prism 1.8.1 → 2.0.0.rc1

data/include/prism/util/pm_string.h

@@ -1,200 +0,0 @@
-/**
- * @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 <errno.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>
-#elif defined(_POSIX_MAPPED_FILES)
-#include <fcntl.h>
-#include <sys/mman.h>
-#include <sys/stat.h>
-#elif defined(PRISM_HAS_FILESYSTEM)
-#include <fcntl.h>
-#include <sys/stat.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,
-
-#ifdef PRISM_HAS_MMAP
-        /** This string is a memory-mapped file, and should be freed using `pm_string_free()`. */
-        PM_STRING_MAPPED
-#endif
-    } 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);
-
-/**
- * Represents the result of calling pm_string_mapped_init or
- * pm_string_file_init. We need this additional information because there is
- * not a platform-agnostic way to indicate that the file that was attempted to
- * be opened was a directory.
- */
-typedef enum {
-    /** Indicates that the string was successfully initialized. */
-    PM_STRING_INIT_SUCCESS = 0,
-    /**
-     * Indicates a generic error from a string_*_init function, where the type
-     * of error should be read from `errno` or `GetLastError()`.
-     */
-    PM_STRING_INIT_ERROR_GENERIC = 1,
-    /**
-     * Indicates that the file that was attempted to be opened was a directory.
-     */
-    PM_STRING_INIT_ERROR_DIRECTORY = 2
-} pm_string_init_result_t;
-
-/**
- * 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 The success of the read, indicated by the value of the enum.
- *
- * \public \memberof pm_string_t
- */
-PRISM_EXPORTED_FUNCTION pm_string_init_result_t pm_string_mapped_init(pm_string_t *string, const char *filepath);
-
-/**
- * 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.
- *
- * @param string The string to initialize.
- * @param filepath The filepath to read.
- * @return The success of the read, indicated by the value of the enum.
- *
- * \public \memberof pm_string_t
- */
-PRISM_EXPORTED_FUNCTION pm_string_init_result_t pm_string_file_init(pm_string_t *string, const char *filepath);
-
-/**
- * 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);
-
-/**
- * Compare the underlying lengths and bytes of two strings. Returns 0 if the
- * strings are equal, a negative number if the left string is less than the
- * right string, and a positive number if the left string is greater than the
- * right string.
- *
- * @param left The left string to compare.
- * @param right The right string to compare.
- * @return The comparison result.
- */
-int pm_string_compare(const pm_string_t *left, const pm_string_t *right);
-
-/**
- * Returns the length associated with the string.
- *
- * @param string The string to get the length of.
- * @return The length of the string.
- *
- * \public \memberof pm_string_t
- */
-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.
- *
- * \public \memberof pm_string_t
- */
-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.
- *
- * \public \memberof pm_string_t
- */
-PRISM_EXPORTED_FUNCTION void pm_string_free(pm_string_t *string);
-
-#endif

data/lib/prism/lex_ripper.rb

@@ -1,64 +0,0 @@
-# frozen_string_literal: true
-# :markup: markdown
-
-require "ripper"
-
-module Prism
-  # This is a class that wraps the Ripper lexer to produce almost exactly the
-  # same tokens.
-  class LexRipper # :nodoc:
-    attr_reader :source
-
-    def initialize(source)
-      @source = source
-    end
-
-    def result
-      previous = [] #: [[Integer, Integer], Symbol, String, untyped] | []
-      results = [] #: Array[[[Integer, Integer], Symbol, String, untyped]]
-
-      lex(source).each do |token|
-        case token[1]
-        when :on_sp
-          # skip
-        when :on_tstring_content
-          if previous[1] == :on_tstring_content && (token[2].start_with?("\#$") || token[2].start_with?("\#@"))
-            previous[2] << token[2]
-          else
-            results << token
-            previous = token
-          end
-        when :on_words_sep
-          if previous[1] == :on_words_sep
-            previous[2] << token[2]
-          else
-            results << token
-            previous = token
-          end
-        else
-          results << token
-          previous = token
-        end
-      end
-
-      results
-    end
-
-    private
-
-    if Ripper.method(:lex).parameters.assoc(:keyrest)
-      def lex(source)
-        Ripper.lex(source, raise_errors: true)
-      end
-    else
-      def lex(source)
-        ripper = Ripper::Lexer.new(source)
-        ripper.lex.tap do |result|
-          raise SyntaxError, ripper.errors.map(&:message).join(' ;') if ripper.errors.any?
-        end
-      end
-    end
-  end
-
-  private_constant :LexRipper
-end

data/lib/prism/pack.rb

@@ -1,230 +0,0 @@
-# frozen_string_literal: true
-# :markup: markdown
-# typed: ignore
-
-#
-module Prism
-  # A parser for the pack template language.
-  module Pack
-    %i[
-      SPACE
-      COMMENT
-      INTEGER
-      UTF8
-      BER
-      FLOAT
-      STRING_SPACE_PADDED
-      STRING_NULL_PADDED
-      STRING_NULL_TERMINATED
-      STRING_MSB
-      STRING_LSB
-      STRING_HEX_HIGH
-      STRING_HEX_LOW
-      STRING_UU
-      STRING_MIME
-      STRING_BASE64
-      STRING_FIXED
-      STRING_POINTER
-      MOVE
-      BACK
-      NULL
-
-      UNSIGNED
-      SIGNED
-      SIGNED_NA
-
-      AGNOSTIC_ENDIAN
-      LITTLE_ENDIAN
-      BIG_ENDIAN
-      NATIVE_ENDIAN
-      ENDIAN_NA
-
-      SIZE_SHORT
-      SIZE_INT
-      SIZE_LONG
-      SIZE_LONG_LONG
-      SIZE_8
-      SIZE_16
-      SIZE_32
-      SIZE_64
-      SIZE_P
-      SIZE_NA
-
-      LENGTH_FIXED
-      LENGTH_MAX
-      LENGTH_RELATIVE
-      LENGTH_NA
-    ].each do |const|
-      const_set(const, const)
-    end
-
-    # A directive in the pack template language.
-    class Directive
-      # A symbol representing the version of Ruby.
-      attr_reader :version
-
-      # A symbol representing whether or not we are packing or unpacking.
-      attr_reader :variant
-
-      # A byteslice of the source string that this directive represents.
-      attr_reader :source
-
-      # The type of the directive.
-      attr_reader :type
-
-      # The type of signedness of the directive.
-      attr_reader :signed
-
-      # The type of endianness of the directive.
-      attr_reader :endian
-
-      # The size of the directive.
-      attr_reader :size
-
-      # The length type of this directive (used for integers).
-      attr_reader :length_type
-
-      # The length of this directive (used for integers).
-      attr_reader :length
-
-      # Initialize a new directive with the given values.
-      def initialize(version, variant, source, type, signed, endian, size, length_type, length)
-        @version = version
-        @variant = variant
-        @source = source
-        @type = type
-        @signed = signed
-        @endian = endian
-        @size = size
-        @length_type = length_type
-        @length = length
-      end
-
-      # The descriptions of the various types of endianness.
-      ENDIAN_DESCRIPTIONS = {
-        AGNOSTIC_ENDIAN: "agnostic",
-        LITTLE_ENDIAN: "little-endian (VAX)",
-        BIG_ENDIAN: "big-endian (network)",
-        NATIVE_ENDIAN: "native-endian",
-        ENDIAN_NA: "n/a"
-      }
-
-      # The descriptions of the various types of signedness.
-      SIGNED_DESCRIPTIONS = {
-        UNSIGNED: "unsigned",
-        SIGNED: "signed",
-        SIGNED_NA: "n/a"
-      }
-
-      # The descriptions of the various types of sizes.
-      SIZE_DESCRIPTIONS = {
-        SIZE_SHORT: "short",
-        SIZE_INT: "int-width",
-        SIZE_LONG: "long",
-        SIZE_LONG_LONG: "long long",
-        SIZE_8: "8-bit",
-        SIZE_16: "16-bit",
-        SIZE_32: "32-bit",
-        SIZE_64: "64-bit",
-        SIZE_P: "pointer-width"
-      }
-
-      # Provide a human-readable description of the directive.
-      def describe
-        case type
-        when SPACE
-          "whitespace"
-        when COMMENT
-          "comment"
-        when INTEGER
-          if size == SIZE_8
-            base = "#{SIGNED_DESCRIPTIONS[signed]} #{SIZE_DESCRIPTIONS[size]} integer"
-          else
-            base = "#{SIGNED_DESCRIPTIONS[signed]} #{SIZE_DESCRIPTIONS[size]} #{ENDIAN_DESCRIPTIONS[endian]} integer"
-          end
-          case length_type
-          when LENGTH_FIXED
-            if length > 1
-              base + ", x#{length}"
-            else
-              base
-            end
-          when LENGTH_MAX
-            base + ", as many as possible"
-          else
-            raise
-          end
-        when UTF8
-          "UTF-8 character"
-        when BER
-          "BER-compressed integer"
-        when FLOAT
-          "#{SIZE_DESCRIPTIONS[size]} #{ENDIAN_DESCRIPTIONS[endian]} float"
-        when STRING_SPACE_PADDED
-          "arbitrary binary string (space padded)"
-        when STRING_NULL_PADDED
-          "arbitrary binary string (null padded, count is width)"
-        when STRING_NULL_TERMINATED
-          "arbitrary binary string (null padded, count is width), except that null is added with *"
-        when STRING_MSB
-          "bit string (MSB first)"
-        when STRING_LSB
-          "bit string (LSB first)"
-        when STRING_HEX_HIGH
-          "hex string (high nibble first)"
-        when STRING_HEX_LOW
-          "hex string (low nibble first)"
-        when STRING_UU
-          "UU-encoded string"
-        when STRING_MIME
-          "quoted printable, MIME encoding"
-        when STRING_BASE64
-          "base64 encoded string"
-        when STRING_FIXED
-          "pointer to a structure (fixed-length string)"
-        when STRING_POINTER
-          "pointer to a null-terminated string"
-        when MOVE
-          "move to absolute position"
-        when BACK
-          "back up a byte"
-        when NULL
-          "null byte"
-        else
-          raise
-        end
-      end
-    end
-
-    # The result of parsing a pack template.
-    class Format
-      # A list of the directives in the template.
-      attr_reader :directives
-
-      # The encoding of the template.
-      attr_reader :encoding
-
-      # Create a new Format with the given directives and encoding.
-      def initialize(directives, encoding)
-        @directives = directives
-        @encoding = encoding
-      end
-
-      # Provide a human-readable description of the format.
-      def describe
-        source_width = directives.map { |d| d.source.inspect.length }.max
-        directive_lines = directives.map do |directive|
-          if directive.type == SPACE
-            source = directive.source.inspect
-          else
-            source = directive.source
-          end
-          # @type var source_width: Integer
-          "  #{source.ljust(source_width)}  #{directive.describe}"
-        end
-
-        (["Directives:"] + directive_lines + ["Encoding:", "  #{encoding}"]).join("\n")
-      end
-    end
-  end
-end

data/lib/prism/parse_result/comments.rb

@@ -1,188 +0,0 @@
-# frozen_string_literal: true
-# :markup: markdown
-
-module Prism
-  class ParseResult < Result
-    # When we've parsed the source, we have both the syntax tree and the list of
-    # comments that we found in the source. This class is responsible for
-    # walking the tree and finding the nearest location to attach each comment.
-    #
-    # It does this by first finding the nearest locations to each comment.
-    # Locations can either come from nodes directly or from location fields on
-    # nodes. For example, a `ClassNode` has an overall location encompassing the
-    # entire class, but it also has a location for the `class` keyword.
-    #
-    # Once the nearest locations are found, it determines which one to attach
-    # to. If it's a trailing comment (a comment on the same line as other source
-    # code), it will favor attaching to the nearest location that occurs before
-    # the comment. Otherwise it will favor attaching to the nearest location
-    # that is after the comment.
-    class Comments
-      # A target for attaching comments that is based on a specific node's
-      # location.
-      class NodeTarget # :nodoc:
-        attr_reader :node
-
-        def initialize(node)
-          @node = node
-        end
-
-        def start_offset
-          node.start_offset
-        end
-
-        def end_offset
-          node.end_offset
-        end
-
-        def encloses?(comment)
-          start_offset <= comment.location.start_offset &&
-            comment.location.end_offset <= end_offset
-        end
-
-        def leading_comment(comment)
-          node.location.leading_comment(comment)
-        end
-
-        def trailing_comment(comment)
-          node.location.trailing_comment(comment)
-        end
-      end
-
-      # A target for attaching comments that is based on a location field on a
-      # node. For example, the `end` token of a ClassNode.
-      class LocationTarget # :nodoc:
-        attr_reader :location
-
-        def initialize(location)
-          @location = location
-        end
-
-        def start_offset
-          location.start_offset
-        end
-
-        def end_offset
-          location.end_offset
-        end
-
-        def encloses?(comment)
-          false
-        end
-
-        def leading_comment(comment)
-          location.leading_comment(comment)
-        end
-
-        def trailing_comment(comment)
-          location.trailing_comment(comment)
-        end
-      end
-
-      # The parse result that we are attaching comments to.
-      attr_reader :parse_result
-
-      # Create a new Comments object that will attach comments to the given
-      # parse result.
-      def initialize(parse_result)
-        @parse_result = parse_result
-      end
-
-      # Attach the comments to their respective locations in the tree by
-      # mutating the parse result.
-      def attach!
-        parse_result.comments.each do |comment|
-          preceding, enclosing, following = nearest_targets(parse_result.value, comment)
-
-          if comment.trailing?
-            if preceding
-              preceding.trailing_comment(comment)
-            else
-              (following || enclosing || NodeTarget.new(parse_result.value)).leading_comment(comment)
-            end
-          else
-            # If a comment exists on its own line, prefer a leading comment.
-            if following
-              following.leading_comment(comment)
-            elsif preceding
-              preceding.trailing_comment(comment)
-            else
-              (enclosing || NodeTarget.new(parse_result.value)).leading_comment(comment)
-            end
-          end
-        end
-      end
-
-      private
-
-      # Responsible for finding the nearest targets to the given comment within
-      # the context of the given encapsulating node.
-      def nearest_targets(node, comment)
-        comment_start = comment.location.start_offset
-        comment_end = comment.location.end_offset
-
-        targets = [] #: Array[_Target]
-        node.comment_targets.map do |value|
-          case value
-          when StatementsNode
-            targets.concat(value.body.map { |node| NodeTarget.new(node) })
-          when Node
-            targets << NodeTarget.new(value)
-          when Location
-            targets << LocationTarget.new(value)
-          end
-        end
-
-        targets.sort_by!(&:start_offset)
-        preceding = nil #: _Target?
-        following = nil #: _Target?
-
-        left = 0
-        right = targets.length
-
-        # This is a custom binary search that finds the nearest nodes to the
-        # given comment. When it finds a node that completely encapsulates the
-        # comment, it recurses downward into the tree.
-        while left < right
-          middle = (left + right) / 2
-          target = targets[middle]
-
-          target_start = target.start_offset
-          target_end = target.end_offset
-
-          if target.encloses?(comment)
-            # @type var target: NodeTarget
-            # The comment is completely contained by this target. Abandon the
-            # binary search at this level.
-            return nearest_targets(target.node, comment)
-          end
-
-          if target_end <= comment_start
-            # This target falls completely before the comment. Because we will
-            # never consider this target or any targets before it again, this
-            # target must be the closest preceding target we have encountered so
-            # far.
-            preceding = target
-            left = middle + 1
-            next
-          end
-
-          if comment_end <= target_start
-            # This target falls completely after the comment. Because we will
-            # never consider this target or any targets after it again, this
-            # target must be the closest following target we have encountered so
-            # far.
-            following = target
-            right = middle
-            next
-          end
-
-          # This should only happen if there is a bug in this parser.
-          raise "Comment location overlaps with a target location"
-        end
-
-        [preceding, NodeTarget.new(node), following]
-      end
-    end
-  end
-end

data/rbi/prism/compiler.rbi

@@ -1,12 +0,0 @@
-# typed: strict
-
-class Prism::Compiler
-  sig { params(node: T.nilable(Prism::Node)).returns(T.untyped) }
-  def visit(node); end
-
-  sig { params(nodes: T::Array[T.nilable(Prism::Node)]).returns(T::Array[T.untyped]) }
-  def visit_all(nodes); end
-
-  sig { params(node: Prism::Node).returns(T::Array[T.untyped]) }
-  def visit_child_nodes(node); end
-end