type_registry.c

Go to the documentation of this file.

 
#include "common/infix_internals.h"
#include <ctype.h>
#include <string.h>
 
extern INFIX_TLS const char * g_infix_last_signature_context;
 
#define INITIAL_REGISTRY_BUCKETS 61
 
typedef struct resolve_memo_node_t {
    infix_type * src;                  
    struct resolve_memo_node_t * next; 
} resolve_memo_node_t;
 
// Hash Table Implementation
 
static uint64_t _registry_hash_string(const char * str) {
    uint64_t hash = 5381;
    int c;
    while ((c = *str++))
        hash = ((hash << 5) + hash) + c;  // hash * 33 + c
    return hash;
}
 
static _infix_registry_entry_t * _registry_lookup(infix_registry_t * registry, const char * name) {
    if (!registry || !name)
        return nullptr;
    size_t index = _registry_hash_string(name) % registry->num_buckets;
    // Traverse the linked list (chain) at the computed bucket index.
    for (_infix_registry_entry_t * current = registry->buckets[index]; current; current = current->next)
        if (strcmp(current->name, name) == 0)
            return current;
    return nullptr;
}
 
static _infix_registry_entry_t * _registry_insert(infix_registry_t * registry, const char * name) {
    size_t index = _registry_hash_string(name) % registry->num_buckets;
    _infix_registry_entry_t * new_entry =
        infix_arena_alloc(registry->arena, sizeof(_infix_registry_entry_t), _Alignof(_infix_registry_entry_t));
    if (!new_entry) {
        _infix_set_error(INFIX_CATEGORY_ALLOCATION, INFIX_CODE_OUT_OF_MEMORY, 0);
        return nullptr;
    }
    size_t name_len = strlen(name) + 1;
    char * name_copy = infix_arena_alloc(registry->arena, name_len, 1);
    if (!name_copy) {
        _infix_set_error(INFIX_CATEGORY_ALLOCATION, INFIX_CODE_OUT_OF_MEMORY, 0);
        return nullptr;
    }
    infix_memcpy((void *)name_copy, name, name_len);
    new_entry->name = name_copy;
    new_entry->type = nullptr;
    new_entry->is_forward_declaration = false;
    // Prepend to the linked list in the bucket.
    new_entry->next = registry->buckets[index];
    registry->buckets[index] = new_entry;
    registry->num_items++;
    return new_entry;
}
 
// Public API: Registry Lifecycle
 
c23_nodiscard infix_registry_t * infix_registry_create(void) {
    _infix_clear_error();
    infix_registry_t * registry = infix_malloc(sizeof(infix_registry_t));
    if (!registry) {
        _infix_set_error(INFIX_CATEGORY_ALLOCATION, INFIX_CODE_OUT_OF_MEMORY, 0);
        return nullptr;
    }
    registry->arena = infix_arena_create(16384);  // Default initial size
    if (!registry->arena) {
        infix_free(registry);
        _infix_set_error(INFIX_CATEGORY_ALLOCATION, INFIX_CODE_OUT_OF_MEMORY, 0);
        return nullptr;
    }
    registry->is_external_arena = false;  // Mark that this arena is owned by the registry.
    registry->num_buckets = INITIAL_REGISTRY_BUCKETS;
    registry->buckets = infix_arena_calloc(
        registry->arena, registry->num_buckets, sizeof(_infix_registry_entry_t *), _Alignof(_infix_registry_entry_t *));
    if (!registry->buckets) {
        infix_arena_destroy(registry->arena);
        infix_free(registry);
        _infix_set_error(INFIX_CATEGORY_ALLOCATION, INFIX_CODE_OUT_OF_MEMORY, 0);
        return nullptr;
    }
    registry->num_items = 0;
    return registry;
}
 
c23_nodiscard infix_registry_t * infix_registry_create_in_arena(infix_arena_t * arena) {
    _infix_clear_error();
    if (!arena) {
        _infix_set_error(INFIX_CATEGORY_GENERAL, INFIX_CODE_UNKNOWN, 0);
        return nullptr;
    }
    infix_registry_t * registry = infix_malloc(sizeof(infix_registry_t));
    if (!registry) {
        _infix_set_error(INFIX_CATEGORY_ALLOCATION, INFIX_CODE_OUT_OF_MEMORY, 0);
        return nullptr;
    }
    registry->arena = arena;
    registry->is_external_arena = true;  // Mark the arena as user-owned
    registry->num_buckets = INITIAL_REGISTRY_BUCKETS;
    registry->buckets = infix_arena_calloc(
        registry->arena, registry->num_buckets, sizeof(_infix_registry_entry_t *), _Alignof(_infix_registry_entry_t *));
    if (!registry->buckets) {
        infix_free(registry);
        _infix_set_error(INFIX_CATEGORY_ALLOCATION, INFIX_CODE_OUT_OF_MEMORY, 0);
        return nullptr;
    }
    registry->num_items = 0;
    return registry;
}
 
void infix_registry_destroy(infix_registry_t * registry) {
    if (!registry)
        return;
    // Only destroy the arena if it was created internally by `infix_registry_create`.
    if (!registry->is_external_arena)
        infix_arena_destroy(registry->arena);
    // Always free the registry struct itself.
    infix_free(registry);
}
 
// Internal Type Graph Resolution
 
static infix_status _resolve_type_graph_inplace_recursive(infix_arena_t * temp_arena,
                                                          infix_type ** type_ptr,
                                                          infix_registry_t * registry,
                                                          resolve_memo_node_t ** memo_head) {
    if (!type_ptr || !*type_ptr || !(*type_ptr)->is_arena_allocated)
        return INFIX_SUCCESS;
 
    infix_type * type = *type_ptr;
 
    // Cycle detection: If we've seen this node before, we're in a cycle.
    // Return success to break the loop.
    for (resolve_memo_node_t * node = *memo_head; node != nullptr; node = node->next)
        if (node->src == type)
            return INFIX_SUCCESS;
 
    // Allocate the memoization node from the stable temporary arena.
    resolve_memo_node_t * memo_node =
        infix_arena_alloc(temp_arena, sizeof(resolve_memo_node_t), _Alignof(resolve_memo_node_t));
    if (!memo_node) {
        _infix_set_error(INFIX_CATEGORY_ALLOCATION, INFIX_CODE_OUT_OF_MEMORY, 0);
        return INFIX_ERROR_ALLOCATION_FAILED;
    }
    memo_node->src = type;
    memo_node->next = *memo_head;
    *memo_head = memo_node;
 
    if (type->category == INFIX_TYPE_NAMED_REFERENCE) {
        if (!registry) {
            _infix_set_error(INFIX_CATEGORY_PARSER, INFIX_CODE_UNRESOLVED_NAMED_TYPE, 0);
            return INFIX_ERROR_INVALID_ARGUMENT;
        }
        const char * name = type->meta.named_reference.name;
        _infix_registry_entry_t * entry = _registry_lookup(registry, name);
        if (!entry || !entry->type) {
            _infix_set_error(INFIX_CATEGORY_PARSER, INFIX_CODE_UNRESOLVED_NAMED_TYPE, 0);
            return INFIX_ERROR_INVALID_ARGUMENT;
        }
        *type_ptr = entry->type;
        return INFIX_SUCCESS;
    }
 
    infix_status status = INFIX_SUCCESS;
    switch (type->category) {
    case INFIX_TYPE_POINTER:
        status = _resolve_type_graph_inplace_recursive(
            temp_arena, &type->meta.pointer_info.pointee_type, registry, memo_head);
        break;
    case INFIX_TYPE_ARRAY:
        status =
            _resolve_type_graph_inplace_recursive(temp_arena, &type->meta.array_info.element_type, registry, memo_head);
        break;
    case INFIX_TYPE_STRUCT:
    case INFIX_TYPE_UNION:
        for (size_t i = 0; i < type->meta.aggregate_info.num_members; ++i) {
            status = _resolve_type_graph_inplace_recursive(
                temp_arena, &type->meta.aggregate_info.members[i].type, registry, memo_head);
            if (status != INFIX_SUCCESS)
                break;
        }
        break;
    case INFIX_TYPE_REVERSE_TRAMPOLINE:
        status = _resolve_type_graph_inplace_recursive(
            temp_arena, &type->meta.func_ptr_info.return_type, registry, memo_head);
        if (status != INFIX_SUCCESS)
            break;
        for (size_t i = 0; i < type->meta.func_ptr_info.num_args; ++i) {
            status = _resolve_type_graph_inplace_recursive(
                temp_arena, &type->meta.func_ptr_info.args[i].type, registry, memo_head);
            if (status != INFIX_SUCCESS)
                break;
        }
        break;
    case INFIX_TYPE_ENUM:
        status = _resolve_type_graph_inplace_recursive(
            temp_arena, &type->meta.enum_info.underlying_type, registry, memo_head);
        break;
    case INFIX_TYPE_COMPLEX:
        status =
            _resolve_type_graph_inplace_recursive(temp_arena, &type->meta.complex_info.base_type, registry, memo_head);
        break;
    case INFIX_TYPE_VECTOR:
        status = _resolve_type_graph_inplace_recursive(
            temp_arena, &type->meta.vector_info.element_type, registry, memo_head);
        break;
    default:
        break;
    }
 
    return status;
}
 
c23_nodiscard infix_status _infix_resolve_type_graph_inplace(infix_type ** type_ptr, infix_registry_t * registry) {
    // Create a temporary arena solely for the visited list's lifetime.
    infix_arena_t * temp_arena = infix_arena_create(1024);
    if (!temp_arena) {
        _infix_set_error(INFIX_CATEGORY_ALLOCATION, INFIX_CODE_OUT_OF_MEMORY, 0);
        return INFIX_ERROR_ALLOCATION_FAILED;
    }
 
    resolve_memo_node_t * memo_head = nullptr;
    infix_status status = _resolve_type_graph_inplace_recursive(temp_arena, type_ptr, registry, &memo_head);
 
    infix_arena_destroy(temp_arena);
    return status;
}
 
// Public API: Type Registration
 
typedef struct {
    const char * p;     
    const char * start; 
} _registry_parser_state_t;
 
static void _registry_parser_skip_whitespace(_registry_parser_state_t * state) {
    while (1) {
        while (isspace((unsigned char)*state->p))
            state->p++;
        if (*state->p == '#')  // Skip comments
            while (*state->p != '\n' && *state->p != '\0')
                state->p++;
        else
            break;
    }
}
 
static char * _registry_parser_parse_name(_registry_parser_state_t * state, char * buffer, size_t buf_size) {
    _registry_parser_skip_whitespace(state);
    const char * name_start = state->p;
    while (isalnum((unsigned char)*state->p) || *state->p == '_' || *state->p == ':') {
        if (*state->p == ':' && state->p[1] != ':')
            break;  // Handle single colon as non-identifier char.
        if (*state->p == ':')
            state->p++;  // Skip the first ':' of '::'
        state->p++;
    }
    size_t len = state->p - name_start;
    if (len == 0 || len >= buf_size)
        return nullptr;
    infix_memcpy(buffer, name_start, len);
    buffer[len] = '\0';
    return buffer;
}
 
c23_nodiscard infix_status infix_register_types(infix_registry_t * registry, const char * definitions) {
    _infix_clear_error();
    if (!registry || !definitions) {
        _infix_set_error(INFIX_CATEGORY_GENERAL, INFIX_CODE_UNKNOWN, 0);
        return INFIX_ERROR_INVALID_ARGUMENT;
    }
 
    _registry_parser_state_t state = {.p = definitions, .start = definitions};
    g_infix_last_signature_context = definitions;
 
    // A temporary structure to hold information about each definition found in Pass 1.
    struct def_info {
        _infix_registry_entry_t * entry;
        const char * def_body_start;
        size_t def_body_len;
    };
 
    size_t defs_capacity = 64;  // Start with an initial capacity.
    struct def_info * defs_found = infix_malloc(sizeof(struct def_info) * defs_capacity);
    if (!defs_found) {
        _infix_set_error(INFIX_CATEGORY_ALLOCATION, INFIX_CODE_OUT_OF_MEMORY, 0);
        return INFIX_ERROR_ALLOCATION_FAILED;
    }
    size_t num_defs_found = 0;
    infix_status final_status = INFIX_SUCCESS;
 
    // Pass 1: Scan & Index all names and their definition bodies.
    while (*state.p != '\0') {
        _registry_parser_skip_whitespace(&state);
        if (*state.p == '\0')
            break;
 
        if (*state.p != '@') {
            _infix_set_error(INFIX_CATEGORY_PARSER, INFIX_CODE_UNEXPECTED_TOKEN, state.p - state.start);
            final_status = INFIX_ERROR_INVALID_ARGUMENT;
            goto cleanup;
        }
        state.p++;
 
        char name_buffer[256];
        if (!_registry_parser_parse_name(&state, name_buffer, sizeof(name_buffer))) {
            _infix_set_error(INFIX_CATEGORY_PARSER, INFIX_CODE_UNEXPECTED_TOKEN, state.p - state.start);
            final_status = INFIX_ERROR_INVALID_ARGUMENT;
            goto cleanup;
        }
        _infix_registry_entry_t * entry = _registry_lookup(registry, name_buffer);
        _registry_parser_skip_whitespace(&state);
 
        if (*state.p == '=') {  // This is a full definition.
            state.p++;
            _registry_parser_skip_whitespace(&state);
            // It's an error to redefine a type that wasn't a forward declaration.
            if (entry && !entry->is_forward_declaration) {
                _infix_set_error(INFIX_CATEGORY_PARSER, INFIX_CODE_UNEXPECTED_TOKEN, state.p - state.start);
                final_status = INFIX_ERROR_INVALID_ARGUMENT;
                goto cleanup;
            }
            if (!entry) {  // If it doesn't exist, create it.
                entry = _registry_insert(registry, name_buffer);
                if (!entry) {
                    final_status = INFIX_ERROR_ALLOCATION_FAILED;
                    goto cleanup;
                }
            }
            if (num_defs_found >= defs_capacity) {
                size_t new_capacity = defs_capacity * 2;
                struct def_info * new_defs = infix_realloc(defs_found, sizeof(struct def_info) * new_capacity);
                if (!new_defs) {
                    _infix_set_error(INFIX_CATEGORY_ALLOCATION, INFIX_CODE_OUT_OF_MEMORY, state.p - state.start);
                    final_status = INFIX_ERROR_ALLOCATION_FAILED;
                    goto cleanup;
                }
                defs_found = new_defs;
                defs_capacity = new_capacity;
            }
 
            // Find the end of the type definition body (the next top-level ';').
            defs_found[num_defs_found].entry = entry;
            defs_found[num_defs_found].def_body_start = state.p;
            int nest_level = 0;
            const char * body_end = state.p;
            while (*body_end != '\0' &&
                   !(*body_end == ';' &&
                     nest_level == 0)) {  // Explicitly check for and skip over the '->' token as a single unit.
                if (*body_end == '-' && body_end[1] == '>') {
                    body_end += 2;  // Advance the pointer past the entire token.
                    continue;       // Continue to the next character in the loop.
                }
                if (*body_end == '{' || *body_end == '<' || *body_end == '(' || *body_end == '[')
                    nest_level++;
                if (*body_end == '}' || *body_end == '>' || *body_end == ')' || *body_end == ']')
                    nest_level--;
                body_end++;
            }
            defs_found[num_defs_found].def_body_len = body_end - state.p;
            state.p = body_end;
            num_defs_found++;
        }
        else if (*state.p == ';') {  // This is a forward declaration.
            if (entry) {             // Cannot forward-declare an already-defined type.
                _infix_set_error(INFIX_CATEGORY_PARSER, INFIX_CODE_UNEXPECTED_TOKEN, state.p - state.start);
                final_status = INFIX_ERROR_INVALID_ARGUMENT;
                goto cleanup;
            }
            entry = _registry_insert(registry, name_buffer);
            if (!entry) {
                final_status = INFIX_ERROR_ALLOCATION_FAILED;
                goto cleanup;
            }
            entry->is_forward_declaration = true;
        }
        else {
            _infix_set_error(INFIX_CATEGORY_PARSER, INFIX_CODE_UNEXPECTED_TOKEN, state.p - state.start);
            final_status = INFIX_ERROR_INVALID_ARGUMENT;
            goto cleanup;
        }
 
        if (*state.p == ';')
            state.p++;
    }
 
    // Pass 2: Parse the bodies of all found definitions into the registry.
    for (size_t i = 0; i < num_defs_found; ++i) {
        _infix_registry_entry_t * entry = defs_found[i].entry;
        // Make a temporary, null-terminated copy of the definition body substring.
        char * body_copy = infix_malloc(defs_found[i].def_body_len + 1);
        if (!body_copy) {
            _infix_set_error(INFIX_CATEGORY_ALLOCATION, INFIX_CODE_OUT_OF_MEMORY, 0);
            final_status = INFIX_ERROR_ALLOCATION_FAILED;
            goto cleanup;
        }
        infix_memcpy(body_copy, defs_found[i].def_body_start, defs_found[i].def_body_len);
        body_copy[defs_found[i].def_body_len] = '\0';
 
        // "Parse" step: parse into a temporary arena.
        infix_type * raw_type = nullptr;
        infix_arena_t * parser_arena = nullptr;
        infix_status status = _infix_parse_type_internal(&raw_type, &parser_arena, body_copy);
        infix_free(body_copy);
        if (status != INFIX_SUCCESS) {
            // Adjust the error position to be relative to the full definition string.
            infix_error_details_t err = infix_get_last_error();
            _infix_set_error(err.category, err.code, (defs_found[i].def_body_start - definitions) + err.position);
            final_status = INFIX_ERROR_INVALID_ARGUMENT;
            infix_arena_destroy(parser_arena);
            goto cleanup;
        }
 
        const infix_type * type_to_alias = raw_type;
 
        // If the RHS is another named type (e.g., @MyAlias = @ExistingType),
        // we need to resolve it first to get the actual type we're aliasing.
        if (raw_type->category == INFIX_TYPE_NAMED_REFERENCE) {
            _infix_registry_entry_t * existing_entry = _registry_lookup(registry, raw_type->meta.named_reference.name);
            if (existing_entry && existing_entry->type)
                type_to_alias = existing_entry->type;
            else {
                _infix_set_error(INFIX_CATEGORY_PARSER,
                                 INFIX_CODE_UNRESOLVED_NAMED_TYPE,
                                 (size_t)(defs_found[i].def_body_start - definitions));
                final_status = INFIX_ERROR_INVALID_ARGUMENT;
                infix_arena_destroy(parser_arena);
                goto cleanup;
            }
        }
 
        if (!type_to_alias->is_arena_allocated) {
            // This is a static type (e.g., primitive). We MUST create a mutable
            // copy in the registry's arena before we can attach a name to it.
            // This prevents corrupting the global static singletons.
            infix_type * prim_copy = infix_arena_alloc(registry->arena, sizeof(infix_type), _Alignof(infix_type));
            if (!prim_copy) {
                final_status = INFIX_ERROR_ALLOCATION_FAILED;
                infix_arena_destroy(parser_arena);
                goto cleanup;
            }
            *prim_copy = *type_to_alias;
            prim_copy->is_arena_allocated = true;
            prim_copy->arena = registry->arena;
            entry->type = prim_copy;
        }
        else
            entry->type = _copy_type_graph_to_arena(registry->arena, type_to_alias);
 
        infix_arena_destroy(parser_arena);  // The temporary raw_type is no longer needed.
 
        if (!entry->type) {
            _infix_set_error(INFIX_CATEGORY_ALLOCATION, INFIX_CODE_OUT_OF_MEMORY, 0);
            final_status = INFIX_ERROR_ALLOCATION_FAILED;
            goto cleanup;
        }
 
        // Attach the new alias name to the copied type. This is the key to preserving semantic names.
        entry->type->name = entry->name;
        entry->is_forward_declaration = false;
    }
 
    // Pass 3: Resolve and layout all the newly defined types.
    for (size_t i = 0; i < num_defs_found; ++i) {
        _infix_registry_entry_t * entry = defs_found[i].entry;
        if (entry->type) {
            // "Resolve" and "Layout" steps.
            if (_infix_resolve_type_graph_inplace(&entry->type, registry) != INFIX_SUCCESS) {
                final_status = INFIX_ERROR_INVALID_ARGUMENT;
                goto cleanup;
            }
            _infix_type_recalculate_layout(entry->type);
        }
    }
 
cleanup:
    infix_free(defs_found);
    return final_status;
}
 
// Registry Introspection API Implementation
c23_nodiscard infix_registry_iterator_t infix_registry_iterator_begin(const infix_registry_t * registry) {
    // Return an iterator positioned before the first element.
    // The first call to next() will advance it to the first valid element.
    return (infix_registry_iterator_t){registry, 0, nullptr};
}
 
c23_nodiscard bool infix_registry_iterator_next(infix_registry_iterator_t * iter) {
    if (!iter || !iter->registry)
        return false;
 
    const _infix_registry_entry_t * entry = iter->current_entry;
 
    // If we have a current entry, start from the next one in the chain.
    if (iter->current_entry)
        entry = entry->next;
 
    // Otherwise, if we are starting, begin with the head of the current bucket.
    else if (iter->current_bucket < iter->registry->num_buckets)
        entry = iter->registry->buckets[iter->current_bucket];
 
    while (true) {
        // Traverse the current chain looking for a valid entry.
        while (entry) {
            if (entry->type && !entry->is_forward_declaration) {
                // Found one. Update the iterator and return successfully.
                iter->current_entry = entry;
                return true;
            }
            entry = entry->next;
        }
 
        // If we're here, the current chain is exhausted. Move to the next bucket.
        iter->current_bucket++;
 
        // If there are no more buckets, we're done.
        if (iter->current_bucket >= iter->registry->num_buckets) {
            iter->current_entry = nullptr;
            return false;
        }
 
        // Start the search from the head of the new bucket.
        entry = iter->registry->buckets[iter->current_bucket];
    }
}
 
c23_nodiscard const char * infix_registry_iterator_get_name(const infix_registry_iterator_t * iter) {
    if (!iter || !iter->current_entry)
        return nullptr;
    return iter->current_entry->name;
}
 
c23_nodiscard const infix_type * infix_registry_iterator_get_type(const infix_registry_iterator_t * iter) {
    if (!iter || !iter->current_entry)
        return nullptr;
    return iter->current_entry->type;
}
 
c23_nodiscard bool infix_registry_is_defined(const infix_registry_t * registry, const char * name) {
    if (!registry || !name)
        return false;
    _infix_registry_entry_t * entry = _registry_lookup((infix_registry_t *)registry, name);
    // It's defined if an entry exists, it has a type, and it's not a lingering forward declaration.
    return entry != nullptr && entry->type != nullptr && !entry->is_forward_declaration;
}
 
c23_nodiscard const infix_type * infix_registry_lookup_type(const infix_registry_t * registry, const char * name) {
    if (!registry || !name)
        return nullptr;
    _infix_registry_entry_t * entry = _registry_lookup((infix_registry_t *)registry, name);
    if (entry && entry->type && !entry->is_forward_declaration)
        return entry->type;
    return nullptr;
}