Last active
August 10, 2023 13:33
-
-
Save Egor-Skriptunoff/e4ab3bfc777faf4482a1b3f3ae19181b to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| #!/usr/bin/env lua | |
| --------------------------------------------------------------------------------------------------------------------------------------- | |
| -- Display list of globals used by your Lua script | |
| --------------------------------------------------------------------------------------------------------------------------------------- | |
| -- Version: 2019-03-28 | |
| -- License: MIT (see at the end of this file) | |
| -- | |
| -- Reads your Lua script from STDIN | |
| -- Writes list of globals to STDOUT (if the script is syntactically correct) | |
| -- Writes parsing error to STDERR (if the script is not syntactically correct) | |
| -- | |
| -- Usage examples (both Windows and Linux): | |
| -- | |
| -- How to display all the globals access (both read and write) | |
| -- lua show_globals.lua RW < your_script.lua | |
| -- | |
| -- How to display only write access to globals | |
| -- lua show_globals.lua W < your_script.lua | |
| local Lua_version = "Lua 5.3" -- set here Lua version "your_script.lua" was written for | |
| local read_write_access = (arg[1] or "RW"):upper() -- what type of access (read/write/both) should be displayed | |
| read_write_access = {R = read_write_access:find"R", W = read_write_access:find"W"} | |
| assert(read_write_access.R or read_write_access.W, "First argument must be R, W or RW") | |
| local parser | |
| do | |
| local | |
| byte, sub, find, char, rep, match, gsub, upper, reverse, pairs, ipairs, tonumber, tostring, type, unpack, table_concat, math_min, huge | |
| = | |
| string.byte, string.sub, string.find, string.char, string.rep, string.match, string.gsub, string.upper, string.reverse, | |
| pairs, ipairs, tonumber, tostring, type, table.unpack or unpack, table.concat, math.min, math.huge | |
| local scanner | |
| do | |
| local keyword_types = { | |
| ["nil"] = "literal value", | |
| ["false"] = "literal value", | |
| ["true"] = "literal value", | |
| ["not"] = "operator", | |
| ["and"] = "operator", | |
| ["or"] = "operator", | |
| ["repeat"] = "operator", | |
| ["until"] = "operator", | |
| ["while"] = "operator", | |
| ["do"] = "operator", | |
| ["end"] = "operator", | |
| ["if"] = "operator", | |
| ["then"] = "operator", | |
| ["elseif"] = "operator", | |
| ["else"] = "operator", | |
| ["for"] = "operator", | |
| ["in"] = "operator", | |
| ["local"] = "operator", | |
| ["break"] = "operator", | |
| ["return"] = "operator", | |
| ["function"] = "operator", | |
| } | |
| local keyword_values = { ["false"] = false, ["true"] = true } -- ["nil"] = nil | |
| local escapes = { a = "\a", b = "\b", f = "\f", n = "\n", r = "\r", t = "\t", v = "\v", ["\\"] = "\\", ['"'] = '"', ["'"] = "'", ["\n"] = "\n" } | |
| local type_of_characters = { -- some characters are omitted intentionally | |
| [" "] = "lexem delimiter", | |
| ["\n"] = "lexem delimiter", | |
| ["\t"] = "lexem delimiter", | |
| ["\f"] = "lexem delimiter", | |
| ["\v"] = "lexem delimiter", | |
| ["#"] = "operator", | |
| ["("] = "operator", [")"] = "operator", ["{"] = "operator", ["}"] = "operator", ["]"] = "operator", | |
| ["+"] = "operator", ["*"] = "operator", ["%"] = "operator", ["^"] = "operator", | |
| [","] = "operator", [";"] = "operator", | |
| ["&"] = "bitwise operator", | |
| ["|"] = "bitwise operator", | |
| ["'"] = "quote", ['"'] = "quote", | |
| ["<"] = "comparison", [">"] = "comparison", ["="] = "comparison", ["~"] = "comparison", | |
| _ = "alpha", | |
| } | |
| for j = 1, 26 do | |
| type_of_characters[char(64 + j)] = "alpha" -- A-Z | |
| type_of_characters[char(96 + j)] = "alpha" -- a-z | |
| end | |
| for j = 0, 9 do | |
| type_of_characters[char(48 + j)] = "digit" -- 0-9 | |
| end | |
| local all_scanners = {} | |
| local all_supported_versions = { ["51"] = true, ["52"] = true, ["53"] = true, ["54"] = true } | |
| function scanner(version) | |
| version = gsub(version, "%D", "") | |
| local scanner_for_version = all_scanners[version] | |
| if not scanner_for_version then | |
| if not all_supported_versions[version] then | |
| error("There is no scanner for this version available", 2) | |
| end | |
| local version_53_plus = version == "53" or version == "54" | |
| local version_52_plus = version_53_plus or version == "52" | |
| local feature_floating_point_hex_numbers = version_52_plus | |
| local feature_goto_is_keyword = version_52_plus | |
| local feature_prohibit_unknown_escapes = version_52_plus | |
| local feature_backslash_z = version_52_plus | |
| local feature_backslash_x = version_52_plus | |
| local feature_backslash_u = version_53_plus | |
| local feature_int64 = version_53_plus | |
| function scanner_for_version(lua_code) | |
| local pos = 1 -- position of first unread character | |
| local current_line_no = 1 | |
| local current_line_start_pos = 1 | |
| local function get_current_line_col() | |
| return | |
| current_line_no, | |
| pos - current_line_start_pos + 1, | |
| pos | |
| end | |
| local function read_character() | |
| -- returns next_character (any variant of newline is converted to "\n") | |
| -- returns nil (when EOF) | |
| local character = sub(lua_code, pos, pos) | |
| if pos == 1 and character == "#" then | |
| repeat -- skip shebang | |
| pos = pos + 1 | |
| character = sub(lua_code, pos, pos) | |
| until character == "\n" or character == "\r" or character == "" | |
| end | |
| if character ~= "" then | |
| pos = pos + 1 | |
| if character == "\n" or character == "\r" then | |
| local newline = character | |
| character = sub(lua_code, pos, pos) | |
| if (character == "\n" or character == "\r") and newline ~= character then | |
| pos = pos + 1 | |
| end | |
| current_line_no = current_line_no + 1 | |
| current_line_start_pos = pos | |
| character = "\n" | |
| end | |
| return character | |
| end | |
| end | |
| local function read_string_literal(line_no, col_no, quote, is_comment) | |
| -- current position is just after opening quote | |
| -- returns true | |
| -- returns nil, error_message (in case of error) | |
| if quote == '"' or quote == "'" then | |
| local unfinished | |
| while true do | |
| local character = read_character() | |
| if not character or character == "\n" then | |
| unfinished = true | |
| break | |
| end | |
| if character == quote then | |
| break | |
| end | |
| if character == "\\" then | |
| character = read_character() | |
| local line_no, col_no = get_current_line_col() | |
| col_no = col_no - 2 | |
| if not character then | |
| unfinished = true | |
| break | |
| end | |
| local code = byte(character) | |
| if code >= 48 and code <= 57 then | |
| character = code - 48 | |
| for _ = 1, 2 do | |
| code = byte(lua_code, pos) | |
| if not code or code < 48 or code > 57 then | |
| break | |
| end | |
| read_character() | |
| character = character * 10 + code - 48 | |
| end | |
| if character > 255 then | |
| return nil, "Syntax error: decimal escape too large at line "..line_no.." col "..col_no | |
| end | |
| else | |
| local unescaped = escapes[character] | |
| if character == "z" and feature_backslash_z then | |
| while find(lua_code, "^%s", pos) do | |
| read_character() | |
| end | |
| unescaped = true | |
| elseif character == "x" and feature_backslash_x then | |
| local character = match(lua_code, "^%x%x", pos) | |
| if not character then | |
| return nil, "Syntax error: two hexadecimal digits expected after '\\x' at line "..line_no.." col "..col_no | |
| end | |
| unescaped = true | |
| read_character() | |
| read_character() | |
| elseif character == "u" and feature_backslash_u then | |
| if read_character() ~= "{" then | |
| return nil, "Syntax error: '{' is expected after '\\u' at line "..line_no.." col "..col_no | |
| end | |
| local digit_code | |
| unescaped = 0 | |
| repeat | |
| character = read_character() | |
| if character ~= "}" then | |
| digit_code = character and byte(upper(character)) or 0 | |
| if digit_code < 48 or digit_code > 90 or digit_code > 57 and digit_code < 65 then | |
| return nil, "Syntax error: '{XXX}' is expected after '\\u' at line "..line_no.." col "..col_no | |
| end | |
| unescaped = unescaped * 16 + tonumber(character, 16) | |
| if unescaped > 0x10FFFF then | |
| return nil, "Syntax error: UTF-8 value too large at line "..line_no.." col "..col_no | |
| end | |
| end | |
| until character == "}" | |
| if not digit_code then | |
| return nil, "Syntax error: hexadecimal digit is expected after '\\u{' at line "..line_no.." col "..col_no | |
| end | |
| end | |
| if feature_prohibit_unknown_escapes and not unescaped then | |
| return nil, "Syntax error: invalid escape sequence at line "..line_no.." col "..col_no | |
| end | |
| end | |
| end | |
| end | |
| if unfinished then | |
| return nil, "Syntax error: unfinished string literal at line "..line_no.." col "..col_no | |
| end | |
| else | |
| quote = "]"..rep("=", #quote - 2).."]" | |
| local next_character = sub(lua_code, pos, pos) | |
| if next_character == "\n" or next_character == "\r" then | |
| read_character() | |
| end | |
| local quote_start_pos, quote_end_pos = find(lua_code, quote, pos, true) | |
| if not quote_start_pos then | |
| return nil, "Syntax error: unfinished long "..(is_comment and "comment" or "string literal").." at line "..line_no.." col "..col_no | |
| end | |
| while pos < quote_start_pos do | |
| read_character() | |
| end | |
| pos = quote_end_pos + 1 | |
| end | |
| return true | |
| end | |
| local function scan_next_lexem() | |
| -- returns {lexem.type == "EOF"} (in case of EOF) | |
| -- returns nil, error_message (in case of error) | |
| -- returns lexem record consisting of the following fields: | |
| -- type subtype additional fields | |
| -- ---------------------------- --------------------------------- ----------------- | |
| -- "identifier" line col pos | |
| -- "literal value" "boolean"/"nil"/"string"/"number" line col pos | |
| -- "comment" "short"/"long" line col pos | |
| -- (all operators and keywords) line col pos | |
| -- "EOF" line col pos | |
| local line_no, col_no, lexem_start_pos, character, character_type | |
| repeat | |
| line_no, col_no, lexem_start_pos = get_current_line_col() | |
| character = read_character() | |
| character_type = type_of_characters[character] | |
| until character_type ~= "lexem delimiter" | |
| if character then | |
| local next_character = sub(lua_code, pos, pos) | |
| local next_next_character = sub(lua_code, pos + 1, pos + 1) | |
| local code = byte(character) | |
| if character_type == "alpha" then -- A-Za-z_ | |
| character, pos = match(lua_code, "^([A-Za-z_%d]+)()", pos - 1) | |
| local keyword_type = keyword_types[character] | |
| if keyword_type == "literal value" then | |
| -- constants of types "boolean"/"nil" | |
| local value = keyword_values[character] | |
| return { type = keyword_type, subtype = type(value), line = line_no, col = col_no, pos = lexem_start_pos } | |
| elseif keyword_type or character == "goto" and feature_goto_is_keyword then | |
| -- keywords | |
| return { type = character, line = line_no, col = col_no, pos = lexem_start_pos } | |
| else | |
| -- identifiers | |
| return { type = "identifier", value = character, line = line_no, col = col_no, pos = lexem_start_pos } | |
| end | |
| elseif character_type == "operator" or character_type == "bitwise operator" and feature_int64 then | |
| -- punctuation | |
| return { type = character, line = line_no, col = col_no, pos = lexem_start_pos } | |
| elseif character_type == "quote" then | |
| local ok, error_message = read_string_literal(line_no, col_no, character) | |
| if ok then | |
| return { type = "literal value", subtype = "string", line = line_no, col = col_no, pos = lexem_start_pos } | |
| else | |
| return nil, error_message | |
| end | |
| elseif character_type == "comparison" then | |
| if next_character == "=" or character == next_character and character ~= "~" and feature_int64 then | |
| character = character..next_character | |
| pos = pos + 1 | |
| end | |
| if character == "~" and not feature_int64 then | |
| return nil, "Syntax error at line "..line_no.." col "..col_no | |
| end | |
| return { type = character, line = line_no, col = col_no, pos = lexem_start_pos } | |
| elseif character == "0" and (next_character == "x" or next_character == "X") then | |
| local hex_pos = pos + 1 | |
| if feature_floating_point_hex_numbers then | |
| character, pos = match(lua_code, "^([%.%x]*)()", hex_pos) | |
| local ok = find(character, "^%x*%.?%x*$") and character ~= "." and character ~= "" | |
| if upper(sub(lua_code, pos, pos)) == "P" then | |
| pos = match(lua_code, "^[%+%-]?%d+()", pos + 1) | |
| end | |
| if not (ok and pos) then | |
| return nil, "Syntax error: malformed hexadecimal number at line "..line_no.." col "..col_no | |
| end | |
| return { type = "literal value", subtype = "number", line = line_no, col = col_no, pos = lexem_start_pos } | |
| else | |
| pos = match(lua_code, "^%x+()", hex_pos) | |
| if not pos then | |
| return nil, "Syntax error: malformed hexadecimal number at line "..line_no.." col "..col_no | |
| end | |
| return { type = "literal value", subtype = "number", line = line_no, col = col_no, pos = lexem_start_pos } | |
| end | |
| elseif character == "." and next_character == "." then | |
| if next_next_character == "." then | |
| pos = pos + 2 | |
| return { type = "...", line = line_no, col = col_no, pos = lexem_start_pos } | |
| else | |
| pos = pos + 1 | |
| return { type = "..", line = line_no, col = col_no, pos = lexem_start_pos } | |
| end | |
| elseif character == "." or character_type == "digit" then | |
| local starting_pos = pos - 1 | |
| character, pos = match(lua_code, "^([%.%d]+)()", starting_pos) | |
| if character == "." then | |
| return { type = character, line = line_no, col = col_no, pos = lexem_start_pos } | |
| else | |
| local ok = match(character, "^%d*%.?%d*$") | |
| if upper(sub(lua_code, pos, pos)) == "E" then | |
| pos = match(lua_code, "^[%+%-]?%d+()", pos + 1) | |
| end | |
| if not (ok and pos) then | |
| return nil, "Syntax error: malformed number at line "..line_no.." col "..col_no | |
| end | |
| return { type = "literal value", subtype = "number", line = line_no, col = col_no, pos = lexem_start_pos } | |
| end | |
| elseif character == "[" then | |
| if next_character == "[" or next_character == "=" then | |
| local equal_ctr = -1 | |
| repeat | |
| equal_ctr = equal_ctr + 1 | |
| character = read_character() | |
| if character ~= "[" and character ~= "=" then | |
| return nil, "Syntax error at line "..line_no.." col "..col_no | |
| end | |
| until character == "[" | |
| local ok, error_message = read_string_literal(line_no, col_no, "["..rep("=", equal_ctr).."[") | |
| if ok then | |
| return { type = "literal value", subtype = "string", line = line_no, col = col_no, pos = lexem_start_pos } | |
| else | |
| return nil, error_message | |
| end | |
| else | |
| return { type = character, line = line_no, col = col_no, pos = lexem_start_pos } | |
| end | |
| elseif character == "-" then | |
| if next_character == "-" then | |
| pos = pos + 1 | |
| local equal_ctr | |
| if next_next_character == "[" then | |
| equal_ctr = -1 | |
| repeat | |
| equal_ctr = equal_ctr + 1 | |
| local next_pos = pos + 1 + equal_ctr | |
| character = sub(lua_code, next_pos, next_pos) | |
| if character ~= "[" and character ~= "=" then | |
| equal_ctr = nil | |
| end | |
| until not equal_ctr or character == "[" | |
| end | |
| if equal_ctr then | |
| pos = pos + 2 + equal_ctr | |
| local ok, error_message = read_string_literal(line_no, col_no, rep("=", equal_ctr + 2), true) | |
| if ok then | |
| return { type = "comment", subtype = "long", line = line_no, col = col_no, pos = lexem_start_pos } | |
| else | |
| return nil, error_message | |
| end | |
| else | |
| pos = pos + #match(lua_code, "[^\r\n]*", pos) | |
| return { type = "comment", subtype = "short", line = line_no, col = col_no, pos = lexem_start_pos } | |
| end | |
| end | |
| return { type = character, line = line_no, col = col_no, pos = lexem_start_pos } | |
| elseif character == "/" then | |
| if next_character == "/" and feature_int64 then | |
| pos = pos + 1 | |
| character = "//" | |
| end | |
| return { type = character, line = line_no, col = col_no, pos = lexem_start_pos } | |
| elseif character == ":" then | |
| if next_character == ":" and feature_goto_is_keyword then | |
| pos = pos + 1 | |
| character = "::" | |
| end | |
| return { type = character, line = line_no, col = col_no, pos = lexem_start_pos } | |
| else | |
| return nil, "Syntax error at line "..line_no.." col "..col_no | |
| end | |
| else | |
| return { type = "EOF", line = line_no, col = col_no, pos = lexem_start_pos } | |
| end | |
| end | |
| return scan_next_lexem | |
| end | |
| all_scanners[version] = scanner_for_version | |
| end | |
| return scanner_for_version | |
| end | |
| end | |
| local function find_def(vars, name) | |
| -- returns subtype("local"/"upvalue"/"global"), def(only for "local"/"upvalue") | |
| -- vars[1] = parent_vars/nil, vars[2] = is_function_body, vars[name] = def | |
| local subtype = "local" | |
| repeat | |
| local def = vars[name] | |
| if def then | |
| return subtype, def | |
| end | |
| if vars[2] then | |
| subtype = "upvalue" | |
| end | |
| vars = vars[1] | |
| until not vars | |
| return "global" | |
| end | |
| local binary_operator_predecessors = { | |
| -- binary operators are preceded in Lua grammar only by the following lexems (ignoring comments): | |
| ["identifier"] = true, | |
| ["literal value"] = true, | |
| ["..."] = true, | |
| ["end"] = true, | |
| [")"] = true, | |
| ["}"] = true, | |
| ["]"] = true, | |
| } | |
| -- Priorities are positive numbers | |
| -- The same priority level must not be shared between unary and binary operators | |
| -- The same priority level must not be shared between left-associative and right-associative binary operators | |
| local binary_operators = { | |
| -- priorities of binary operators | |
| ["or"] = 1, | |
| ["and"] = 2, | |
| ["<"] = 3, [">"] = 3, ["<="] = 3, [">="] = 3, ["~="] = 3, ["=="] = 3, | |
| ["|"] = 4, | |
| ["~"] = 5, | |
| ["&"] = 6, | |
| ["<<"] = 7, [">>"] = 7, | |
| [".."] = 8, | |
| ["+"] = 9, ["-"] = 9, | |
| ["*"] = 10, ["/"] = 10, ["//"] = 10, ["%"] = 10, | |
| ["^"] = 12 | |
| } | |
| local unary_operators = { | |
| -- priorities of unary operators | |
| ["not"] = 11, ["#"] = 11, ["unary -"] = 11, ["unary ~"] = 11 | |
| } | |
| local right_associative_priorities = { | |
| -- precedences of right associative binary operators: | |
| [8] = true, -- .. concatenation | |
| [12] = true, -- ^ exponentiation | |
| } | |
| local all_parsers = {} | |
| local all_supported_versions = { ["51"] = true, ["52"] = true, ["53"] = true, ["54"] = true } | |
| function parser(version) | |
| version = gsub(version, "%D", "") | |
| local parser_for_version = all_parsers[version] | |
| if not parser_for_version then | |
| if not all_supported_versions[version] then | |
| error("There is no parser for this version available", 2) | |
| end | |
| local version_52_plus = version == "52" or version == "53" or version == "54" | |
| local feature_break_is_last_statement = version == "51" | |
| local feature_empty_statement = version_52_plus | |
| local feature_ENV = version_52_plus | |
| local lua_scanner = scanner(version) | |
| function parser_for_version(lua_code, global_catcher) | |
| -- returns true | |
| -- returns nil, error_message (in case of parsing error) | |
| local get_next_lexem_from_scanner = lua_scanner(lua_code) | |
| local error_message | |
| local next_lexem, next_next_lexem | |
| local last_lexem, before_last_lexem | |
| local prev_non_comment_lexem_type | |
| local function read_lexem() | |
| -- returns nil in case of error (error_message is set) | |
| -- lexem.type == "EOF" (when EOF) | |
| local lexem | |
| if next_lexem then | |
| lexem, next_lexem, next_next_lexem = next_lexem, next_next_lexem | |
| else | |
| repeat | |
| lexem, error_message = get_next_lexem_from_scanner() | |
| if not lexem then | |
| return | |
| end | |
| local lexem_type = lexem.type | |
| local is_comment = lexem_type == "comment" | |
| if not is_comment then | |
| if (lexem_type == "-" or lexem_type == "~") and not binary_operator_predecessors[prev_non_comment_lexem_type] then | |
| lexem_type = "unary "..lexem_type | |
| lexem.type = lexem_type | |
| end | |
| prev_non_comment_lexem_type = lexem_type | |
| end | |
| until not is_comment | |
| end | |
| before_last_lexem, last_lexem = last_lexem, lexem | |
| return lexem | |
| end | |
| local function unread_last_lexem() | |
| last_lexem, next_lexem, next_next_lexem = before_last_lexem, last_lexem, next_lexem | |
| end | |
| local function get_lexem_coordinates_as_text(lexem) | |
| if lexem.type == "EOF" then | |
| return "EOF" | |
| else | |
| return "line "..lexem.line.." col "..lexem.col | |
| end | |
| end | |
| local read_chunk, read_exp, read_explist | |
| -------------------------------------------------------------------------------------------------------------------------------------------- | |
| local function read_table_constructor(vars) | |
| -------------------------------------------------------------------------------------------------------------------------------------------- | |
| -- opening curly brace is already read | |
| unread_last_lexem() | |
| local result = read_lexem() -- this is the opening curly brace | |
| result.type = "table constructor" | |
| local array_of_table_items = {} | |
| result.items = array_of_table_items | |
| local lexem = read_lexem() | |
| if not lexem then | |
| return | |
| end | |
| local lexem_type = lexem.type | |
| while lexem_type ~= "}" do | |
| local index_exp, field_exp, value_exp | |
| if lexem_type == "[" then | |
| index_exp = read_exp(vars) | |
| if not index_exp then | |
| return | |
| end | |
| lexem = read_lexem() | |
| if not lexem then | |
| return | |
| end | |
| if lexem.type ~= "]" then | |
| error_message = "Syntax error: ']' is expected at "..get_lexem_coordinates_as_text(lexem) | |
| return | |
| end | |
| lexem = read_lexem() | |
| if not lexem then | |
| return | |
| end | |
| if lexem.type ~= "=" then | |
| error_message = "Syntax error: '=' is expected at "..get_lexem_coordinates_as_text(lexem) | |
| return | |
| end | |
| value_exp = read_exp(vars) | |
| if not value_exp then | |
| return | |
| end | |
| else | |
| if lexem_type == "identifier" then | |
| local next_lexem = read_lexem() | |
| if not next_lexem then | |
| return | |
| end | |
| if next_lexem.type == "=" then | |
| field_exp = lexem | |
| field_exp.subtype = "field" | |
| else | |
| unread_last_lexem() | |
| end | |
| end | |
| if not field_exp then | |
| unread_last_lexem() | |
| end | |
| value_exp = read_exp(vars) | |
| if not value_exp then | |
| return | |
| end | |
| end | |
| array_of_table_items[#array_of_table_items + 1] = { index = index_exp, field = field_exp, value = value_exp } | |
| lexem = read_lexem() | |
| if not lexem then | |
| return | |
| end | |
| lexem_type = lexem.type | |
| if lexem_type == "," or lexem_type == ";" then | |
| lexem = read_lexem() | |
| if not lexem then | |
| return | |
| end | |
| lexem_type = lexem.type | |
| elseif lexem_type ~= "}" then | |
| error_message = "Syntax error: ',' or ';' or '}' is expected at "..get_lexem_coordinates_as_text(lexem) | |
| return | |
| end | |
| end | |
| return result | |
| end | |
| -------------------------------------------------------------------------------------------------------------------------------------------- | |
| local function read_funcbody(vars, invisible_self_argument) | |
| -------------------------------------------------------------------------------------------------------------------------------------------- | |
| local result = read_lexem() | |
| if not result then | |
| return | |
| end | |
| if result.type ~= "(" then | |
| error_message = "Syntax error: '(' is expected at "..get_lexem_coordinates_as_text(result) | |
| return | |
| end | |
| result.type = "function definition" | |
| local arguments = {} | |
| if invisible_self_argument then | |
| arguments[1] = { type = "identifier", subtype = "definition", value = "self", line = result.line, col = result.col, pos = result.pos } | |
| end | |
| result.arguments = arguments | |
| local is_vararg = false | |
| local after_separator = true | |
| local visible_argument | |
| while true do | |
| local lexem = read_lexem() | |
| if not lexem then | |
| return | |
| end | |
| local lexem_type = lexem.type | |
| if lexem_type == ")" and not (after_separator and visible_argument) then | |
| local body = read_chunk(vars, false, arguments, true) | |
| if not body then | |
| return | |
| end | |
| result.body = body | |
| local closing_lexem = read_lexem() | |
| if not closing_lexem then | |
| return | |
| end | |
| if closing_lexem.type ~= "end" then | |
| error_message = "Syntax error: 'end' is expected at "..get_lexem_coordinates_as_text(closing_lexem) | |
| return | |
| end | |
| result.is_vararg = is_vararg | |
| return result | |
| elseif lexem_type == "," and not after_separator and not is_vararg then | |
| after_separator = true | |
| elseif (lexem_type == "identifier" or lexem_type == "...") and after_separator then | |
| arguments[#arguments + 1] = lexem | |
| visible_argument = true | |
| after_separator = false | |
| is_vararg = lexem_type == "..." | |
| else | |
| error_message = "Syntax error in function argument list at "..get_lexem_coordinates_as_text(lexem) | |
| return | |
| end | |
| end | |
| end | |
| -------------------------------------------------------------------------------------------------------------------------------------------- | |
| local function read_prefixexp(vars, pure_global_LHS) | |
| -------------------------------------------------------------------------------------------------------------------------------------------- | |
| -- returns prefixexp_subtree, exp_type, first_lexem_of_prefixexp | |
| -- exp_type: 1 = rvalue (neither lvalue nor funccall) | |
| -- 3 = funccall | |
| -- 5 = lvalue | |
| local exp_type, write_access | |
| local first_lexem = read_lexem() | |
| if not first_lexem then | |
| return | |
| end | |
| local lexem_type = first_lexem.type | |
| if lexem_type == "(" then | |
| exp_type = 1 | |
| first_lexem.type = "(expression)" | |
| local expression = read_exp(vars) | |
| if not expression then | |
| return | |
| end | |
| first_lexem.expression = expression | |
| local lexem = read_lexem() | |
| if not lexem then | |
| return | |
| end | |
| if lexem.type ~= ")" then | |
| error_message = "Syntax error: ')' is expected at "..get_lexem_coordinates_as_text(lexem) | |
| return | |
| end | |
| elseif lexem_type == "identifier" then | |
| exp_type = 5 | |
| first_lexem.subtype, first_lexem.def = find_def(vars, first_lexem.value) | |
| if global_catcher and first_lexem.subtype == "global" then | |
| global_catcher(first_lexem) | |
| if pure_global_LHS then | |
| write_access = true | |
| end | |
| end | |
| else | |
| error_message = "Syntax error at "..get_lexem_coordinates_as_text(first_lexem) | |
| return | |
| end | |
| local result = first_lexem | |
| while true do | |
| local lexem = read_lexem() | |
| if not lexem then | |
| return | |
| end | |
| lexem_type = lexem.type | |
| if lexem_type == "[" then | |
| write_access = false | |
| -- [index] | |
| exp_type = 5 | |
| lexem.type = "table[index]" | |
| lexem.table = result | |
| local exp = read_exp(vars) | |
| if not exp then | |
| return | |
| end | |
| lexem.index = exp | |
| local next_lexem = read_lexem() | |
| if not next_lexem then | |
| return | |
| end | |
| if next_lexem.type ~= "]" then | |
| error_message = "Syntax error: ']' is expected at "..get_lexem_coordinates_as_text(next_lexem) | |
| end | |
| result = lexem | |
| elseif lexem_type == "." then | |
| write_access = false | |
| -- .field | |
| exp_type = 5 | |
| lexem.type = "object.field" | |
| lexem.object = result | |
| local next_lexem = read_lexem() | |
| if not next_lexem then | |
| return | |
| end | |
| if next_lexem.type ~= "identifier" then | |
| error_message = "Syntax error: an identifier is expected at"..get_lexem_coordinates_as_text(next_lexem) | |
| return | |
| end | |
| next_lexem.subtype = "field" | |
| lexem.field = next_lexem | |
| result = lexem | |
| elseif lexem_type == ":" or lexem_type == "(" or lexem_type == "{" or lexem_type == "literal value" and lexem.subtype == "string" then | |
| write_access = false | |
| -- function call | |
| exp_type = 3 | |
| if lexem_type == ":" then | |
| lexem.object = result | |
| result = lexem | |
| lexem = read_lexem() | |
| if not lexem then | |
| return | |
| end | |
| if lexem.type ~= "identifier" then | |
| error_message = "Syntax error: an identifier is expected after ':' at "..get_lexem_coordinates_as_text(lexem) | |
| return | |
| end | |
| lexem.subtype = "field" | |
| result.method = lexem | |
| lexem = read_lexem() | |
| if not lexem then | |
| return | |
| end | |
| lexem_type = lexem.type | |
| if not (lexem_type == "(" or lexem_type == "{" or lexem_type == "literal value" and lexem.subtype == "string") then | |
| error_message = "Syntax error: '(' is expected at "..get_lexem_coordinates_as_text(lexem) | |
| return | |
| end | |
| else | |
| if lexem_type == "(" then | |
| lexem.object = result | |
| result = lexem | |
| else | |
| result = { object = result, line = lexem.line, col = lexem.col } | |
| end | |
| end | |
| result.type = "funccall" | |
| if lexem_type == "(" then | |
| lexem = read_lexem() | |
| if not lexem then | |
| return | |
| end | |
| if lexem.type == ")" then | |
| result.arguments = {} | |
| else | |
| unread_last_lexem() | |
| local arguments = read_explist(vars) | |
| if not arguments then | |
| return | |
| end | |
| result.arguments = arguments | |
| lexem = read_lexem() | |
| if not lexem then | |
| return | |
| end | |
| if lexem.type ~= ")" then | |
| error_message = "Syntax error: ')' is expected at "..get_lexem_coordinates_as_text(lexem) | |
| return | |
| end | |
| end | |
| elseif lexem_type == "{" then | |
| -- read table constructor | |
| local table_constructor_exp = read_table_constructor(vars) | |
| if not table_constructor_exp then | |
| return | |
| end | |
| result.arguments = { table_constructor_exp } | |
| else -- literal string as single argument | |
| result.arguments = { lexem } | |
| end | |
| else | |
| unread_last_lexem() | |
| if write_access then | |
| global_catcher() -- mark write access | |
| end | |
| return result, exp_type, first_lexem | |
| end | |
| end | |
| end | |
| -------------------------------------------------------------------------------------------------------------------------------------------- | |
| local function read_term(vars) | |
| -------------------------------------------------------------------------------------------------------------------------------------------- | |
| -- term is an expression having root operator (last operator to be executed) different from "exp ::= exp binop exp" | |
| local lexem = read_lexem() | |
| if not lexem then | |
| return | |
| end | |
| local lexem_type = lexem.type | |
| if lexem_type == "literal value" then | |
| return lexem | |
| elseif lexem_type == "..." then | |
| lexem.subtype, lexem.def = find_def(vars, lexem_type) | |
| return lexem | |
| elseif lexem_type == "{" then | |
| return read_table_constructor(vars) | |
| elseif lexem_type == "function" then | |
| return read_funcbody(vars) | |
| else | |
| local unary_oper_priority = unary_operators[lexem_type] | |
| if unary_oper_priority then | |
| local argument, ends_with_prefixexp = read_exp(vars, unary_oper_priority) | |
| if not argument then | |
| return | |
| end | |
| lexem.argument = argument | |
| return lexem, ends_with_prefixexp | |
| else | |
| unread_last_lexem() | |
| return read_prefixexp(vars) -- second returned value is always truthy | |
| end | |
| end | |
| end | |
| -------------------------------------------------------------------------------------------------------------------------------------------- | |
| local function read_chain_of_binary_operators_with_specified_priority(vars, first_term, first_binary_operator, chain_priority) | |
| -------------------------------------------------------------------------------------------------------------------------------------------- | |
| -- parse expression consisting of binary operators with specified priority | |
| first_binary_operator.left_argument = first_term | |
| local is_right_associative = right_associative_priorities[chain_priority] | |
| local term, ends_with_prefixexp = read_term(vars) | |
| if not term then | |
| return | |
| end | |
| first_binary_operator.right_argument = term | |
| local last_binary_operator = first_binary_operator | |
| while true do | |
| local binary_operator = read_lexem() | |
| if not binary_operator then | |
| return | |
| end | |
| local priority = binary_operators[binary_operator.type] | |
| if not priority or priority < chain_priority then | |
| unread_last_lexem() | |
| return is_right_associative and first_binary_operator or last_binary_operator, ends_with_prefixexp | |
| elseif priority > chain_priority then | |
| term, ends_with_prefixexp = read_chain_of_binary_operators_with_specified_priority(vars, last_binary_operator.right_argument, binary_operator, priority) | |
| if not term then | |
| return | |
| end | |
| last_binary_operator.right_argument = term | |
| else | |
| term, ends_with_prefixexp = read_term(vars) | |
| if not term then | |
| return | |
| end | |
| binary_operator.right_argument = term | |
| if is_right_associative then | |
| binary_operator.left_argument = last_binary_operator.right_argument | |
| last_binary_operator.right_argument = binary_operator | |
| else | |
| binary_operator.left_argument = last_binary_operator | |
| end | |
| last_binary_operator = binary_operator | |
| end | |
| end | |
| end | |
| -------------------------------------------------------------------------------------------------------------------------------------------- | |
| function read_exp(vars, outer_priority) | |
| -------------------------------------------------------------------------------------------------------------------------------------------- | |
| local fake_operator, ends_with_prefixexp = read_chain_of_binary_operators_with_specified_priority(vars, nil, {}, outer_priority or 0) | |
| if not fake_operator then | |
| return | |
| end | |
| return fake_operator.right_argument, ends_with_prefixexp | |
| end | |
| -------------------------------------------------------------------------------------------------------------------------------------------- | |
| function read_explist(vars, min_qty, max_qty) | |
| -------------------------------------------------------------------------------------------------------------------------------------------- | |
| min_qty, max_qty = min_qty or 1, max_qty or huge | |
| local rvalues = { } | |
| local exp, ends_with_prefixexp | |
| repeat | |
| exp, ends_with_prefixexp = read_exp(vars) | |
| if not exp then | |
| return | |
| end | |
| rvalues[#rvalues + 1] = exp | |
| local qty = #rvalues | |
| local lexem = read_lexem() | |
| if not lexem then | |
| return | |
| end | |
| local is_comma = qty < max_qty and lexem.type == "," | |
| if qty < min_qty and not is_comma then | |
| error_message = "Syntax error: ',' is expected at "..get_lexem_coordinates_as_text(lexem) | |
| return | |
| end | |
| until not is_comma | |
| unread_last_lexem() | |
| return rvalues, ends_with_prefixexp | |
| end | |
| -------------------------------------------------------------------------------------------------------------------------------------------- | |
| local function read_namelist(vars, name_separator) | |
| -------------------------------------------------------------------------------------------------------------------------------------------- | |
| name_separator = name_separator or "," | |
| local identifiers, global_write_access = {} | |
| repeat | |
| local lexem = read_lexem() | |
| if not lexem then | |
| return | |
| end | |
| if lexem.type ~= "identifier" then | |
| error_message = "Syntax error: an identifier is expected at "..get_lexem_coordinates_as_text(lexem) | |
| return | |
| end | |
| if name_separator == "," then | |
| if vars then | |
| vars[lexem.value] = lexem | |
| end | |
| lexem.subtype = "definition" | |
| elseif identifiers[1] then | |
| lexem.subtype = "field" | |
| global_write_access = false | |
| else | |
| lexem.subtype, lexem.def = find_def(vars, lexem.value) | |
| if global_catcher and lexem.subtype == "global" then | |
| global_catcher(lexem) | |
| global_write_access = true | |
| end | |
| end | |
| identifiers[#identifiers + 1] = lexem | |
| lexem = read_lexem() | |
| if not lexem then | |
| return | |
| end | |
| until lexem.type ~= name_separator | |
| unread_last_lexem() | |
| if global_write_access then | |
| global_catcher() -- mark write access | |
| end | |
| return identifiers | |
| end | |
| -------------------------------------------------------------------------------------------------------------------------------------------- | |
| local function read_statement(vars, inside_loop) | |
| -------------------------------------------------------------------------------------------------------------------------------------------- | |
| local lexem = read_lexem() | |
| if not lexem then | |
| return | |
| end | |
| local lexem_type = lexem.type | |
| if lexem_type == "break" then | |
| if not inside_loop then | |
| error_message = "Syntax error: 'break' must be inside loop at "..get_lexem_coordinates_as_text(next_lexem) | |
| return | |
| end | |
| return lexem, feature_break_is_last_statement and "Syntax error: statement after 'break' at " | |
| elseif lexem_type == "do" then | |
| local chunk = read_chunk(vars, inside_loop) | |
| if not chunk then | |
| return | |
| end | |
| lexem.body = chunk | |
| local next_lexem = read_lexem() | |
| if not next_lexem then | |
| return | |
| end | |
| if next_lexem.type ~= "end" then | |
| error_message = "Syntax error: 'end' is expected at "..get_lexem_coordinates_as_text(next_lexem) | |
| return | |
| end | |
| return lexem | |
| elseif lexem_type == "for" then | |
| local names = read_namelist() | |
| if not names then | |
| return | |
| end | |
| local next_lexem = read_lexem() | |
| if not next_lexem then | |
| return | |
| end | |
| lexem_type = next_lexem.type | |
| if lexem_type == "in" or lexem_type == "=" and not names[2] then | |
| -- for namelist in explist do chunk end | |
| -- for Name `=` exp `,` exp [`,` exp] do chunk end | |
| lexem.variables = names | |
| local numeric_for = lexem_type == "=" | |
| lexem.subtype = numeric_for and "numeric" or "generic" | |
| local expressions = read_explist(vars, numeric_for and 2, 3) | |
| if not expressions then | |
| return | |
| end | |
| lexem.expressions = expressions | |
| next_lexem = read_lexem() | |
| if not next_lexem then | |
| return | |
| end | |
| if next_lexem.type ~= "do" then | |
| error_message = "Syntax error: 'do' is expected at "..get_lexem_coordinates_as_text(next_lexem) | |
| return | |
| end | |
| local body = read_chunk(vars, true, names) | |
| if not body then | |
| return | |
| end | |
| lexem.body = body | |
| next_lexem = read_lexem() | |
| if not next_lexem then | |
| return | |
| end | |
| if next_lexem.type ~= "end" then | |
| error_message = "Syntax error: 'end' is expected at "..get_lexem_coordinates_as_text(next_lexem) | |
| return | |
| end | |
| return lexem | |
| end | |
| error_message = "Syntax error in 'for' loop at "..get_lexem_coordinates_as_text(next_lexem) | |
| return | |
| elseif lexem_type == "if" then | |
| -- if exp then chunk {elseif exp then chunk} [else chunk] end | |
| local conditions = {} | |
| lexem.conditions = conditions | |
| local chunks = {} -- chunks array is one element longer than conditions array when "else" clause is present | |
| lexem.chunks = chunks | |
| while true do | |
| local exp = read_exp(vars) | |
| if not exp then | |
| return | |
| end | |
| conditions[#conditions + 1] = exp | |
| local next_lexem = read_lexem() | |
| if not next_lexem then | |
| return | |
| end | |
| if next_lexem.type ~= "then" then | |
| error_message = "Syntax error: 'then' is expected at "..get_lexem_coordinates_as_text(next_lexem) | |
| return | |
| end | |
| local chunk = read_chunk(vars, inside_loop) | |
| if not chunk then | |
| return | |
| end | |
| chunks[#conditions] = chunk | |
| next_lexem = read_lexem() | |
| if not next_lexem then | |
| return | |
| end | |
| local lexem_type = next_lexem.type | |
| if lexem_type == "end" or lexem_type == "else" then | |
| if lexem_type == "else" then | |
| chunk = read_chunk(vars, inside_loop) | |
| if not chunk then | |
| return | |
| end | |
| chunks[#chunks + 1] = chunk | |
| next_lexem = read_lexem() | |
| if not next_lexem then | |
| return | |
| end | |
| if next_lexem.type ~= "end" then | |
| error_message = "Syntax error: 'end' is expected at "..get_lexem_coordinates_as_text(next_lexem) | |
| return | |
| end | |
| end | |
| return lexem | |
| elseif lexem_type ~= "elseif" then | |
| error_message = "Syntax error: 'end' or 'else' or 'elseif' is expected at "..get_lexem_coordinates_as_text(next_lexem) | |
| return | |
| end | |
| end | |
| elseif lexem_type == "local" then | |
| local next_lexem = read_lexem() | |
| if not next_lexem then | |
| return | |
| end | |
| if next_lexem.type == "function" then | |
| -- local function Name funcbody | |
| lexem.type = "local function" | |
| next_lexem = read_lexem() | |
| if not next_lexem then | |
| return | |
| end | |
| if next_lexem.type ~= "identifier" then | |
| error_message = "Syntax error: an identifier is expected at "..get_lexem_coordinates_as_text(next_lexem) | |
| return | |
| end | |
| next_lexem.subtype = "definition" | |
| vars[next_lexem.value] = next_lexem | |
| lexem.variable = next_lexem | |
| local func_def = read_funcbody(vars) | |
| if not func_def then | |
| return | |
| end | |
| lexem.func_def = func_def | |
| return lexem | |
| else | |
| unread_last_lexem() | |
| -- local namelist [`=` explist] | |
| local new_vars = {} | |
| local names = read_namelist(new_vars) | |
| if not names then | |
| return | |
| end | |
| lexem.variables = names | |
| next_lexem = read_lexem() | |
| if not next_lexem then | |
| return | |
| end | |
| local expressions, ends_with_prefixexp | |
| if next_lexem.type == "=" then | |
| expressions, ends_with_prefixexp = read_explist(vars) | |
| if not expressions then | |
| return | |
| end | |
| else | |
| unread_last_lexem() | |
| end | |
| for k, v in pairs(new_vars) do | |
| vars[k] = v | |
| end | |
| lexem.expressions = expressions -- lexem.expressions may be nil for local variables without initialization | |
| lexem.ends_with_prefixexp = ends_with_prefixexp | |
| return lexem | |
| end | |
| elseif lexem_type == "repeat" then | |
| -- repeat chunk until exp | |
| local body, chunk_vars = read_chunk(vars, true) | |
| if not body then | |
| return | |
| end | |
| lexem.body = body | |
| local next_lexem = read_lexem() | |
| if not next_lexem then | |
| return | |
| end | |
| if next_lexem.type ~= "until" then | |
| error_message = "Syntax error: 'until' is expected at "..get_lexem_coordinates_as_text(next_lexem) | |
| return | |
| end | |
| local expression, ends_with_prefixexp = read_exp(chunk_vars) | |
| if not expression then | |
| return | |
| end | |
| lexem.expression = expression | |
| lexem.ends_with_prefixexp = ends_with_prefixexp | |
| return lexem | |
| elseif lexem_type == "return" then | |
| local next_lexem = read_lexem() | |
| if not next_lexem then | |
| return | |
| end | |
| lexem_type = next_lexem.type | |
| unread_last_lexem() | |
| if lexem_type == "EOF" or lexem_type == "end" or lexem_type == "until" or lexem_type == "elseif" or lexem_type == "else" then | |
| lexem.expressions = {} | |
| else | |
| local expressions = read_explist(vars) | |
| if not expressions then | |
| return | |
| end | |
| lexem.expressions = expressions | |
| end | |
| return lexem, "Syntax error: statement after 'return' at " | |
| elseif lexem_type == "while" then | |
| -- while exp do chunk end | |
| local expression = read_exp(vars) | |
| if not expression then | |
| return | |
| end | |
| lexem.expression = expression | |
| local next_lexem = read_lexem() | |
| if not next_lexem then | |
| return | |
| end | |
| if next_lexem.type ~= "do" then | |
| error_message = "Syntax error: 'do' is expected at "..get_lexem_coordinates_as_text(next_lexem) | |
| return | |
| end | |
| local body = read_chunk(vars, true) | |
| if not body then | |
| return | |
| end | |
| lexem.body = body | |
| next_lexem = read_lexem() | |
| if not next_lexem then | |
| return | |
| end | |
| if next_lexem.type ~= "end" then | |
| error_message = "Syntax error: 'end' is expected at "..get_lexem_coordinates_as_text(next_lexem) | |
| return | |
| end | |
| return lexem | |
| elseif lexem_type == "function" then | |
| -- function Name {`.` Name} [`:` Name] funcbody | |
| local invisible_self_argument | |
| local names = read_namelist(vars, ".") | |
| if not names then | |
| return | |
| end | |
| lexem.name_chain = names | |
| local next_lexem = read_lexem() | |
| if not next_lexem then | |
| return | |
| end | |
| if next_lexem.type == ":" then | |
| local method = read_lexem() | |
| if not method then | |
| return | |
| end | |
| if method.type ~= "identifier" then | |
| error_message = "Syntax error: an identifier is expected at "..get_lexem_coordinates_as_text(method) | |
| return | |
| end | |
| method.subtype = "field" | |
| lexem.method = method | |
| invisible_self_argument = true | |
| else | |
| unread_last_lexem() | |
| end | |
| local func_def = read_funcbody(vars, invisible_self_argument) | |
| if not func_def then | |
| return | |
| end | |
| lexem.func_def = func_def | |
| return lexem | |
| elseif lexem_type == "::" then | |
| local next_lexem = read_lexem() | |
| if not next_lexem then | |
| return | |
| end | |
| if next_lexem.type ~= "identifier" then | |
| error_message = "Syntax error: label name is expected after '::' at "..get_lexem_coordinates_as_text(next_lexem) | |
| return | |
| end | |
| lexem.label_identifier = next_lexem | |
| next_lexem = read_lexem() | |
| if not next_lexem then | |
| return | |
| end | |
| if next_lexem.type ~= "::" then | |
| error_message = "Syntax error: '::' is expected after label name at "..get_lexem_coordinates_as_text(next_lexem) | |
| return | |
| end | |
| return lexem | |
| elseif lexem_type == "goto" then | |
| local next_lexem = read_lexem() | |
| if not next_lexem then | |
| return | |
| end | |
| if next_lexem.type ~= "identifier" then | |
| error_message = "Syntax error: label name is expected after 'goto' at "..get_lexem_coordinates_as_text(next_lexem) | |
| return | |
| end | |
| lexem.label_identifier = next_lexem | |
| return lexem | |
| elseif lexem_type == ";" and feature_empty_statement then | |
| return lexem | |
| else | |
| local starts_with_parenthesis = lexem_type == "(" | |
| unread_last_lexem() | |
| local prefixexp, exp_type, first_lexem = read_prefixexp(vars, true) -- either lvalue or funccall | |
| if not prefixexp then | |
| return | |
| end | |
| if exp_type > 4 then | |
| -- lvalue found | |
| local lvalues = { prefixexp } | |
| while true do | |
| lexem = read_lexem() | |
| if not lexem then | |
| return | |
| end | |
| lexem_type = lexem.type | |
| if lexem_type == "=" then | |
| local rvalues, ends_with_prefixexp = read_explist(vars) | |
| if not rvalues then | |
| return | |
| end | |
| lexem.lvalues = lvalues | |
| lexem.rvalues = rvalues | |
| lexem.starts_with_parenthesis = starts_with_parenthesis | |
| lexem.ends_with_prefixexp = ends_with_prefixexp | |
| return lexem | |
| elseif lexem_type == "," then | |
| prefixexp, exp_type, first_lexem = read_prefixexp(vars, true) | |
| if not prefixexp then | |
| return | |
| end | |
| if exp_type < 4 then | |
| error_message = "Syntax error: L-value is expected at "..get_lexem_coordinates_as_text(first_lexem) | |
| return | |
| end | |
| lvalues[#lvalues + 1] = prefixexp | |
| else | |
| error_message = "Syntax error: '=' or ',' is expected at "..get_lexem_coordinates_as_text(lexem) | |
| return | |
| end | |
| end | |
| elseif exp_type > 2 then | |
| -- funccall found | |
| prefixexp.starts_with_parenthesis = starts_with_parenthesis | |
| prefixexp.ends_with_prefixexp = true | |
| return prefixexp | |
| end | |
| error_message = "Syntax error: L-value or function call is expected at "..get_lexem_coordinates_as_text(first_lexem) | |
| return | |
| end | |
| end | |
| -------------------------------------------------------------------------------------------------------------------------------------------- | |
| function read_chunk(vars, inside_loop, additional_variables, is_function) | |
| -------------------------------------------------------------------------------------------------------------------------------------------- | |
| vars = { vars, is_function } | |
| if additional_variables then | |
| for _, def in ipairs(additional_variables) do | |
| vars[def.value or "..."] = def | |
| end | |
| end | |
| local statements = {} | |
| local after_last_statement | |
| while true do | |
| local lexem = read_lexem() | |
| if not lexem then | |
| return | |
| end | |
| unread_last_lexem() | |
| local lexem_type = lexem.type | |
| if lexem_type == "EOF" or lexem_type == "end" or lexem_type == "until" or lexem_type == "elseif" or lexem_type == "else" then | |
| return statements, vars | |
| end | |
| if after_last_statement then | |
| error_message = after_last_statement..get_lexem_coordinates_as_text(lexem) | |
| return | |
| end | |
| lexem, after_last_statement = read_statement(vars, inside_loop) | |
| if not lexem then | |
| return | |
| end | |
| if lexem.type ~= ";" then | |
| statements[#statements + 1] = lexem | |
| end | |
| lexem = read_lexem() | |
| if not lexem then | |
| return | |
| end | |
| if lexem.type ~= ";" then | |
| unread_last_lexem() | |
| end | |
| end | |
| end | |
| local vars = | |
| feature_ENV | |
| and { _ENV = { type = "identifier", subtype = "definition", value = "_ENV", line = 1, col = 1, pos = 1 } } | |
| or nil | |
| local array_of_statements = read_chunk(vars, false, {{ type = "...", subtype = "definition", line = 1, col = 1, pos = 1 }}, true) | |
| if not array_of_statements then | |
| return nil, error_message | |
| end | |
| local lexem = read_lexem() | |
| if not lexem then | |
| return nil, error_message | |
| end | |
| if lexem.type ~= "EOF" then | |
| return nil, "EOF is expected at "..get_lexem_coordinates_as_text(lexem) | |
| end | |
| return true | |
| end | |
| all_parsers[version] = parser_for_version | |
| end | |
| return parser_for_version | |
| end | |
| end | |
| local program = io.read"*a" | |
| local globals = {} | |
| local global_names = {} | |
| do | |
| local last_global_name, last_global_coord | |
| local function flush_last_global_name(write_access) | |
| if last_global_name then | |
| table.insert(globals[last_global_name][write_access and "W" or "R"], last_global_coord) | |
| last_global_name, last_global_coord = nil | |
| end | |
| end | |
| local function global_catcher(lexem) | |
| if not lexem then | |
| flush_last_global_name(true) | |
| else | |
| flush_last_global_name() | |
| last_global_name = lexem.value | |
| if not globals[last_global_name] then | |
| globals[last_global_name] = {R = {}, W = {}} | |
| table.insert(global_names, last_global_name) | |
| end | |
| last_global_coord = lexem.line..":"..lexem.col | |
| end | |
| end | |
| assert(parser(Lua_version)(program, global_catcher)) | |
| flush_last_global_name() | |
| end | |
| table.sort(global_names) | |
| for _, name in ipairs(global_names) do | |
| for _, access_type in ipairs{"W", "R"} do | |
| if read_write_access[access_type] then | |
| local list_line_col = table.concat(globals[name][access_type], ", ") | |
| if list_line_col ~= "" then | |
| print(({R = "read", W = "write"})[access_type], name, list_line_col) | |
| end | |
| end | |
| end | |
| end | |
| --[[ | |
| MIT License | |
| Copyright (c) 2019 Egor Skriptunoff | |
| Permission is hereby granted, free of charge, to any person obtaining a copy | |
| of this software and associated documentation files (the "Software"), to deal | |
| in the Software without restriction, including without limitation the rights | |
| to use, copy, modify, merge, publish, distribute, sublicense, and/or sell | |
| copies of the Software, and to permit persons to whom the Software is | |
| furnished to do so, subject to the following conditions: | |
| The above copyright notice and this permission notice shall be included in all | |
| copies or substantial portions of the Software. | |
| THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR | |
| IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, | |
| FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE | |
| AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER | |
| LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, | |
| OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE | |
| SOFTWARE. | |
| ]] |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Thank you so much for the show-globals script (although, frankly, 'script' is quite an understatement)!
It's so awesome I dared to reformat this program to a LuaRocks package for mere convenience, additionally extending its cli capabilities: https://github.com/Warhammer-Mods/lua-globals
Thanks heaps again!