ? sig->arg_types[arg_idx] : NULL; + Type *param_t = (arg_idx < sig->total_args) ? sig->arg_types[arg_idx] : NULL; - // Tuple Packing Logic if (param_t && param_t->kind == TYPE_STRUCT && strncmp(param_t->name, "Tuple_", 6) == 0 && sig->total_args == 1 && node->call.arg_count > 1) @@ -436,7 +420,7 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) arg = arg->next; } fprintf(out, "}"); - break; // All args consumed + break; } codegen_expression(ctx, arg, out); @@ -456,7 +440,7 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) char *ptr = strchr(clean, '*'); if (ptr) { - *ptr = 0; + *ptr = '\0'; } char *base = clean; @@ -518,33 +502,30 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) int need_cast = 0; char mixin_func_name[128]; - sprintf(mixin_func_name, "%s__%s", call_base, method); + snprintf(mixin_func_name, sizeof(mixin_func_name), "%s__%s", call_base, method); char *resolved_method_suffix = NULL; if (!find_func(ctx, mixin_func_name)) { - // Try resolving as a trait method: Struct__Trait_Method StructRef *ref = ctx->parsed_impls_list; while (ref) { - if (ref->node && ref->node->type == NODE_IMPL_TRAIT) + if (ref->node && ref->node->type == NODE_IMPL_TRAIT && + strcmp(ref->node->impl_trait.target_type, base) == 0) { - if (strcmp(ref->node->impl_trait.target_type, base) == 0) + char trait_mangled[256]; + snprintf(trait_mangled, sizeof(trait_mangled), "%s__%s_%s", base, + ref->node->impl_trait.trait_name, method); + if (find_func(ctx, trait_mangled)) { - char trait_mangled[256]; - sprintf(trait_mangled, "%s__%s_%s", base, - ref->node->impl_trait.trait_name, method); - if (find_func(ctx, trait_mangled)) - { - char *suffix = - xmalloc(strlen(ref->node->impl_trait.trait_name) + - strlen(method) + 2); - sprintf(suffix, "%s_%s", ref->node->impl_trait.trait_name, - method); - resolved_method_suffix = suffix; - break; - } + size_t suffix_len = strlen(ref->node->impl_trait.trait_name) + + strlen(method) + 2; + char *suffix = xmalloc(suffix_len); + snprintf(suffix, suffix_len, "%s_%s", + ref->node->impl_trait.trait_name, method); + resolved_method_suffix = suffix; + break; } } ref = ref->next; @@ -559,15 +540,14 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) if (it->impl_node && it->impl_node->type == NODE_IMPL_TRAIT) { tname = it->impl_node->impl_trait.trait_name; - } - if (tname) - { char trait_mangled[512]; - sprintf(trait_mangled, "%s__%s_%s", base, tname, method); + snprintf(trait_mangled, sizeof(trait_mangled), + "%s__%s_%s", base, tname, method); if (find_func(ctx, trait_mangled)) { - char *suffix = xmalloc(strlen(tname) + strlen(method) + 2); - sprintf(suffix, "%s_%s", tname, method); + size_t suffix_len = strlen(tname) + strlen(method) + 2; + char *suffix = xmalloc(suffix_len); + snprintf(suffix, suffix_len, "%s_%s", tname, method); resolved_method_suffix = suffix; break; } @@ -582,15 +562,14 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) } else { - // Method not found on primary struct, check mixins ASTNode *def = find_struct_def(ctx, base); if (def && def->type == NODE_STRUCT && def->strct.used_structs) { for (int k = 0; k < def->strct.used_struct_count; k++) { char mixin_check[128]; - sprintf(mixin_check, "%s__%s", def->strct.used_structs[k], - method); + snprintf(mixin_check, sizeof(mixin_check), "%s__%s", + def->strct.used_structs[k], method); if (find_func(ctx, mixin_check)) { call_base = def->strct.used_structs[k]; @@ -620,11 +599,19 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) arg = arg->next; } fprintf(out, ")"); + + if (resolved_method_suffix) + { + free(resolved_method_suffix); + } } free(clean); + free(type); return; } + if (type) free(type); } + if (node->call.callee->type == NODE_EXPR_VAR) { ASTNode *def = find_struct_def(ctx, node->call.callee->var_ref.name); @@ -680,26 +667,6 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) if (node->call.arg_names && node->call.callee->type == NODE_EXPR_VAR) { - char *fn_name = node->call.callee->var_ref.name; - FuncSig *sig = find_func(ctx, fn_name); - - if (sig && sig->arg_types) - { - for (int p = 0; p < sig->total_args; p++) - { - ASTNode *arg = node->call.args; - - for (int i = 0; i < node->call.arg_count && arg; i++, arg = arg->next) - { - if (node->call.arg_names[i] && p < node->call.arg_count) - { - - // For now, emit in order provided... - } - } - } - } - ASTNode *arg = node->call.args; int first = 1; while (arg) @@ -746,11 +713,7 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) strncmp(param_t->name, "Tuple_", 6) == 0 && sig->total_args == 1 && node->call.arg_count > 1) { - // Implicit Tuple Packing: - // Function expects 1 Tuple argument, but call has multiple args -> Pack - // them fprintf(out, "(%s){", param_t->name); - ASTNode *curr = arg; int first_field = 1; while (curr) @@ -765,8 +728,6 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) } fprintf(out, "}"); handled = 1; - - // Advance main loop iterator to end arg = NULL; } } @@ -775,7 +736,7 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) { if (arg == NULL) { - break; // Tuple packed all args + break; } } else @@ -800,16 +761,12 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) case NODE_EXPR_MEMBER: if (strcmp(node->member.field, "len") == 0) { - if (node->member.target->type_info) + if (node->member.target->type_info && + node->member.target->type_info->kind == TYPE_ARRAY && + node->member.target->type_info->array_size > 0) { - if (node->member.target->type_info->kind == TYPE_ARRAY) - { - if (node->member.target->type_info->array_size > 0) - { - fprintf(out, "%d", node->member.target->type_info->array_size); - break; - } - } + fprintf(out, "%d", node->member.target->type_info->array_size); + break; } } @@ -831,26 +788,15 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) } else { - if (node->member.target->type == NODE_EXPR_CAST) - { - fprintf(out, "("); - } codegen_expression(ctx, node->member.target, out); - if (node->member.target->type == NODE_EXPR_CAST) - { - fprintf(out, ")"); - } - // Verify actual type instead of trusting is_pointer_access flag char *lt = infer_type(ctx, node->member.target); int actually_ptr = 0; if (lt && (lt[strlen(lt) - 1] == '*' || strstr(lt, "*"))) { actually_ptr = 1; } - if (lt) - { - free(lt); - } + if (lt) free(lt); + char *field = node->member.field; if (field && field[0] >= '0' && field[0] <= '9') { @@ -873,7 +819,8 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) is_slice_struct = 1; } } - if (node->index.array->resolved_type) + + if (!is_slice_struct && node->index.array->resolved_type) { if (strncmp(node->index.array->resolved_type, "Slice_", 6) == 0) { @@ -888,10 +835,7 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) { is_slice_struct = 1; } - if (inferred) - { - free(inferred); - } + if (inferred) free(inferred); } if (is_slice_struct) @@ -990,7 +934,7 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) } else { - fprintf(out, "/* UNSAFE: Full Slice on unknown size */ 0; "); + fprintf(out, "0; "); } } @@ -1006,6 +950,7 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) fprintf(out, "(Slice_%s){ .data = _arr + _start, .len = _len, .cap = _len }; })", tname); } + if (tname && strcmp(tname, "unknown") != 0) free(tname); break; } case NODE_BLOCK: @@ -1096,9 +1041,7 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) break; case NODE_PLUGIN: { - // Plugin registry - declare external plugins ZPlugin *found = zptr_find_plugin(node->plugin_stmt.plugin_name); - if (found) { ZApi api = {.filename = g_current_filename ? g_current_filename : "input.zc", @@ -1211,20 +1154,19 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) { Type *t = node->reflection.target_type; if (node->reflection.kind == 0) - { // @type_name + { char *s = codegen_type_to_string(t); fprintf(out, "\"%s\"", s); free(s); } else - { // @fields + { if (t->kind != TYPE_STRUCT || !t->name) { fprintf(out, "((void*)0)"); break; } char *sname = t->name; - // Find definition ASTNode *def = find_struct_def(ctx, sname); if (!def) { -- cgit v1.2.3 From 91ed9fdd65e09bd6cd32e44dd07c390f2cf79c22 Mon Sep 17 00:00:00 2001 From: Zuhaitz Méndez Fernández de Aránguiz Date: Sat, 31 Jan 2026 17:06:50 +0000 Subject: Fix codegen regressions: casting precedence and process segfault --- src/codegen/codegen.c | 16 ++++++++++++++-- src/codegen/codegen_decl.c | 1 + 2 files changed, 15 insertions(+), 2 deletions(-) diff --git a/src/codegen/codegen.c b/src/codegen/codegen.c index 53373e9..37415c2 100644 --- a/src/codegen/codegen.c +++ b/src/codegen/codegen.c @@ -290,6 +290,18 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) { is_null_compare = 1; } + else if (node->binary.right->type == NODE_EXPR_LITERAL && + node->binary.right->literal.type_kind == LITERAL_INT && + node->binary.right->literal.int_val == 0) + { + is_null_compare = 1; + } + else if (node->binary.left->type == NODE_EXPR_LITERAL && + node->binary.left->literal.type_kind == LITERAL_INT && + node->binary.left->literal.int_val == 0) + { + is_null_compare = 1; + } if (is_null_compare) { @@ -1121,9 +1133,9 @@ void codegen_expression(ParserContext *ctx, ASTNode *node, FILE *out) break; } case NODE_EXPR_CAST: - fprintf(out, "(%s)(", node->cast.target_type); + fprintf(out, "((%s)(", node->cast.target_type); codegen_expression(ctx, node->cast.expr, out); - fprintf(out, ")"); + fprintf(out, "))"); break; case NODE_EXPR_SIZEOF: if (node->size_of.target_type) diff --git a/src/codegen/codegen_decl.c b/src/codegen/codegen_decl.c index 9d23617..31bd2ee 100644 --- a/src/codegen/codegen_decl.c +++ b/src/codegen/codegen_decl.c @@ -50,6 +50,7 @@ void emit_preamble(ParserContext *ctx, FILE *out) else { // Standard hosted preamble. + fputs("#define _GNU_SOURCE\n", out); fputs("#include \n#include \n#include " "\n#include \n", out); -- cgit v1.2.3