@shd101wyy/yo 0.1.5 → 0.1.6

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@shd101wyy/yo",
   "displayName": "Yo",
-  "version": "0.1.5",
+  "version": "0.1.6",
   "main": "./out/cjs/index.cjs",
   "module": "./out/esm/index.mjs",
   "types": "./out/types/src/index.d.ts",

package/std/build.yo

@@ -57,6 +57,7 @@ CompilationTarget :: {
   X86_64_Linux_Gnu: "x86_64-linux-gnu",
   X86_64_Linux_Musl: "x86_64-linux-musl",
   Aarch64_Linux_Gnu: "aarch64-linux-gnu",
+  Aarch64_Linux_Musl: "aarch64-linux-musl",
   Aarch64_Macos: "aarch64-macos",
   X86_64_Macos: "x86_64-macos",
   X86_64_Windows_Msvc: "x86_64-windows-msvc",
@@ -80,7 +81,7 @@ Executable :: struct(
   root : comptime_string,
   (target : comptime_string) ?= __yo_build_target_host(),
   (optimize : Optimize) ?= Optimize.Debug,
-  (allocator : Allocator) ?= Allocator.Mimalloc,
+  (allocator : Allocator) ?= Allocator.Libc,
   (sanitize : Sanitize) ?= Sanitize.None
 );
 export Executable;

package/std/collections/array_list.yo

@@ -8,7 +8,7 @@
  * - RAII for automatic cleanup
  */
 { GlobalAllocator, AllocError } :: import "../allocator.yo";
-{ memmove } :: import "../libc/string.yo";
+{ memmove, memcpy, memset } :: import "../libc/string.yo";
 { malloc, calloc, realloc, free, aligned_alloc } :: GlobalAllocator;
 
 /**
@@ -415,12 +415,144 @@ impl(forall(T : Type), ArrayList(T),
     )
   ),
 
+  /**
+    * Ensure the ArrayList can hold at least `min_cap` total elements
+    * without further reallocation.
+    */
+  ensure_total_capacity : (fn(self: Self, min_cap: usize) -> unit)({
+    cond(
+      (min_cap <= self._capacity) => (),
+      true => {
+        new_capacity := cond(
+          (self._capacity == usize(0)) => min_cap,
+          true => {
+            cap := self._capacity;
+            while (cap < min_cap), {
+              cap = (cap * usize(2));
+            };
+            cap
+          }
+        );
+        new_some_ptr := match(self._ptr,
+          .None => GlobalAllocator.malloc((sizeof(T) * new_capacity)),
+          .Some(old_ptr) => GlobalAllocator.realloc(
+            .Some((*(void))(old_ptr)),
+            (sizeof(T) * new_capacity)
+          )
+        );
+        match(new_some_ptr,
+          .Some(new_ptr) => {
+            self._ptr = .Some((*(T))(new_ptr));
+            self._capacity = new_capacity;
+          },
+          .None => panic("ArrayList.ensure_total_capacity: allocation failed")
+        );
+      }
+    );
+  }),
+
+  /**
+    * Append `count` elements from a raw pointer using memcpy.
+    * The caller must ensure `src` points to at least `count` valid elements.
+    */
+  extend_from_ptr : (fn(self: Self, src: *(T), count: usize) -> unit)({
+    cond(
+      (count == usize(0)) => (),
+      true => {
+        self.ensure_total_capacity((self._length + count));
+        match(self._ptr,
+          .Some(dst_base) => {
+            dst := (*(void))((dst_base &+ self._length));
+            _ := memcpy(dst, (*(void))(src), (count * sizeof(T)));
+            self._length = (self._length + count);
+          },
+          .None => panic("ArrayList.extend_from_ptr: no ptr after ensure_total_capacity")
+        );
+      }
+    );
+  }),
+
   /**
     * Clear all elements but keep capacity
     */
   clear : (fn(self: Self) -> unit)({
     Self._free_elements(self);
     self._length = usize(0);
+  }),
+
+  /**
+    * Get element at index without bounds checking.
+    * Caller must ensure index < len.
+    */
+  get_unchecked : (fn(self: Self, index: usize) -> T)(
+    match(self._ptr,
+      .None => panic("ArrayList.get_unchecked: no ptr"),
+      .Some(_ptr) => (_ptr &+ index).*
+    )
+  ),
+
+  /**
+    * Get a pointer to element at index without bounds checking or copying.
+    * Caller must ensure index < len. The pointer is valid until the list is modified.
+    */
+  get_ptr : (fn(self: Self, index: usize) -> *(T))(
+    match(self._ptr,
+      .None => panic("ArrayList.get_ptr: no ptr"),
+      .Some(_ptr) => (_ptr &+ index)
+    )
+  ),
+
+  /**
+    * Set element at index without bounds checking.
+    * Caller must ensure index < len.
+    */
+  set_unchecked : (fn(self: Self, index: usize, value: T) -> unit)(
+    match(self._ptr,
+      .None => panic("ArrayList.set_unchecked: no ptr"),
+      .Some(_ptr) => {
+        target_ptr := (_ptr &+ index);
+        target_ptr.* = value;
+      }
+    )
+  ),
+
+  /**
+    * Fill all elements with a byte pattern using memset.
+    * Useful for zeroing bool/integer arrays in O(1).
+    * Only safe for types without RC (e.g., bool, u8, usize).
+    */
+  fill_with_byte : (fn(self: Self, byte_val: int) -> unit)(
+    match(self._ptr,
+      .None => (),
+      .Some(_ptr) => {
+        _ := memset((*(void))(_ptr), byte_val, (self._length * sizeof(T)));
+      }
+    )
+  ),
+
+  /**
+    * Resize ArrayList to exactly `new_len` elements, filling new slots with
+    * a byte pattern via memset. Does not call destructors on removed elements.
+    * Only safe for trivial types (bool, u8, usize, etc.).
+    */
+  resize_with_byte : (fn(self: Self, new_len: usize, byte_val: int) -> unit)({
+    cond(
+      (new_len <= self._length) => {
+        self._length = new_len;
+      },
+      true => {
+        self.ensure_total_capacity(new_len);
+        match(self._ptr,
+          .Some(_ptr) => {
+            start := (*(void))((_ptr &+ self._length));
+            fill_count := ((new_len - self._length) * sizeof(T));
+            _ := memset(start, byte_val, fill_count);
+            self._length = new_len;
+          },
+          .None => panic("ArrayList.resize_with_byte: no ptr after ensure")
+        );
+      }
+    );
   })
 );
 impl(forall(T : Type), ArrayList(T), Dispose(

package/std/encoding/html.yo

@@ -0,0 +1,283 @@
+// HTML entity decoding
+//
+// Decodes named (&), decimal (&), and hex (&) HTML character references.
+// Uses Legacy mode — entities without trailing semicolon are also decoded.
+//
+// Example:
+//   { decode_html } :: import "std/encoding/html";
+//
+//   result := decode_html(`& < & &`);
+//   assert((result == `& < & &`), "decoded entities");
+
+open import "../string";
+{ HashMap } :: import "../collections/hash_map";
+{ HashSet } :: import "../collections/hash_set";
+{ is_valid_entity_code, from_code_point } :: import "./html_char_utils";
+{ _build_entity_map, _build_legacy_set } :: import "./html_entities";
+
+// Module-level state: lazily initialized entity map and legacy set.
+_state_initialized := false;
+_entity_map := HashMap(String, String).new();
+_legacy_set := HashSet(String).new();
+
+_ensure_init :: (fn() -> unit)({
+  if(!((_state_initialized)), {
+    _entity_map = _build_entity_map();
+    _legacy_set = _build_legacy_set();
+    _state_initialized = true;
+  });
+});
+
+// Parse a hex string to i32
+_parse_hex :: (fn(s: String) -> i32)({
+  (result : i32) = i32(0);
+  (i : usize) = usize(0);
+  while ((i < s.len())), {
+    c := s.at(i).unwrap();
+    result = (result * i32(16));
+    if(((c >= rune(u32('0'))) && (c <= rune(u32('9')))), {
+      result = (result + (i32(c.to_u32()) - i32(48)));
+    }, if(((c >= rune(u32('a'))) && (c <= rune(u32('f')))), {
+      result = (result + ((i32(c.to_u32()) - i32(97)) + i32(10)));
+    }, if(((c >= rune(u32('A'))) && (c <= rune(u32('F')))), {
+      result = (result + ((i32(c.to_u32()) - i32(65)) + i32(10)));
+    })));
+    i = (i + usize(1));
+  };
+  result
+});
+
+// Parse a decimal string to i32
+_parse_dec :: (fn(s: String) -> i32)({
+  (result : i32) = i32(0);
+  (i : usize) = usize(0);
+  while ((i < s.len())), {
+    c := s.at(i).unwrap();
+    result = ((result * i32(10)) + (i32(c.to_u32()) - i32(48)));
+    i = (i + usize(1));
+  };
+  result
+});
+
+// Check if a character is an ASCII alphanumeric
+_is_alpha_numeric :: (fn(c: rune) -> bool)(
+  ((((c >= rune(u32('a'))) && (c <= rune(u32('z')))) || ((c >= rune(u32('A'))) && (c <= rune(u32('Z'))))) || ((c >= rune(u32('0'))) && (c <= rune(u32('9')))))
+);
+
+// Decode HTML entities in a string (Legacy mode — entities without ; are also decoded).
+decode_html :: (fn(input: String) -> String)({
+  _ensure_init();
+
+  (len : usize) = input.len();
+  if(((len == usize(0))), {
+    return input;
+  });
+
+  // Quick check: if no '&', return as-is
+  if(!(input.contains(`&`)), {
+    return input;
+  });
+
+  (result : String) = ``;
+  (i : usize) = usize(0);
+
+  while ((i < len)), {
+    c := input.at(i).unwrap();
+
+    if((c != rune(u32('&'))), {
+      // Not an entity start, just append the character
+      result = `${result}${from_code_point(i32(c.to_u32()))}`;
+      i = (i + usize(1));
+    }, {
+      // Found '&' — try to decode entity
+      (start : usize) = i;
+      i = (i + usize(1));
+
+      if(((i >= len)), {
+        result = `${result}&`;
+      }, {
+        next := input.at(i).unwrap();
+
+        if(((next == rune(u32('#')))), {
+          // Numeric entity: &#N; or &#xN;
+          i = (i + usize(1));
+          if(((i >= len)), {
+            result = `${result}&#`;
+          }, {
+            hex_char := input.at(i).unwrap();
+            if(((hex_char == rune(u32('x'))) || (hex_char == rune(u32('X')))), {
+              // Hex: &#xHH;
+              (digit_start : usize) = (i + usize(1));
+              (digit_end : usize) = digit_start;
+              while (((digit_end < len))), {
+                dc := input.at(digit_end).unwrap();
+                if(((((dc >= rune(u32('0'))) && (dc <= rune(u32('9')))) || (((dc >= rune(u32('a'))) && (dc <= rune(u32('f')))) || ((dc >= rune(u32('A'))) && (dc <= rune(u32('F'))))))), {
+                  digit_end = (digit_end + usize(1));
+                }, {
+                  // Done with hex digits, break out
+                  digit_end = (digit_end + usize(0));
+                  // Use a flag to break
+                  return_early := true;
+                  // TODO: proper break
+                  digit_end = (len + usize(1));
+                });
+              };
+              // Fix digit_end if it overflowed
+              if(((digit_end > len)), {
+                // We used the overflow trick - find actual end
+                digit_end = digit_start;
+                while ((digit_end < len)), {
+                  dc2 := input.at(digit_end).unwrap();
+                  if(((((dc2 >= rune(u32('0'))) && (dc2 <= rune(u32('9')))) || (((dc2 >= rune(u32('a'))) && (dc2 <= rune(u32('f')))) || ((dc2 >= rune(u32('A'))) && (dc2 <= rune(u32('F'))))))), {
+                    digit_end = (digit_end + usize(1));
+                  }, {
+                    digit_end = ((len + digit_end) + usize(1));
+                  });
+                };
+                if(((digit_end > len)), {
+                  digit_end = ((digit_end - len) - usize(1));
+                });
+              });
+
+              if(((digit_end > digit_start)), {
+                hex_str := input.substring(digit_start, digit_end);
+                (code : i32) = _parse_hex(hex_str);
+
+                // Check for semicolon
+                if((((digit_end < len) && (input.at(digit_end).unwrap() == rune(u32(';'))))), {
+                  i = (digit_end + usize(1));
+                }, {
+                  i = digit_end;
+                });
+
+                if(is_valid_entity_code(code), {
+                  result = `${result}${from_code_point(code)}`;
+                }, {
+                  // Invalid code (e.g., surrogates) — keep original entity text
+                  (orig_hex : String) = input.substring(start, i);
+                  result = `${result}${orig_hex}`;
+                });
+              }, {
+                // No hex digits — output literally
+                result = `${result}&#${from_code_point(i32(hex_char.to_u32()))}`;
+                i = (i + usize(1));
+              });
+            }, {
+              // Decimal: &#DD;
+              (digit_start : usize) = i;
+              (digit_end : usize) = digit_start;
+              while ((digit_end < len)), {
+                dc := input.at(digit_end).unwrap();
+                if((((dc >= rune(u32('0'))) && (dc <= rune(u32('9'))))), {
+                  digit_end = (digit_end + usize(1));
+                }, {
+                  digit_end = ((len + digit_end) + usize(1));
+                });
+              };
+              if(((digit_end > len)), {
+                digit_end = ((digit_end - len) - usize(1));
+              });
+
+              if(((digit_end > digit_start)), {
+                dec_str := input.substring(digit_start, digit_end);
+                (code : i32) = _parse_dec(dec_str);
+
+                // Check for semicolon
+                if((((digit_end < len) && (input.at(digit_end).unwrap() == rune(u32(';'))))), {
+                  i = (digit_end + usize(1));
+                }, {
+                  i = digit_end;
+                });
+
+                if(is_valid_entity_code(code), {
+                  result = `${result}${from_code_point(code)}`;
+                }, {
+                  // Invalid code (e.g., surrogates) — keep original entity text
+                  (orig_dec : String) = input.substring(start, i);
+                  result = `${result}${orig_dec}`;
+                });
+              }, {
+                // No decimal digits
+                result = `${result}&#`;
+              });
+            });
+          });
+        }, {
+          // Named entity: &name; or &name (legacy)
+          (name_start : usize) = i;
+          (name_end : usize) = name_start;
+          while ((name_end < len)), {
+            nc := input.at(name_end).unwrap();
+            if(((nc == rune(u32(';')))), {
+              // Found semicolon — end of entity name
+              name_end = ((len + name_end) + usize(1));
+            }, if(_is_alpha_numeric(nc), {
+              name_end = (name_end + usize(1));
+            }, {
+              // Non-alphanumeric, non-semicolon — end of potential entity
+              name_end = ((len + name_end) + usize(1));
+            }));
+          };
+          // Decode the overflow trick
+          (found_end : bool) = false;
+          if(((name_end > len)), {
+            name_end = ((name_end - len) - usize(1));
+            found_end = true;
+          });
+
+          name_str := input.substring(name_start, name_end);
+
+          // Check for semicolon at name_end
+          (has_semi : bool) = (((name_end < len) && (input.at(name_end).unwrap() == rune(u32(';')))));
+
+          if(has_semi, {
+            // Try exact match with semicolon
+            match(_entity_map.get(name_str),
+              .Some(decoded) => {
+                result = `${result}${decoded}`;
+                i = (name_end + usize(1));
+              },
+              .None => {
+                // Unknown entity — output literally
+                result = `${result}&${name_str};`;
+                i = (name_end + usize(1));
+              }
+            );
+          }, {
+            // Legacy mode: try progressively shorter names
+            (matched : bool) = false;
+            (try_end : usize) = name_end;
+
+            while ((((try_end > name_start) && !(matched)))), {
+              try_name := input.substring(name_start, try_end);
+              if(_legacy_set.contains(try_name), {
+                match(_entity_map.get(try_name),
+                  .Some(decoded) => {
+                    result = `${result}${decoded}`;
+                    i = try_end;
+                    matched = true;
+                  },
+                  .None => {
+                    try_end = (try_end - usize(1));
+                  }
+                );
+              }, {
+                try_end = (try_end - usize(1));
+              });
+            };
+
+            if(!(matched), {
+              // No legacy match — output '&' literally and continue
+              result = `${result}&`;
+              i = name_start;
+            });
+          });
+        });
+      });
+    });
+  };
+
+  result
+});
+
+export decode_html, is_valid_entity_code, from_code_point;

package/std/encoding/html_char_utils.yo

@@ -0,0 +1,36 @@
+// HTML character utility functions
+//
+// Provides Unicode codepoint validation and conversion for HTML entity processing.
+//
+// Example:
+//   { is_valid_entity_code, from_code_point } :: import "std/encoding/html_char_utils";
+//
+//   assert(is_valid_entity_code(i32(65)), "A is valid");
+//   s := from_code_point(i32(65));  // "A"
+
+open import "../string";
+
+// Check if a Unicode codepoint is a valid HTML entity value.
+is_valid_entity_code :: (fn(c: i32) -> bool)(
+  cond(
+    ((c >= i32(0xD800)) && (c <= i32(0xDFFF))) => false,
+    ((c >= i32(0xFDD0)) && (c <= i32(0xFDEF))) => false,
+    (((c & i32(0xFFFF)) == i32(0xFFFF)) || ((c & i32(0xFFFF)) == i32(0xFFFE))) => false,
+    ((c >= i32(0x00)) && (c <= i32(0x08))) => false,
+    (c == i32(0x0B)) => false,
+    ((c >= i32(0x0E)) && (c <= i32(0x1F))) => false,
+    ((c >= i32(0x7F)) && (c <= i32(0x9F))) => false,
+    (c > i32(0x10FFFF)) => false,
+    true => true
+  )
+);
+
+// Convert a Unicode codepoint to a String.
+from_code_point :: (fn(c: i32) -> String)(
+  {
+    (r : rune) = rune(u32(c));
+    `${r}`
+  }
+);
+
+export is_valid_entity_code, from_code_point;