aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/LuaSL/testLua/yueliang-0.4.1/nat-5.1.3/llex_mk2.lua
diff options
context:
space:
mode:
Diffstat (limited to 'LuaSL/testLua/yueliang-0.4.1/nat-5.1.3/llex_mk2.lua')
-rw-r--r--LuaSL/testLua/yueliang-0.4.1/nat-5.1.3/llex_mk2.lua314
1 files changed, 0 insertions, 314 deletions
diff --git a/LuaSL/testLua/yueliang-0.4.1/nat-5.1.3/llex_mk2.lua b/LuaSL/testLua/yueliang-0.4.1/nat-5.1.3/llex_mk2.lua
deleted file mode 100644
index dae57f1..0000000
--- a/LuaSL/testLua/yueliang-0.4.1/nat-5.1.3/llex_mk2.lua
+++ /dev/null
@@ -1,314 +0,0 @@
1--[[--------------------------------------------------------------------
2
3 llex.lua
4 Lua 5.1 lexical analyzer in Lua
5 This file is part of Yueliang.
6
7 Copyright (c) 2008 Kein-Hong Man <khman@users.sf.net>
8 The COPYRIGHT file describes the conditions
9 under which this software may be distributed.
10
11 See the ChangeLog for more information.
12
13----------------------------------------------------------------------]]
14
15--[[--------------------------------------------------------------------
16-- Notes:
17-- * takes in the entire source at once
18-- * greatly simplified chunkid, error handling
19-- * NO shbang handling (it's done elsewhere in Lua 5.1)
20-- * NO localized decimal point replacement magic
21-- * NO limit to number of lines (MAX_INT = 2147483645)
22-- * NO support for compatible long strings (LUA_COMPAT_LSTR)
23-- * NO next(), lookahead() because I want next() to set tok and
24-- seminfo that are locals, and that can only be done easily in
25-- lparser, not llex. lastline would be handled in lparser too.
26--
27-- Usage example:
28-- local llex = require("llex_mk2")
29-- llex.init(source_code, source_code_name)
30-- repeat
31-- local token, seminfo = llex.llex()
32-- until token == "<eof>"
33--
34----------------------------------------------------------------------]]
35
36local base = _G
37local string = require "string"
38module "llex"
39
40----------------------------------------------------------------------
41-- initialize keyword list
42----------------------------------------------------------------------
43local kw = {}
44for v in string.gmatch([[
45and break do else elseif end false for function if in
46local nil not or repeat return then true until while]], "%S+") do
47 kw[v] = true
48end
49
50----------------------------------------------------------------------
51-- initialize lexer for given source _z and source name _sourceid
52----------------------------------------------------------------------
53local z, sourceid, I
54local find = string.find
55local match = string.match
56local sub = string.sub
57
58function init(_z, _sourceid)
59 z = _z -- source
60 sourceid = _sourceid -- name of source
61 I = 1 -- lexer's position in source
62 ln = 1 -- line number
63end
64
65----------------------------------------------------------------------
66-- returns a chunk name or id, no truncation for long names
67----------------------------------------------------------------------
68function chunkid()
69 if sourceid and match(sourceid, "^[=@]") then
70 return sub(sourceid, 2) -- remove first char
71 end
72 return "[string]"
73end
74
75----------------------------------------------------------------------
76-- formats error message and throws error
77-- * a simplified version, does not report what token was responsible
78----------------------------------------------------------------------
79function errorline(s, line)
80 base.error(string.format("%s:%d: %s", chunkid(), line or ln, s))
81end
82
83----------------------------------------------------------------------
84-- handles line number incrementation and end-of-line characters
85----------------------------------------------------------------------
86
87local function inclinenumber(i)
88 local sub = sub
89 local old = sub(z, i, i)
90 i = i + 1 -- skip '\n' or '\r'
91 local c = sub(z, i, i)
92 if (c == "\n" or c == "\r") and (c ~= old) then
93 i = i + 1 -- skip '\n\r' or '\r\n'
94 end
95 ln = ln + 1
96 I = i
97 return i
98end
99
100------------------------------------------------------------------------
101-- count separators ("=") in a long string delimiter
102------------------------------------------------------------------------
103local function skip_sep(i)
104 local sub = sub
105 local s = sub(z, i, i)
106 i = i + 1
107 local count = #match(z, "=*", i) -- note, take the length
108 i = i + count
109 I = i
110 return (sub(z, i, i) == s) and count or (-count) - 1
111end
112
113----------------------------------------------------------------------
114-- reads a long string or long comment
115----------------------------------------------------------------------
116
117local function read_long_string(is_str, sep)
118 local i = I + 1 -- skip 2nd '['
119 local sub = sub
120 local buff = ""
121 local c = sub(z, i, i)
122 if c == "\r" or c == "\n" then -- string starts with a newline?
123 i = inclinenumber(i) -- skip it
124 end
125 local j = i
126 while true do
127 local p, q, r = find(z, "([\r\n%]])", i) -- (long range)
128 if not p then
129 errorline(is_str and "unfinished long string" or
130 "unfinished long comment")
131 end
132 if is_str then
133 buff = buff..sub(z, i, p - 1) -- save string portions
134 end
135 i = p
136 if r == "]" then -- delimiter test
137 if skip_sep(i) == sep then
138 i = I + 1 -- skip 2nd ']'
139 break
140 end
141 buff = buff..sub(z, i, I - 1)
142 i = I
143 else -- newline
144 buff = buff.."\n"
145 i = inclinenumber(i)
146 end
147 end--while
148 I = i
149 return buff
150end
151
152----------------------------------------------------------------------
153-- reads a string
154----------------------------------------------------------------------
155local function read_string(del)
156 local i = I
157 local find = find
158 local sub = sub
159 local buff = ""
160 while true do
161 local p, q, r = find(z, "([\n\r\\\"\'])", i) -- (long range)
162 if p then
163 if r == "\n" or r == "\r" then
164 errorline("unfinished string")
165 end
166 buff = buff..sub(z, i, p - 1) -- normal portions
167 i = p
168 if r == "\\" then -- handle escapes
169 i = i + 1
170 r = sub(z, i, i)
171 if r == "" then break end -- (EOZ error)
172 p = find("abfnrtv\n\r", r, 1, true)
173 ------------------------------------------------------
174 if p then -- special escapes
175 if p > 7 then
176 r = "\n"
177 i = inclinenumber(i)
178 else
179 r = sub("\a\b\f\n\r\t\v", p, p)
180 i = i + 1
181 end
182 ------------------------------------------------------
183 elseif find(r, "%D") then -- other non-digits
184 i = i + 1
185 ------------------------------------------------------
186 else -- \xxx sequence
187 local p, q, s = find(z, "^(%d%d?%d?)", i)
188 i = q + 1
189 if s + 1 > 256 then -- UCHAR_MAX
190 errorline("escape sequence too large")
191 end
192 r = string.char(s)
193 ------------------------------------------------------
194 end--if p
195 else
196 i = i + 1
197 if r == del then -- ending delimiter
198 I = i; return buff -- return string
199 end
200 end--if r
201 buff = buff..r -- handled escapes falls through to here
202 else
203 break -- (error)
204 end--if p
205 end--while
206 errorline("unfinished string")
207end
208
209------------------------------------------------------------------------
210-- main lexer function
211------------------------------------------------------------------------
212function llex()
213 local find = find
214 local match = match
215 while true do--outer
216 local i = I
217 -- inner loop allows break to be used to nicely section tests
218 while true do--inner
219 ----------------------------------------------------------------
220 local p, _, r = find(z, "^([_%a][_%w]*)", i)
221 if p then
222 I = i + #r
223 if kw[r] then return r end -- reserved word (keyword)
224 return "<name>", r -- identifier
225 end
226 ----------------------------------------------------------------
227 local p, _, r = find(z, "^(%.?)%d", i)
228 if p then -- numeral
229 if r == "." then i = i + 1 end
230 local _, q, r = find(z, "^%d*[%.%d]*([eE]?)", i)
231 i = q + 1
232 if #r == 1 then -- optional exponent
233 if match(z, "^[%+%-]", i) then -- optional sign
234 i = i + 1
235 end
236 end
237 local _, q = find(z, "^[_%w]*", i)
238 I = q + 1
239 local v = base.tonumber(sub(z, p, q)) -- handles hex also
240 if not v then errorline("malformed number") end
241 return "<number>", v
242 end
243 ----------------------------------------------------------------
244 local p, q, r = find(z, "^(%s)[ \t]*", i)
245 if p then
246 if r == "\n" or r == "\r" then -- newline
247 inclinenumber(i)
248 else
249 I = q + 1 -- whitespace
250 end
251 break -- (continue)
252 end
253 ----------------------------------------------------------------
254 local r = match(z, "^%p", i)
255 if r then
256 local p = find("-[\"\'.=<>~", r, 1, true)
257 if p then
258 -- two-level if block for punctuation/symbols
259 --------------------------------------------------------
260 if p <= 2 then
261 if p == 1 then -- minus
262 local c = match(z, "^%-%-(%[?)", i)
263 if c then
264 i = i + 2
265 local sep = -1
266 if c == "[" then
267 sep = skip_sep(i)
268 end
269 if sep >= 0 then -- long comment
270 read_long_string(false, sep)
271 else -- short comment
272 I = find(z, "[\n\r]", i) or (#z + 1)
273 end
274 break -- (continue)
275 end
276 -- (fall through for "-")
277 else -- [ or long string
278 local sep = skip_sep(i)
279 if sep >= 0 then
280 return "<string>", read_long_string(true, sep)
281 elseif sep == -1 then
282 return "["
283 else
284 errorline("invalid long string delimiter")
285 end
286 end
287 --------------------------------------------------------
288 elseif p <= 5 then
289 if p < 5 then -- strings
290 I = i + 1
291 return "<string>", read_string(r)
292 end
293 r = match(z, "^%.%.?%.?", i) -- .|..|... dots
294 -- (fall through)
295 --------------------------------------------------------
296 else -- relational
297 r = match(z, "^%p=?", i)
298 -- (fall through)
299 end
300 end
301 I = i + #r; return r -- for other symbols, fall through
302 end
303 ----------------------------------------------------------------
304 local r = sub(z, i, i)
305 if r ~= "" then
306 I = i + 1; return r -- other single-char tokens
307 end
308 return "<eof>" -- end of stream
309 ----------------------------------------------------------------
310 end--while inner
311 end--while outer
312end
313
314return base.getfenv()
generated by cgit v1.1 at 2024-07-06 14:25:31 +0000