jruby-prism-parser 0.24.0-java → 1.4.0-java

data/include/prism/defines.h

@@ -10,6 +10,8 @@
 #define PRISM_DEFINES_H
 
 #include <ctype.h>
+#include <limits.h>
+#include <math.h>
 #include <stdarg.h>
 #include <stddef.h>
 #include <stdint.h>
@@ -21,9 +23,20 @@
  * 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
@@ -48,7 +61,11 @@
  * compiler-agnostic way.
  */
 #if defined(__GNUC__)
-#   define PRISM_ATTRIBUTE_FORMAT(string_index, argument_index) __attribute__((format(printf, string_index, argument_index)))
+#   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
@@ -99,4 +116,145 @@
 #   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/diagnostic.h

@@ -1,3 +1,11 @@
+/*----------------------------------------------------------------------------*/
+/* This file is generated by the templates/template.rb script and should not  */
+/* be modified manually. See                                                  */
+/* templates/include/prism/diagnostic.h.erb                                   */
+/* if you are looking to modify the                                           */
+/* template                                                                   */
+/*----------------------------------------------------------------------------*/
+
 /**
  * @file diagnostic.h
  *
@@ -14,83 +22,32 @@
 #include <stdlib.h>
 #include <assert.h>
 
-/**
- * The levels of errors generated during parsing.
- */
-typedef enum {
-    /** For errors that cannot be recovered from. */
-    PM_ERROR_LEVEL_FATAL = 0,
-
-    /** For errors that should raise an argument error. */
-    PM_ERROR_LEVEL_ARGUMENT = 1
-} pm_error_level_t;
-
-/**
- * The levels of warnings generated during parsing.
- */
-typedef enum {
-    /** For warnings which should be emitted if $VERBOSE != nil. */
-    PM_WARNING_LEVEL_DEFAULT = 0,
-
-    /** For warnings which should be emitted if $VERBOSE == true. */
-    PM_WARNING_LEVEL_VERBOSE = 1
-} pm_warning_level_t;
-
-/**
- * This struct represents a diagnostic generated during parsing.
- *
- * @extends pm_list_node_t
- */
-typedef struct {
-    /** The embedded base node. */
-    pm_list_node_t node;
-
-    /** The location of the diagnostic in the source. */
-    pm_location_t location;
-
-    /** The message associated with the diagnostic. */
-    const char *message;
-
-    /**
-     * Whether or not the memory related to the message of this diagnostic is
-     * owned by this diagnostic. If it is, it needs to be freed when the
-     * diagnostic is freed.
-     */
-    bool owned;
-
-    /**
-     * The level of the diagnostic, see `pm_error_level_t` and
-     * `pm_warning_level_t` for possible values.
-     */
-    uint8_t level;
-} pm_diagnostic_t;
-
 /**
  * The diagnostic IDs of all of the diagnostics, used to communicate the types
  * of errors between the parser and the user.
  */
 typedef enum {
-    // This is a special error that we can potentially replace by others. For
-    // an example of how this is used, see parse_expression_prefix.
-    PM_ERR_CANNOT_PARSE_EXPRESSION,
-
-    // These are the error codes.
+    // These are the error diagnostics.
     PM_ERR_ALIAS_ARGUMENT,
+    PM_ERR_ALIAS_ARGUMENT_NUMBERED_REFERENCE,
     PM_ERR_AMPAMPEQ_MULTI_ASSIGN,
     PM_ERR_ARGUMENT_AFTER_BLOCK,
     PM_ERR_ARGUMENT_AFTER_FORWARDING_ELLIPSES,
     PM_ERR_ARGUMENT_BARE_HASH,
     PM_ERR_ARGUMENT_BLOCK_FORWARDING,
     PM_ERR_ARGUMENT_BLOCK_MULTI,
+    PM_ERR_ARGUMENT_CONFLICT_AMPERSAND,
+    PM_ERR_ARGUMENT_CONFLICT_STAR,
+    PM_ERR_ARGUMENT_CONFLICT_STAR_STAR,
     PM_ERR_ARGUMENT_FORMAL_CLASS,
     PM_ERR_ARGUMENT_FORMAL_CONSTANT,
     PM_ERR_ARGUMENT_FORMAL_GLOBAL,
     PM_ERR_ARGUMENT_FORMAL_IVAR,
     PM_ERR_ARGUMENT_FORWARDING_UNBOUND,
-    PM_ERR_ARGUMENT_IN,
-    PM_ERR_ARGUMENT_NO_FORWARDING_AMP,
+    PM_ERR_ARGUMENT_NO_FORWARDING_AMPERSAND,
     PM_ERR_ARGUMENT_NO_FORWARDING_ELLIPSES,
     PM_ERR_ARGUMENT_NO_FORWARDING_STAR,
+    PM_ERR_ARGUMENT_NO_FORWARDING_STAR_STAR,
     PM_ERR_ARGUMENT_SPLAT_AFTER_ASSOC_SPLAT,
     PM_ERR_ARGUMENT_SPLAT_AFTER_SPLAT,
     PM_ERR_ARGUMENT_TERM_PAREN,
@@ -109,6 +66,7 @@ typedef enum {
     PM_ERR_BLOCK_PARAM_PIPE_TERM,
     PM_ERR_BLOCK_TERM_BRACE,
     PM_ERR_BLOCK_TERM_END,
+    PM_ERR_CANNOT_PARSE_EXPRESSION,
     PM_ERR_CANNOT_PARSE_STRING_PART,
     PM_ERR_CASE_EXPRESSION_AFTER_CASE,
     PM_ERR_CASE_EXPRESSION_AFTER_WHEN,
@@ -120,6 +78,7 @@ typedef enum {
     PM_ERR_CLASS_SUPERCLASS,
     PM_ERR_CLASS_TERM,
     PM_ERR_CLASS_UNEXPECTED_END,
+    PM_ERR_CLASS_VARIABLE_BARE,
     PM_ERR_CONDITIONAL_ELSIF_PREDICATE,
     PM_ERR_CONDITIONAL_IF_PREDICATE,
     PM_ERR_CONDITIONAL_PREDICATE_TERM,
@@ -132,7 +91,6 @@ typedef enum {
     PM_ERR_DEF_ENDLESS,
     PM_ERR_DEF_ENDLESS_SETTER,
     PM_ERR_DEF_NAME,
-    PM_ERR_DEF_NAME_AFTER_RECEIVER,
     PM_ERR_DEF_PARAMS_TERM,
     PM_ERR_DEF_PARAMS_TERM_PAREN,
     PM_ERR_DEF_RECEIVER,
@@ -151,57 +109,95 @@ typedef enum {
     PM_ERR_ESCAPE_INVALID_META_REPEAT,
     PM_ERR_ESCAPE_INVALID_UNICODE,
     PM_ERR_ESCAPE_INVALID_UNICODE_CM_FLAGS,
+    PM_ERR_ESCAPE_INVALID_UNICODE_LIST,
     PM_ERR_ESCAPE_INVALID_UNICODE_LITERAL,
     PM_ERR_ESCAPE_INVALID_UNICODE_LONG,
+    PM_ERR_ESCAPE_INVALID_UNICODE_SHORT,
     PM_ERR_ESCAPE_INVALID_UNICODE_TERM,
     PM_ERR_EXPECT_ARGUMENT,
     PM_ERR_EXPECT_EOL_AFTER_STATEMENT,
     PM_ERR_EXPECT_EXPRESSION_AFTER_AMPAMPEQ,
-    PM_ERR_EXPECT_EXPRESSION_AFTER_PIPEPIPEEQ,
     PM_ERR_EXPECT_EXPRESSION_AFTER_COMMA,
     PM_ERR_EXPECT_EXPRESSION_AFTER_EQUAL,
     PM_ERR_EXPECT_EXPRESSION_AFTER_LESS_LESS,
     PM_ERR_EXPECT_EXPRESSION_AFTER_LPAREN,
-    PM_ERR_EXPECT_EXPRESSION_AFTER_QUESTION,
     PM_ERR_EXPECT_EXPRESSION_AFTER_OPERATOR,
+    PM_ERR_EXPECT_EXPRESSION_AFTER_PIPEPIPEEQ,
+    PM_ERR_EXPECT_EXPRESSION_AFTER_QUESTION,
     PM_ERR_EXPECT_EXPRESSION_AFTER_SPLAT,
     PM_ERR_EXPECT_EXPRESSION_AFTER_SPLAT_HASH,
     PM_ERR_EXPECT_EXPRESSION_AFTER_STAR,
+    PM_ERR_EXPECT_FOR_DELIMITER,
     PM_ERR_EXPECT_IDENT_REQ_PARAMETER,
+    PM_ERR_EXPECT_IN_DELIMITER,
     PM_ERR_EXPECT_LPAREN_REQ_PARAMETER,
+    PM_ERR_EXPECT_MESSAGE,
     PM_ERR_EXPECT_RBRACKET,
     PM_ERR_EXPECT_RPAREN,
     PM_ERR_EXPECT_RPAREN_AFTER_MULTI,
     PM_ERR_EXPECT_RPAREN_REQ_PARAMETER,
+    PM_ERR_EXPECT_SINGLETON_CLASS_DELIMITER,
     PM_ERR_EXPECT_STRING_CONTENT,
     PM_ERR_EXPECT_WHEN_DELIMITER,
     PM_ERR_EXPRESSION_BARE_HASH,
+    PM_ERR_EXPRESSION_NOT_WRITABLE,
+    PM_ERR_EXPRESSION_NOT_WRITABLE_ENCODING,
+    PM_ERR_EXPRESSION_NOT_WRITABLE_FALSE,
+    PM_ERR_EXPRESSION_NOT_WRITABLE_FILE,
+    PM_ERR_EXPRESSION_NOT_WRITABLE_LINE,
+    PM_ERR_EXPRESSION_NOT_WRITABLE_NIL,
+    PM_ERR_EXPRESSION_NOT_WRITABLE_NUMBERED,
+    PM_ERR_EXPRESSION_NOT_WRITABLE_SELF,
+    PM_ERR_EXPRESSION_NOT_WRITABLE_TRUE,
+    PM_ERR_FLOAT_PARSE,
     PM_ERR_FOR_COLLECTION,
     PM_ERR_FOR_IN,
     PM_ERR_FOR_INDEX,
     PM_ERR_FOR_TERM,
+    PM_ERR_GLOBAL_VARIABLE_BARE,
     PM_ERR_HASH_EXPRESSION_AFTER_LABEL,
     PM_ERR_HASH_KEY,
     PM_ERR_HASH_ROCKET,
     PM_ERR_HASH_TERM,
     PM_ERR_HASH_VALUE,
+    PM_ERR_HEREDOC_IDENTIFIER,
     PM_ERR_HEREDOC_TERM,
     PM_ERR_INCOMPLETE_QUESTION_MARK,
     PM_ERR_INCOMPLETE_VARIABLE_CLASS,
+    PM_ERR_INCOMPLETE_VARIABLE_CLASS_3_3,
     PM_ERR_INCOMPLETE_VARIABLE_INSTANCE,
+    PM_ERR_INCOMPLETE_VARIABLE_INSTANCE_3_3,
+    PM_ERR_INSTANCE_VARIABLE_BARE,
+    PM_ERR_INVALID_BLOCK_EXIT,
+    PM_ERR_INVALID_CHARACTER,
+    PM_ERR_INVALID_COMMA,
     PM_ERR_INVALID_ENCODING_MAGIC_COMMENT,
+    PM_ERR_INVALID_ESCAPE_CHARACTER,
     PM_ERR_INVALID_FLOAT_EXPONENT,
+    PM_ERR_INVALID_LOCAL_VARIABLE_READ,
+    PM_ERR_INVALID_LOCAL_VARIABLE_WRITE,
+    PM_ERR_INVALID_MULTIBYTE_CHAR,
+    PM_ERR_INVALID_MULTIBYTE_CHARACTER,
+    PM_ERR_INVALID_MULTIBYTE_ESCAPE,
     PM_ERR_INVALID_NUMBER_BINARY,
     PM_ERR_INVALID_NUMBER_DECIMAL,
+    PM_ERR_INVALID_NUMBER_FRACTION,
     PM_ERR_INVALID_NUMBER_HEXADECIMAL,
     PM_ERR_INVALID_NUMBER_OCTAL,
-    PM_ERR_INVALID_NUMBER_UNDERSCORE,
-    PM_ERR_INVALID_CHARACTER,
-    PM_ERR_INVALID_MULTIBYTE_CHARACTER,
-    PM_ERR_INVALID_PRINTABLE_CHARACTER,
+    PM_ERR_INVALID_NUMBER_UNDERSCORE_INNER,
+    PM_ERR_INVALID_NUMBER_UNDERSCORE_TRAILING,
     PM_ERR_INVALID_PERCENT,
+    PM_ERR_INVALID_PERCENT_EOF,
+    PM_ERR_INVALID_PRINTABLE_CHARACTER,
+    PM_ERR_INVALID_RETRY_AFTER_ELSE,
+    PM_ERR_INVALID_RETRY_AFTER_ENSURE,
+    PM_ERR_INVALID_RETRY_WITHOUT_RESCUE,
+    PM_ERR_INVALID_SYMBOL,
     PM_ERR_INVALID_VARIABLE_GLOBAL,
-    PM_ERR_IT_NOT_ALLOWED,
+    PM_ERR_INVALID_VARIABLE_GLOBAL_3_3,
+    PM_ERR_INVALID_YIELD,
+    PM_ERR_IT_NOT_ALLOWED_NUMBERED,
+    PM_ERR_IT_NOT_ALLOWED_ORDINARY,
     PM_ERR_LAMBDA_OPEN,
     PM_ERR_LAMBDA_TERM_BRACE,
     PM_ERR_LAMBDA_TERM_END,
@@ -220,19 +216,24 @@ typedef enum {
     PM_ERR_MODULE_TERM,
     PM_ERR_MULTI_ASSIGN_MULTI_SPLATS,
     PM_ERR_MULTI_ASSIGN_UNEXPECTED_REST,
-    PM_ERR_NOT_EXPRESSION,
+    PM_ERR_NESTING_TOO_DEEP,
     PM_ERR_NO_LOCAL_VARIABLE,
+    PM_ERR_NON_ASSOCIATIVE_OPERATOR,
+    PM_ERR_NOT_EXPRESSION,
     PM_ERR_NUMBER_LITERAL_UNDERSCORE,
-    PM_ERR_NUMBERED_PARAMETER_NOT_ALLOWED,
-    PM_ERR_NUMBERED_PARAMETER_OUTER_SCOPE,
+    PM_ERR_NUMBERED_PARAMETER_INNER_BLOCK,
+    PM_ERR_NUMBERED_PARAMETER_IT,
+    PM_ERR_NUMBERED_PARAMETER_ORDINARY,
+    PM_ERR_NUMBERED_PARAMETER_OUTER_BLOCK,
     PM_ERR_OPERATOR_MULTI_ASSIGN,
     PM_ERR_OPERATOR_WRITE_ARGUMENTS,
     PM_ERR_OPERATOR_WRITE_BLOCK,
     PM_ERR_PARAMETER_ASSOC_SPLAT_MULTI,
     PM_ERR_PARAMETER_BLOCK_MULTI,
     PM_ERR_PARAMETER_CIRCULAR,
+    PM_ERR_PARAMETER_FORWARDING_AFTER_REST,
     PM_ERR_PARAMETER_METHOD_NAME,
-    PM_ERR_PARAMETER_NAME_REPEAT,
+    PM_ERR_PARAMETER_NAME_DUPLICATED,
     PM_ERR_PARAMETER_NO_DEFAULT,
     PM_ERR_PARAMETER_NO_DEFAULT_KW,
     PM_ERR_PARAMETER_NUMBERED_RESERVED,
@@ -240,10 +241,13 @@ typedef enum {
     PM_ERR_PARAMETER_SPLAT_MULTI,
     PM_ERR_PARAMETER_STAR,
     PM_ERR_PARAMETER_UNEXPECTED_FWD,
+    PM_ERR_PARAMETER_UNEXPECTED_NO_KW,
     PM_ERR_PARAMETER_WILD_LOOSE_COMMA,
+    PM_ERR_PATTERN_ARRAY_MULTIPLE_RESTS,
+    PM_ERR_PATTERN_CAPTURE_DUPLICATE,
     PM_ERR_PATTERN_EXPRESSION_AFTER_BRACKET,
-    PM_ERR_PATTERN_EXPRESSION_AFTER_HROCKET,
     PM_ERR_PATTERN_EXPRESSION_AFTER_COMMA,
+    PM_ERR_PATTERN_EXPRESSION_AFTER_HROCKET,
     PM_ERR_PATTERN_EXPRESSION_AFTER_IN,
     PM_ERR_PATTERN_EXPRESSION_AFTER_KEY,
     PM_ERR_PATTERN_EXPRESSION_AFTER_PAREN,
@@ -251,8 +255,13 @@ typedef enum {
     PM_ERR_PATTERN_EXPRESSION_AFTER_PIPE,
     PM_ERR_PATTERN_EXPRESSION_AFTER_RANGE,
     PM_ERR_PATTERN_EXPRESSION_AFTER_REST,
+    PM_ERR_PATTERN_FIND_MISSING_INNER,
+    PM_ERR_PATTERN_HASH_IMPLICIT,
     PM_ERR_PATTERN_HASH_KEY,
+    PM_ERR_PATTERN_HASH_KEY_DUPLICATE,
+    PM_ERR_PATTERN_HASH_KEY_INTERPOLATED,
     PM_ERR_PATTERN_HASH_KEY_LABEL,
+    PM_ERR_PATTERN_HASH_KEY_LOCALS,
     PM_ERR_PATTERN_IDENT_AFTER_HROCKET,
     PM_ERR_PATTERN_LABEL_AFTER_COMMA,
     PM_ERR_PATTERN_REST,
@@ -260,12 +269,20 @@ typedef enum {
     PM_ERR_PATTERN_TERM_BRACKET,
     PM_ERR_PATTERN_TERM_PAREN,
     PM_ERR_PIPEPIPEEQ_MULTI_ASSIGN,
+    PM_ERR_REGEXP_ENCODING_OPTION_MISMATCH,
+    PM_ERR_REGEXP_INCOMPAT_CHAR_ENCODING,
+    PM_ERR_REGEXP_INVALID_UNICODE_RANGE,
+    PM_ERR_REGEXP_NON_ESCAPED_MBC,
+    PM_ERR_REGEXP_PARSE_ERROR,
     PM_ERR_REGEXP_TERM,
+    PM_ERR_REGEXP_UNKNOWN_OPTIONS,
+    PM_ERR_REGEXP_UTF8_CHAR_NON_UTF8_REGEXP,
     PM_ERR_RESCUE_EXPRESSION,
     PM_ERR_RESCUE_MODIFIER_VALUE,
     PM_ERR_RESCUE_TERM,
     PM_ERR_RESCUE_VARIABLE,
     PM_ERR_RETURN_INVALID,
+    PM_ERR_SCRIPT_NOT_FOUND,
     PM_ERR_SINGLETON_FOR_LITERALS,
     PM_ERR_STATEMENT_ALIAS,
     PM_ERR_STATEMENT_POSTEXE_END,
@@ -281,10 +298,18 @@ typedef enum {
     PM_ERR_TERNARY_COLON,
     PM_ERR_TERNARY_EXPRESSION_FALSE,
     PM_ERR_TERNARY_EXPRESSION_TRUE,
+    PM_ERR_UNARY_DISALLOWED,
     PM_ERR_UNARY_RECEIVER,
+    PM_ERR_UNDEF_ARGUMENT,
+    PM_ERR_UNEXPECTED_BLOCK_ARGUMENT,
+    PM_ERR_UNEXPECTED_INDEX_BLOCK,
+    PM_ERR_UNEXPECTED_INDEX_KEYWORDS,
+    PM_ERR_UNEXPECTED_LABEL,
+    PM_ERR_UNEXPECTED_MULTI_WRITE,
+    PM_ERR_UNEXPECTED_RANGE_OPERATOR,
+    PM_ERR_UNEXPECTED_SAFE_NAVIGATION,
     PM_ERR_UNEXPECTED_TOKEN_CLOSE_CONTEXT,
     PM_ERR_UNEXPECTED_TOKEN_IGNORE,
-    PM_ERR_UNDEF_ARGUMENT,
     PM_ERR_UNTIL_TERM,
     PM_ERR_VOID_EXPRESSION,
     PM_ERR_WHILE_TERM,
@@ -293,17 +318,104 @@ typedef enum {
     PM_ERR_WRITE_TARGET_UNEXPECTED,
     PM_ERR_XSTRING_TERM,
 
-    // These are the warning codes.
+    // These are the warning diagnostics.
+    PM_WARN_AMBIGUOUS_BINARY_OPERATOR,
     PM_WARN_AMBIGUOUS_FIRST_ARGUMENT_MINUS,
     PM_WARN_AMBIGUOUS_FIRST_ARGUMENT_PLUS,
+    PM_WARN_AMBIGUOUS_PREFIX_AMPERSAND,
     PM_WARN_AMBIGUOUS_PREFIX_STAR,
+    PM_WARN_AMBIGUOUS_PREFIX_STAR_STAR,
     PM_WARN_AMBIGUOUS_SLASH,
+    PM_WARN_COMPARISON_AFTER_COMPARISON,
+    PM_WARN_DOT_DOT_DOT_EOL,
+    PM_WARN_EQUAL_IN_CONDITIONAL,
+    PM_WARN_EQUAL_IN_CONDITIONAL_3_3,
     PM_WARN_END_IN_METHOD,
-
-    // This is the number of diagnostic codes.
-    PM_DIAGNOSTIC_ID_LEN,
+    PM_WARN_DUPLICATED_HASH_KEY,
+    PM_WARN_DUPLICATED_WHEN_CLAUSE,
+    PM_WARN_FLOAT_OUT_OF_RANGE,
+    PM_WARN_IGNORED_FROZEN_STRING_LITERAL,
+    PM_WARN_INDENTATION_MISMATCH,
+    PM_WARN_INTEGER_IN_FLIP_FLOP,
+    PM_WARN_INVALID_CHARACTER,
+    PM_WARN_INVALID_MAGIC_COMMENT_VALUE,
+    PM_WARN_INVALID_NUMBERED_REFERENCE,
+    PM_WARN_KEYWORD_EOL,
+    PM_WARN_LITERAL_IN_CONDITION_DEFAULT,
+    PM_WARN_LITERAL_IN_CONDITION_VERBOSE,
+    PM_WARN_SHAREABLE_CONSTANT_VALUE_LINE,
+    PM_WARN_SHEBANG_CARRIAGE_RETURN,
+    PM_WARN_UNEXPECTED_CARRIAGE_RETURN,
+    PM_WARN_UNREACHABLE_STATEMENT,
+    PM_WARN_UNUSED_LOCAL_VARIABLE,
+    PM_WARN_VOID_STATEMENT,
 } pm_diagnostic_id_t;
 
+/**
+ * This struct represents a diagnostic generated during parsing.
+ *
+ * @extends pm_list_node_t
+ */
+typedef struct {
+    /** The embedded base node. */
+    pm_list_node_t node;
+
+    /** The location of the diagnostic in the source. */
+    pm_location_t location;
+
+    /** The ID of the diagnostic. */
+    pm_diagnostic_id_t diag_id;
+
+    /** The message associated with the diagnostic. */
+    const char *message;
+
+    /**
+     * Whether or not the memory related to the message of this diagnostic is
+     * owned by this diagnostic. If it is, it needs to be freed when the
+     * diagnostic is freed.
+     */
+    bool owned;
+
+    /**
+     * The level of the diagnostic, see `pm_error_level_t` and
+     * `pm_warning_level_t` for possible values.
+     */
+    uint8_t level;
+} pm_diagnostic_t;
+
+/**
+ * The levels of errors generated during parsing.
+ */
+typedef enum {
+    /** For errors that should raise a syntax error. */
+    PM_ERROR_LEVEL_SYNTAX = 0,
+
+    /** For errors that should raise an argument error. */
+    PM_ERROR_LEVEL_ARGUMENT = 1,
+
+    /** For errors that should raise a load error. */
+    PM_ERROR_LEVEL_LOAD = 2
+} pm_error_level_t;
+
+/**
+ * The levels of warnings generated during parsing.
+ */
+typedef enum {
+    /** For warnings which should be emitted if $VERBOSE != nil. */
+    PM_WARNING_LEVEL_DEFAULT = 0,
+
+    /** For warnings which should be emitted if $VERBOSE == true. */
+    PM_WARNING_LEVEL_VERBOSE = 1
+} pm_warning_level_t;
+
+/**
+ * Get the human-readable name of the given diagnostic ID.
+ *
+ * @param diag_id The diagnostic ID.
+ * @return The human-readable name of the diagnostic ID.
+ */
+const char * pm_diagnostic_id_human(pm_diagnostic_id_t diag_id);
+
 /**
  * Append a diagnostic to the given list of diagnostics that is using shared
  * memory for its message.

data/include/prism/encoding.h

@@ -135,7 +135,14 @@ extern const uint8_t pm_encoding_unicode_table[256];
  */
 typedef enum {
     PM_ENCODING_UTF_8 = 0,
+    PM_ENCODING_US_ASCII,
     PM_ENCODING_ASCII_8BIT,
+    PM_ENCODING_EUC_JP,
+    PM_ENCODING_WINDOWS_31J,
+
+// We optionally support excluding the full set of encodings to only support the
+// minimum necessary to process Ruby code without encoding comments.
+#ifndef PRISM_ENCODING_EXCLUDE_FULL
     PM_ENCODING_BIG5,
     PM_ENCODING_BIG5_HKSCS,
     PM_ENCODING_BIG5_UAO,
@@ -148,7 +155,6 @@ typedef enum {
     PM_ENCODING_CP950,
     PM_ENCODING_CP951,
     PM_ENCODING_EMACS_MULE,
-    PM_ENCODING_EUC_JP,
     PM_ENCODING_EUC_JP_MS,
     PM_ENCODING_EUC_JIS_2004,
     PM_ENCODING_EUC_KR,
@@ -208,7 +214,6 @@ typedef enum {
     PM_ENCODING_STATELESS_ISO_2022_JP,
     PM_ENCODING_STATELESS_ISO_2022_JP_KDDI,
     PM_ENCODING_TIS_620,
-    PM_ENCODING_US_ASCII,
     PM_ENCODING_UTF8_MAC,
     PM_ENCODING_UTF8_DOCOMO,
     PM_ENCODING_UTF8_KDDI,
@@ -222,8 +227,9 @@ typedef enum {
     PM_ENCODING_WINDOWS_1256,
     PM_ENCODING_WINDOWS_1257,
     PM_ENCODING_WINDOWS_1258,
-    PM_ENCODING_WINDOWS_31J,
     PM_ENCODING_WINDOWS_874,
+#endif
+
     PM_ENCODING_MAXIMUM
 } pm_encoding_type_t;
 
@@ -248,10 +254,22 @@ extern const pm_encoding_t pm_encodings[PM_ENCODING_MAXIMUM];
 /**
  * This is the ASCII-8BIT encoding. We need a reference to it so that pm_strpbrk
  * can compare against it because invalid multibyte characters are not a thing
- * in this encoding.
+ * in this encoding. It is also needed for handling Regexp encoding flags.
  */
 #define PM_ENCODING_ASCII_8BIT_ENTRY (&pm_encodings[PM_ENCODING_ASCII_8BIT])
 
+/**
+ * This is the EUC-JP encoding. We need a reference to it to quickly process
+ * regular expression modifiers.
+ */
+#define PM_ENCODING_EUC_JP_ENTRY (&pm_encodings[PM_ENCODING_EUC_JP])
+
+/**
+ * This is the Windows-31J encoding. We need a reference to it to quickly
+ * process regular expression modifiers.
+ */
+#define PM_ENCODING_WINDOWS_31J_ENTRY (&pm_encodings[PM_ENCODING_WINDOWS_31J])
+
 /**
  * Parse the given name of an encoding and return a pointer to the corresponding
  * encoding struct if one can be found, otherwise return NULL.