jruby-prism-parser 0.23.0.pre.SNAPSHOT-java

data/include/prism/util/pm_list.h

@@ -0,0 +1,97 @@
+/**
+ * @file pm_list.h
+ *
+ * An abstract linked list.
+ */
+#ifndef PRISM_LIST_H
+#define PRISM_LIST_H
+
+#include "prism/defines.h"
+
+#include <stdbool.h>
+#include <stddef.h>
+#include <stdint.h>
+#include <stdlib.h>
+
+/**
+ * This struct represents an abstract linked list that provides common
+ * functionality. It is meant to be used any time a linked list is necessary to
+ * store data.
+ *
+ * The linked list itself operates off a set of pointers. Because the pointers
+ * are not necessarily sequential, they can be of any size. We use this fact to
+ * allow the consumer of this linked list to extend the node struct to include
+ * any data they want. This is done by using the pm_list_node_t as the first
+ * member of the struct.
+ *
+ * For example, if we want to store a list of integers, we can do the following:
+ *
+ * ```c
+ * typedef struct {
+ *     pm_list_node_t node;
+ *     int value;
+ * } pm_int_node_t;
+ *
+ * pm_list_t list = { 0 };
+ * pm_int_node_t *node = malloc(sizeof(pm_int_node_t));
+ * node->value = 5;
+ *
+ * pm_list_append(&list, &node->node);
+ * ```
+ *
+ * The pm_list_t struct is used to represent the overall linked list. It
+ * contains a pointer to the head and tail of the list. This allows for easy
+ * iteration and appending of new nodes.
+ */
+typedef struct pm_list_node {
+    /** A pointer to the next node in the list. */
+    struct pm_list_node *next;
+} pm_list_node_t;
+
+/**
+ * This represents the overall linked list. It keeps a pointer to the head and
+ * tail so that iteration is easy and pushing new nodes is easy.
+ */
+typedef struct {
+    /** The size of the list. */
+    size_t size;
+
+    /** A pointer to the head of the list. */
+    pm_list_node_t *head;
+
+    /** A pointer to the tail of the list. */
+    pm_list_node_t *tail;
+} pm_list_t;
+
+/**
+ * Returns true if the given list is empty.
+ *
+ * @param list The list to check.
+ * @return True if the given list is empty, otherwise false.
+ */
+PRISM_EXPORTED_FUNCTION bool pm_list_empty_p(pm_list_t *list);
+
+/**
+ * Returns the size of the list.
+ *
+ * @param list The list to check.
+ * @return The size of the list.
+ */
+PRISM_EXPORTED_FUNCTION size_t pm_list_size(pm_list_t *list);
+
+/**
+ * Append a node to the given list.
+ *
+ * @param list The list to append to.
+ * @param node The node to append.
+ */
+void pm_list_append(pm_list_t *list, pm_list_node_t *node);
+
+/**
+ * Deallocate the internal state of the given list.
+ *
+ * @param list The list to free.
+ */
+PRISM_EXPORTED_FUNCTION void pm_list_free(pm_list_t *list);
+
+#endif

data/include/prism/util/pm_memchr.h

@@ -0,0 +1,29 @@
+/**
+ * @file pm_memchr.h
+ *
+ * A custom memchr implementation.
+ */
+#ifndef PRISM_MEMCHR_H
+#define PRISM_MEMCHR_H
+
+#include "prism/defines.h"
+#include "prism/encoding.h"
+
+#include <stddef.h>
+
+/**
+ * We need to roll our own memchr to handle cases where the encoding changes and
+ * we need to search for a character in a buffer that could be the trailing byte
+ * of a multibyte character.
+ *
+ * @param source The source string.
+ * @param character The character to search for.
+ * @param number The maximum number of bytes to search.
+ * @param encoding_changed Whether the encoding changed.
+ * @param encoding A pointer to the encoding.
+ * @return A pointer to the first occurrence of the character in the source
+ *     string, or NULL if no such character exists.
+ */
+void * pm_memchr(const void *source, int character, size_t number, bool encoding_changed, const pm_encoding_t *encoding);
+
+#endif

data/include/prism/util/pm_newline_list.h

@@ -0,0 +1,93 @@
+/**
+ * @file pm_newline_list.h
+ *
+ * A list of byte offsets of newlines in a string.
+ *
+ * When compiling the syntax tree, it's necessary to know the line and column
+ * of many nodes. This is necessary to support things like error messages,
+ * tracepoints, etc.
+ *
+ * It's possible that we could store the start line, start column, end line, and
+ * end column on every node in addition to the offsets that we already store,
+ * but that would be quite a lot of memory overhead.
+ */
+#ifndef PRISM_NEWLINE_LIST_H
+#define PRISM_NEWLINE_LIST_H
+
+#include "prism/defines.h"
+
+#include <assert.h>
+#include <stdbool.h>
+#include <stddef.h>
+#include <stdlib.h>
+
+/**
+ * A list of offsets of newlines in a string. The offsets are assumed to be
+ * sorted/inserted in ascending order.
+ */
+typedef struct {
+    /** A pointer to the start of the source string. */
+    const uint8_t *start;
+
+    /** The number of offsets in the list. */
+    size_t size;
+
+    /** The capacity of the list that has been allocated. */
+    size_t capacity;
+
+    /** The list of offsets. */
+    size_t *offsets;
+} pm_newline_list_t;
+
+/**
+ * A line and column in a string.
+ */
+typedef struct {
+    /** The line number. */
+    size_t line;
+
+    /** The column number. */
+    size_t column;
+} pm_line_column_t;
+
+/**
+ * Initialize a new newline list with the given capacity. Returns true if the
+ * allocation of the offsets succeeds, otherwise returns false.
+ *
+ * @param list The list to initialize.
+ * @param start A pointer to the start of the source string.
+ * @param capacity The initial capacity of the list.
+ * @return True if the allocation of the offsets succeeds, otherwise false.
+ */
+bool pm_newline_list_init(pm_newline_list_t *list, const uint8_t *start, size_t capacity);
+
+/**
+ * Append a new offset to the newline list. Returns true if the reallocation of
+ * the offsets succeeds (if one was necessary), otherwise returns false.
+ *
+ * @param list The list to append to.
+ * @param cursor A pointer to the offset to append.
+ * @return True if the reallocation of the offsets succeeds (if one was
+ *     necessary), otherwise false.
+ */
+bool pm_newline_list_append(pm_newline_list_t *list, const uint8_t *cursor);
+
+/**
+ * Returns the line and column of the given offset. If the offset is not in the
+ * list, the line and column of the closest offset less than the given offset
+ * are returned.
+ *
+ * @param list The list to search.
+ * @param cursor A pointer to the offset to search for.
+ * @return The line and column of the given offset.
+ */
+pm_line_column_t pm_newline_list_line_column(const pm_newline_list_t *list, const uint8_t *cursor);
+
+/**
+ * Free the internal memory allocated for the newline list.
+ *
+ * @param list The list to free.
+ */
+void pm_newline_list_free(pm_newline_list_t *list);
+
+#endif

data/include/prism/util/pm_state_stack.h

@@ -0,0 +1,42 @@
+/**
+ * @file pm_state_stack.h
+ *
+ * A stack of boolean values.
+ */
+#ifndef PRISM_STATE_STACK_H
+#define PRISM_STATE_STACK_H
+
+#include "prism/defines.h"
+
+#include <stdbool.h>
+#include <stdint.h>
+
+/**
+ * A struct that represents a stack of boolean values.
+ */
+typedef uint32_t pm_state_stack_t;
+
+/**
+ * Pushes a value onto the stack.
+ *
+ * @param stack The stack to push the value onto.
+ * @param value The value to push onto the stack.
+ */
+void pm_state_stack_push(pm_state_stack_t *stack, bool value);
+
+/**
+ * Pops a value off the stack.
+ *
+ * @param stack The stack to pop the value off of.
+ */
+void pm_state_stack_pop(pm_state_stack_t *stack);
+
+/**
+ * Returns the value at the top of the stack.
+ *
+ * @param stack The stack to get the value from.
+ * @return The value at the top of the stack.
+ */
+bool pm_state_stack_p(pm_state_stack_t *stack);
+
+#endif

data/include/prism/util/pm_string.h

@@ -0,0 +1,150 @@
+/**
+ * @file pm_string.h
+ *
+ * A generic string type that can have various ownership semantics.
+ */
+#ifndef PRISM_STRING_H
+#define PRISM_STRING_H
+
+#include "prism/defines.h"
+
+#include <assert.h>
+#include <stdbool.h>
+#include <stddef.h>
+#include <stdlib.h>
+#include <string.h>
+
+// The following headers are necessary to read files using demand paging.
+#ifdef _WIN32
+#include <windows.h>
+#else
+#include <fcntl.h>
+#include <sys/mman.h>
+#include <sys/stat.h>
+#include <unistd.h>
+#endif
+
+/**
+ * A generic string type that can have various ownership semantics.
+ */
+typedef struct {
+    /** A pointer to the start of the string. */
+    const uint8_t *source;
+
+    /** The length of the string in bytes of memory. */
+    size_t length;
+
+    /** The type of the string. This field determines how the string should be freed. */
+    enum {
+        /** This string is a constant string, and should not be freed. */
+        PM_STRING_CONSTANT,
+
+        /** This is a slice of another string, and should not be freed. */
+        PM_STRING_SHARED,
+
+        /** This string owns its memory, and should be freed using `pm_string_free`. */
+        PM_STRING_OWNED,
+
+        /** This string is a memory-mapped file, and should be freed using `pm_string_free`. */
+        PM_STRING_MAPPED
+    } type;
+} pm_string_t;
+
+/**
+ * Returns the size of the pm_string_t struct. This is necessary to allocate the
+ * correct amount of memory in the FFI backend.
+ *
+ * @return The size of the pm_string_t struct.
+ */
+PRISM_EXPORTED_FUNCTION size_t pm_string_sizeof(void);
+
+/**
+ * Defines an empty string. This is useful for initializing a string that will
+ * be filled in later.
+ */
+#define PM_STRING_EMPTY ((pm_string_t) { .type = PM_STRING_CONSTANT, .source = NULL, .length = 0 })
+
+/**
+ * Initialize a shared string that is based on initial input.
+ *
+ * @param string The string to initialize.
+ * @param start The start of the string.
+ * @param end The end of the string.
+ */
+void pm_string_shared_init(pm_string_t *string, const uint8_t *start, const uint8_t *end);
+
+/**
+ * Initialize an owned string that is responsible for freeing allocated memory.
+ *
+ * @param string The string to initialize.
+ * @param source The source of the string.
+ * @param length The length of the string.
+ */
+void pm_string_owned_init(pm_string_t *string, uint8_t *source, size_t length);
+
+/**
+ * Initialize a constant string that doesn't own its memory source.
+ *
+ * @param string The string to initialize.
+ * @param source The source of the string.
+ * @param length The length of the string.
+ */
+void pm_string_constant_init(pm_string_t *string, const char *source, size_t length);
+
+/**
+ * Read the file indicated by the filepath parameter into source and load its
+ * contents and size into the given `pm_string_t`. The given `pm_string_t`
+ * should be freed using `pm_string_free` when it is no longer used.
+ *
+ * We want to use demand paging as much as possible in order to avoid having to
+ * read the entire file into memory (which could be detrimental to performance
+ * for large files). This means that if we're on windows we'll use
+ * `MapViewOfFile`, on POSIX systems that have access to `mmap` we'll use
+ * `mmap`, and on other POSIX systems we'll use `read`.
+ *
+ * @param string The string to initialize.
+ * @param filepath The filepath to read.
+ * @return Whether or not the file was successfully mapped.
+ */
+PRISM_EXPORTED_FUNCTION bool pm_string_mapped_init(pm_string_t *string, const char *filepath);
+
+/**
+ * Returns the memory size associated with the string.
+ *
+ * @param string The string to get the memory size of.
+ * @return The size of the memory associated with the string.
+ */
+size_t pm_string_memsize(const pm_string_t *string);
+
+/**
+ * Ensure the string is owned. If it is not, then reinitialize it as owned and
+ * copy over the previous source.
+ *
+ * @param string The string to ensure is owned.
+ */
+void pm_string_ensure_owned(pm_string_t *string);
+
+/**
+ * Returns the length associated with the string.
+ *
+ * @param string The string to get the length of.
+ * @return The length of the string.
+ */
+PRISM_EXPORTED_FUNCTION size_t pm_string_length(const pm_string_t *string);
+
+/**
+ * Returns the start pointer associated with the string.
+ *
+ * @param string The string to get the start pointer of.
+ * @return The start pointer of the string.
+ */
+PRISM_EXPORTED_FUNCTION const uint8_t * pm_string_source(const pm_string_t *string);
+
+/**
+ * Free the associated memory of the given string.
+ *
+ * @param string The string to free.
+ */
+PRISM_EXPORTED_FUNCTION void pm_string_free(pm_string_t *string);
+
+#endif

data/include/prism/util/pm_string_list.h

@@ -0,0 +1,44 @@
+/**
+ * @file pm_string_list.h
+ *
+ * A list of strings.
+ */
+#ifndef PRISM_STRING_LIST_H
+#define PRISM_STRING_LIST_H
+
+#include "prism/defines.h"
+#include "prism/util/pm_string.h"
+
+#include <stddef.h>
+#include <stdlib.h>
+
+/**
+ * A list of strings.
+ */
+typedef struct {
+    /** The length of the string list. */
+    size_t length;
+
+    /** The capacity of the string list that has been allocated. */
+    size_t capacity;
+
+    /** A pointer to the start of the string list. */
+    pm_string_t *strings;
+} pm_string_list_t;
+
+/**
+ * Append a pm_string_t to the given string list.
+ *
+ * @param string_list The string list to append to.
+ * @param string The string to append.
+ */
+void pm_string_list_append(pm_string_list_t *string_list, pm_string_t *string);
+
+/**
+ * Free the memory associated with the string list.
+ *
+ * @param string_list The string list to free.
+ */
+PRISM_EXPORTED_FUNCTION void pm_string_list_free(pm_string_list_t *string_list);
+
+#endif

data/include/prism/util/pm_strncasecmp.h

@@ -0,0 +1,32 @@
+/**
+ * @file pm_strncasecmp.h
+ *
+ * A custom strncasecmp implementation.
+ */
+#ifndef PRISM_STRNCASECMP_H
+#define PRISM_STRNCASECMP_H
+
+#include "prism/defines.h"
+
+#include <ctype.h>
+#include <stddef.h>
+#include <stdint.h>
+
+/**
+ * Compare two strings, ignoring case, up to the given length. Returns 0 if the
+ * strings are equal, a negative number if string1 is less than string2, or a
+ * positive number if string1 is greater than string2.
+ *
+ * Note that this is effectively our own implementation of strncasecmp, but it's
+ * not available on all of the platforms we want to support so we're rolling it
+ * here.
+ *
+ * @param string1 The first string to compare.
+ * @param string2 The second string to compare
+ * @param length The maximum number of characters to compare.
+ * @return 0 if the strings are equal, a negative number if string1 is less than
+ *     string2, or a positive number if string1 is greater than string2.
+ */
+int pm_strncasecmp(const uint8_t *string1, const uint8_t *string2, size_t length);
+
+#endif

data/include/prism/util/pm_strpbrk.h

@@ -0,0 +1,46 @@
+/**
+ * @file pm_strpbrk.h
+ *
+ * A custom strpbrk implementation.
+ */
+#ifndef PRISM_STRPBRK_H
+#define PRISM_STRPBRK_H
+
+#include "prism/defines.h"
+#include "prism/diagnostic.h"
+#include "prism/parser.h"
+
+#include <stddef.h>
+#include <string.h>
+
+/**
+ * Here we have rolled our own version of strpbrk. The standard library strpbrk
+ * has undefined behavior when the source string is not null-terminated. We want
+ * to support strings that are not null-terminated because pm_parse does not
+ * have the contract that the string is null-terminated. (This is desirable
+ * because it means the extension can call pm_parse with the result of a call to
+ * mmap).
+ *
+ * The standard library strpbrk also does not support passing a maximum length
+ * to search. We want to support this for the reason mentioned above, but we
+ * also don't want it to stop on null bytes. Ruby actually allows null bytes
+ * within strings, comments, regular expressions, etc. So we need to be able to
+ * skip past them.
+ *
+ * Finally, we want to support encodings wherein the charset could contain
+ * characters that are trailing bytes of multi-byte characters. For example, in
+ * Shift-JIS, the backslash character can be a trailing byte. In that case we
+ * need to take a slower path and iterate one multi-byte character at a time.
+ *
+ * @param parser The parser.
+ * @param source The source to search.
+ * @param charset The charset to search for.
+ * @param length The maximum number of bytes to search.
+ * @param validate Whether to validate that the source string is valid in the
+ *     current encoding of the parser.
+ * @return A pointer to the first character in the source string that is in the
+ *     charset, or NULL if no such character exists.
+ */
+const uint8_t * pm_strpbrk(pm_parser_t *parser, const uint8_t *source, const uint8_t *charset, ptrdiff_t length, bool validate);
+
+#endif

data/include/prism/version.h

@@ -0,0 +1,29 @@
+/**
+ * @file version.h
+ *
+ * The version of the Prism library.
+ */
+#ifndef PRISM_VERSION_H
+#define PRISM_VERSION_H
+
+/**
+ * The major version of the Prism library as an int.
+ */
+#define PRISM_VERSION_MAJOR 0
+
+/**
+ * The minor version of the Prism library as an int.
+ */
+#define PRISM_VERSION_MINOR 23
+
+/**
+ * The patch version of the Prism library as an int.
+ */
+#define PRISM_VERSION_PATCH 0
+
+/**
+ * The version of the Prism library as a constant string.
+ */
+#define PRISM_VERSION "0.23.0"
+
+#endif