prism 1.9.0 → 2.0.0.rc1

data/include/prism/defines.h

@@ -1,260 +0,0 @@
-/**
- * @file defines.h
- *
- * Macro definitions used throughout the prism library.
- *
- * This file should be included first by any *.h or *.c in prism for consistency
- * and to ensure that the macros are defined before they are used.
- */
-#ifndef PRISM_DEFINES_H
-#define PRISM_DEFINES_H
-
-#include <ctype.h>
-#include <limits.h>
-#include <math.h>
-#include <stdarg.h>
-#include <stddef.h>
-#include <stdint.h>
-#include <stdio.h>
-#include <string.h>
-
-/**
- * We want to be able to use the PRI* macros for printing out integers, but on
- * some platforms they aren't included unless this is already defined.
- */
-#define __STDC_FORMAT_MACROS
-// Include sys/types.h before inttypes.h to work around issue with
-// certain versions of GCC and newlib which causes omission of PRIx64
-#include <sys/types.h>
-#include <inttypes.h>
-
-/**
- * When we are parsing using recursive descent, we want to protect against
- * malicious payloads that could attempt to crash our parser. We do this by
- * specifying a maximum depth to which we are allowed to recurse.
- */
-#ifndef PRISM_DEPTH_MAXIMUM
-    #define PRISM_DEPTH_MAXIMUM 10000
-#endif
-
-/**
- * By default, we compile with -fvisibility=hidden. When this is enabled, we
- * need to mark certain functions as being publically-visible. This macro does
- * that in a compiler-agnostic way.
- */
-#ifndef PRISM_EXPORTED_FUNCTION
-#   ifdef PRISM_EXPORT_SYMBOLS
-#       ifdef _WIN32
-#          define PRISM_EXPORTED_FUNCTION __declspec(dllexport) extern
-#       else
-#          define PRISM_EXPORTED_FUNCTION __attribute__((__visibility__("default"))) extern
-#       endif
-#   else
-#       define PRISM_EXPORTED_FUNCTION
-#   endif
-#endif
-
-/**
- * Certain compilers support specifying that a function accepts variadic
- * parameters that look like printf format strings to provide a better developer
- * experience when someone is using the function. This macro does that in a
- * compiler-agnostic way.
- */
-#if defined(__GNUC__)
-#   if defined(__MINGW_PRINTF_FORMAT)
-#       define PRISM_ATTRIBUTE_FORMAT(string_index, argument_index) __attribute__((format(__MINGW_PRINTF_FORMAT, string_index, argument_index)))
-#   else
-#       define PRISM_ATTRIBUTE_FORMAT(string_index, argument_index) __attribute__((format(printf, string_index, argument_index)))
-#   endif
-#elif defined(__clang__)
-#   define PRISM_ATTRIBUTE_FORMAT(string_index, argument_index) __attribute__((__format__(__printf__, string_index, argument_index)))
-#else
-#   define PRISM_ATTRIBUTE_FORMAT(string_index, argument_index)
-#endif
-
-/**
- * GCC will warn if you specify a function or parameter that is unused at
- * runtime. This macro allows you to mark a function or parameter as unused in a
- * compiler-agnostic way.
- */
-#if defined(__GNUC__)
-#   define PRISM_ATTRIBUTE_UNUSED __attribute__((unused))
-#else
-#   define PRISM_ATTRIBUTE_UNUSED
-#endif
-
-/**
- * Old Visual Studio versions do not support the inline keyword, so we need to
- * define it to be __inline.
- */
-#if defined(_MSC_VER) && !defined(inline)
-#   define inline __inline
-#endif
-
-/**
- * Old Visual Studio versions before 2015 do not implement sprintf, but instead
- * implement _snprintf. We standard that here.
- */
-#if !defined(snprintf) && defined(_MSC_VER) && (_MSC_VER < 1900)
-#   define snprintf _snprintf
-#endif
-
-/**
- * A simple utility macro to concatenate two tokens together, necessary when one
- * of the tokens is itself a macro.
- */
-#define PM_CONCATENATE(left, right) left ## right
-
-/**
- * We want to be able to use static assertions, but they weren't standardized
- * until C11. As such, we polyfill it here by making a hacky typedef that will
- * fail to compile due to a negative array size if the condition is false.
- */
-#if defined(_Static_assert)
-#   define PM_STATIC_ASSERT(line, condition, message) _Static_assert(condition, message)
-#else
-#   define PM_STATIC_ASSERT(line, condition, message) typedef char PM_CONCATENATE(static_assert_, line)[(condition) ? 1 : -1]
-#endif
-
-/**
- * In general, libc for embedded systems does not support memory-mapped files.
- * If the target platform is POSIX or Windows, we can map a file in memory and
- * read it in a more efficient manner.
- */
-#ifdef _WIN32
-#   define PRISM_HAS_MMAP
-#else
-#   include <unistd.h>
-#   ifdef _POSIX_MAPPED_FILES
-#       define PRISM_HAS_MMAP
-#   endif
-#endif
-
-/**
- * If PRISM_HAS_NO_FILESYSTEM is defined, then we want to exclude all filesystem
- * related code from the library. All filesystem related code should be guarded
- * by PRISM_HAS_FILESYSTEM.
- */
-#ifndef PRISM_HAS_NO_FILESYSTEM
-#   define PRISM_HAS_FILESYSTEM
-#endif
-
-/**
- * isinf on POSIX systems it accepts a float, a double, or a long double.
- * But mingw didn't provide an isinf macro, only an isinf function that only
- * accepts floats, so we need to use _finite instead.
- */
-#ifdef __MINGW64__
-    #include <float.h>
-    #define PRISM_ISINF(x) (!_finite(x))
-#else
-    #define PRISM_ISINF(x) isinf(x)
-#endif
-
-/**
- * If you build prism with a custom allocator, configure it with
- * "-D PRISM_XALLOCATOR" to use your own allocator that defines xmalloc,
- * xrealloc, xcalloc, and xfree.
- *
- * For example, your `prism_xallocator.h` file could look like this:
- *
- * ```
- * #ifndef PRISM_XALLOCATOR_H
- * #define PRISM_XALLOCATOR_H
- * #define xmalloc      my_malloc
- * #define xrealloc     my_realloc
- * #define xcalloc      my_calloc
- * #define xfree        my_free
- * #endif
- * ```
- */
-#ifdef PRISM_XALLOCATOR
-    #include "prism_xallocator.h"
-#else
-    #ifndef xmalloc
-        /**
-         * The malloc function that should be used. This can be overridden with
-         * the PRISM_XALLOCATOR define.
-         */
-        #define xmalloc malloc
-    #endif
-
-    #ifndef xrealloc
-        /**
-         * The realloc function that should be used. This can be overridden with
-         * the PRISM_XALLOCATOR define.
-         */
-        #define xrealloc realloc
-    #endif
-
-    #ifndef xcalloc
-        /**
-         * The calloc function that should be used. This can be overridden with
-         * the PRISM_XALLOCATOR define.
-         */
-        #define xcalloc calloc
-    #endif
-
-    #ifndef xfree
-        /**
-         * The free function that should be used. This can be overridden with the
-         * PRISM_XALLOCATOR define.
-         */
-        #define xfree free
-    #endif
-#endif
-
-/**
- * If PRISM_BUILD_MINIMAL is defined, then we're going to define every possible
- * switch that will turn off certain features of prism.
- */
-#ifdef PRISM_BUILD_MINIMAL
-    /** Exclude the serialization API. */
-    #define PRISM_EXCLUDE_SERIALIZATION
-
-    /** Exclude the JSON serialization API. */
-    #define PRISM_EXCLUDE_JSON
-
-    /** Exclude the Array#pack parser API. */
-    #define PRISM_EXCLUDE_PACK
-
-    /** Exclude the prettyprint API. */
-    #define PRISM_EXCLUDE_PRETTYPRINT
-
-    /** Exclude the full set of encodings, using the minimal only. */
-    #define PRISM_ENCODING_EXCLUDE_FULL
-#endif
-
-/**
- * Support PRISM_LIKELY and PRISM_UNLIKELY to help the compiler optimize its
- * branch predication.
- */
-#if defined(__GNUC__) || defined(__clang__)
-    /** The compiler should predicate that this branch will be taken. */
-    #define PRISM_LIKELY(x) __builtin_expect(!!(x), 1)
-
-    /** The compiler should predicate that this branch will not be taken. */
-    #define PRISM_UNLIKELY(x) __builtin_expect(!!(x), 0)
-#else
-    /** Void because this platform does not support branch prediction hints. */
-    #define PRISM_LIKELY(x)   (x)
-
-    /** Void because this platform does not support branch prediction hints. */
-    #define PRISM_UNLIKELY(x) (x)
-#endif
-
-/**
- * We use -Wimplicit-fallthrough to guard potentially unintended fall-through between cases of a switch.
- * Use PRISM_FALLTHROUGH to explicitly annotate cases where the fallthrough is intentional.
- */
-#if defined(__STDC_VERSION__) && __STDC_VERSION__ >= 202311L // C23 or later
-    #define PRISM_FALLTHROUGH [[fallthrough]];
-#elif defined(__GNUC__) || defined(__clang__)
-    #define PRISM_FALLTHROUGH __attribute__((fallthrough));
-#elif defined(_MSC_VER)
-    #define PRISM_FALLTHROUGH __fallthrough;
-#else
-    #define PRISM_FALLTHROUGH
-#endif
-
-#endif

data/include/prism/pack.h

@@ -1,163 +0,0 @@
-/**
- * @file pack.h
- *
- * A pack template string parser.
- */
-#ifndef PRISM_PACK_H
-#define PRISM_PACK_H
-
-#include "prism/defines.h"
-
-// We optionally support parsing String#pack templates. For systems that don't
-// want or need this functionality, it can be turned off with the
-// PRISM_EXCLUDE_PACK define.
-#ifdef PRISM_EXCLUDE_PACK
-
-void pm_pack_parse(void);
-
-#else
-
-#include <stdint.h>
-#include <stdlib.h>
-
-/** The version of the pack template language that we are parsing. */
-typedef enum pm_pack_version {
-    PM_PACK_VERSION_3_2_0
-} pm_pack_version;
-
-/** The type of pack template we are parsing. */
-typedef enum pm_pack_variant {
-    PM_PACK_VARIANT_PACK,
-    PM_PACK_VARIANT_UNPACK
-} pm_pack_variant;
-
-/** A directive within the pack template. */
-typedef enum pm_pack_type {
-    PM_PACK_SPACE,
-    PM_PACK_COMMENT,
-    PM_PACK_INTEGER,
-    PM_PACK_UTF8,
-    PM_PACK_BER,
-    PM_PACK_FLOAT,
-    PM_PACK_STRING_SPACE_PADDED,
-    PM_PACK_STRING_NULL_PADDED,
-    PM_PACK_STRING_NULL_TERMINATED,
-    PM_PACK_STRING_MSB,
-    PM_PACK_STRING_LSB,
-    PM_PACK_STRING_HEX_HIGH,
-    PM_PACK_STRING_HEX_LOW,
-    PM_PACK_STRING_UU,
-    PM_PACK_STRING_MIME,
-    PM_PACK_STRING_BASE64,
-    PM_PACK_STRING_FIXED,
-    PM_PACK_STRING_POINTER,
-    PM_PACK_MOVE,
-    PM_PACK_BACK,
-    PM_PACK_NULL,
-    PM_PACK_END
-} pm_pack_type;
-
-/** The signness of a pack directive. */
-typedef enum pm_pack_signed {
-    PM_PACK_UNSIGNED,
-    PM_PACK_SIGNED,
-    PM_PACK_SIGNED_NA
-} pm_pack_signed;
-
-/** The endianness of a pack directive. */
-typedef enum pm_pack_endian {
-    PM_PACK_AGNOSTIC_ENDIAN,
-    PM_PACK_LITTLE_ENDIAN,      // aka 'VAX', or 'V'
-    PM_PACK_BIG_ENDIAN,         // aka 'network', or 'N'
-    PM_PACK_NATIVE_ENDIAN,
-    PM_PACK_ENDIAN_NA
-} pm_pack_endian;
-
-/** The size of an integer pack directive. */
-typedef enum pm_pack_size {
-    PM_PACK_SIZE_SHORT,
-    PM_PACK_SIZE_INT,
-    PM_PACK_SIZE_LONG,
-    PM_PACK_SIZE_LONG_LONG,
-    PM_PACK_SIZE_8,
-    PM_PACK_SIZE_16,
-    PM_PACK_SIZE_32,
-    PM_PACK_SIZE_64,
-    PM_PACK_SIZE_P,
-    PM_PACK_SIZE_NA
-} pm_pack_size;
-
-/** The type of length of a pack directive. */
-typedef enum pm_pack_length_type {
-    PM_PACK_LENGTH_FIXED,
-    PM_PACK_LENGTH_MAX,
-    PM_PACK_LENGTH_RELATIVE,  // special case for unpack @*
-    PM_PACK_LENGTH_NA
-} pm_pack_length_type;
-
-/** The type of encoding for a pack template string. */
-typedef enum pm_pack_encoding {
-    PM_PACK_ENCODING_START,
-    PM_PACK_ENCODING_ASCII_8BIT,
-    PM_PACK_ENCODING_US_ASCII,
-    PM_PACK_ENCODING_UTF_8
-} pm_pack_encoding;
-
-/** The result of parsing a pack template. */
-typedef enum pm_pack_result {
-    PM_PACK_OK,
-    PM_PACK_ERROR_UNSUPPORTED_DIRECTIVE,
-    PM_PACK_ERROR_UNKNOWN_DIRECTIVE,
-    PM_PACK_ERROR_LENGTH_TOO_BIG,
-    PM_PACK_ERROR_BANG_NOT_ALLOWED,
-    PM_PACK_ERROR_DOUBLE_ENDIAN
-} pm_pack_result;
-
-/**
- * Parse a single directive from a pack or unpack format string.
- *
- * @param variant (in) pack or unpack
- * @param format (in, out) the start of the next directive to parse on calling,
- *     and advanced beyond the parsed directive on return, or as much of it as
- *     was consumed until an error was encountered
- * @param format_end (in) the end of the format string
- * @param type (out) the type of the directive
- * @param signed_type (out) whether the value is signed
- * @param endian (out) the endianness of the value
- * @param size (out) the size of the value
- * @param length_type (out) what kind of length is specified
- * @param length (out) the length of the directive
- * @param encoding (in, out) takes the current encoding of the string which
- *     would result from parsing the whole format string, and returns a possibly
- *     changed directive - the encoding should be `PM_PACK_ENCODING_START` when
- *     pm_pack_parse is called for the first directive in a format string
- *
- * @return `PM_PACK_OK` on success or `PM_PACK_ERROR_*` on error
- * @note Consult Ruby documentation for the meaning of directives.
- */
-PRISM_EXPORTED_FUNCTION pm_pack_result
-pm_pack_parse(
-    pm_pack_variant variant,
-    const char **format,
-    const char *format_end,
-    pm_pack_type *type,
-    pm_pack_signed *signed_type,
-    pm_pack_endian *endian,
-    pm_pack_size *size,
-    pm_pack_length_type *length_type,
-    uint64_t *length,
-    pm_pack_encoding *encoding
-);
-
-/**
- * Prism abstracts sizes away from the native system - this converts an abstract
- * size to a native size.
- *
- * @param size The abstract size to convert.
- * @return The native size.
- */
-PRISM_EXPORTED_FUNCTION size_t pm_size_to_native(pm_pack_size size);
-
-#endif
-
-#endif

data/include/prism/regexp.h

@@ -1,43 +0,0 @@
-/**
- * @file regexp.h
- *
- * A regular expression parser.
- */
-#ifndef PRISM_REGEXP_H
-#define PRISM_REGEXP_H
-
-#include "prism/defines.h"
-#include "prism/parser.h"
-#include "prism/encoding.h"
-#include "prism/util/pm_memchr.h"
-#include "prism/util/pm_string.h"
-
-#include <stdbool.h>
-#include <stddef.h>
-#include <string.h>
-
-/**
- * This callback is called by pm_regexp_parse() when a named capture group is found.
- */
-typedef void (*pm_regexp_name_callback_t)(const pm_string_t *name, void *data);
-
-/**
- * This callback is called by pm_regexp_parse() when a parse error is found.
- */
-typedef void (*pm_regexp_error_callback_t)(const uint8_t *start, const uint8_t *end, const char *message, void *data);
-
-/**
- * Parse a regular expression.
- *
- * @param parser The parser that is currently being used.
- * @param source The source code to parse.
- * @param size The size of the source code.
- * @param extended_mode Whether to parse the regular expression in extended mode.
- * @param name_callback The optional callback to call when a named capture group is found.
- * @param name_data The optional data to pass to the name callback.
- * @param error_callback The callback to call when a parse error is found.
- * @param error_data The data to pass to the error callback.
- */
-PRISM_EXPORTED_FUNCTION void pm_regexp_parse(pm_parser_t *parser, const uint8_t *source, size_t size, bool extended_mode, pm_regexp_name_callback_t name_callback, void *name_data, pm_regexp_error_callback_t error_callback, void *error_data);
-
-#endif

data/include/prism/util/pm_buffer.h

@@ -1,236 +0,0 @@
-/**
- * @file pm_buffer.h
- *
- * A wrapper around a contiguous block of allocated memory.
- */
-#ifndef PRISM_BUFFER_H
-#define PRISM_BUFFER_H
-
-#include "prism/defines.h"
-#include "prism/util/pm_char.h"
-
-#include <assert.h>
-#include <stdbool.h>
-#include <stdint.h>
-#include <stdlib.h>
-#include <string.h>
-
-/**
- * A pm_buffer_t is a simple memory buffer that stores data in a contiguous
- * block of memory.
- */
-typedef struct {
-    /** The length of the buffer in bytes. */
-    size_t length;
-
-    /** The capacity of the buffer in bytes that has been allocated. */
-    size_t capacity;
-
-    /** A pointer to the start of the buffer. */
-    char *value;
-} pm_buffer_t;
-
-/**
- * Return the size of the pm_buffer_t struct.
- *
- * @returns The size of the pm_buffer_t struct.
- */
-PRISM_EXPORTED_FUNCTION size_t pm_buffer_sizeof(void);
-
-/**
- * Initialize a pm_buffer_t with the given capacity.
- *
- * @param buffer The buffer to initialize.
- * @param capacity The capacity of the buffer.
- * @returns True if the buffer was initialized successfully, false otherwise.
- */
-bool pm_buffer_init_capacity(pm_buffer_t *buffer, size_t capacity);
-
-/**
- * Initialize a pm_buffer_t with its default values.
- *
- * @param buffer The buffer to initialize.
- * @returns True if the buffer was initialized successfully, false otherwise.
- *
- * \public \memberof pm_buffer_t
- */
-PRISM_EXPORTED_FUNCTION bool pm_buffer_init(pm_buffer_t *buffer);
-
-/**
- * Return the value of the buffer.
- *
- * @param buffer The buffer to get the value of.
- * @returns The value of the buffer.
- *
- * \public \memberof pm_buffer_t
- */
-PRISM_EXPORTED_FUNCTION char * pm_buffer_value(const pm_buffer_t *buffer);
-
-/**
- * Return the length of the buffer.
- *
- * @param buffer The buffer to get the length of.
- * @returns The length of the buffer.
- *
- * \public \memberof pm_buffer_t
- */
-PRISM_EXPORTED_FUNCTION size_t pm_buffer_length(const pm_buffer_t *buffer);
-
-/**
- * Append the given amount of space as zeroes to the buffer.
- *
- * @param buffer The buffer to append to.
- * @param length The amount of space to append and zero.
- */
-void pm_buffer_append_zeroes(pm_buffer_t *buffer, size_t length);
-
-/**
- * Append a formatted string to the buffer.
- *
- * @param buffer The buffer to append to.
- * @param format The format string to append.
- * @param ... The arguments to the format string.
- */
-void pm_buffer_append_format(pm_buffer_t *buffer, const char *format, ...) PRISM_ATTRIBUTE_FORMAT(2, 3);
-
-/**
- * Append a string to the buffer.
- *
- * @param buffer The buffer to append to.
- * @param value The string to append.
- * @param length The length of the string to append.
- */
-void pm_buffer_append_string(pm_buffer_t *buffer, const char *value, size_t length);
-
-/**
- * Append a list of bytes to the buffer.
- *
- * @param buffer The buffer to append to.
- * @param value The bytes to append.
- * @param length The length of the bytes to append.
- */
-void pm_buffer_append_bytes(pm_buffer_t *buffer, const uint8_t *value, size_t length);
-
-/**
- * Append a single byte to the buffer.
- *
- * @param buffer The buffer to append to.
- * @param value The byte to append.
- */
-void pm_buffer_append_byte(pm_buffer_t *buffer, uint8_t value);
-
-/**
- * Append a 32-bit unsigned integer to the buffer as a variable-length integer.
- *
- * @param buffer The buffer to append to.
- * @param value The integer to append.
- */
-void pm_buffer_append_varuint(pm_buffer_t *buffer, uint32_t value);
-
-/**
- * Append a 32-bit signed integer to the buffer as a variable-length integer.
- *
- * @param buffer The buffer to append to.
- * @param value The integer to append.
- */
-void pm_buffer_append_varsint(pm_buffer_t *buffer, int32_t value);
-
-/**
- * Append a double to the buffer.
- *
- * @param buffer The buffer to append to.
- * @param value The double to append.
- */
-void pm_buffer_append_double(pm_buffer_t *buffer, double value);
-
-/**
- * Append a unicode codepoint to the buffer.
- *
- * @param buffer The buffer to append to.
- * @param value The character to append.
- * @returns True if the codepoint was valid and appended successfully, false
- *   otherwise.
- */
-bool pm_buffer_append_unicode_codepoint(pm_buffer_t *buffer, uint32_t value);
-
-/**
- * The different types of escaping that can be performed by the buffer when
- * appending a slice of Ruby source code.
- */
-typedef enum {
-    PM_BUFFER_ESCAPING_RUBY,
-    PM_BUFFER_ESCAPING_JSON
-} pm_buffer_escaping_t;
-
-/**
- * Append a slice of source code to the buffer.
- *
- * @param buffer The buffer to append to.
- * @param source The source code to append.
- * @param length The length of the source code to append.
- * @param escaping The type of escaping to perform.
- */
-void pm_buffer_append_source(pm_buffer_t *buffer, const uint8_t *source, size_t length, pm_buffer_escaping_t escaping);
-
-/**
- * Prepend the given string to the buffer.
- *
- * @param buffer The buffer to prepend to.
- * @param value The string to prepend.
- * @param length The length of the string to prepend.
- */
-void pm_buffer_prepend_string(pm_buffer_t *buffer, const char *value, size_t length);
-
-/**
- * Concatenate one buffer onto another.
- *
- * @param destination The buffer to concatenate onto.
- * @param source The buffer to concatenate.
- */
-void pm_buffer_concat(pm_buffer_t *destination, const pm_buffer_t *source);
-
-/**
- * Clear the buffer by reducing its size to 0. This does not free the allocated
- * memory, but it does allow the buffer to be reused.
- *
- * @param buffer The buffer to clear.
- */
-void pm_buffer_clear(pm_buffer_t *buffer);
-
-/**
- * Strip the whitespace from the end of the buffer.
- *
- * @param buffer The buffer to strip.
- */
-void pm_buffer_rstrip(pm_buffer_t *buffer);
-
-/**
- * Checks if the buffer includes the given value.
- *
- * @param buffer The buffer to check.
- * @param value The value to check for.
- * @returns The index of the first occurrence of the value in the buffer, or
- *   SIZE_MAX if the value is not found.
- */
-size_t pm_buffer_index(const pm_buffer_t *buffer, char value);
-
-/**
- * Insert the given string into the buffer at the given index.
- *
- * @param buffer The buffer to insert into.
- * @param index The index to insert at.
- * @param value The string to insert.
- * @param length The length of the string to insert.
- */
-void pm_buffer_insert(pm_buffer_t *buffer, size_t index, const char *value, size_t length);
-
-/**
- * Free the memory associated with the buffer.
- *
- * @param buffer The buffer to free.
- *
- * \public \memberof pm_buffer_t
- */
-PRISM_EXPORTED_FUNCTION void pm_buffer_free(pm_buffer_t *buffer);
-
-#endif