Mercurial Hosting > luan
comparison website/src/manual.html.luan @ 1651:5b8f056527a3
docs work
author | Franklin Schmidt <fschmidt@gmail.com> |
---|---|
date | Thu, 31 Mar 2022 16:13:32 -0600 |
parents | website/src/manual.html@0af6a9d6d12f |
children | d5779a264a4a |
comparison
equal
deleted
inserted
replaced
1650:cc3b10a94612 | 1651:5b8f056527a3 |
---|---|
1 local Luan = require "luan:Luan.luan" | |
2 local error = Luan.error | |
3 local Io = require "luan:Io.luan" | |
4 local Http = require "luan:http/Http.luan" | |
5 local Shared = require "site:/lib/Shared.luan" | |
6 local head = Shared.head or error() | |
7 local header = Shared.header or error() | |
8 | |
9 | |
10 return function() | |
11 Io.stdout = Http.response.text_writer() | |
12 %> | |
13 <!doctype html> | |
14 <html> | |
15 <head> | |
16 <% head() %> | |
17 <title>Luan Reference Manual</title> | |
18 <style> | |
19 p[keywords] { | |
20 font-family: monospace; | |
21 margin-left: 40px; | |
22 max-width: 700px; | |
23 } | |
24 p[keywords] span { | |
25 display: inline-block; | |
26 width: 100px; | |
27 } | |
28 </style> | |
29 </head> | |
30 <body> | |
31 <% header{[[<a href="docs.html">Documentation</a>]]} %> | |
32 <div content> | |
33 | |
34 <h1><a href="manual.html">Luan Reference Manual</a></h1> | |
35 | |
36 <p small> | |
37 Original copyright © 2015 Lua.org, PUC-Rio. | |
38 Freely available under the terms of the | |
39 <a href="http://www.lua.org/license.html">Lua license</a>. | |
40 Modified for Luan. | |
41 </p> | |
42 | |
43 <hr/> | |
44 | |
45 <h2>Contents</h2> | |
46 | |
47 <div contents><a href="#intro">Introduction</a></div> | |
48 | |
49 <div contents> | |
50 <a href="#basic">Basic Concepts</a> | |
51 <ul> | |
52 <li><a href="#types">Values and Types</a></li> | |
53 <li><a href="#env">Environments</a></li> | |
54 <li><a href="#error">Error Handling</a></li> | |
55 <li><a href="#meta">Metatables and Metamethods</a></li> | |
56 <li><a href="#gc">Garbage Collection</a></li> | |
57 </ul> | |
58 </div> | |
59 | |
60 <div contents> | |
61 <a href="#lang">The Language</a> | |
62 <ul> | |
63 <li><a href="#lex">Lexical Conventions</a></li> | |
64 <li><a href="#vars">Variables</a></li> | |
65 <li> | |
66 <a href="#stmts">Statements</a> | |
67 <ul> | |
68 <li><a href="#blocks">Blocks</a></li> | |
69 <li><a href="#chunks">Chunks</a></li> | |
70 <li><a href="#assignment">Assignment</a></li> | |
71 <li><a href="#control">Control Structures</a></li> | |
72 <li><a href="#for">For Statement</a></li> | |
73 <li><a href="#try">Try Statement</a></li> | |
74 <li><a href="#fn_stmt">Function Calls as Statements</a></li> | |
75 <li><a href="#local_stmt">Local Declarations</a></li> | |
76 <li><a href="#template_stmt">Template Statements</a></li> | |
77 </ul> | |
78 </li> | |
79 <li> | |
80 <a href="#expressions">Expressions</a> | |
81 <ul> | |
82 <li><a href="#arithmetic">Arithmetic Operators</a></li> | |
83 <li><a href="#conversions">Coercions and Conversions</a></li> | |
84 <li><a href="#relational">Relational Operators</a></li> | |
85 <li><a href="#logical_ops">Logical Operators</a></li> | |
86 <li><a href="#concatenation">Concatenation</a></li> | |
87 <li><a href="#length">The Length Operator</a></li> | |
88 <li><a href="#precedence">Precedence</a></li> | |
89 <li><a href="#constructors">Table Constructors</a></li> | |
90 <li><a href="#fn_calls">Function Calls</a></li> | |
91 <li><a href="#fn_def">Function Definitions</a></li> | |
92 </ul> | |
93 </li> | |
94 <li><a href="#visibility">Visibility Rules</a></li> | |
95 </ul> | |
96 </div> | |
97 | |
98 <div contents> | |
99 <a href="#libs">Standard Libraries</a> | |
100 <ul> | |
101 <li><a href="#default_lib">Default Environment</a></li> | |
102 <li><a href="#luan_lib">Basic Functions</a></li> | |
103 <li><a href="#package_lib">Modules</a></li> | |
104 <li><a href="#string_lib">String Manipulation</a></li> | |
105 <li><a href="#binary_lib">Binary Manipulation</a></li> | |
106 <li><a href="#table_lib">Table Manipulation</a></li> | |
107 <li><a href="#number_lib">Number Manipulation</a></li> | |
108 <li><a href="#math_lib">Mathematical Functions</a></li> | |
109 </ul> | |
110 </div> | |
111 | |
112 <hr/> | |
113 | |
114 | |
115 <h2 heading><a name="intro" href="#intro">Introduction</a></h2> | |
116 | |
117 <p>Luan is a high level programming language based on <a href="http://www.lua.org">Lua</a>. A great strength of Lua is its simplicity and Luan takes this even further, being even simpler than Lua. The goal is to provide a simple programming language for the casual programmer with as few concepts as possible so that one can quickly learn the language and then easily understand any code written in Luan.</p> | |
118 | |
119 <p>Luan is implemented in Java and is tightly coupled with Java. So it makes a great scripting language for Java programmers.</p> | |
120 | |
121 <p>Unlike Lua which is meant to be embedded, Luan is meant to be a full scripting language. This done not by adding feature to Luan, but rather by providing a complete set of libraries.</p> | |
122 | |
123 | |
124 <h2 heading><a name="basic" href="#basic">Basic Concepts</a></h2> | |
125 | |
126 <p>This section describes the basic concepts of the language.</p> | |
127 | |
128 <h3 heading><a name="types" href="#types">Values and Types</a></h3> | |
129 | |
130 <p> | |
131 Luan is a <em>dynamically typed language</em>. | |
132 This means that | |
133 variables do not have types; only values do. | |
134 There are no type definitions in the language. | |
135 All values carry their own type. | |
136 | |
137 | |
138 <p> | |
139 All values in Luan are <em>first-class values</em>. | |
140 This means that all values can be stored in variables, | |
141 passed as arguments to other functions, and returned as results. | |
142 | |
143 | |
144 <p> | |
145 There are eight basic types in Luan: | |
146 <em>nil</em>, <em>boolean</em>, <em>number</em>, | |
147 <em>string</em>, <em>binary</em>, <em>function</em>, <em>java</em>, | |
148 and <em>table</em>. | |
149 <em>Nil</em> is the type of the value <b>nil</b>, | |
150 whose main property is to be different from any other value; | |
151 it usually represents the absence of a useful value. | |
152 <em>Nil</em> is implemented as the Java value <em>null</em>. | |
153 <em>Boolean</em> is the type of the values <b>false</b> and <b>true</b>. | |
154 <em>Boolean</em> is implemented as the Java class <em>Boolean</em>. | |
155 <em>Number</em> represents both | |
156 integer numbers and real (floating-point) numbers. | |
157 <em>Number</em> is implemented as the Java class <em>Number</em>. Any Java subclass of <em>Number</em> is allowed and this is invisible to the Luan user. Operations on numbers follow the same rules of | |
158 the underlying Java implementation. | |
159 | |
160 <em>String</em> is implemented as the Java class <em>String</em>. | |
161 <em>Binary</em> is implemented as the Java type <em>byte[]</em>. | |
162 | |
163 | |
164 <p> | |
165 Luan can call (and manipulate) functions written in Luan and | |
166 functions written in Java (see <a href="#fn_calls">Function Calls</a>). | |
167 Both are represented by the type <em>function</em>. | |
168 | |
169 | |
170 <p> | |
171 The type <em>java</em> is provided to allow arbitrary Java objects to | |
172 be stored in Luan variables. | |
173 A <em>java</em> value is a Java object that isn't one of the standard Luan types. | |
174 Java values have no predefined operations in Luan, | |
175 except assignment and identity test. | |
176 Java values are useful when Java access is enabled in Luan | |
177 | |
178 | |
179 | |
180 <p> | |
181 The type <em>table</em> implements associative arrays, | |
182 that is, arrays that can be indexed not only with numbers, | |
183 but with any Luan value except <b>nil</b>. | |
184 Tables can be <em>heterogeneous</em>; | |
185 that is, they can contain values of all types (except <b>nil</b>). | |
186 Any key with value <b>nil</b> is not considered part of the table. | |
187 Conversely, any key that is not part of a table has | |
188 an associated value <b>nil</b>. | |
189 | |
190 | |
191 <p> | |
192 Tables are the sole data-structuring mechanism in Luan; | |
193 they can be used to represent ordinary arrays, sequences, | |
194 symbol tables, sets, records, graphs, trees, etc. | |
195 To represent records, Luan uses the field name as an index. | |
196 The language supports this representation by | |
197 providing <code>a.name</code> as syntactic sugar for <code>a["name"]</code>. | |
198 There are several convenient ways to create tables in Luan | |
199 (see <a href="#constructors">Table Constructors</a>). | |
200 | |
201 | |
202 <p> | |
203 We use the term <em>sequence</em> to denote a table where | |
204 the set of all positive numeric keys is equal to {1..<em>n</em>} | |
205 for some non-negative integer <em>n</em>, | |
206 which is called the length of the sequence (see <a href="#length">The Length Operator</a>). | |
207 | |
208 | |
209 <p> | |
210 Like indices, | |
211 the values of table fields can be of any type. | |
212 In particular, | |
213 because functions are first-class values, | |
214 table fields can contain functions. | |
215 Thus tables can also carry <em>methods</em> (see <a href="#fn_def">Function Definitions</a>). | |
216 | |
217 | |
218 <p> | |
219 The indexing of tables follows | |
220 the definition of raw equality in the language. | |
221 The expressions <code>a[i]</code> and <code>a[j]</code> | |
222 denote the same table element | |
223 if and only if <code>i</code> and <code>j</code> are raw equal | |
224 (that is, equal without metamethods). | |
225 In particular, floats with integral values | |
226 are equal to their respective integers | |
227 (e.g., <code>1.0 == 1</code>). | |
228 | |
229 | |
230 <p> | |
231 Luan values are <em>objects</em>: | |
232 variables do not actually <em>contain</em> values, | |
233 only <em>references</em> to them. | |
234 Assignment, parameter passing, and function returns | |
235 always manipulate references to values; | |
236 these operations do not imply any kind of copy. | |
237 | |
238 | |
239 <p> | |
240 The library function <a href="#Luan.type"><code>Luan.type</code></a> returns a string describing the type | |
241 of a given value. | |
242 | |
243 | |
244 | |
245 | |
246 | |
247 <h3 heading><a name="env" href="#env">Environments</a></h3> | |
248 | |
249 <p> | |
250 The environment of a chunk starts with only one local variable: <code><a href="#require">require</a></code>. This function is used to load and access libraries and other modules. All other variables must be added to the environment using <a href="http://localhost:8080/manual.html#local_stmt">local declarations</a>. | |
251 | |
252 <p> | |
253 As will be discussed in <a href="#vars">Variables</a> and <a href=#assignment">Assignment</a>, | |
254 any reference to a free name | |
255 (that is, a name not bound to any declaration) <code>var</code> | |
256 can be syntactically translated to <code>_ENV.var</code> if <code>_ENV</code> is defined. | |
257 | |
258 | |
259 <h3 heading><a name="error" href="#error">Error Handling</a></h3> | |
260 | |
261 <p> | |
262 Luan code can explicitly generate an error by calling the | |
263 <a href="#Luan.error"><code>error</code></a> function. | |
264 If you need to catch errors in Luan, | |
265 you can use the <a href="#try">Try Statement</code></a>. | |
266 | |
267 | |
268 <p> | |
269 Whenever there is an error, | |
270 an <em>error table</em> | |
271 is propagated with information about the error. | |
272 See <a href="#Luan.new_error"><code>Luan.new_error</code></a>. | |
273 | |
274 | |
275 | |
276 <h3 heading><a name="meta" href="#meta">Metatables and Metamethods</a></h3> | |
277 | |
278 <p> | |
279 Every table in Luan can have a <em>metatable</em>. | |
280 This <em>metatable</em> is an ordinary Luan table | |
281 that defines the behavior of the original value | |
282 under certain special operations. | |
283 You can change several aspects of the behavior | |
284 of operations over a value by setting specific fields in its metatable. | |
285 For instance, when a table is the operand of an addition, | |
286 Luan checks for a function in the field "<code>__add</code>" of the table's metatable. | |
287 If it finds one, | |
288 Luan calls this function to perform the addition. | |
289 | |
290 | |
291 <p> | |
292 The keys in a metatable are derived from the <em>event</em> names; | |
293 the corresponding values are called <ii>metamethods</em>. | |
294 In the previous example, the event is <code>"add"</code> | |
295 and the metamethod is the function that performs the addition. | |
296 | |
297 | |
298 <p> | |
299 You can query the metatable of any table | |
300 using the <a href="#Luan.get_metatable"><code>get_metatable</code></a> function. | |
301 | |
302 | |
303 <p> | |
304 You can replace the metatable of tables | |
305 using the <a href="#Luan.set_metatable"><code>set_metatable</code></a> function. | |
306 | |
307 | |
308 <p> | |
309 A metatable controls how a table behaves in | |
310 arithmetic operations, bitwise operations, | |
311 order comparisons, concatenation, length operation, calls, and indexing. | |
312 | |
313 | |
314 <p> | |
315 A detailed list of events controlled by metatables is given next. | |
316 Each operation is identified by its corresponding event name. | |
317 The key for each event is a string with its name prefixed by | |
318 two underscores, '<code>__</code>'; | |
319 for instance, the key for operation "add" is the | |
320 string "<code>__add</code>". | |
321 Note that queries for metamethods are always raw; | |
322 the access to a metamethod does not invoke other metamethods. | |
323 You can emulate how Luan queries a metamethod for an object <code>obj</code> | |
324 with the following code: | |
325 | |
326 <pre> | |
327 raw_get(get_metatable(obj) or {}, "__" .. event_name) | |
328 </pre> | |
329 | |
330 <p> | |
331 Here are the events: | |
332 | |
333 <ul> | |
334 | |
335 <li><p><b>"add": </b> | |
336 the <code>+</code> operation. | |
337 | |
338 If any operand for an addition is a table, | |
339 Luan will try to call a metamethod. | |
340 First, Luan will check the first operand (even if it is valid). | |
341 If that operand does not define a metamethod for the "<code>__add</code>" event, | |
342 then Luan will check the second operand. | |
343 If Luan can find a metamethod, | |
344 it calls the metamethod with the two operands as arguments, | |
345 and the result of the call | |
346 (adjusted to one value) | |
347 is the result of the operation. | |
348 Otherwise, | |
349 it raises an error. | |
350 </li> | |
351 | |
352 <li><p><b>"sub": </b> | |
353 the <code>-</code> operation. | |
354 | |
355 Behavior similar to the "add" operation. | |
356 </li> | |
357 | |
358 <li><p><b>"mul": </b> | |
359 the <code>*</code> operation. | |
360 | |
361 Behavior similar to the "add" operation. | |
362 </li> | |
363 | |
364 <li><p><b>"div": </b> | |
365 the <code>/</code> operation. | |
366 | |
367 Behavior similar to the "add" operation. | |
368 </li> | |
369 | |
370 <li><p><b>"mod": </b> | |
371 the <code>%</code> operation. | |
372 | |
373 Behavior similar to the "add" operation. | |
374 </li> | |
375 | |
376 <li><p><b>"pow": </b> | |
377 the <code>^</code> (exponentiation) operation. | |
378 | |
379 Behavior similar to the "add" operation. | |
380 </li> | |
381 | |
382 <li><p><b>"unm": </b> | |
383 the <code>-</code> (unary minus) operation. | |
384 | |
385 Behavior similar to the "add" operation. | |
386 </li> | |
387 | |
388 <li><p><b>"concat": </b> | |
389 the <code>..</code> (concatenation) operation. | |
390 | |
391 Behavior similar to the "add" operation. | |
392 </li> | |
393 | |
394 <li><p><b>"len": </b> | |
395 the <code>#</code> (length) operation. | |
396 | |
397 If there is a metamethod, | |
398 Luan calls it with the object as argument, | |
399 and the result of the call | |
400 (always adjusted to one value) | |
401 is the result of the operation. | |
402 If there is no metamethod but the object is a table, | |
403 then Luan uses the table length operation (see <a href="#length">The Length Operator</a>). | |
404 Otherwise, Luan raises an error. | |
405 </li> | |
406 | |
407 <li><p><b>"eq": </b> | |
408 the <code>==</code> (equal) operation. | |
409 | |
410 Behavior similar to the "add" operation, | |
411 except that Luan will try a metamethod only when the values | |
412 being compared are both tables | |
413 and they are not primitively equal. | |
414 The result of the call is always converted to a boolean. | |
415 </li> | |
416 | |
417 <li><p><b>"lt": </b> | |
418 the <code><</code> (less than) operation. | |
419 | |
420 Behavior similar to the "add" operation. | |
421 The result of the call is always converted to a boolean. | |
422 </li> | |
423 | |
424 <li><p><b>"le": </b> | |
425 the <code><=</code> (less equal) operation. | |
426 | |
427 Unlike other operations, | |
428 The less-equal operation can use two different events. | |
429 First, Luan looks for the "<code>__le</code>" metamethod in both operands, | |
430 like in the "lt" operation. | |
431 If it cannot find such a metamethod, | |
432 then it will try the "<code>__lt</code>" event, | |
433 assuming that <code>a <= b</code> is equivalent to <code>not (b < a)</code>. | |
434 As with the other comparison operators, | |
435 the result is always a boolean. | |
436 </li> | |
437 | |
438 <li><p><b>"index": </b> | |
439 The indexing access <code>table[key]</code>. | |
440 | |
441 This event happens | |
442 when <code>key</code> is not present in <code>table</code>. | |
443 The metamethod is looked up in <code>table</code>. | |
444 | |
445 | |
446 <p> | |
447 Despite the name, | |
448 the metamethod for this event can be any type. | |
449 If it is a function, | |
450 it is called with <code>table</code> and <code>key</code> as arguments. | |
451 Otherwise | |
452 the final result is the result of indexing this metamethod object with <code>key</code>. | |
453 (This indexing is regular, not raw, | |
454 and therefore can trigger another metamethod if the metamethod object is a table.) | |
455 </li> | |
456 | |
457 <li><p><b>"new_index": </b> | |
458 The indexing assignment <code>table[key] = value</code>. | |
459 | |
460 Like the index event, | |
461 this event happens when | |
462 when <code>key</code> is not present in <code>table</code>. | |
463 The metamethod is looked up in <code>table</code>. | |
464 | |
465 | |
466 <p> | |
467 Like with indexing, | |
468 the metamethod for this event can be either a function or a table. | |
469 If it is a function, | |
470 it is called with <code>table</code>, <code>key</code>, and <code>value</code> as arguments. | |
471 If it is a table, | |
472 Luan does an indexing assignment to this table with the same key and value. | |
473 (This assignment is regular, not raw, | |
474 and therefore can trigger another metamethod.) | |
475 | |
476 | |
477 <p> | |
478 Whenever there is a "new_index" metamethod, | |
479 Luan does not perform the primitive assignment. | |
480 (If necessary, | |
481 the metamethod itself can call <a href="#Luan.raw_set"><code>raw_set</code></a> | |
482 to do the assignment.) | |
483 </li> | |
484 | |
485 <li><p><b>"gc":</b> | |
486 This is when a table is garbage collected. When the table's <a href="https://docs.oracle.com/javase/7/docs/api/java/lang/Object.html#finalize()">finalize</a> method is called by the Java garbage collector, if there is a "<code>__gc</code>" metamethod then it is called with the table as a parameter. | |
487 | |
488 </li> | |
489 | |
490 </ul> | |
491 | |
492 | |
493 | |
494 | |
495 <h3 heading><a name="gc" href="#gc">Garbage Collection</a></h3> | |
496 | |
497 <p> | |
498 Luan uses Java's garbage collection. | |
499 | |
500 | |
501 | |
502 | |
503 | |
504 <h2 heading><a name="lang" href="#lang">The Language</a></h2> | |
505 | |
506 <p> | |
507 This section describes the lexis, the syntax, and the semantics of Luan. | |
508 In other words, | |
509 this section describes | |
510 which tokens are valid, | |
511 how they can be combined, | |
512 and what their combinations mean. | |
513 | |
514 | |
515 <p> | |
516 Language constructs will be explained using the usual extended BNF notation, | |
517 in which | |
518 {<em>a</em>} means 0 or more <em>a</em>'s, and | |
519 [<em>a</em>] means an optional <em>a</em>. | |
520 Non-terminals are shown like non-terminal, | |
521 keywords are shown like <b>kword</b>, | |
522 and other terminal symbols are shown like ‘<b>=</b>’. | |
523 The complete syntax of Luan can be found in <a href="#9">§9</a> | |
524 at the end of this manual. | |
525 | |
526 | |
527 | |
528 <h3 heading><a name="lex" href="#lex">Lexical Conventions</a></h3> | |
529 | |
530 <p> | |
531 Luan ignores spaces and comments | |
532 between lexical elements (tokens), | |
533 except as delimiters between names and keywords. | |
534 Luan considers the end of a line to be the end of a statement. This catches errors and encourages readability. If you want to continue a statement on another line, you can use a backslash followed by a newline which will be treated as white space. | |
535 | |
536 <p> | |
537 <em>Names</em> | |
538 (also called <em>identifiers</em>) | |
539 in Luan can be any string of letters, | |
540 digits, and underscores, | |
541 not beginning with a digit. | |
542 Identifiers are used to name variables, table fields, and labels. | |
543 | |
544 | |
545 <p> | |
546 The following <em>keywords</em> are reserved | |
547 and cannot be used as names: | |
548 | |
549 | |
550 <p keywords> | |
551 <span>and</span> | |
552 <span>break</span> | |
553 <span>catch</span> | |
554 <span>continue</span> | |
555 <span>do</span> | |
556 <span>else</span> | |
557 <span>elseif</span> | |
558 <span>end_do</span> | |
559 <span>end_for</span> | |
560 <span>end_function</span> | |
561 <span>end_if</span> | |
562 <span>end_try</span> | |
563 <span>end_while</span> | |
564 <span>false</span> | |
565 <span>finally</span> | |
566 <span>for</span> | |
567 <span>function</span> | |
568 <span>if</span> | |
569 <span>in</span> | |
570 <span>local</span> | |
571 <span>nil</span> | |
572 <span>not</span> | |
573 <span>or</span> | |
574 <span>repeat</span> | |
575 <span>return</span> | |
576 <span>then</span> | |
577 <span>true</span> | |
578 <span>try</span> | |
579 <span>until</span> | |
580 <span>while</span> | |
581 </p> | |
582 | |
583 <p> | |
584 Luan is a case-sensitive language: | |
585 <code>and</code> is a reserved word, but <code>And</code> and <code>AND</code> | |
586 are two different, valid names. | |
587 | |
588 | |
589 <p> | |
590 The following strings denote other tokens: | |
591 | |
592 <pre> | |
593 + - * / % ^ # | |
594 & ~ | << >> // | |
595 == ~= <= >= < > = | |
596 ( ) { } [ ] :: | |
597 ; : , . .. ... | |
598 </pre> | |
599 | |
600 <p> | |
601 <em>Literal strings</em> | |
602 can be delimited by matching single or double quotes, | |
603 and can contain the following C-like escape sequences: | |
604 '<code>\a</code>' (bell), | |
605 '<code>\b</code>' (backspace), | |
606 '<code>\f</code>' (form feed), | |
607 '<code>\n</code>' (newline), | |
608 '<code>\r</code>' (carriage return), | |
609 '<code>\t</code>' (horizontal tab), | |
610 '<code>\v</code>' (vertical tab), | |
611 '<code>\\</code>' (backslash), | |
612 '<code>\"</code>' (quotation mark [double quote]), | |
613 and '<code>\'</code>' (apostrophe [single quote]). | |
614 A backslash followed by a real newline | |
615 results in a newline in the string. | |
616 The escape sequence '<code>\z</code>' skips the following span | |
617 of white-space characters, | |
618 including line breaks; | |
619 it is particularly useful to break and indent a long literal string | |
620 into multiple lines without adding the newlines and spaces | |
621 into the string contents. | |
622 | |
623 | |
624 <p> | |
625 Luan can specify any character in a literal string by its numerical value. | |
626 This can be done | |
627 with the escape sequence <code>\x<em>XX</em></code>, | |
628 where <em>XX</em> is a sequence of exactly two hexadecimal digits, | |
629 or with the escape sequence <code>\u<em>XXXX</em></code>, | |
630 where <em>XXXX</em> is a sequence of exactly four hexadecimal digits, | |
631 or with the escape sequence <code>\<em>ddd</em></code>, | |
632 where <em>ddd</em> is a sequence of up to three decimal digits. | |
633 (Note that if a decimal escape sequence is to be followed by a digit, | |
634 it must be expressed using exactly three digits.) | |
635 | |
636 | |
637 <p> | |
638 Literal strings can also be defined using a long format | |
639 enclosed by <em>long brackets</em>. | |
640 We define an <em>opening long bracket of level <em>n</em></em> as an opening | |
641 square bracket followed by <em>n</em> equal signs followed by another | |
642 opening square bracket. | |
643 So, an opening long bracket of level 0 is written as <code>[[</code>, | |
644 an opening long bracket of level 1 is written as <code>[=[</code>, | |
645 and so on. | |
646 A <em>closing long bracket</em> is defined similarly; | |
647 for instance, | |
648 a closing long bracket of level 4 is written as <code>]====]</code>. | |
649 A <em>long literal</em> starts with an opening long bracket of any level and | |
650 ends at the first closing long bracket of the same level. | |
651 It can contain any text except a closing bracket of the same level. | |
652 Literals in this bracketed form can run for several lines, | |
653 do not interpret any escape sequences, | |
654 and ignore long brackets of any other level. | |
655 Any kind of end-of-line sequence | |
656 (carriage return, newline, carriage return followed by newline, | |
657 or newline followed by carriage return) | |
658 is converted to a simple newline. | |
659 | |
660 | |
661 <p> | |
662 Any character in a literal string not | |
663 explicitly affected by the previous rules represents itself. | |
664 However, Luan opens files for parsing in text mode, | |
665 and the system file functions may have problems with | |
666 some control characters. | |
667 So, it is safer to represent | |
668 non-text data as a quoted literal with | |
669 explicit escape sequences for non-text characters. | |
670 | |
671 | |
672 <p> | |
673 For convenience, | |
674 when the opening long bracket is immediately followed by a newline, | |
675 the newline is not included in the string. | |
676 As an example | |
677 the five literal strings below denote the same string: | |
678 | |
679 <pre> | |
680 a = 'alo\n123"' | |
681 a = "alo\n123\"" | |
682 a = '\97lo\10\04923"' | |
683 a = [[alo | |
684 123"]] | |
685 a = [==[ | |
686 alo | |
687 123"]==] | |
688 </pre> | |
689 | |
690 <p> | |
691 A <em>numerical constant</em> (or <em>numeral</em>) | |
692 can be written with an optional fractional part | |
693 and an optional decimal exponent, | |
694 marked by a letter '<code>e</code>' or '<code>E</code>'. | |
695 Luan also accepts hexadecimal constants, | |
696 which start with <code>0x</code> or <code>0X</code>. | |
697 Hexadecimal constants also accept an optional fractional part | |
698 plus an optional binary exponent, | |
699 marked by a letter '<code>p</code>' or '<code>P</code>'. | |
700 A numeric constant with a fractional dot or an exponent | |
701 denotes a float; | |
702 otherwise it denotes an integer. | |
703 Examples of valid integer constants are | |
704 | |
705 <pre> | |
706 3 345 0xff 0xBEBADA | |
707 </pre> | |
708 | |
709 <p> | |
710 Examples of valid float constants are | |
711 | |
712 <pre> | |
713 3.0 3.1416 314.16e-2 0.31416E1 34e1 | |
714 0x0.1E 0xA23p-4 0X1.921FB54442D18P+1 | |
715 </pre> | |
716 | |
717 <p> | |
718 A <em>comment</em> starts with a double hyphen (<code>--</code>) | |
719 anywhere outside a string. | |
720 If the text immediately after <code>--</code> is not an opening long bracket, | |
721 the comment is a <em>short comment</em>, | |
722 which runs until the end of the line. | |
723 Otherwise, it is a <em>long comment</em>, | |
724 which runs until the corresponding closing long bracket. | |
725 Long comments are frequently used to disable code temporarily. | |
726 | |
727 | |
728 | |
729 | |
730 | |
731 <h3 heading><a name="vars" href="#vars">Variables</a></h3> | |
732 | |
733 <p> | |
734 Variables are places that store values. | |
735 There are three kinds of variables in Luan: | |
736 global variables, local variables, and table fields. | |
737 | |
738 <p> | |
739 A single name can denote a global variable or a local variable | |
740 (or a function's formal parameter, | |
741 which is a particular kind of local variable): | |
742 | |
743 <pre> | |
744 var ::= Name | |
745 </pre> | |
746 | |
747 <p> | |
748 Name denotes identifiers, as defined in <a href="#lex">Lexical Conventions</a>. | |
749 | |
750 <p> | |
751 Local variables are <em>lexically scoped</em>: | |
752 local variables can be freely accessed by functions | |
753 defined inside their scope (see <a href="#visibility">Visibility Rules</a>). | |
754 | |
755 | |
756 <p> | |
757 Before the first assignment to a variable, its value is <b>nil</b>. | |
758 | |
759 <p> | |
760 Square brackets are used to index a table: | |
761 | |
762 <pre> | |
763 var ::= prefixexp ‘<b>[</b>’ exp ‘<b>]</b>’ | |
764 </pre> | |
765 | |
766 <p> | |
767 The meaning of accesses to table fields can be changed via metatables. | |
768 An access to an indexed variable <code>t[i]</code> is equivalent to | |
769 a call <code>gettable_event(t,i)</code>. | |
770 (See <a href="#meta">Metatables and Metamethods</a> for a complete description of the | |
771 <code>gettable_event</code> function. | |
772 This function is not defined or callable in Luan. | |
773 We use it here only for explanatory purposes.) | |
774 | |
775 | |
776 <p> | |
777 The syntax <code>var.Name</code> is just syntactic sugar for | |
778 <code>var["Name"]</code>: | |
779 | |
780 <pre> | |
781 var ::= prefixexp ‘<b>.</b>’ Name | |
782 </pre> | |
783 | |
784 <p> | |
785 Global variables are not available by default. To enable global variable, you must define <code>_ENV</code> as a local variable whose value is a table. If <code>_ENV</code> is not defined, then an unrecognized variable name will produce a compile error. If <code>_ENV</code> is defined then an access to an unrecognized variable name will be consider a global variable. So then an acces to global variable <code>x</code> | |
786 is equivalent to <code>_ENV.x</code>. | |
787 Due to the way that chunks are compiled, | |
788 <code>_ENV</code> is never a global name (see <a href="#env">Environments</a>). | |
789 | |
790 | |
791 | |
792 | |
793 | |
794 <h3 heading><a name="stmts" href="#stmts">Statements</a></h3> | |
795 | |
796 <p> | |
797 Luan supports an almost conventional set of statements, | |
798 similar to those in Pascal or C. | |
799 This set includes | |
800 assignments, control structures, function calls, | |
801 and variable declarations. | |
802 | |
803 | |
804 | |
805 <h4 heading><a name="blocks" href="#blocks">Blocks</a></h4> | |
806 | |
807 <p> | |
808 A block is a list of statements, | |
809 which are executed sequentially: | |
810 | |
811 <pre> | |
812 block ::= {stat} | |
813 </pre> | |
814 | |
815 <p> | |
816 Luan has <em>empty statements</em> | |
817 that allow you to separate statements with semicolons, | |
818 start a block with a semicolon | |
819 or write two semicolons in sequence: | |
820 | |
821 <pre> | |
822 stat ::= ‘<b>;</b>’ | |
823 </pre> | |
824 | |
825 <p> | |
826 A block can be explicitly delimited to produce a single statement: | |
827 | |
828 <pre> | |
829 stat ::= <b>do</b> block end_do | |
830 end_do ::= <b>end_do</b> | <b>end</b> | |
831 </pre> | |
832 | |
833 <p> | |
834 Explicit blocks are useful | |
835 to control the scope of variable declarations. | |
836 Explicit blocks are also sometimes used to | |
837 add a <b>return</b> statement in the middle | |
838 of another block (see <a href="#control">Control Structures</a>). | |
839 | |
840 | |
841 | |
842 | |
843 | |
844 <h4 heading><a name="chunks" href="#chunks">Chunks</a></h4> | |
845 | |
846 <p> | |
847 The unit of compilation of Luan is called a <em>chunk</em>. | |
848 Syntactically, | |
849 a chunk is simply a block: | |
850 | |
851 <pre> | |
852 chunk ::= block | |
853 </pre> | |
854 | |
855 <p> | |
856 Luan handles a chunk as the body of an anonymous function | |
857 with a variable number of arguments | |
858 (see <a href="#fn_def">Function Definitions</a>). | |
859 As such, chunks can define local variables, | |
860 receive arguments, and return values. | |
861 | |
862 | |
863 <p> | |
864 A chunk can be stored in a file or in a string inside the host program. | |
865 To execute a chunk, | |
866 Luan first <em>loads</em> it, | |
867 compiling the chunk's code, | |
868 and then Luan executes the compiled code. | |
869 | |
870 | |
871 | |
872 | |
873 | |
874 <h4 heading><a name="assignment" href="#assignment">Assignment</a></h4> | |
875 | |
876 <p> | |
877 Luan allows multiple assignments. | |
878 Therefore, the syntax for assignment | |
879 defines a list of variables on the left side | |
880 and a list of expressions on the right side. | |
881 The elements in both lists are separated by commas: | |
882 | |
883 <pre> | |
884 stat ::= varlist ‘<b>=</b>’ explist | |
885 varlist ::= var {‘<b>,</b>’ var} | |
886 explist ::= exp {‘<b>,</b>’ exp} | |
887 </pre> | |
888 | |
889 <p> | |
890 Expressions are discussed in <a href="#expressions">Expressions</a>. | |
891 | |
892 | |
893 <p> | |
894 Before the assignment, | |
895 the list of values is <em>adjusted</em> to the length of | |
896 the list of variables. | |
897 If there are more values than needed, | |
898 the excess values are thrown away. | |
899 If there are fewer values than needed, | |
900 the list is extended with as many <b>nil</b>'s as needed. | |
901 If the list of expressions ends with a function call, | |
902 then all values returned by that call enter the list of values, | |
903 before the adjustment | |
904 (except when the call is enclosed in parentheses; see <a href="#expressions">Expressions</a>). | |
905 | |
906 | |
907 <p> | |
908 The assignment statement first evaluates all its expressions | |
909 and only then the assignments are performed. | |
910 Thus the code | |
911 | |
912 <pre> | |
913 i = 3 | |
914 i, a[i] = i+1, 20 | |
915 </pre> | |
916 | |
917 <p> | |
918 sets <code>a[3]</code> to 20, without affecting <code>a[4]</code> | |
919 because the <code>i</code> in <code>a[i]</code> is evaluated (to 3) | |
920 before it is assigned 4. | |
921 Similarly, the line | |
922 | |
923 <pre> | |
924 x, y = y, x | |
925 </pre> | |
926 | |
927 <p> | |
928 exchanges the values of <code>x</code> and <code>y</code>, | |
929 and | |
930 | |
931 <pre> | |
932 x, y, z = y, z, x | |
933 </pre> | |
934 | |
935 <p> | |
936 cyclically permutes the values of <code>x</code>, <code>y</code>, and <code>z</code>. | |
937 | |
938 | |
939 <p> | |
940 The meaning of assignments to global variables | |
941 and table fields can be changed via metatables. | |
942 An assignment to an indexed variable <code>t[i] = val</code> is equivalent to | |
943 <code>settable_event(t,i,val)</code>. | |
944 (See <a href="#meta">Metatables and Metamethods</a> for a complete description of the | |
945 <code>settable_event</code> function. | |
946 This function is not defined or callable in Luan. | |
947 We use it here only for explanatory purposes.) | |
948 | |
949 | |
950 <p> | |
951 An assignment to a global name <code>x = val</code> | |
952 is equivalent to the assignment | |
953 <code>_ENV.x = val</code> (see <a href="#env">Environments</a>). | |
954 Global names are only available when <code>_ENV</code> is defined. | |
955 | |
956 | |
957 | |
958 <h4 heading><a name="control" href="#control">Control Structures</a></h4> | |
959 | |
960 <p> | |
961 The control structures | |
962 <b>if</b>, <b>while</b>, and <b>repeat</b> have the usual meaning and | |
963 familiar syntax: | |
964 | |
965 <pre> | |
966 stat ::= <b>while</b> exp <b>do</b> block end_while | |
967 stat ::= <b>repeat</b> block <b>until</b> exp | |
968 stat ::= <b>if</b> exp <b>then</b> block {<b>elseif</b> exp <b>then</b> block} [<b>else</b> block] end_if | |
969 end_while ::= <b>end_while</b> | <b>end</b> | |
970 end_if ::= <b>end_if</b> | <b>end</b> | |
971 </pre> | |
972 | |
973 <p> | |
974 Luan also has a <b>for</b> statement (see <a href="#for">For Statement</a>). | |
975 | |
976 | |
977 <p> | |
978 The condition expression of a | |
979 control structure must be a boolean. | |
980 Any other value type will produce an error. | |
981 This helps catch errors and makes code more readable. | |
982 | |
983 | |
984 <p> | |
985 In the <b>repeat</b>–<b>until</b> loop, | |
986 the inner block does not end at the <b>until</b> keyword, | |
987 but only after the condition. | |
988 So, the condition can refer to local variables | |
989 declared inside the loop block. | |
990 | |
991 | |
992 <p> | |
993 The <b>break</b> statement terminates the execution of a | |
994 <b>while</b>, <b>repeat</b>, or <b>for</b> loop, | |
995 skipping to the next statement after the loop: | |
996 | |
997 <pre> | |
998 stat ::= <b>break</b> | |
999 </pre> | |
1000 | |
1001 <p> | |
1002 A <b>break</b> ends the innermost enclosing loop. | |
1003 | |
1004 | |
1005 <p> | |
1006 The <b>continue</b> statement jumps to the beginning of a | |
1007 <b>while</b>, <b>repeat</b>, or <b>for</b> loop for next iteration, | |
1008 skipping the execution of statements inside the body of loop for the current iteration: | |
1009 | |
1010 <pre> | |
1011 stat ::= <b>continue</b> | |
1012 </pre> | |
1013 | |
1014 | |
1015 <p> | |
1016 The <b>return</b> statement is used to return values | |
1017 from a function or a chunk | |
1018 (which is an anonymous function). | |
1019 | |
1020 Functions can return more than one value, | |
1021 so the syntax for the <b>return</b> statement is | |
1022 | |
1023 <pre> | |
1024 stat ::= <b>return</b> [explist] [‘<b>;</b>’] | |
1025 </pre> | |
1026 | |
1027 | |
1028 | |
1029 | |
1030 <h4 heading><a name="for" href="#for">For Statement</a></h4> | |
1031 | |
1032 <p> | |
1033 The <b>for</b> statement works over functions, | |
1034 called <em>iterators</em>. | |
1035 On each iteration, the iterator function is called to produce a new value, | |
1036 stopping when this new value is <b>nil</b>. | |
1037 The <b>for</b> loop has the following syntax: | |
1038 | |
1039 <pre> | |
1040 stat ::= <b>for</b> namelist <b>in</b> exp <b>do</b> block end_for | |
1041 namelist ::= Name {‘<b>,</b>’ Name} | |
1042 end_for ::= <b>end_for</b> | <b>end</b> | |
1043 </pre> | |
1044 | |
1045 <p> | |
1046 A <b>for</b> statement like | |
1047 | |
1048 <pre> | |
1049 for <em>var_1</em>, ···, <em>var_n</em> in <em>exp</em> do <em>block</em> end | |
1050 </pre> | |
1051 | |
1052 <p> | |
1053 is equivalent to the code: | |
1054 | |
1055 <pre> | |
1056 do | |
1057 local <em>f</em> = <em>exp</em> | |
1058 while true do | |
1059 local <em>var_1</em>, ···, <em>var_n</em> = <em>f</em>() | |
1060 if <em>var_1</em> == nil then break end | |
1061 <em>block</em> | |
1062 end | |
1063 end | |
1064 </pre> | |
1065 | |
1066 <p> | |
1067 Note the following: | |
1068 | |
1069 <ul> | |
1070 | |
1071 <li> | |
1072 <code><em>exp</em></code> is evaluated only once. | |
1073 Its result is an <em>iterator</em> function. | |
1074 </li> | |
1075 | |
1076 <li> | |
1077 <code><em>f</em></code> is an invisible variable. | |
1078 The name is here for explanatory purposes only. | |
1079 </li> | |
1080 | |
1081 <li> | |
1082 You can use <b>break</b> to exit a <b>for</b> loop. | |
1083 </li> | |
1084 | |
1085 <li> | |
1086 The loop variables <code><em>var_i</em></code> are local to the loop; | |
1087 you cannot use their values after the <b>for</b> ends. | |
1088 If you need these values, | |
1089 then assign them to other variables before breaking or exiting the loop. | |
1090 </li> | |
1091 | |
1092 </ul> | |
1093 | |
1094 | |
1095 | |
1096 <h4 heading><a name="try" href="#for">Try Statement</a></h4> | |
1097 | |
1098 <p>The <b>try</b> statement has the same semantics as in Java.</p> | |
1099 | |
1100 <pre> | |
1101 stat ::= <b>try</b> block [<b>catch</b> Name block] [<b>finally</b> block] end_try | |
1102 end_try ::= <b>end_try</b> | <b>end</b> | |
1103 </pre> | |
1104 | |
1105 | |
1106 | |
1107 | |
1108 <h4 heading><a name="fn_stmt" href="#fn_stmt">Function Calls as Statements</a></h4> | |
1109 | |
1110 <p> | |
1111 To allow possible side-effects, | |
1112 function calls can be executed as statements: | |
1113 | |
1114 <pre> | |
1115 stat ::= functioncall | |
1116 </pre> | |
1117 | |
1118 <p> | |
1119 In this case, all returned values are thrown away. | |
1120 Function calls are explained in <a href="#fn_calls">Function Calls</a>. | |
1121 | |
1122 | |
1123 | |
1124 <h4 heading><a name="local_stmt" href="#local_stmt">Local Declarations</a></h4> | |
1125 | |
1126 <p> | |
1127 Local variables can be declared anywhere inside a block. | |
1128 The declaration can include an initial assignment: | |
1129 | |
1130 <pre> | |
1131 stat ::= <b>local</b> namelist [‘<b>=</b>’ explist] | |
1132 </pre> | |
1133 | |
1134 <p> | |
1135 If present, an initial assignment has the same semantics | |
1136 of a multiple assignment (see <a href="#assignment">Assignment</a>). | |
1137 Otherwise, all variables are initialized with <b>nil</b>. | |
1138 | |
1139 | |
1140 <p> | |
1141 A chunk is also a block (see <a href="#chunks">Chunks</a>), | |
1142 and so local variables can be declared in a chunk outside any explicit block. | |
1143 | |
1144 | |
1145 <p> | |
1146 The visibility rules for local variables are explained in <a href="#visibility">Visibility Rules</a>. | |
1147 | |
1148 | |
1149 <h4 heading><a name="template_stmt" href="#template_stmt">Template Statements</a></h4> | |
1150 | |
1151 <p>Template statements provide the full equivalent of <a href="http://en.wikipedia.org/wiki/JavaServer_Pages">JSP</a> but in a general way. Template statements write to standard output. For example:</p> | |
1152 | |
1153 <pre> | |
1154 local name = "Bob" | |
1155 %> | |
1156 Hello <%= name %>! | |
1157 Bye <%= name %>. | |
1158 <% | |
1159 </pre> | |
1160 | |
1161 <p>is equivalent to the code:</p> | |
1162 | |
1163 <pre> | |
1164 local name = "Bob" | |
1165 require("luan:Io.luan").stdout.write( "Hello ", name , "!\nBye ", name , ".\n" ) | |
1166 </pre> | |
1167 | |
1168 | |
1169 | |
1170 <h3 heading><a name="expressions" href="#expressions">Expressions</a></h3> | |
1171 | |
1172 <p> | |
1173 The basic expressions in Luan are the following: | |
1174 | |
1175 <pre> | |
1176 exp ::= prefixexp | |
1177 exp ::= <b>nil</b> | <b>false</b> | <b>true</b> | |
1178 exp ::= Numeral | |
1179 exp ::= LiteralString | |
1180 exp ::= functiondef | |
1181 exp ::= tableconstructor | |
1182 exp ::= ‘<b>...</b>’ | |
1183 exp ::= exp binop exp | |
1184 exp ::= unop exp | |
1185 prefixexp ::= var | functioncall | ‘<b>(</b>’ exp ‘<b>)</b>’ | |
1186 </pre> | |
1187 | |
1188 <p> | |
1189 Numerals and literal strings are explained in <a href="#lex">Lexical Conventions</a>; | |
1190 variables are explained in <a href="#vars">Variables</a>; | |
1191 function definitions are explained in <a href="#fn_def">Function Definitions</a>; | |
1192 function calls are explained in <a href="#fn_calls">Function Calls</a>; | |
1193 table constructors are explained in <a href="#constructors">Table Constructors</a>. | |
1194 Vararg expressions, | |
1195 denoted by three dots ('<code>...</code>'), can only be used when | |
1196 directly inside a vararg function; | |
1197 they are explained in <a href="#fn_def">Function Definitions</a>. | |
1198 | |
1199 | |
1200 <p> | |
1201 Binary operators comprise arithmetic operators (see <a href="#arithmetic">Arithmetic Operators</a>), | |
1202 relational operators (see <a href="#relational">Relational Operators</a>), logical operators (see <a href="#logical_ops">Logical Operators</a>), | |
1203 and the concatenation operator (see <a href="#concatenation">Concatenation</a>). | |
1204 Unary operators comprise the unary minus (see <a href="#arithmetic">Arithmetic Operators</a>), | |
1205 the unary logical <b>not</b> (see <a href="#logical_ops">Logical Operators</a>), | |
1206 and the unary <em>length operator</em> (see <a href="#length">The Length Operator</a>). | |
1207 | |
1208 | |
1209 <p> | |
1210 Both function calls and vararg expressions can result in multiple values. | |
1211 If a function call is used as a statement (see <a href="#fn_stmt">Function Calls as Statements</a>), | |
1212 then its return list is adjusted to zero elements, | |
1213 thus discarding all returned values. | |
1214 If an expression is used as the last (or the only) element | |
1215 of a list of expressions, | |
1216 then no adjustment is made | |
1217 (unless the expression is enclosed in parentheses). | |
1218 In all other contexts, | |
1219 Luan adjusts the result list to one element, | |
1220 either discarding all values except the first one | |
1221 or adding a single <b>nil</b> if there are no values. | |
1222 | |
1223 | |
1224 <p> | |
1225 Here are some examples: | |
1226 | |
1227 <pre> | |
1228 f() -- adjusted to 0 results | |
1229 g(f(), x) -- f() is adjusted to 1 result | |
1230 g(x, f()) -- g gets x plus all results from f() | |
1231 a,b,c = f(), x -- f() is adjusted to 1 result (c gets nil) | |
1232 a,b = ... -- a gets the first vararg parameter, b gets | |
1233 -- the second (both a and b can get nil if there | |
1234 -- is no corresponding vararg parameter) | |
1235 | |
1236 a,b,c = x, f() -- f() is adjusted to 2 results | |
1237 a,b,c = f() -- f() is adjusted to 3 results | |
1238 return f() -- returns all results from f() | |
1239 return ... -- returns all received vararg parameters | |
1240 return x,y,f() -- returns x, y, and all results from f() | |
1241 {f()} -- creates a list with all results from f() | |
1242 {...} -- creates a list with all vararg parameters | |
1243 {f(), nil} -- f() is adjusted to 1 result | |
1244 </pre> | |
1245 | |
1246 <p> | |
1247 Any expression enclosed in parentheses always results in only one value. | |
1248 Thus, | |
1249 <code>(f(x,y,z))</code> is always a single value, | |
1250 even if <code>f</code> returns several values. | |
1251 (The value of <code>(f(x,y,z))</code> is the first value returned by <code>f</code> | |
1252 or <b>nil</b> if <code>f</code> does not return any values.) | |
1253 | |
1254 | |
1255 | |
1256 <h4 heading><a name="arithmetic" href="#arithmetic">Arithmetic Operators</a></h4> | |
1257 | |
1258 <p> | |
1259 Luan supports the following arithmetic operators: | |
1260 | |
1261 <ul> | |
1262 <li><b><code>+</code>: </b>addition</li> | |
1263 <li><b><code>-</code>: </b>subtraction</li> | |
1264 <li><b><code>*</code>: </b>multiplication</li> | |
1265 <li><b><code>/</code>: </b>division</li> | |
1266 <li><b><code>%</code>: </b>modulo</li> | |
1267 <li><b><code>^</code>: </b>exponentiation</li> | |
1268 <li><b><code>-</code>: </b>unary minus</li> | |
1269 </ul> | |
1270 | |
1271 <p> | |
1272 Addition, subtraction, multiplication, division, and unary minus are the same as these operators in Java. Exponentiation uses Java's <a href="http://docs.oracle.com/javase/7/docs/api/java/lang/Math.html#pow(double,%20double)">Math.pow</a> function. | |
1273 | |
1274 <p> | |
1275 Modulo is defined as the remainder of a division | |
1276 that rounds the quotient towards minus infinite (floor division). | |
1277 (The Java modulo operator is not used.) | |
1278 | |
1279 | |
1280 | |
1281 <h4 heading><a name="conversions" href="#conversions">Coercions and Conversions</a></h4> | |
1282 | |
1283 <p> | |
1284 Luan generally avoids automatic conversions. | |
1285 String concatenation automatically converts all of its arguments to strings. | |
1286 | |
1287 <p> | |
1288 Luan provides library functions for explicit type conversions. | |
1289 | |
1290 | |
1291 | |
1292 | |
1293 <h4 heading><a name="relational" href="#relational">Relational Operators</a></h4> | |
1294 | |
1295 <p> | |
1296 Luan supports the following relational operators: | |
1297 | |
1298 <ul> | |
1299 <li><b><code>==</code>: </b>equality</li> | |
1300 <li><b><code>~=</code>: </b>inequality</li> | |
1301 <li><b><code><</code>: </b>less than</li> | |
1302 <li><b><code>></code>: </b>greater than</li> | |
1303 <li><b><code><=</code>: </b>less or equal</li> | |
1304 <li><b><code>>=</code>: </b>greater or equal</li> | |
1305 </ul><p> | |
1306 These operators always result in <b>false</b> or <b>true</b>. | |
1307 | |
1308 | |
1309 <p> | |
1310 Equality (<code>==</code>) first compares the type of its operands. | |
1311 If the types are different, then the result is <b>false</b>. | |
1312 Otherwise, the values of the operands are compared. | |
1313 Strings, numbers, and binary values are compared in the obvious way (by value). | |
1314 | |
1315 <p> | |
1316 Tables | |
1317 are compared by reference: | |
1318 two objects are considered equal only if they are the same object. | |
1319 Every time you create a new table, | |
1320 it is different from any previously existing table. | |
1321 Closures are also compared by reference. | |
1322 | |
1323 <p> | |
1324 You can change the way that Luan compares tables | |
1325 by using the "eq" metamethod (see <a href="#meta">Metatables and Metamethods</a>). | |
1326 | |
1327 <p> | |
1328 Java values are compared for equality with the Java <a href="http://docs.oracle.com/javase/7/docs/api/java/lang/Object.html#equals(java.lang.Object)"><code>equals</code></a> method. | |
1329 | |
1330 <p> | |
1331 Equality comparisons do not convert strings to numbers | |
1332 or vice versa. | |
1333 Thus, <code>"0"==0</code> evaluates to <b>false</b>, | |
1334 and <code>t[0]</code> and <code>t["0"]</code> denote different | |
1335 entries in a table. | |
1336 | |
1337 | |
1338 <p> | |
1339 The operator <code>~=</code> is exactly the negation of equality (<code>==</code>). | |
1340 | |
1341 | |
1342 <p> | |
1343 The order operators work as follows. | |
1344 | |
1345 If both arguments are numbers, | |
1346 then they are compared following | |
1347 the usual rule for binary operations. | |
1348 Otherwise, if both arguments are strings, | |
1349 then their values are compared according to the current locale. | |
1350 Otherwise, Luan tries to call the "lt" or the "le" | |
1351 metamethod (see <a href="#meta">Metatables and Metamethods</a>). | |
1352 A comparison <code>a > b</code> is translated to <code>b < a</code> | |
1353 and <code>a >= b</code> is translated to <code>b <= a</code>. | |
1354 | |
1355 | |
1356 | |
1357 | |
1358 | |
1359 <h4 heading><a name="logical_ops" href="#logical_ops">Logical Operators</a></h4> | |
1360 | |
1361 <p> | |
1362 The logical operators in Luan are | |
1363 <b>and</b>, <b>or</b>, and <b>not</b>. | |
1364 The <b>and</b> and <b>or</b> operators consider both <b>false</b> and <b>nil</b> as false | |
1365 and anything else as true. | |
1366 Like the control structures (see <a href="#control">Control Structures</a>), | |
1367 the <b>not</b> operator requires a boolean value. | |
1368 | |
1369 <p> | |
1370 The negation operator <b>not</b> always returns <b>false</b> or <b>true</b>. | |
1371 The conjunction operator <b>and</b> returns its first argument | |
1372 if this value is <b>false</b> or <b>nil</b>; | |
1373 otherwise, <b>and</b> returns its second argument. | |
1374 The disjunction operator <b>or</b> returns its first argument | |
1375 if this value is different from <b>nil</b> and <b>false</b>; | |
1376 otherwise, <b>or</b> returns its second argument. | |
1377 Both <b>and</b> and <b>or</b> use short-circuit evaluation; | |
1378 that is, | |
1379 the second operand is evaluated only if necessary. | |
1380 Here are some examples: | |
1381 | |
1382 <pre> | |
1383 10 or 20 --> 10 | |
1384 10 or error() --> 10 | |
1385 nil or "a" --> "a" | |
1386 nil and 10 --> nil | |
1387 false and error() --> false | |
1388 false and nil --> false | |
1389 false or nil --> nil | |
1390 10 and 20 --> 20 | |
1391 </pre> | |
1392 | |
1393 <p> | |
1394 (In this manual, | |
1395 <code>--></code> indicates the result of the preceding expression.) | |
1396 | |
1397 | |
1398 | |
1399 <h4 heading><a name="concatenation" href="#concatenation">Concatenation</a></h4> | |
1400 | |
1401 <p> | |
1402 The string concatenation operator in Luan is | |
1403 denoted by two dots ('<code>..</code>'). | |
1404 All operands are converted to strings. | |
1405 | |
1406 | |
1407 | |
1408 <h4 heading><a name="length" href="#length">The Length Operator</a></h4> | |
1409 | |
1410 <p> | |
1411 The length operator is denoted by the unary prefix operator <code>#</code>. | |
1412 The length of a string is its number of characters. | |
1413 The length of a binary is its number of bytes. | |
1414 | |
1415 | |
1416 <p> | |
1417 A program can modify the behavior of the length operator for | |
1418 any table through the <code>__len</code> metamethod (see <a href="#meta">Metatables and Metamethods</a>). | |
1419 | |
1420 | |
1421 <p> | |
1422 Unless a <code>__len</code> metamethod is given, | |
1423 the length of a table <code>t</code> is defined | |
1424 as the number of elements in <em>sequence</em>, | |
1425 that is, | |
1426 the size of the set of its positive numeric keys is equal to <em>{1..n}</em> | |
1427 for some non-negative integer <em>n</em>. | |
1428 In that case, <em>n</em> is its length. | |
1429 Note that a table like | |
1430 | |
1431 <pre> | |
1432 {10, 20, nil, 40} | |
1433 </pre> | |
1434 | |
1435 <p> | |
1436 has a length of <code>2</code>, because that is the last key in sequence. | |
1437 | |
1438 | |
1439 | |
1440 | |
1441 | |
1442 <h4 heading><a name="precedence" href="#precedence">Precedence</a></h4> | |
1443 | |
1444 <p> | |
1445 Operator precedence in Luan follows the table below, | |
1446 from lower to higher priority: | |
1447 | |
1448 <pre> | |
1449 or | |
1450 and | |
1451 < > <= >= ~= == | |
1452 .. | |
1453 + - | |
1454 * / % | |
1455 unary operators (not # -) | |
1456 ^ | |
1457 </pre> | |
1458 | |
1459 <p> | |
1460 As usual, | |
1461 you can use parentheses to change the precedences of an expression. | |
1462 The concatenation ('<code>..</code>') and exponentiation ('<code>^</code>') | |
1463 operators are right associative. | |
1464 All other binary operators are left associative. | |
1465 | |
1466 | |
1467 | |
1468 | |
1469 | |
1470 <h4 heading><a name="constructors" href="#constructors">Table Constructors</a></h4> | |
1471 | |
1472 <p> | |
1473 Table constructors are expressions that create tables. | |
1474 Every time a constructor is evaluated, a new table is created. | |
1475 A constructor can be used to create an empty table | |
1476 or to create a table and initialize some of its fields. | |
1477 The general syntax for constructors is | |
1478 | |
1479 <pre> | |
1480 tableconstructor ::= ‘<b>{</b>’ fieldlist ‘<b>}</b>’ | |
1481 fieldlist ::= [field] {fieldsep [field]} | |
1482 field ::= ‘<b>[</b>’ exp ‘<b>]</b>’ ‘<b>=</b>’ exp | Name ‘<b>=</b>’ exp | exp | |
1483 fieldsep ::= ‘<b>,</b>’ | ‘<b>;</b>’ | <b>end_of_line</b> | |
1484 </pre> | |
1485 | |
1486 <p> | |
1487 Each field of the form <code>[exp1] = exp2</code> adds to the new table an entry | |
1488 with key <code>exp1</code> and value <code>exp2</code>. | |
1489 A field of the form <code>name = exp</code> is equivalent to | |
1490 <code>["name"] = exp</code>. | |
1491 Finally, fields of the form <code>exp</code> are equivalent to | |
1492 <code>[i] = exp</code>, where <code>i</code> are consecutive integers | |
1493 starting with 1. | |
1494 Fields in the other formats do not affect this counting. | |
1495 For example, | |
1496 | |
1497 <pre> | |
1498 a = { [f(1)] = g; "x", "y"; x = 1, f(x), [30] = 23; 45 } | |
1499 </pre> | |
1500 | |
1501 <p> | |
1502 is equivalent to | |
1503 | |
1504 <pre> | |
1505 do | |
1506 local t = {} | |
1507 t[f(1)] = g | |
1508 t[1] = "x" -- 1st exp | |
1509 t[2] = "y" -- 2nd exp | |
1510 t.x = 1 -- t["x"] = 1 | |
1511 t[3] = f(x) -- 3rd exp | |
1512 t[30] = 23 | |
1513 t[4] = 45 -- 4th exp | |
1514 a = t | |
1515 end | |
1516 </pre> | |
1517 | |
1518 <p> | |
1519 The order of the assignments in a constructor is undefined. | |
1520 (This order would be relevant only when there are repeated keys.) | |
1521 | |
1522 | |
1523 <p> | |
1524 If the last field in the list has the form <code>exp</code> | |
1525 and the expression is a function call or a vararg expression, | |
1526 then all values returned by this expression enter the list consecutively | |
1527 (see <a href="#fn_calls">Function Calls</a>). | |
1528 | |
1529 | |
1530 <p> | |
1531 The field list can have an optional trailing separator, | |
1532 as a convenience for machine-generated code. | |
1533 | |
1534 | |
1535 | |
1536 | |
1537 | |
1538 <h4 heading><a name="fn_calls" href="#fn_calls">Function Calls</a></h4> | |
1539 | |
1540 <p> | |
1541 A function call in Luan has the following syntax: | |
1542 | |
1543 <pre> | |
1544 functioncall ::= prefixexp args | |
1545 </pre> | |
1546 | |
1547 <p> | |
1548 In a function call, | |
1549 first prefixexp and args are evaluated. | |
1550 The value of prefixexp must have type <em>function</em>. | |
1551 This function is called | |
1552 with the given arguments. | |
1553 | |
1554 | |
1555 <p> | |
1556 Arguments have the following syntax: | |
1557 | |
1558 <pre> | |
1559 args ::= ‘<b>(</b>’ [explist] ‘<b>)</b>’ | |
1560 args ::= tableconstructor | |
1561 args ::= LiteralString | |
1562 </pre> | |
1563 | |
1564 <p> | |
1565 All argument expressions are evaluated before the call. | |
1566 A call of the form <code>f{<em>fields</em>}</code> is | |
1567 syntactic sugar for <code>f({<em>fields</em>})</code>; | |
1568 that is, the argument list is a single new table. | |
1569 A call of the form <code>f'<em>string</em>'</code> | |
1570 (or <code>f"<em>string</em>"</code> or <code>f[[<em>string</em>]]</code>) | |
1571 is syntactic sugar for <code>f('<em>string</em>')</code>; | |
1572 that is, the argument list is a single literal string. | |
1573 | |
1574 | |
1575 | |
1576 | |
1577 <h4 heading><a name="fn_def" href="#fn_def">Function Definitions</a></h4> | |
1578 | |
1579 <p> | |
1580 The syntax for function definition is | |
1581 | |
1582 <pre> | |
1583 functiondef ::= <b>function</b> funcbody | |
1584 funcbody ::= ‘<b>(</b>’ [parlist] ‘<b>)</b>’ block end_function | |
1585 end_function ::= <b>end_function</b> | <b>end</b> | |
1586 </pre> | |
1587 | |
1588 <p> | |
1589 The following syntactic sugar simplifies function definitions: | |
1590 | |
1591 <pre> | |
1592 stat ::= <b>function</b> funcname funcbody | |
1593 stat ::= <b>local</b> <b>function</b> Name funcbody | |
1594 funcname ::= Name {‘<b>.</b>’ Name} [‘<b>:</b>’ Name] | |
1595 </pre> | |
1596 | |
1597 <p> | |
1598 The statement | |
1599 | |
1600 <pre> | |
1601 function f () <em>body</em> end | |
1602 </pre> | |
1603 | |
1604 <p> | |
1605 translates to | |
1606 | |
1607 <pre> | |
1608 f = function () <em>body</em> end | |
1609 </pre> | |
1610 | |
1611 <p> | |
1612 The statement | |
1613 | |
1614 <pre> | |
1615 function t.a.b.c.f () <em>body</em> end | |
1616 </pre> | |
1617 | |
1618 <p> | |
1619 translates to | |
1620 | |
1621 <pre> | |
1622 t.a.b.c.f = function () <em>body</em> end | |
1623 </pre> | |
1624 | |
1625 <p> | |
1626 The statement | |
1627 | |
1628 <pre> | |
1629 local function f () <em>body</em> end | |
1630 </pre> | |
1631 | |
1632 <p> | |
1633 translates to | |
1634 | |
1635 <pre> | |
1636 local f; f = function () <em>body</em> end | |
1637 </pre> | |
1638 | |
1639 <p> | |
1640 not to | |
1641 | |
1642 <pre> | |
1643 local f = function () <em>body</em> end | |
1644 </pre> | |
1645 | |
1646 <p> | |
1647 (This only makes a difference when the body of the function | |
1648 contains references to <code>f</code>.) | |
1649 | |
1650 | |
1651 <p> | |
1652 A function definition is an executable expression, | |
1653 whose value has type <em>function</em>. | |
1654 When Luan precompiles a chunk, | |
1655 all its function bodies are precompiled too. | |
1656 Then, whenever Luan executes the function definition, | |
1657 the function is <em>instantiated</em> (or <em>closed</em>). | |
1658 This function instance (or <em>closure</em>) | |
1659 is the final value of the expression. | |
1660 | |
1661 | |
1662 <p> | |
1663 Parameters act as local variables that are | |
1664 initialized with the argument values: | |
1665 | |
1666 <pre> | |
1667 parlist ::= namelist [‘<b>,</b>’ ‘<b>...</b>’] | ‘<b>...</b>’ | |
1668 </pre> | |
1669 | |
1670 <p> | |
1671 When a function is called, | |
1672 the list of arguments is adjusted to | |
1673 the length of the list of parameters if the list is too short, | |
1674 unless the function is a <em>vararg function</em>, | |
1675 which is indicated by three dots ('<code>...</code>') | |
1676 at the end of its parameter list. | |
1677 A vararg function does not adjust its argument list; | |
1678 instead, it collects all extra arguments and supplies them | |
1679 to the function through a <em>vararg expression</em>, | |
1680 which is also written as three dots. | |
1681 The value of this expression is a list of all actual extra arguments, | |
1682 similar to a function with multiple results. | |
1683 If a vararg expression is used inside another expression | |
1684 or in the middle of a list of expressions, | |
1685 then its return list is adjusted to one element. | |
1686 If the expression is used as the last element of a list of expressions, | |
1687 then no adjustment is made | |
1688 (unless that last expression is enclosed in parentheses). | |
1689 | |
1690 | |
1691 <p> | |
1692 As an example, consider the following definitions: | |
1693 | |
1694 <pre> | |
1695 function f(a, b) end | |
1696 function g(a, b, ...) end | |
1697 function r() return 1,2,3 end | |
1698 </pre> | |
1699 | |
1700 <p> | |
1701 Then, we have the following mapping from arguments to parameters and | |
1702 to the vararg expression: | |
1703 | |
1704 <pre> | |
1705 CALL PARAMETERS | |
1706 | |
1707 f(3) a=3, b=nil | |
1708 f(3, 4) a=3, b=4 | |
1709 f(3, 4, 5) runtime error | |
1710 f(r(), 10) runtime error | |
1711 f(r()) runtime error | |
1712 | |
1713 g(3) a=3, b=nil, ... --> (nothing) | |
1714 g(3, 4) a=3, b=4, ... --> (nothing) | |
1715 g(3, 4, 5, 8) a=3, b=4, ... --> 5 8 | |
1716 g(5, r()) a=5, b=1, ... --> 2 3 | |
1717 </pre> | |
1718 | |
1719 <p> | |
1720 Results are returned using the <b>return</b> statement (see <a href="#control">Control Structures</a>). | |
1721 If control reaches the end of a function | |
1722 without encountering a <b>return</b> statement, | |
1723 then the function returns with no results. | |
1724 | |
1725 | |
1726 <h3 heading><a name="visibility" href="#visibility">Visibility Rules</a></h3> | |
1727 | |
1728 <p> | |
1729 Luan is a lexically scoped language. | |
1730 The scope of a local variable begins at the first statement after | |
1731 its declaration and lasts until the last non-void statement | |
1732 of the innermost block that includes the declaration. | |
1733 Consider the following example: | |
1734 | |
1735 <pre> | |
1736 x = 10 -- global variable | |
1737 do -- new block | |
1738 local x = x -- new 'x', with value 10 | |
1739 print(x) --> 10 | |
1740 x = x+1 | |
1741 do -- another block | |
1742 local x = x+1 -- another 'x' | |
1743 print(x) --> 12 | |
1744 end | |
1745 print(x) --> 11 | |
1746 end | |
1747 print(x) --> 10 (the global one) | |
1748 </pre> | |
1749 | |
1750 <p> | |
1751 Notice that, in a declaration like <code>local x = x</code>, | |
1752 the new <code>x</code> being declared is not in scope yet, | |
1753 and so the second <code>x</code> refers to the outside variable. | |
1754 | |
1755 | |
1756 <p> | |
1757 Because of the lexical scoping rules, | |
1758 local variables can be freely accessed by functions | |
1759 defined inside their scope. | |
1760 A local variable used by an inner function is called | |
1761 an <em>upvalue</em>, or <em>external local variable</em>, | |
1762 inside the inner function. | |
1763 | |
1764 | |
1765 <p> | |
1766 Notice that each execution of a <b>local</b> statement | |
1767 defines new local variables. | |
1768 Consider the following example: | |
1769 | |
1770 <pre> | |
1771 a = {} | |
1772 local x = 20 | |
1773 for i=1,10 do | |
1774 local y = 0 | |
1775 a[i] = function () y=y+1; return x+y end | |
1776 end | |
1777 </pre> | |
1778 | |
1779 <p> | |
1780 The loop creates ten closures | |
1781 (that is, ten instances of the anonymous function). | |
1782 Each of these closures uses a different <code>y</code> variable, | |
1783 while all of them share the same <code>x</code>. | |
1784 | |
1785 | |
1786 | |
1787 | |
1788 | |
1789 <h2 heading><a name="libs" href="#libs">Standard Libraries</a></h2> | |
1790 | |
1791 <p> | |
1792 The standard Luan libraries provide useful functions | |
1793 that are implemented both in Java and in Luan itself. | |
1794 How each function is implemented shouldn't matter to the user. | |
1795 Some of these functions provide essential services to the language | |
1796 (e.g., <a href="#Luan.type"><code>type</code></a> and <a href="#Luan.get_metatable"><code>get_metatable</code></a>); | |
1797 others provide access to "outside" services (e.g., I/O). | |
1798 | |
1799 | |
1800 <h3 heading><a name="default_lib" href="#default_lib">Default Environment</a></h3> | |
1801 | |
1802 <p> | |
1803 This is provided by default as a local variable for any Luan code as described in <a href="#env">Environments</a>. | |
1804 | |
1805 | |
1806 <h4 heading><a name="require" href="#require"><code>require (mod_uri)</code></a></h4> | |
1807 | |
1808 <p> | |
1809 Example use: | |
1810 | |
1811 <pre> | |
1812 local Table = require "luan:Table.luan" | |
1813 </pre> | |
1814 | |
1815 <p> | |
1816 Could be defined as: | |
1817 | |
1818 <pre> | |
1819 local function require(mod_name) | |
1820 return <a href="#Package.load">Package.load</a>(mod_name) or <a href="#Luan.error">Luan.error</a>("module '"..mod_name.."' not found") | |
1821 end | |
1822 </pre> | |
1823 | |
1824 <p> | |
1825 A special case is: | |
1826 | |
1827 <pre> | |
1828 require "java" | |
1829 </pre> | |
1830 | |
1831 <p> | |
1832 This enables Java in the current chunk if that chunk has permission to use Java. If the chunk doesn't have permission to use Java, then an error is thrown. | |
1833 | |
1834 | |
1835 <h3 heading><a name="luan_lib" href="#luan_lib">Basic Functions</a></h3> | |
1836 | |
1837 <p> | |
1838 Include this library by: | |
1839 | |
1840 <pre> | |
1841 local Luan = require "luan:Luan.luan" | |
1842 </pre> | |
1843 | |
1844 <p> | |
1845 The basic library provides basic functions to Luan that don't depend on other libaries. | |
1846 | |
1847 | |
1848 <h4 heading><a name="Luan.do_file" href="#Luan.do_file"><code>Luan.do_file ([uri])</code></a></h4> | |
1849 | |
1850 <p> | |
1851 Could be defined as: | |
1852 | |
1853 <pre> | |
1854 function Luan.do_file(uri) | |
1855 local fn = <a href="#Luan.load_file">Luan.load_file</a>(uri) or <a href="#Luan.error">Luan.error</a>("file '"..uri.."' not found") | |
1856 return fn() | |
1857 end | |
1858 </pre> | |
1859 | |
1860 | |
1861 | |
1862 <h4 heading><a name="Luan.error" href="#Luan.error"><code>Luan.error (message)</code></a></h4> | |
1863 | |
1864 <p> | |
1865 Throws an error containing the message. | |
1866 | |
1867 <p> | |
1868 Could be defined as: | |
1869 | |
1870 <pre> | |
1871 function Luan.error(message) | |
1872 <a href="#Luan.new_error">Luan.new_error</a>(message).throw() | |
1873 end | |
1874 </pre> | |
1875 | |
1876 | |
1877 | |
1878 <h4 heading><a name="Luan.eval" href="#Luan.eval"><code>Luan.eval (text [, source_name [, env]])</code></a></h4> | |
1879 | |
1880 <p> | |
1881 Evaluates <code>text</code> as a Luan expression. | |
1882 | |
1883 <p> | |
1884 Could be defined as: | |
1885 | |
1886 <pre> | |
1887 function Luan.eval(text,source_name, env) | |
1888 return <a href="#Luan.load">Luan.load</a>( "return "..text, source_name or "eval", env )() | |
1889 end | |
1890 </pre> | |
1891 | |
1892 | |
1893 | |
1894 <h4 heading><a name="Luan.get_metatable" href="#Luan.get_metatable"><code>Luan.get_metatable (table)</code></a></h4> | |
1895 | |
1896 <p> | |
1897 If <code>table</code> does not have a metatable, returns <b>nil</b>. | |
1898 Otherwise, | |
1899 if the table's metatable has a <code>"__metatable"</code> field, | |
1900 returns the associated value. | |
1901 Otherwise, returns the metatable of the given table. | |
1902 | |
1903 | |
1904 <h4 heading><a name="Luan.hash_code" href="#Luan.ipairs"><code>Luan.hash_code (v)</code></a></h4> | |
1905 | |
1906 <p> | |
1907 Returns the hash code of <code>v</code>. | |
1908 | |
1909 | |
1910 <h4 heading><a name="Luan.ipairs" href="#Luan.ipairs"><code>Luan.ipairs (t)</code></a></h4> | |
1911 | |
1912 <p> | |
1913 Returns an iterator function | |
1914 so that the construction | |
1915 | |
1916 <pre> | |
1917 for i,v in ipairs(t) do <em>body</em> end | |
1918 </pre> | |
1919 | |
1920 <p> | |
1921 will iterate over the key–value pairs | |
1922 (<code>1,t[1]</code>), (<code>2,t[2]</code>), ..., | |
1923 up to the first nil value. | |
1924 | |
1925 <p> | |
1926 Could be defined as: | |
1927 | |
1928 <pre> | |
1929 function Luan.ipairs(t) | |
1930 local i = 0 | |
1931 return function() | |
1932 if i < #t then | |
1933 i = i + 1 | |
1934 return i, t[i] | |
1935 end | |
1936 end | |
1937 end | |
1938 </pre> | |
1939 | |
1940 | |
1941 | |
1942 <h4 heading><a name="Luan.load" href="#Luan.load"><code>Luan.load (text, [source_name [, env [, persist]]])</code></a></h4> | |
1943 | |
1944 <p> | |
1945 Loads a chunk. | |
1946 | |
1947 <p> | |
1948 The <code>text</code> is compiled. | |
1949 If there are no syntactic errors, | |
1950 returns the compiled chunk as a function; | |
1951 otherwise, throws an error. | |
1952 | |
1953 <p> | |
1954 The <code>source_name</code> parameter is a string saying where the text came from. It is used to produce error messages. Defaults to "load". | |
1955 | |
1956 <p> | |
1957 If the <code>env</code> parameter is supplied, it becomes the <code>_ENV</code> of the chunk. | |
1958 | |
1959 <p> | |
1960 The <code>persist</code> parameter is a boolean which determines if the compiled code is persistently cached to a temporary file. Defaults to <code>false</code>. | |
1961 | |
1962 | |
1963 <h4 heading><a name="Luan.load_file" href="#Luan.load_file"><code>Luan.load_file (file_uri)</code></a></h4> | |
1964 | |
1965 <p> | |
1966 Similar to <a href="#Luan.load"><code>load</code></a>, | |
1967 but gets the chunk from file <code>file_uri</code>. | |
1968 <code>file_uri</code> can be a string or a uri table. | |
1969 | |
1970 | |
1971 <h4 heading><a name="Luan.new_error" href="#Luan.new_error"><code>Luan.new_error (message)</code></a></h4> | |
1972 | |
1973 <p> | |
1974 Creates a new error table containing the message assigned to "<code>message</code>". The error table also contains a <code>throw</code> function which throws the error. The table also contains a list of stack trace elements where each stack trace element is a table containing "<code>source</code>", "<code>line</code>", and possible "<code>call_to</code>". The table also has a metatable containing "<code>__to_string</code>" to render the error. | |
1975 | |
1976 <p> | |
1977 To print the current stack trace, you could do: | |
1978 | |
1979 <pre> | |
1980 Io.print( Luan.new_error "stack" ) | |
1981 </pre> | |
1982 | |
1983 | |
1984 <h4 heading><a name="Luan.pairs" href="#Luan.pairs"><code>Luan.pairs (t)</code></a></h4> | |
1985 | |
1986 <p> | |
1987 If <code>t</code> has a metamethod <code>__pairs</code>, | |
1988 calls it with <code>t</code> as argument and returns the | |
1989 result from the call. | |
1990 | |
1991 | |
1992 <p> | |
1993 Otherwise, | |
1994 returns a function | |
1995 so that the construction | |
1996 | |
1997 <pre> | |
1998 for k,v in pairs(t) do <em>body</em> end | |
1999 </pre> | |
2000 | |
2001 <p> | |
2002 will iterate over all key–value pairs of table <code>t</code>. | |
2003 | |
2004 | |
2005 | |
2006 <p> | |
2007 <hr><h3><a name="pdf-print"><code>print (···)</code></a></h3> | |
2008 Receives any number of arguments | |
2009 and prints their values to <code>stdout</code>, | |
2010 using the <a href="#pdf-tostring"><code>tostring</code></a> function to convert each argument to a string. | |
2011 <code>print</code> is not intended for formatted output, | |
2012 but only as a quick way to show a value, | |
2013 for instance for debugging. | |
2014 For complete control over the output, | |
2015 use <a href="#pdf-string.format"><code>string.format</code></a> and <a href="#pdf-io.write"><code>io.write</code></a>. | |
2016 | |
2017 | |
2018 | |
2019 <h4 heading><a name="Luan.range" href="#Luan.range"><code>Luan.range (start, stop [, step])</code></a></h4> | |
2020 | |
2021 <p> | |
2022 Based on <a href="https://docs.python.org/2/library/functions.html#range">the Python range() function</a>, this lets one iterate through a sequence of numbers. | |
2023 | |
2024 <p> | |
2025 Example use: | |
2026 | |
2027 <pre> | |
2028 for i in range(1,10) do | |
2029 Io.print("count up:",i) | |
2030 end | |
2031 for i in range(10,0,-1) do | |
2032 Io.print("count down:",i) | |
2033 end | |
2034 </pre> | |
2035 | |
2036 <p> | |
2037 Could be defined as: | |
2038 | |
2039 <pre> | |
2040 function Luan.range(start, stop, step) | |
2041 step = step or 1 | |
2042 step == 0 and <a href="#Luan.error">Luan.error</a> "bad argument #3 (step may not be zero)" | |
2043 local i = start | |
2044 return function() | |
2045 if step > 0 and i <= stop or step < 0 and i >= stop then | |
2046 local rtn = i | |
2047 i = i + step | |
2048 return rtn | |
2049 end | |
2050 end | |
2051 end | |
2052 </pre> | |
2053 | |
2054 | |
2055 | |
2056 <h4 heading><a name="Luan.raw_equal" href="#Luan.raw_equal"><code>Luan.raw_equal (v1, v2)</code></a></h4> | |
2057 | |
2058 <p> | |
2059 Checks whether <code>v1</code> is equal to <code>v2</code>, | |
2060 without invoking any metamethod. | |
2061 Returns a boolean. | |
2062 | |
2063 | |
2064 | |
2065 <h4 heading><a name="Luan.raw_get" href="#Luan.raw_get"><code>Luan.raw_get (table, index)</code></a></h4> | |
2066 | |
2067 <p> | |
2068 Gets the real value of <code>table[index]</code>, | |
2069 without invoking any metamethod. | |
2070 <code>table</code> must be a table; | |
2071 <code>index</code> may be any value. | |
2072 | |
2073 | |
2074 | |
2075 <h4 heading><a name="Luan.raw_len" href="#Luan.raw_len"><code>Luan.raw_len (v)</code></a></h4> | |
2076 | |
2077 <p> | |
2078 Returns the length of the object <code>v</code>, | |
2079 which must be a table or a string, | |
2080 without invoking any metamethod. | |
2081 Returns an integer. | |
2082 | |
2083 | |
2084 | |
2085 <h4 heading><a name="Luan.raw_set" href="#Luan.raw_set"><code>Luan.raw_set (table, index, value)</code></a></h4> | |
2086 | |
2087 <p> | |
2088 Sets the real value of <code>table[index]</code> to <code>value</code>, | |
2089 without invoking any metamethod. | |
2090 <code>table</code> must be a table, | |
2091 <code>index</code> any value different from <b>nil</b>, | |
2092 and <code>value</code> any Lua value. | |
2093 | |
2094 | |
2095 <h4 heading><a name="Luan.set_metatable" href="#Luan.set_metatable"><code>Luan.set_metatable (table, metatable)</code></a></h4> | |
2096 | |
2097 <p> | |
2098 Sets the metatable for the given table. | |
2099 If <code>metatable</code> is <b>nil</b>, | |
2100 removes the metatable of the given table. | |
2101 If the original metatable has a <code>"__metatable"</code> field, | |
2102 raises an error. | |
2103 | |
2104 | |
2105 <h4 heading><a name="Luan.stringify" href="#Luan.stringify"><code>Luan.stringify (v [,options])</code></a></h4> | |
2106 | |
2107 <p> | |
2108 Receives a value of any type and converts it to a string that is a Luan expression. <code>options</code> is a table. If <code>options.strict==true</code> then invalid types throw an error. Otherwise invalid types are represented but the resulting expression is invalid. If <code>options.number_types==true</code> then numbers will be wrapped in functions for their type. | |
2109 | |
2110 | |
2111 <h4 heading><a name="Luan.to_string" href="#Luan.to_string"><code>Luan.to_string (v)</code></a></h4> | |
2112 | |
2113 <p> | |
2114 Receives a value of any type and | |
2115 converts it to a string in a human-readable format. | |
2116 | |
2117 <p> | |
2118 If the metatable of <code>v</code> has a <code>"__to_string"</code> field, | |
2119 then <code>to_string</code> calls the corresponding value | |
2120 with <code>v</code> as argument, | |
2121 and uses the result of the call as its result. | |
2122 | |
2123 | |
2124 | |
2125 <h4 heading><a name="Luan.type" href="#Luan.type"><code>Luan.type (v)</code></a></h4> | |
2126 | |
2127 <p> | |
2128 Returns the type of its only argument, coded as a string. | |
2129 The possible results of this function are | |
2130 "<code>nil</code>" (a string, not the value <b>nil</b>), | |
2131 "<code>number</code>", | |
2132 "<code>string</code>", | |
2133 "<code>binary</code>", | |
2134 "<code>boolean</code>", | |
2135 "<code>table</code>", | |
2136 "<code>function</code>", | |
2137 and "<code>java</code>". | |
2138 | |
2139 | |
2140 <h4 heading><a name="Luan.values" href="#Luan.values"><code>Luan.values (···)</code></a></h4> | |
2141 | |
2142 <p> | |
2143 Returns a function so that the construction | |
2144 | |
2145 <pre> | |
2146 for i, v in Luan.values(···) do <em>body</em> end | |
2147 </pre> | |
2148 | |
2149 <p> | |
2150 will iterate over all values of <code>···</code>. | |
2151 | |
2152 | |
2153 | |
2154 <h4 heading><a name="Luan.VERSION" href="#Luan.VERSION"><code>Luan.VERSION</code></a></h4> | |
2155 | |
2156 <p> | |
2157 A global variable (not a function) that | |
2158 holds a string containing the current Luan version. | |
2159 | |
2160 | |
2161 | |
2162 | |
2163 | |
2164 | |
2165 <h3 heading><a name="package_lib" href="#package_lib">Modules</a></h3> | |
2166 | |
2167 <p> | |
2168 Include this library by: | |
2169 | |
2170 <pre> | |
2171 local Package = require "luan:Package.luan" | |
2172 </pre> | |
2173 | |
2174 <p> | |
2175 The package library provides basic | |
2176 facilities for loading modules in Luan. | |
2177 | |
2178 | |
2179 <h4 heading><a name="Package.load" href="#Package.load"><code>Package.load (mod_uri)</code></a></h4> | |
2180 | |
2181 <p> | |
2182 Loads the given module. | |
2183 The function starts by looking into the <a href="#Package.loaded"><code>Package.loaded</code></a> table | |
2184 to determine whether <code>mod_uri</code> is already loaded. | |
2185 If it is, then <code>Package.load</code> returns the value stored | |
2186 at <code>Package.loaded[mod_uri]</code>. | |
2187 Otherwise, it tries to load a new value for the module. | |
2188 | |
2189 <p> | |
2190 To load a new value, <code>Package.load</code> first checks if <code>mod_uri</code> starts with "<b>java:</b>". If yes, then this is a Java class which is loaded by special Java code. | |
2191 | |
2192 <p> | |
2193 Otherwise <code>Package.load</code> tries to read the text of the file referred to by <code>mod_uri</code>. If the file doesn't exist, then <code>Package.load</code> returns <b>false</b>. If the file exists, then its content is compiled into a chunk by calling <a href="#Luan.load"><code>Luan.load</code></a>. This chunk is run passing in <code>mod_uri</code> as an argument. The value returned by the chunk must not be <b>nil</b> and is loaded. | |
2194 | |
2195 <p> | |
2196 If a new value for the module successful loaded, then it is stored in <code>Package.loaded[mod_uri]</code>. The value is returned. | |
2197 | |
2198 | |
2199 | |
2200 | |
2201 <h4 heading><a name="Package.loaded" href="#Package.loaded"><code>Package.loaded</code></a></h4> | |
2202 | |
2203 | |
2204 <p> | |
2205 A table used by <a href="#Package.load"><code>Package.load</code></a> to control which | |
2206 modules are already loaded. | |
2207 When you load a module <code>mod_uri</code> and | |
2208 <code>Package.loaded[mod_uri]</code> is not <b>nil</b>, | |
2209 <a href="#Package.load"><code>Package.load</code></a> simply returns the value stored there. | |
2210 | |
2211 | |
2212 <p> | |
2213 This variable is only a reference to the real table; | |
2214 assignments to this variable do not change the | |
2215 table used by <a href="#Package.load"><code>Package.load</code></a>. | |
2216 | |
2217 | |
2218 | |
2219 | |
2220 | |
2221 | |
2222 <h3 heading><a name="string_lib" href="#string_lib">String Manipulation</a></h3> | |
2223 | |
2224 <p> | |
2225 Include this library by: | |
2226 | |
2227 <pre> | |
2228 local String = require "luan:String.luan" | |
2229 </pre> | |
2230 | |
2231 <p> | |
2232 This library provides generic functions for string manipulation, | |
2233 such as finding and extracting substrings, and pattern matching. | |
2234 When indexing a string in Luan, the first character is at position 1 | |
2235 (not at 0, as in Java). | |
2236 Indices are allowed to be negative and are interpreted as indexing backwards, | |
2237 from the end of the string. | |
2238 Thus, the last character is at position -1, and so on. | |
2239 | |
2240 | |
2241 | |
2242 <h4 heading><a name="String.char" href="#String.char"><code>String.char (···)</code></a></h4> | |
2243 | |
2244 <p> | |
2245 Receives zero or more integers. | |
2246 Returns a string with length equal to the number of arguments, | |
2247 in which each character has the internal numerical code equal | |
2248 to its corresponding argument. | |
2249 | |
2250 | |
2251 <h4 heading><a name="String.encode" href="#String.encode"><code>String.encode (s)</code></a></h4> | |
2252 | |
2253 <p> | |
2254 Encodes argument <code>s</code> into a string that can be placed in quotes so as to return the original value of the string. | |
2255 | |
2256 | |
2257 | |
2258 | |
2259 <h4 heading><a name="String.find" href="#String.find"><code>String.find (s, pattern [, init [, plain]])</code></a></h4> | |
2260 | |
2261 <p> | |
2262 Looks for the first match of | |
2263 <code>pattern</code> (see <a href="http://docs.oracle.com/javase/7/docs/api/java/util/regex/Pattern.html">Pattern</a>) in the string <code>s</code>. | |
2264 If it finds a match, then <code>find</code> returns the indices of <code>s</code> | |
2265 where this occurrence starts and ends; | |
2266 otherwise, it returns <b>nil</b>. | |
2267 A third, optional numerical argument <code>init</code> specifies | |
2268 where to start the search; | |
2269 its default value is 1 and can be negative. | |
2270 A value of <b>true</b> as a fourth, optional argument <code>plain</code> | |
2271 turns off the pattern matching facilities, | |
2272 so the function does a plain "find substring" operation, | |
2273 with no characters in <code>pattern</code> being considered magic. | |
2274 Note that if <code>plain</code> is given, then <code>init</code> must be given as well. | |
2275 | |
2276 <p> | |
2277 If the pattern has captures, | |
2278 then in a successful match | |
2279 the captured values are also returned, | |
2280 after the two indices. | |
2281 | |
2282 | |
2283 | |
2284 | |
2285 <h4 heading><a name="String.format" href="#String.format"><code>String.format (formatstring, ···)</code></a></h4> | |
2286 | |
2287 | |
2288 <p> | |
2289 Returns a formatted version of its variable number of arguments | |
2290 following the description given in its first argument (which must be a string). | |
2291 The format string follows the same rules as the Java function <a href="http://docs.oracle.com/javase/7/docs/api/java/lang/String.html#format(java.lang.String,%20java.lang.Object...)"><code>String.format</code></a> because Luan calls this internally. | |
2292 | |
2293 <p> | |
2294 Note that Java's <code>String.format</code> is too stupid to convert between ints and floats, so you must provide the right kind of number. | |
2295 | |
2296 | |
2297 | |
2298 <h4 heading><a name="String.gmatch" href="#String.gmatch"><code>String.gmatch (s, pattern)</code></a></h4> | |
2299 | |
2300 <p> | |
2301 Returns an iterator function that, | |
2302 each time it is called, | |
2303 returns the next captures from <code>pattern</code> (see <a href="http://docs.oracle.com/javase/7/docs/api/java/util/regex/Pattern.html">Pattern</a>) | |
2304 over the string <code>s</code>. | |
2305 If <code>pattern</code> specifies no captures, | |
2306 then the whole match is produced in each call. | |
2307 | |
2308 | |
2309 <p> | |
2310 As an example, the following loop | |
2311 will iterate over all the words from string <code>s</code>, | |
2312 printing one per line: | |
2313 | |
2314 <pre> | |
2315 local s = "hello world from Lua" | |
2316 for w in String.gmatch(s, [[\w+]]) do | |
2317 print(w) | |
2318 end | |
2319 </pre> | |
2320 | |
2321 <p> | |
2322 The next example collects all pairs <code>key=value</code> from the | |
2323 given string into a table: | |
2324 | |
2325 <pre> | |
2326 local t = {} | |
2327 local s = "from=world, to=Lua" | |
2328 for k, v in String.gmatch(s, [[(\w+)=(\w+)]]) do | |
2329 t[k] = v | |
2330 end | |
2331 </pre> | |
2332 | |
2333 <p> | |
2334 For this function, a caret '<code>^</code>' at the start of a pattern does not | |
2335 work as an anchor, as this would prevent the iteration. | |
2336 | |
2337 | |
2338 | |
2339 <h4 heading><a name="String.gsub" href="#String.gsub"><code>String.gsub (s, pattern, repl [, n])</code></a></h4> | |
2340 | |
2341 <p> | |
2342 Returns a copy of <code>s</code> | |
2343 in which all (or the first <code>n</code>, if given) | |
2344 occurrences of the <code>pattern</code> (see <a href="http://docs.oracle.com/javase/7/docs/api/java/util/regex/Pattern.html">Pattern</a>) have been | |
2345 replaced by a replacement string specified by <code>repl</code>, | |
2346 which can be a string, a table, or a function. | |
2347 <code>gsub</code> also returns, as its second value, | |
2348 the total number of matches that occurred. | |
2349 The name <code>gsub</code> comes from <em>Global SUBstitution</em>. | |
2350 | |
2351 | |
2352 <p> | |
2353 If <code>repl</code> is a string, then its value is used for replacement. | |
2354 The character <code>\</code> works as an escape character. | |
2355 Any sequence in <code>repl</code> of the form <code>$<em>d</em></code>, | |
2356 with <em>d</em> between 1 and 9, | |
2357 stands for the value of the <em>d</em>-th captured substring. | |
2358 The sequence <code>$0</code> stands for the whole match. | |
2359 | |
2360 | |
2361 <p> | |
2362 If <code>repl</code> is a table, then the table is queried for every match, | |
2363 using the first capture as the key. | |
2364 | |
2365 | |
2366 <p> | |
2367 If <code>repl</code> is a function, then this function is called every time a | |
2368 match occurs, with all captured substrings passed as arguments, | |
2369 in order. | |
2370 | |
2371 | |
2372 <p> | |
2373 In any case, | |
2374 if the pattern specifies no captures, | |
2375 then it behaves as if the whole pattern was inside a capture. | |
2376 | |
2377 | |
2378 <p> | |
2379 If the value returned by the table query or by the function call | |
2380 is not <b>nil</b>, | |
2381 then it is used as the replacement string; | |
2382 otherwise, if it is <b>nil</b>, | |
2383 then there is no replacement | |
2384 (that is, the original match is kept in the string). | |
2385 | |
2386 | |
2387 <p> | |
2388 Here are some examples: | |
2389 | |
2390 <pre> | |
2391 x = String.gsub("hello world", [[(\w+)]], "$1 $1") | |
2392 --> x="hello hello world world" | |
2393 | |
2394 x = String.gsub("hello world", [[\w+]], "$0 $0", 1) | |
2395 --> x="hello hello world" | |
2396 | |
2397 x = String.gsub("hello world from Luan", [[(\w+)\s*(\w+)]], "$2 $1") | |
2398 --> x="world hello Luan from" | |
2399 | |
2400 x = String.gsub("4+5 = $return 4+5$", [[\$(.*?)\$]], function (s) | |
2401 return load(s)() | |
2402 end) | |
2403 --> x="4+5 = 9" | |
2404 | |
2405 local t = {name="lua", version="5.3"} | |
2406 x = String.gsub("$name-$version.tar.gz", [[\$(\w+)]], t) | |
2407 --> x="lua-5.3.tar.gz" | |
2408 </pre> | |
2409 | |
2410 | |
2411 <h4 heading><a name="String.lower" href="#String.lower"><code>String.lower (s)</code></a></h4> | |
2412 <p> | |
2413 Receives a string and returns a copy of this string with all | |
2414 uppercase letters changed to lowercase. | |
2415 All other characters are left unchanged. | |
2416 | |
2417 | |
2418 | |
2419 <h4 heading><a name="String.match" href="#String.match"><code>String.match (s, pattern [, init])</code></a></h4> | |
2420 | |
2421 <p> | |
2422 Looks for the first <em>match</em> of | |
2423 <code>pattern</code> (see <a href="http://docs.oracle.com/javase/7/docs/api/java/util/regex/Pattern.html">Pattern</a>) in the string <code>s</code>. | |
2424 If it finds one, then <code>match</code> returns | |
2425 the captures from the pattern; | |
2426 otherwise it returns <b>nil</b>. | |
2427 If <code>pattern</code> specifies no captures, | |
2428 then the whole match is returned. | |
2429 A third, optional numerical argument <code>init</code> specifies | |
2430 where to start the search; | |
2431 its default value is 1 and can be negative. | |
2432 | |
2433 | |
2434 <h4 heading><a name="String.matches" href="#String.matches"><code>String.matches (s, pattern)</code></a></h4> | |
2435 <p> | |
2436 Returns a boolean indicating whether the <code>pattern</code> can be found in string <code>s</code>. | |
2437 This function is equivalent to | |
2438 | |
2439 <pre> | |
2440 return String.match(s,pattern) ~= nil | |
2441 </pre> | |
2442 | |
2443 | |
2444 <h4 heading><a name="String.regex_quote" href="#String.regex_quote"><code>String.regex_quote (s)</code></a></h4> | |
2445 <p> | |
2446 Returns a string which matches the literal string <code>s</code> in a regular expression. This function is simply the Java method <a href="http://docs.oracle.com/javase/7/docs/api/java/util/regex/Pattern.html#quote(java.lang.String)"><code>Pattern.quote</code></a>. | |
2447 | |
2448 | |
2449 <h4 heading><a name="String.rep" href="#String.rep"><code>String.rep (s, n [, sep])</code></a></h4> | |
2450 <p> | |
2451 Returns a string that is the concatenation of <code>n</code> copies of | |
2452 the string <code>s</code> separated by the string <code>sep</code>. | |
2453 The default value for <code>sep</code> is the empty string | |
2454 (that is, no separator). | |
2455 Returns the empty string if <code>n</code> is not positive. | |
2456 | |
2457 | |
2458 | |
2459 | |
2460 <h4 heading><a name="String.reverse" href="#String.reverse"><code>String.reverse (s)</code></a></h4> | |
2461 <p> | |
2462 Returns a string that is the string <code>s</code> reversed. | |
2463 | |
2464 | |
2465 | |
2466 <h4 heading><a name="String.split" href="#String.match"><code>String.split (s, pattern [, limit])</code></a></h4> | |
2467 | |
2468 <p> | |
2469 Splits <code>s</code> using regex <code>pattern</code> and returns the results. If <code>limit</code> is positive, then only returns at most that many results. If <code>limit</code> is zero, then remove trailing empty results. | |
2470 | |
2471 | |
2472 | |
2473 <h4 heading><a name="String.sub" href="#String.sub"><code>String.sub (s, i [, j])</code></a></h4> | |
2474 | |
2475 <p> | |
2476 Returns the substring of <code>s</code> that | |
2477 starts at <code>i</code> and continues until <code>j</code>; | |
2478 <code>i</code> and <code>j</code> can be negative. | |
2479 If <code>j</code> is absent, then it is assumed to be equal to -1 | |
2480 (which is the same as the string length). | |
2481 In particular, | |
2482 the call <code>string.sub(s,1,j)</code> returns a prefix of <code>s</code> | |
2483 with length <code>j</code>, | |
2484 and <code>string.sub(s, -i)</code> returns a suffix of <code>s</code> | |
2485 with length <code>i</code>. | |
2486 | |
2487 | |
2488 <p> | |
2489 If, after the translation of negative indices, | |
2490 <code>i</code> is less than 1, | |
2491 it is corrected to 1. | |
2492 If <code>j</code> is greater than the string length, | |
2493 it is corrected to that length. | |
2494 If, after these corrections, | |
2495 <code>i</code> is greater than <code>j</code>, | |
2496 the function returns the empty string. | |
2497 | |
2498 | |
2499 | |
2500 <h4 heading><a name="String.to_binary" href="#String.to_binary"><code>String.to_binary (s)</code></a></h4> | |
2501 | |
2502 <p> | |
2503 Converts a string to a binary by calling the Java method <a href="http://docs.oracle.com/javase/7/docs/api/java/lang/String.html#getBytes()"><code>String.getBytes</code></a>. | |
2504 | |
2505 | |
2506 | |
2507 <h4 heading><a name="String.to_number" href="#String.to_number"><code>String.to_number (s [, base])</code></a></h4> | |
2508 | |
2509 <p> | |
2510 When called with no <code>base</code>, | |
2511 <code>to_number</code> tries to convert its argument to a number. | |
2512 If the argument is | |
2513 a string convertible to a number, | |
2514 then <code>to_number</code> returns this number; | |
2515 otherwise, it returns <b>nil</b>. | |
2516 | |
2517 The conversion of strings can result in integers or floats. | |
2518 | |
2519 | |
2520 <p> | |
2521 When called with <code>base</code>, | |
2522 then <code>s</code> must be a string to be interpreted as | |
2523 an integer numeral in that base. | |
2524 In bases above 10, the letter '<code>A</code>' (in either upper or lower case) | |
2525 represents 10, '<code>B</code>' represents 11, and so forth, | |
2526 with '<code>Z</code>' representing 35. | |
2527 If the string <code>s</code> is not a valid numeral in the given base, | |
2528 the function returns <b>nil</b>. | |
2529 | |
2530 | |
2531 | |
2532 <h4 heading><a name="String.trim" href="#String.trim"><code>String.trim (s)</code></a></h4> | |
2533 | |
2534 <p> | |
2535 Removes the leading and trailing whitespace by calling the Java method <a href="http://docs.oracle.com/javase/7/docs/api/java/lang/String.html#trim()"><code>String.trim</code></a>. | |
2536 | |
2537 | |
2538 | |
2539 | |
2540 <h4 heading><a name="String.unicode" href="#String.unicode"><code>String.unicode (s [, i [, j]])</code></a></h4> | |
2541 | |
2542 <p> | |
2543 Returns the internal numerical codes of the characters <code>s[i]</code>, | |
2544 <code>s[i+1]</code>, ..., <code>s[j]</code>. | |
2545 The default value for <code>i</code> is 1; | |
2546 the default value for <code>j</code> is <code>i</code>. | |
2547 These indices are corrected | |
2548 following the same rules of function <a href="#String.sub"><code>String.sub</code></a>. | |
2549 | |
2550 | |
2551 | |
2552 | |
2553 | |
2554 <h4 heading><a name="String.upper" href="#String.upper"><code>String.upper (s)</code></a></h4> | |
2555 <p> | |
2556 Receives a string and returns a copy of this string with all | |
2557 lowercase letters changed to uppercase. | |
2558 All other characters are left unchanged. | |
2559 The definition of what a lowercase letter is depends on the current locale. | |
2560 | |
2561 | |
2562 | |
2563 | |
2564 | |
2565 <h3 heading><a name="binary_lib" href="#binary_lib">Binary Manipulation</a></h3> | |
2566 | |
2567 <p> | |
2568 Include this library by: | |
2569 | |
2570 <pre> | |
2571 local Binary = require "luan:Binary.luan" | |
2572 </pre> | |
2573 | |
2574 | |
2575 <h4 heading><a name="Binary.binary" href="#Binary.binary"><code>Binary.binary (···)</code></a></h4> | |
2576 | |
2577 <p> | |
2578 Receives zero or more bytes (as integers). | |
2579 Returns a binary with length equal to the number of arguments, | |
2580 in which each byte has the internal numerical code equal | |
2581 to its corresponding argument. | |
2582 | |
2583 | |
2584 <h4 heading><a name="Binary.byte" href="#Binary.byte"><code>Binary.byte (b [, i [, j]])</code></a></h4> | |
2585 | |
2586 <p> | |
2587 Returns the internal numerical codes of the bytes <code>b[i]</code>, | |
2588 <code>b[i+1]</code>, ..., <code>b[j]</code>. | |
2589 The default value for <code>i</code> is 1; | |
2590 the default value for <code>j</code> is <code>i</code>. | |
2591 These indices are corrected | |
2592 following the same rules of function <a href="#String.sub"><code>String.sub</code></a>. | |
2593 | |
2594 | |
2595 <h4 heading><a name="Binary.to_string" href="#Binary.to_string"><code>Binary.to_string (b [,charset])</code></a></h4> | |
2596 <p> | |
2597 If <code>charset</code> is not nil then converts the binary <code>b</code> to a string using the Java <a href="http://docs.oracle.com/javase/7/docs/api/java/lang/String.html#String(byte[],%20java.lang.String)">String constructor</a>, else makes each byte a char. | |
2598 | |
2599 | |
2600 | |
2601 | |
2602 <h3 heading><a name="table_lib" href="#table_lib">Table Manipulation</a></h3> | |
2603 | |
2604 <p> | |
2605 Include this library by: | |
2606 | |
2607 <pre> | |
2608 local Table = require "luan:Table.luan" | |
2609 </pre> | |
2610 | |
2611 <p> | |
2612 This library provides generic functions for table manipulation. | |
2613 It provides all its functions inside the table <code>Table</code>. | |
2614 | |
2615 | |
2616 | |
2617 <h4 heading><a name="Table.clear" href="#Table.clear"><code>Table.clear (tbl)</code></a></h4> | |
2618 | |
2619 <p> | |
2620 Clears the table. | |
2621 | |
2622 | |
2623 <h4 heading><a name="Table.concat" href="#Table.concat"><code>Table.concat (list [, sep [, i [, j]]])</code></a></h4> | |
2624 | |
2625 <p> | |
2626 Given a list, | |
2627 returns the string <code>list[i]..sep..list[i+1] ··· sep..list[j]</code>. | |
2628 The default value for <code>sep</code> is the empty string, | |
2629 the default for <code>i</code> is 1, | |
2630 and the default for <code>j</code> is <code>#list</code>. | |
2631 If <code>i</code> is greater than <code>j</code>, returns the empty string. | |
2632 | |
2633 | |
2634 <h4 heading><a name="Table.copy" href="#Table.copy"><code>Table.copy (tbl [, i [, j]])</code></a></h4> | |
2635 | |
2636 <p> | |
2637 If <code>i</code> is <code>nil</code>, returns a shallow copy of <code>tbl</code>. | |
2638 Otherwise returns a new table which is a list of the elements <code>tbl[i] ··· tbl[j]</code>. | |
2639 By default, <code>j</code> is <code>#tbl</code>. | |
2640 | |
2641 | |
2642 | |
2643 <h4 heading><a name="Table.insert" href="#Table.insert"><code>Table.insert (list, pos, value)</code></a></h4> | |
2644 | |
2645 <p> | |
2646 Inserts element <code>value</code> at position <code>pos</code> in <code>list</code>, | |
2647 shifting up the elements | |
2648 <code>list[pos], list[pos+1], ···, list[#list]</code>. | |
2649 | |
2650 | |
2651 | |
2652 <h4 heading><a name="Table.is_empty" href="#Table.is_empty"><code>Table.is_empty (tbl)</code></a></h4> | |
2653 | |
2654 | |
2655 | |
2656 <h4 heading><a name="Table.pack" href="#Table.pack"><code>Table.pack (···)</code></a></h4> | |
2657 | |
2658 <p> | |
2659 Returns a new table with all parameters stored into keys 1, 2, etc. | |
2660 and with a field "<code>n</code>" with the total number of parameters. | |
2661 Note that the resulting table may not be a sequence. | |
2662 | |
2663 | |
2664 | |
2665 | |
2666 <h4 heading><a name="Table.remove" href="#Table.remove"><code>Table.remove (list, pos)</code></a></h4> | |
2667 | |
2668 | |
2669 <p> | |
2670 Removes from <code>list</code> the element at position <code>pos</code>, | |
2671 returning the value of the removed element. | |
2672 When <code>pos</code> is an integer between 1 and <code>#list</code>, | |
2673 it shifts down the elements | |
2674 <code>list[pos+1], list[pos+2], ···, list[#list]</code> | |
2675 and erases element <code>list[#list]</code>; | |
2676 The index <code>pos</code> can also be 0 when <code>#list</code> is 0, | |
2677 or <code>#list + 1</code>; | |
2678 in those cases, the function erases the element <code>list[pos]</code>. | |
2679 | |
2680 | |
2681 | |
2682 <h4 heading><a name="Table.size" href="#Table.size"><code>Table.size (tbl)</code></a></h4> | |
2683 | |
2684 | |
2685 | |
2686 <h4 heading><a name="Table.sort" href="#Table.sort"><code>Table.sort (list [, comp])</code></a></h4> | |
2687 | |
2688 <p> | |
2689 Sorts list elements in a given order, <em>in-place</em>, | |
2690 from <code>list[1]</code> to <code>list[#list]</code>. | |
2691 If <code>comp</code> is given, | |
2692 then it must be a function that receives two list elements | |
2693 and returns true when the first element must come | |
2694 before the second in the final order | |
2695 (so that <code>not comp(list[i+1],list[i])</code> will be true after the sort). | |
2696 If <code>comp</code> is not given, | |
2697 then the standard Lua operator <code><</code> is used instead. | |
2698 | |
2699 <p> | |
2700 The sort algorithm is not stable; | |
2701 that is, elements considered equal by the given order | |
2702 may have their relative positions changed by the sort. | |
2703 | |
2704 | |
2705 | |
2706 <h4 heading><a name="Table.unpack" href="#Table.unpack"><code>Table.unpack (list [, i [, j]])</code></a></h4> | |
2707 | |
2708 <p> | |
2709 Returns the elements from the given list. | |
2710 This function is equivalent to | |
2711 | |
2712 <pre> | |
2713 return list[i], list[i+1], ···, list[j] | |
2714 </pre> | |
2715 | |
2716 <p> | |
2717 By default, <code>i</code> is 1 and <code>j</code> is <code>list.n or #list</code>. | |
2718 | |
2719 | |
2720 | |
2721 | |
2722 <h3 heading><a name="number_lib" href="#number_lib">Number Manipulation</a></h3> | |
2723 | |
2724 <p> | |
2725 Include this library by: | |
2726 | |
2727 <pre> | |
2728 local Number = require "luan:Number.luan" | |
2729 </pre> | |
2730 | |
2731 | |
2732 <h4 heading><a name="Number.double" href="#Number.double"><code>Number.double (x)</code></a></h4> | |
2733 <p> | |
2734 Returns <code>x</code> as a double. | |
2735 | |
2736 | |
2737 <h4 heading><a name="Number.float" href="#Number.double"><code>Number.float (x)</code></a></h4> | |
2738 <p> | |
2739 Returns <code>x</code> as a float. | |
2740 | |
2741 | |
2742 <h4 heading><a name="Number.integer" href="#Number.integer"><code>Number.integer (x)</code></a></h4> | |
2743 <p> | |
2744 If the value <code>x</code> is convertible to an integer, | |
2745 returns that integer. | |
2746 Otherwise throws an error. | |
2747 | |
2748 | |
2749 <h4 heading><a name="Number.long" href="#Number.long"><code>Number.long (x)</code></a></h4> | |
2750 <p> | |
2751 If the value <code>x</code> is convertible to an long, | |
2752 returns that long. | |
2753 Otherwise throws an error. | |
2754 | |
2755 | |
2756 <h4 heading><a name="Number.long_to_string" href="#Number.long_to_string"><code>Number.long_to_string (i, radix)</code></a></h4> | |
2757 <p> | |
2758 Converts long value <code>i</code> to a string by calling <code><a href="http://docs.oracle.com/javase/7/docs/api/java/lang/Long.html#toString(long,%20int)">Long.toString</a></code>. | |
2759 | |
2760 | |
2761 <h4 heading><a name="Number.type" href="#Number.type"><code>Number.type (x)</code></a></h4> | |
2762 <p> | |
2763 Returns a string for the numeric type of <code>x</code>. Possible return values include "<code>integer</code>", "<code>long</code>", "<code>double</code>", and "<code>float</code>". | |
2764 | |
2765 | |
2766 | |
2767 | |
2768 <h3 heading><a name="math_lib" href="#math_lib">Mathematical Functions</a></h3> | |
2769 | |
2770 <p> | |
2771 Include this library by: | |
2772 | |
2773 <pre> | |
2774 local Math = require "luan:Math.luan" | |
2775 </pre> | |
2776 | |
2777 <p> | |
2778 This library provides basic mathematical functions. | |
2779 It provides all its functions and constants inside the table <code>Math</code>. | |
2780 | |
2781 | |
2782 <h4 heading><a name="Math.abs" href="#Math.abs"><code>Math.abs (x)</code></a></h4> | |
2783 | |
2784 <p> | |
2785 Returns the absolute value of <code>x</code>. | |
2786 | |
2787 | |
2788 | |
2789 <h4 heading><a name="Math.acos" href="#Math.acos"><code>Math.acos (x)</code></a></h4> | |
2790 | |
2791 <p> | |
2792 Returns the arc cosine of <code>x</code> (in radians). | |
2793 | |
2794 | |
2795 | |
2796 | |
2797 <h4 heading><a name="Math.asin" href="#Math.asin"><code>Math.asin (x)</code></a></h4> | |
2798 | |
2799 <p> | |
2800 Returns the arc sine of <code>x</code> (in radians). | |
2801 | |
2802 | |
2803 | |
2804 | |
2805 <h4 heading><a name="Math.atan" href="#Math.atan"><code>Math.atan (y, x)</code></a></h4> | |
2806 | |
2807 <p> | |
2808 Returns the arc tangent of <code>y/x</code> (in radians), | |
2809 but uses the signs of both parameters to find the | |
2810 quadrant of the result. | |
2811 (It also handles correctly the case of <code>x</code> being zero.) | |
2812 | |
2813 | |
2814 | |
2815 | |
2816 <h4 heading><a name="Math.ceil" href="#Math.ceil"><code>Math.ceil (x)</code></a></h4> | |
2817 | |
2818 <p> | |
2819 Returns the smallest integral value larger than or equal to <code>x</code>. | |
2820 | |
2821 | |
2822 | |
2823 | |
2824 <h4 heading><a name="Math.cos" href="#Math.cos"><code>Math.cos (x)</code></a></h4> | |
2825 | |
2826 <p> | |
2827 Returns the cosine of <code>x</code> (assumed to be in radians). | |
2828 | |
2829 | |
2830 | |
2831 | |
2832 <h4 heading><a name="Math.deg" href="#Math.deg"><code>Math.deg (x)</code></a></h4> | |
2833 | |
2834 <p> | |
2835 Converts the angle <code>x</code> from radians to degrees. | |
2836 | |
2837 | |
2838 | |
2839 | |
2840 <h4 heading><a name="Math.exp" href="#Math.exp"><code>Math.exp (x)</code></a></h4> | |
2841 | |
2842 <p> | |
2843 Returns the value <em>e<sup>x</sup></em> | |
2844 (where <code>e</code> is the base of natural logarithms). | |
2845 | |
2846 | |
2847 | |
2848 | |
2849 <h4 heading><a name="Math.floor" href="#Math.floor"><code>Math.floor (x)</code></a></h4> | |
2850 | |
2851 <p> | |
2852 Returns the largest integral value smaller than or equal to <code>x</code>. | |
2853 | |
2854 | |
2855 | |
2856 | |
2857 <h4 heading><a name="Math.fmod" href="#Math.fmod"><code>Math.fmod (x, y)</code></a></h4> | |
2858 | |
2859 <p> | |
2860 Returns the remainder of the division of <code>x</code> by <code>y</code> | |
2861 that rounds the quotient towards zero. | |
2862 | |
2863 | |
2864 | |
2865 | |
2866 <h4 heading><a name="Math.huge" href="#Math.huge"><code>Math.huge</code></a></h4> | |
2867 | |
2868 <p> | |
2869 A value larger than any other numerical value. | |
2870 | |
2871 | |
2872 | |
2873 | |
2874 <h4 heading><a name="Math.log" href="#Math.log"><code>Math.log (x [, base])</code></a></h4> | |
2875 | |
2876 <p> | |
2877 Returns the logarithm of <code>x</code> in the given base. | |
2878 The default for <code>base</code> is <em>e</em> | |
2879 (so that the function returns the natural logarithm of <code>x</code>). | |
2880 | |
2881 | |
2882 | |
2883 | |
2884 <h4 heading><a name="Math.max" href="#Math.max"><code>Math.max (x, ···)</code></a></h4> | |
2885 | |
2886 <p> | |
2887 Returns the argument with the maximum value, | |
2888 according to the Lua operator <code><</code>. | |
2889 | |
2890 | |
2891 | |
2892 | |
2893 <h4 heading><a name="Math.max_integer" href="#Math.max_integer"><code>Math.max_integer</code></a></h4> | |
2894 <p> | |
2895 An integer with the maximum value for an integer. | |
2896 | |
2897 | |
2898 | |
2899 | |
2900 <h4 heading><a name="Math.min" href="#Math.min"><code>Math.min (x, ···)</code></a></h4> | |
2901 | |
2902 <p> | |
2903 Returns the argument with the minimum value, | |
2904 according to the Lua operator <code><</code>. | |
2905 | |
2906 | |
2907 | |
2908 | |
2909 <h4 heading><a name="Math.min_integer" href="#Math.min_integer"><code>Math.min_integer</code></a></h4> | |
2910 <p> | |
2911 An integer with the minimum value for an integer. | |
2912 | |
2913 | |
2914 | |
2915 | |
2916 <h4 heading><a name="Math.modf" href="#Math.modf"><code>Math.modf (x)</code></a></h4> | |
2917 | |
2918 <p> | |
2919 Returns the integral part of <code>x</code> and the fractional part of <code>x</code>. | |
2920 | |
2921 | |
2922 | |
2923 | |
2924 <h4 heading><a name="Math.pi" href="#Math.pi"><code>Math.pi</code></a></h4> | |
2925 | |
2926 <p> | |
2927 The value of <em>π</em>. | |
2928 | |
2929 | |
2930 | |
2931 | |
2932 <h4 heading><a name="Math.rad" href="#Math.rad"><code>Math.rad (x)</code></a></h4> | |
2933 | |
2934 <p> | |
2935 Converts the angle <code>x</code> from degrees to radians. | |
2936 | |
2937 | |
2938 | |
2939 | |
2940 <h4 heading><a name="Math.random" href="#Math.random"><code>Math.random ([m [, n])</code></a></h4> | |
2941 | |
2942 | |
2943 <p> | |
2944 When called without arguments, | |
2945 returns a pseudo-random float with uniform distribution | |
2946 in the range <em>[0,1)</em>. | |
2947 When called with two integers <code>m</code> and <code>n</code>, | |
2948 <code>Math.random</code> returns a pseudo-random integer | |
2949 with uniform distribution in the range <em>[m, n]</em>. | |
2950 (The value <em>m-n</em> cannot be negative and must fit in a Luan integer.) | |
2951 The call <code>Math.random(n)</code> is equivalent to <code>Math.random(1,n)</code>. | |
2952 | |
2953 | |
2954 <p> | |
2955 This function is an interface to the underling | |
2956 pseudo-random generator function provided by Java. | |
2957 No guarantees can be given for its statistical properties. | |
2958 | |
2959 | |
2960 | |
2961 | |
2962 <h4 heading><a name="Math.sin" href="#Math.sin"><code>Math.sin (x)</code></a></h4> | |
2963 | |
2964 <p> | |
2965 Returns the sine of <code>x</code> (assumed to be in radians). | |
2966 | |
2967 | |
2968 | |
2969 | |
2970 <h4 heading><a name="Math.sqrt" href="#Math.sqrt"><code>Math.sqrt (x)</code></a></h4> | |
2971 | |
2972 <p> | |
2973 Returns the square root of <code>x</code>. | |
2974 (You can also use the expression <code>x^0.5</code> to compute this value.) | |
2975 | |
2976 | |
2977 | |
2978 | |
2979 <h4 heading><a name="Math.tan" href="#Math.tan"><code>Math.tan (x)</code></a></h4> | |
2980 | |
2981 <p> | |
2982 Returns the tangent of <code>x</code> (assumed to be in radians). | |
2983 | |
2984 | |
2985 | |
2986 | |
2987 | |
2988 | |
2989 | |
2990 | |
2991 <h2>6.8 – <a name="6.8">Input and Output Facilities</a></h2> | |
2992 | |
2993 <p> | |
2994 The I/O library provides two different styles for file manipulation. | |
2995 The first one uses implicit file handles; | |
2996 that is, there are operations to set a default input file and a | |
2997 default output file, | |
2998 and all input/output operations are over these default files. | |
2999 The second style uses explicit file handles. | |
3000 | |
3001 | |
3002 <p> | |
3003 When using implicit file handles, | |
3004 all operations are supplied by table <a name="pdf-io"><code>io</code></a>. | |
3005 When using explicit file handles, | |
3006 the operation <a href="#pdf-io.open"><code>io.open</code></a> returns a file handle | |
3007 and then all operations are supplied as methods of the file handle. | |
3008 | |
3009 | |
3010 <p> | |
3011 The table <code>io</code> also provides | |
3012 three predefined file handles with their usual meanings from C: | |
3013 <a name="pdf-io.stdin"><code>io.stdin</code></a>, <a name="pdf-io.stdout"><code>io.stdout</code></a>, and <a name="pdf-io.stderr"><code>io.stderr</code></a>. | |
3014 The I/O library never closes these files. | |
3015 | |
3016 | |
3017 <p> | |
3018 Unless otherwise stated, | |
3019 all I/O functions return <b>nil</b> on failure | |
3020 (plus an error message as a second result and | |
3021 a system-dependent error code as a third result) | |
3022 and some value different from <b>nil</b> on success. | |
3023 On non-POSIX systems, | |
3024 the computation of the error message and error code | |
3025 in case of errors | |
3026 may be not thread safe, | |
3027 because they rely on the global C variable <code>errno</code>. | |
3028 | |
3029 | |
3030 <p> | |
3031 <hr><h3><a name="pdf-io.close"><code>io.close ([file])</code></a></h3> | |
3032 | |
3033 | |
3034 <p> | |
3035 Equivalent to <code>file:close()</code>. | |
3036 Without a <code>file</code>, closes the default output file. | |
3037 | |
3038 | |
3039 | |
3040 | |
3041 <p> | |
3042 <hr><h3><a name="pdf-io.flush"><code>io.flush ()</code></a></h3> | |
3043 | |
3044 | |
3045 <p> | |
3046 Equivalent to <code>io.output():flush()</code>. | |
3047 | |
3048 | |
3049 | |
3050 | |
3051 <p> | |
3052 <hr><h3><a name="pdf-io.input"><code>io.input ([file])</code></a></h3> | |
3053 | |
3054 | |
3055 <p> | |
3056 When called with a file name, it opens the named file (in text mode), | |
3057 and sets its handle as the default input file. | |
3058 When called with a file handle, | |
3059 it simply sets this file handle as the default input file. | |
3060 When called without parameters, | |
3061 it returns the current default input file. | |
3062 | |
3063 | |
3064 <p> | |
3065 In case of errors this function raises the error, | |
3066 instead of returning an error code. | |
3067 | |
3068 | |
3069 | |
3070 | |
3071 <p> | |
3072 <hr><h3><a name="pdf-io.lines"><code>io.lines ([filename ···])</code></a></h3> | |
3073 | |
3074 | |
3075 <p> | |
3076 Opens the given file name in read mode | |
3077 and returns an iterator function that | |
3078 works like <code>file:lines(···)</code> over the opened file. | |
3079 When the iterator function detects the end of file, | |
3080 it returns no values (to finish the loop) and automatically closes the file. | |
3081 | |
3082 | |
3083 <p> | |
3084 The call <code>io.lines()</code> (with no file name) is equivalent | |
3085 to <code>io.input():lines("*l")</code>; | |
3086 that is, it iterates over the lines of the default input file. | |
3087 In this case it does not close the file when the loop ends. | |
3088 | |
3089 | |
3090 <p> | |
3091 In case of errors this function raises the error, | |
3092 instead of returning an error code. | |
3093 | |
3094 | |
3095 | |
3096 | |
3097 <p> | |
3098 <hr><h3><a name="pdf-io.open"><code>io.open (filename [, mode])</code></a></h3> | |
3099 | |
3100 | |
3101 <p> | |
3102 This function opens a file, | |
3103 in the mode specified in the string <code>mode</code>. | |
3104 It returns a new file handle, | |
3105 or, in case of errors, <b>nil</b> plus an error message. | |
3106 | |
3107 | |
3108 <p> | |
3109 The <code>mode</code> string can be any of the following: | |
3110 | |
3111 <ul> | |
3112 <li><b>"<code>r</code>": </b> read mode (the default);</li> | |
3113 <li><b>"<code>w</code>": </b> write mode;</li> | |
3114 <li><b>"<code>a</code>": </b> append mode;</li> | |
3115 <li><b>"<code>r+</code>": </b> update mode, all previous data is preserved;</li> | |
3116 <li><b>"<code>w+</code>": </b> update mode, all previous data is erased;</li> | |
3117 <li><b>"<code>a+</code>": </b> append update mode, previous data is preserved, | |
3118 writing is only allowed at the end of file.</li> | |
3119 </ul><p> | |
3120 The <code>mode</code> string can also have a '<code>b</code>' at the end, | |
3121 which is needed in some systems to open the file in binary mode. | |
3122 | |
3123 | |
3124 | |
3125 | |
3126 <p> | |
3127 <hr><h3><a name="pdf-io.output"><code>io.output ([file])</code></a></h3> | |
3128 | |
3129 | |
3130 <p> | |
3131 Similar to <a href="#pdf-io.input"><code>io.input</code></a>, but operates over the default output file. | |
3132 | |
3133 | |
3134 | |
3135 | |
3136 <p> | |
3137 <hr><h3><a name="pdf-io.popen"><code>io.popen (prog [, mode])</code></a></h3> | |
3138 | |
3139 | |
3140 <p> | |
3141 This function is system dependent and is not available | |
3142 on all platforms. | |
3143 | |
3144 | |
3145 <p> | |
3146 Starts program <code>prog</code> in a separated process and returns | |
3147 a file handle that you can use to read data from this program | |
3148 (if <code>mode</code> is <code>"r"</code>, the default) | |
3149 or to write data to this program | |
3150 (if <code>mode</code> is <code>"w"</code>). | |
3151 | |
3152 | |
3153 | |
3154 | |
3155 <p> | |
3156 <hr><h3><a name="pdf-io.read"><code>io.read (···)</code></a></h3> | |
3157 | |
3158 | |
3159 <p> | |
3160 Equivalent to <code>io.input():read(···)</code>. | |
3161 | |
3162 | |
3163 | |
3164 | |
3165 <p> | |
3166 <hr><h3><a name="pdf-io.tmpfile"><code>io.tmpfile ()</code></a></h3> | |
3167 | |
3168 | |
3169 <p> | |
3170 Returns a handle for a temporary file. | |
3171 This file is opened in update mode | |
3172 and it is automatically removed when the program ends. | |
3173 | |
3174 | |
3175 | |
3176 | |
3177 <p> | |
3178 <hr><h3><a name="pdf-io.type"><code>io.type (obj)</code></a></h3> | |
3179 | |
3180 | |
3181 <p> | |
3182 Checks whether <code>obj</code> is a valid file handle. | |
3183 Returns the string <code>"file"</code> if <code>obj</code> is an open file handle, | |
3184 <code>"closed file"</code> if <code>obj</code> is a closed file handle, | |
3185 or <b>nil</b> if <code>obj</code> is not a file handle. | |
3186 | |
3187 | |
3188 | |
3189 | |
3190 <p> | |
3191 <hr><h3><a name="pdf-io.write"><code>io.write (···)</code></a></h3> | |
3192 | |
3193 | |
3194 <p> | |
3195 Equivalent to <code>io.output():write(···)</code>. | |
3196 | |
3197 | |
3198 | |
3199 | |
3200 <p> | |
3201 <hr><h3><a name="pdf-file:close"><code>file:close ()</code></a></h3> | |
3202 | |
3203 | |
3204 <p> | |
3205 Closes <code>file</code>. | |
3206 Note that files are automatically closed when | |
3207 their handles are garbage collected, | |
3208 but that takes an unpredictable amount of time to happen. | |
3209 | |
3210 | |
3211 <p> | |
3212 When closing a file handle created with <a href="#pdf-io.popen"><code>io.popen</code></a>, | |
3213 <a href="#pdf-file:close"><code>file:close</code></a> returns the same values | |
3214 returned by <a href="#pdf-os.execute"><code>os.execute</code></a>. | |
3215 | |
3216 | |
3217 | |
3218 | |
3219 <p> | |
3220 <hr><h3><a name="pdf-file:flush"><code>file:flush ()</code></a></h3> | |
3221 | |
3222 | |
3223 <p> | |
3224 Saves any written data to <code>file</code>. | |
3225 | |
3226 | |
3227 | |
3228 | |
3229 <p> | |
3230 <hr><h3><a name="pdf-file:lines"><code>file:lines (···)</code></a></h3> | |
3231 | |
3232 | |
3233 <p> | |
3234 Returns an iterator function that, | |
3235 each time it is called, | |
3236 reads the file according to the given formats. | |
3237 When no format is given, | |
3238 uses "<code>l</code>" as a default. | |
3239 As an example, the construction | |
3240 | |
3241 <pre> | |
3242 for c in file:lines(1) do <em>body</em> end | |
3243 </pre><p> | |
3244 will iterate over all characters of the file, | |
3245 starting at the current position. | |
3246 Unlike <a href="#pdf-io.lines"><code>io.lines</code></a>, this function does not close the file | |
3247 when the loop ends. | |
3248 | |
3249 | |
3250 <p> | |
3251 In case of errors this function raises the error, | |
3252 instead of returning an error code. | |
3253 | |
3254 | |
3255 | |
3256 | |
3257 <p> | |
3258 <hr><h3><a name="pdf-file:read"><code>file:read (···)</code></a></h3> | |
3259 | |
3260 | |
3261 <p> | |
3262 Reads the file <code>file</code>, | |
3263 according to the given formats, which specify what to read. | |
3264 For each format, | |
3265 the function returns a string or a number with the characters read, | |
3266 or <b>nil</b> if it cannot read data with the specified format. | |
3267 (In this latter case, | |
3268 the function does not read subsequent formats.) | |
3269 When called without formats, | |
3270 it uses a default format that reads the next line | |
3271 (see below). | |
3272 | |
3273 | |
3274 <p> | |
3275 The available formats are | |
3276 | |
3277 <ul> | |
3278 | |
3279 <li><b>"<code>n</code>": </b> | |
3280 reads a numeral and returns it as a float or an integer, | |
3281 following the lexical conventions of Lua. | |
3282 (The numeral may have leading spaces and a sign.) | |
3283 This format always reads the longest input sequence that | |
3284 is a valid prefix for a number; | |
3285 if that prefix does not form a valid number | |
3286 (e.g., an empty string, "<code>0x</code>", or "<code>3.4e-</code>"), | |
3287 it is discarded and the function returns <b>nil</b>. | |
3288 </li> | |
3289 | |
3290 <li><b>"<code>i</code>": </b> | |
3291 reads an integral number and returns it as an integer. | |
3292 </li> | |
3293 | |
3294 <li><b>"<code>a</code>": </b> | |
3295 reads the whole file, starting at the current position. | |
3296 On end of file, it returns the empty string. | |
3297 </li> | |
3298 | |
3299 <li><b>"<code>l</code>": </b> | |
3300 reads the next line skipping the end of line, | |
3301 returning <b>nil</b> on end of file. | |
3302 This is the default format. | |
3303 </li> | |
3304 | |
3305 <li><b>"<code>L</code>": </b> | |
3306 reads the next line keeping the end-of-line character (if present), | |
3307 returning <b>nil</b> on end of file. | |
3308 </li> | |
3309 | |
3310 <li><b><em>number</em>: </b> | |
3311 reads a string with up to this number of bytes, | |
3312 returning <b>nil</b> on end of file. | |
3313 If <code>number</code> is zero, | |
3314 it reads nothing and returns an empty string, | |
3315 or <b>nil</b> on end of file. | |
3316 </li> | |
3317 | |
3318 </ul><p> | |
3319 The formats "<code>l</code>" and "<code>L</code>" should be used only for text files. | |
3320 | |
3321 | |
3322 | |
3323 | |
3324 <p> | |
3325 <hr><h3><a name="pdf-file:seek"><code>file:seek ([whence [, offset]])</code></a></h3> | |
3326 | |
3327 | |
3328 <p> | |
3329 Sets and gets the file position, | |
3330 measured from the beginning of the file, | |
3331 to the position given by <code>offset</code> plus a base | |
3332 specified by the string <code>whence</code>, as follows: | |
3333 | |
3334 <ul> | |
3335 <li><b>"<code>set</code>": </b> base is position 0 (beginning of the file);</li> | |
3336 <li><b>"<code>cur</code>": </b> base is current position;</li> | |
3337 <li><b>"<code>end</code>": </b> base is end of file;</li> | |
3338 </ul><p> | |
3339 In case of success, <code>seek</code> returns the final file position, | |
3340 measured in bytes from the beginning of the file. | |
3341 If <code>seek</code> fails, it returns <b>nil</b>, | |
3342 plus a string describing the error. | |
3343 | |
3344 | |
3345 <p> | |
3346 The default value for <code>whence</code> is <code>"cur"</code>, | |
3347 and for <code>offset</code> is 0. | |
3348 Therefore, the call <code>file:seek()</code> returns the current | |
3349 file position, without changing it; | |
3350 the call <code>file:seek("set")</code> sets the position to the | |
3351 beginning of the file (and returns 0); | |
3352 and the call <code>file:seek("end")</code> sets the position to the | |
3353 end of the file, and returns its size. | |
3354 | |
3355 | |
3356 | |
3357 | |
3358 <p> | |
3359 <hr><h3><a name="pdf-file:setvbuf"><code>file:setvbuf (mode [, size])</code></a></h3> | |
3360 | |
3361 | |
3362 <p> | |
3363 Sets the buffering mode for an output file. | |
3364 There are three available modes: | |
3365 | |
3366 <ul> | |
3367 | |
3368 <li><b>"<code>no</code>": </b> | |
3369 no buffering; the result of any output operation appears immediately. | |
3370 </li> | |
3371 | |
3372 <li><b>"<code>full</code>": </b> | |
3373 full buffering; output operation is performed only | |
3374 when the buffer is full or when | |
3375 you explicitly <code>flush</code> the file (see <a href="#pdf-io.flush"><code>io.flush</code></a>). | |
3376 </li> | |
3377 | |
3378 <li><b>"<code>line</code>": </b> | |
3379 line buffering; output is buffered until a newline is output | |
3380 or there is any input from some special files | |
3381 (such as a terminal device). | |
3382 </li> | |
3383 | |
3384 </ul><p> | |
3385 For the last two cases, <code>size</code> | |
3386 specifies the size of the buffer, in bytes. | |
3387 The default is an appropriate size. | |
3388 | |
3389 | |
3390 | |
3391 | |
3392 <p> | |
3393 <hr><h3><a name="pdf-file:write"><code>file:write (···)</code></a></h3> | |
3394 | |
3395 | |
3396 <p> | |
3397 Writes the value of each of its arguments to <code>file</code>. | |
3398 The arguments must be strings or numbers. | |
3399 | |
3400 | |
3401 <p> | |
3402 In case of success, this function returns <code>file</code>. | |
3403 Otherwise it returns <b>nil</b> plus a string describing the error. | |
3404 | |
3405 | |
3406 | |
3407 | |
3408 | |
3409 | |
3410 | |
3411 <h2>6.9 – <a name="6.9">Operating System Facilities</a></h2> | |
3412 | |
3413 <p> | |
3414 This library is implemented through table <a name="pdf-os"><code>os</code></a>. | |
3415 | |
3416 | |
3417 <p> | |
3418 <hr><h3><a name="pdf-os.clock"><code>os.clock ()</code></a></h3> | |
3419 | |
3420 | |
3421 <p> | |
3422 Returns an approximation of the amount in seconds of CPU time | |
3423 used by the program. | |
3424 | |
3425 | |
3426 | |
3427 | |
3428 <p> | |
3429 <hr><h3><a name="pdf-os.date"><code>os.date ([format [, time]])</code></a></h3> | |
3430 | |
3431 | |
3432 <p> | |
3433 Returns a string or a table containing date and time, | |
3434 formatted according to the given string <code>format</code>. | |
3435 | |
3436 | |
3437 <p> | |
3438 If the <code>time</code> argument is present, | |
3439 this is the time to be formatted | |
3440 (see the <a href="#pdf-os.time"><code>os.time</code></a> function for a description of this value). | |
3441 Otherwise, <code>date</code> formats the current time. | |
3442 | |
3443 | |
3444 <p> | |
3445 If <code>format</code> starts with '<code>!</code>', | |
3446 then the date is formatted in Coordinated Universal Time. | |
3447 After this optional character, | |
3448 if <code>format</code> is the string "<code>*t</code>", | |
3449 then <code>date</code> returns a table with the following fields: | |
3450 <code>year</code> (four digits), <code>month</code> (1–12), <code>day</code> (1–31), | |
3451 <code>hour</code> (0–23), <code>min</code> (0–59), <code>sec</code> (0–61), | |
3452 <code>wday</code> (weekday, Sunday is 1), | |
3453 <code>yday</code> (day of the year), | |
3454 and <code>isdst</code> (daylight saving flag, a boolean). | |
3455 This last field may be absent | |
3456 if the information is not available. | |
3457 | |
3458 | |
3459 <p> | |
3460 If <code>format</code> is not "<code>*t</code>", | |
3461 then <code>date</code> returns the date as a string, | |
3462 formatted according to the same rules as the ISO C function <code>strftime</code>. | |
3463 | |
3464 | |
3465 <p> | |
3466 When called without arguments, | |
3467 <code>date</code> returns a reasonable date and time representation that depends on | |
3468 the host system and on the current locale | |
3469 (that is, <code>os.date()</code> is equivalent to <code>os.date("%c")</code>). | |
3470 | |
3471 | |
3472 <p> | |
3473 On non-POSIX systems, | |
3474 this function may be not thread safe | |
3475 because of its reliance on C function <code>gmtime</code> and C function <code>localtime</code>. | |
3476 | |
3477 | |
3478 | |
3479 | |
3480 <p> | |
3481 <hr><h3><a name="pdf-os.difftime"><code>os.difftime (t2, t1)</code></a></h3> | |
3482 | |
3483 | |
3484 <p> | |
3485 Returns the difference, in seconds, | |
3486 from time <code>t1</code> to time <code>t2</code> | |
3487 (where the times are values returned by <a href="#pdf-os.time"><code>os.time</code></a>). | |
3488 In POSIX, Windows, and some other systems, | |
3489 this value is exactly <code>t2</code><em>-</em><code>t1</code>. | |
3490 | |
3491 | |
3492 | |
3493 | |
3494 <p> | |
3495 <hr><h3><a name="pdf-os.execute"><code>os.execute ([command])</code></a></h3> | |
3496 | |
3497 | |
3498 <p> | |
3499 This function is equivalent to the ISO C function <code>system</code>. | |
3500 It passes <code>command</code> to be executed by an operating system shell. | |
3501 Its first result is <b>true</b> | |
3502 if the command terminated successfully, | |
3503 or <b>nil</b> otherwise. | |
3504 After this first result | |
3505 the function returns a string plus a number, | |
3506 as follows: | |
3507 | |
3508 <ul> | |
3509 | |
3510 <li><b>"<code>exit</code>": </b> | |
3511 the command terminated normally; | |
3512 the following number is the exit status of the command. | |
3513 </li> | |
3514 | |
3515 <li><b>"<code>signal</code>": </b> | |
3516 the command was terminated by a signal; | |
3517 the following number is the signal that terminated the command. | |
3518 </li> | |
3519 | |
3520 </ul> | |
3521 | |
3522 <p> | |
3523 When called without a <code>command</code>, | |
3524 <code>os.execute</code> returns a boolean that is true if a shell is available. | |
3525 | |
3526 | |
3527 | |
3528 | |
3529 <p> | |
3530 <hr><h3><a name="pdf-os.exit"><code>os.exit ([code [, close]])</code></a></h3> | |
3531 | |
3532 | |
3533 <p> | |
3534 Calls the ISO C function <code>exit</code> to terminate the host program. | |
3535 If <code>code</code> is <b>true</b>, | |
3536 the returned status is <code>EXIT_SUCCESS</code>; | |
3537 if <code>code</code> is <b>false</b>, | |
3538 the returned status is <code>EXIT_FAILURE</code>; | |
3539 if <code>code</code> is a number, | |
3540 the returned status is this number. | |
3541 The default value for <code>code</code> is <b>true</b>. | |
3542 | |
3543 | |
3544 <p> | |
3545 If the optional second argument <code>close</code> is true, | |
3546 closes the Lua state before exiting. | |
3547 | |
3548 | |
3549 | |
3550 | |
3551 <p> | |
3552 <hr><h3><a name="pdf-os.getenv"><code>os.getenv (varname)</code></a></h3> | |
3553 | |
3554 | |
3555 <p> | |
3556 Returns the value of the process environment variable <code>varname</code>, | |
3557 or <b>nil</b> if the variable is not defined. | |
3558 | |
3559 | |
3560 | |
3561 | |
3562 <p> | |
3563 <hr><h3><a name="pdf-os.remove"><code>os.remove (filename)</code></a></h3> | |
3564 | |
3565 | |
3566 <p> | |
3567 Deletes the file (or empty directory, on POSIX systems) | |
3568 with the given name. | |
3569 If this function fails, it returns <b>nil</b>, | |
3570 plus a string describing the error and the error code. | |
3571 | |
3572 | |
3573 | |
3574 | |
3575 <p> | |
3576 <hr><h3><a name="pdf-os.rename"><code>os.rename (oldname, newname)</code></a></h3> | |
3577 | |
3578 | |
3579 <p> | |
3580 Renames file or directory named <code>oldname</code> to <code>newname</code>. | |
3581 If this function fails, it returns <b>nil</b>, | |
3582 plus a string describing the error and the error code. | |
3583 | |
3584 | |
3585 | |
3586 | |
3587 <p> | |
3588 <hr><h3><a name="pdf-os.setlocale"><code>os.setlocale (locale [, category])</code></a></h3> | |
3589 | |
3590 | |
3591 <p> | |
3592 Sets the current locale of the program. | |
3593 <code>locale</code> is a system-dependent string specifying a locale; | |
3594 <code>category</code> is an optional string describing which category to change: | |
3595 <code>"all"</code>, <code>"collate"</code>, <code>"ctype"</code>, | |
3596 <code>"monetary"</code>, <code>"numeric"</code>, or <code>"time"</code>; | |
3597 the default category is <code>"all"</code>. | |
3598 The function returns the name of the new locale, | |
3599 or <b>nil</b> if the request cannot be honored. | |
3600 | |
3601 | |
3602 <p> | |
3603 If <code>locale</code> is the empty string, | |
3604 the current locale is set to an implementation-defined native locale. | |
3605 If <code>locale</code> is the string "<code>C</code>", | |
3606 the current locale is set to the standard C locale. | |
3607 | |
3608 | |
3609 <p> | |
3610 When called with <b>nil</b> as the first argument, | |
3611 this function only returns the name of the current locale | |
3612 for the given category. | |
3613 | |
3614 | |
3615 <p> | |
3616 This function may be not thread safe | |
3617 because of its reliance on C function <code>setlocale</code>. | |
3618 | |
3619 | |
3620 | |
3621 | |
3622 <p> | |
3623 <hr><h3><a name="pdf-os.time"><code>os.time ([table])</code></a></h3> | |
3624 | |
3625 | |
3626 <p> | |
3627 Returns the current time when called without arguments, | |
3628 or a time representing the date and time specified by the given table. | |
3629 This table must have fields <code>year</code>, <code>month</code>, and <code>day</code>, | |
3630 and may have fields | |
3631 <code>hour</code> (default is 12), | |
3632 <code>min</code> (default is 0), | |
3633 <code>sec</code> (default is 0), | |
3634 and <code>isdst</code> (default is <b>nil</b>). | |
3635 For a description of these fields, see the <a href="#pdf-os.date"><code>os.date</code></a> function. | |
3636 | |
3637 | |
3638 <p> | |
3639 The returned value is a number, whose meaning depends on your system. | |
3640 In POSIX, Windows, and some other systems, | |
3641 this number counts the number | |
3642 of seconds since some given start time (the "epoch"). | |
3643 In other systems, the meaning is not specified, | |
3644 and the number returned by <code>time</code> can be used only as an argument to | |
3645 <a href="#pdf-os.date"><code>os.date</code></a> and <a href="#pdf-os.difftime"><code>os.difftime</code></a>. | |
3646 | |
3647 | |
3648 | |
3649 | |
3650 <p> | |
3651 <hr><h3><a name="pdf-os.tmpname"><code>os.tmpname ()</code></a></h3> | |
3652 | |
3653 | |
3654 <p> | |
3655 Returns a string with a file name that can | |
3656 be used for a temporary file. | |
3657 The file must be explicitly opened before its use | |
3658 and explicitly removed when no longer needed. | |
3659 | |
3660 | |
3661 <p> | |
3662 On POSIX systems, | |
3663 this function also creates a file with that name, | |
3664 to avoid security risks. | |
3665 (Someone else might create the file with wrong permissions | |
3666 in the time between getting the name and creating the file.) | |
3667 You still have to open the file to use it | |
3668 and to remove it (even if you do not use it). | |
3669 | |
3670 | |
3671 <p> | |
3672 When possible, | |
3673 you may prefer to use <a href="#pdf-io.tmpfile"><code>io.tmpfile</code></a>, | |
3674 which automatically removes the file when the program ends. | |
3675 | |
3676 | |
3677 | |
3678 | |
3679 | |
3680 | |
3681 | |
3682 <h2>6.10 – <a name="6.10">The Debug Library</a></h2> | |
3683 | |
3684 <p> | |
3685 This library provides | |
3686 the functionality of the debug interface (<a href="#4.9">§4.9</a>) to Lua programs. | |
3687 You should exert care when using this library. | |
3688 Several of its functions | |
3689 violate basic assumptions about Lua code | |
3690 (e.g., that variables local to a function | |
3691 cannot be accessed from outside; | |
3692 that userdata metatables cannot be changed by Lua code; | |
3693 that Lua programs do not crash) | |
3694 and therefore can compromise otherwise secure code. | |
3695 Moreover, some functions in this library may be slow. | |
3696 | |
3697 | |
3698 <p> | |
3699 All functions in this library are provided | |
3700 inside the <a name="pdf-debug"><code>debug</code></a> table. | |
3701 All functions that operate over a thread | |
3702 have an optional first argument which is the | |
3703 thread to operate over. | |
3704 The default is always the current thread. | |
3705 | |
3706 | |
3707 <p> | |
3708 <hr><h3><a name="pdf-debug.debug"><code>debug.debug ()</code></a></h3> | |
3709 | |
3710 | |
3711 <p> | |
3712 Enters an interactive mode with the user, | |
3713 running each string that the user enters. | |
3714 Using simple commands and other debug facilities, | |
3715 the user can inspect global and local variables, | |
3716 change their values, evaluate expressions, and so on. | |
3717 A line containing only the word <code>cont</code> finishes this function, | |
3718 so that the caller continues its execution. | |
3719 | |
3720 | |
3721 <p> | |
3722 Note that commands for <code>debug.debug</code> are not lexically nested | |
3723 within any function and so have no direct access to local variables. | |
3724 | |
3725 | |
3726 | |
3727 | |
3728 <p> | |
3729 <hr><h3><a name="pdf-debug.gethook"><code>debug.gethook ([thread])</code></a></h3> | |
3730 | |
3731 | |
3732 <p> | |
3733 Returns the current hook settings of the thread, as three values: | |
3734 the current hook function, the current hook mask, | |
3735 and the current hook count | |
3736 (as set by the <a href="#pdf-debug.sethook"><code>debug.sethook</code></a> function). | |
3737 | |
3738 | |
3739 | |
3740 | |
3741 <p> | |
3742 <hr><h3><a name="pdf-debug.getinfo"><code>debug.getinfo ([thread,] f [, what])</code></a></h3> | |
3743 | |
3744 | |
3745 <p> | |
3746 Returns a table with information about a function. | |
3747 You can give the function directly | |
3748 or you can give a number as the value of <code>f</code>, | |
3749 which means the function running at level <code>f</code> of the call stack | |
3750 of the given thread: | |
3751 level 0 is the current function (<code>getinfo</code> itself); | |
3752 level 1 is the function that called <code>getinfo</code> | |
3753 (except for tail calls, which do not count on the stack); | |
3754 and so on. | |
3755 If <code>f</code> is a number larger than the number of active functions, | |
3756 then <code>getinfo</code> returns <b>nil</b>. | |
3757 | |
3758 | |
3759 <p> | |
3760 The returned table can contain all the fields returned by <a href="#lua_getinfo"><code>lua_getinfo</code></a>, | |
3761 with the string <code>what</code> describing which fields to fill in. | |
3762 The default for <code>what</code> is to get all information available, | |
3763 except the table of valid lines. | |
3764 If present, | |
3765 the option '<code>f</code>' | |
3766 adds a field named <code>func</code> with the function itself. | |
3767 If present, | |
3768 the option '<code>L</code>' | |
3769 adds a field named <code>activelines</code> with the table of | |
3770 valid lines. | |
3771 | |
3772 | |
3773 <p> | |
3774 For instance, the expression <code>debug.getinfo(1,"n").name</code> returns | |
3775 a table with a name for the current function, | |
3776 if a reasonable name can be found, | |
3777 and the expression <code>debug.getinfo(print)</code> | |
3778 returns a table with all available information | |
3779 about the <a href="#pdf-print"><code>print</code></a> function. | |
3780 | |
3781 | |
3782 | |
3783 | |
3784 <p> | |
3785 <hr><h3><a name="pdf-debug.getlocal"><code>debug.getlocal ([thread,] f, local)</code></a></h3> | |
3786 | |
3787 | |
3788 <p> | |
3789 This function returns the name and the value of the local variable | |
3790 with index <code>local</code> of the function at level <code>f</code> of the stack. | |
3791 This function accesses not only explicit local variables, | |
3792 but also parameters, temporaries, etc. | |
3793 | |
3794 | |
3795 <p> | |
3796 The first parameter or local variable has index 1, and so on, | |
3797 following the order that they are declared in the code, | |
3798 counting only the variables that are active | |
3799 in the current scope of the function. | |
3800 Negative indices refer to vararg parameters; | |
3801 -1 is the first vararg parameter. | |
3802 The function returns <b>nil</b> if there is no variable with the given index, | |
3803 and raises an error when called with a level out of range. | |
3804 (You can call <a href="#pdf-debug.getinfo"><code>debug.getinfo</code></a> to check whether the level is valid.) | |
3805 | |
3806 | |
3807 <p> | |
3808 Variable names starting with '<code>(</code>' (open parenthesis) | |
3809 represent variables with no known names | |
3810 (internal variables such as loop control variables, | |
3811 and variables from chunks saved without debug information). | |
3812 | |
3813 | |
3814 <p> | |
3815 The parameter <code>f</code> may also be a function. | |
3816 In that case, <code>getlocal</code> returns only the name of function parameters. | |
3817 | |
3818 | |
3819 | |
3820 | |
3821 <p> | |
3822 <hr><h3><a name="pdf-debug.getmetatable"><code>debug.getmetatable (value)</code></a></h3> | |
3823 | |
3824 | |
3825 <p> | |
3826 Returns the metatable of the given <code>value</code> | |
3827 or <b>nil</b> if it does not have a metatable. | |
3828 | |
3829 | |
3830 | |
3831 | |
3832 <p> | |
3833 <hr><h3><a name="pdf-debug.getregistry"><code>debug.getregistry ()</code></a></h3> | |
3834 | |
3835 | |
3836 <p> | |
3837 Returns the registry table (see <a href="#4.5">§4.5</a>). | |
3838 | |
3839 | |
3840 | |
3841 | |
3842 <p> | |
3843 <hr><h3><a name="pdf-debug.getupvalue"><code>debug.getupvalue (f, up)</code></a></h3> | |
3844 | |
3845 | |
3846 <p> | |
3847 This function returns the name and the value of the upvalue | |
3848 with index <code>up</code> of the function <code>f</code>. | |
3849 The function returns <b>nil</b> if there is no upvalue with the given index. | |
3850 | |
3851 | |
3852 <p> | |
3853 Variable names starting with '<code>(</code>' (open parenthesis) | |
3854 represent variables with no known names | |
3855 (variables from chunks saved without debug information). | |
3856 | |
3857 | |
3858 | |
3859 | |
3860 <p> | |
3861 <hr><h3><a name="pdf-debug.getuservalue"><code>debug.getuservalue (u)</code></a></h3> | |
3862 | |
3863 | |
3864 <p> | |
3865 Returns the Lua value associated to <code>u</code>. | |
3866 If <code>u</code> is not a userdata, | |
3867 returns <b>nil</b>. | |
3868 | |
3869 | |
3870 | |
3871 | |
3872 <p> | |
3873 <hr><h3><a name="pdf-debug.sethook"><code>debug.sethook ([thread,] hook, mask [, count])</code></a></h3> | |
3874 | |
3875 | |
3876 <p> | |
3877 Sets the given function as a hook. | |
3878 The string <code>mask</code> and the number <code>count</code> describe | |
3879 when the hook will be called. | |
3880 The string mask may have any combination of the following characters, | |
3881 with the given meaning: | |
3882 | |
3883 <ul> | |
3884 <li><b>'<code>c</code>': </b> the hook is called every time Lua calls a function;</li> | |
3885 <li><b>'<code>r</code>': </b> the hook is called every time Lua returns from a function;</li> | |
3886 <li><b>'<code>l</code>': </b> the hook is called every time Lua enters a new line of code.</li> | |
3887 </ul><p> | |
3888 Moreover, | |
3889 with a <code>count</code> different from zero, | |
3890 the hook is called also after every <code>count</code> instructions. | |
3891 | |
3892 | |
3893 <p> | |
3894 When called without arguments, | |
3895 <a href="#pdf-debug.sethook"><code>debug.sethook</code></a> turns off the hook. | |
3896 | |
3897 | |
3898 <p> | |
3899 When the hook is called, its first parameter is a string | |
3900 describing the event that has triggered its call: | |
3901 <code>"call"</code> (or <code>"tail call"</code>), | |
3902 <code>"return"</code>, | |
3903 <code>"line"</code>, and <code>"count"</code>. | |
3904 For line events, | |
3905 the hook also gets the new line number as its second parameter. | |
3906 Inside a hook, | |
3907 you can call <code>getinfo</code> with level 2 to get more information about | |
3908 the running function | |
3909 (level 0 is the <code>getinfo</code> function, | |
3910 and level 1 is the hook function). | |
3911 | |
3912 | |
3913 | |
3914 | |
3915 <p> | |
3916 <hr><h3><a name="pdf-debug.setlocal"><code>debug.setlocal ([thread,] level, local, value)</code></a></h3> | |
3917 | |
3918 | |
3919 <p> | |
3920 This function assigns the value <code>value</code> to the local variable | |
3921 with index <code>local</code> of the function at level <code>level</code> of the stack. | |
3922 The function returns <b>nil</b> if there is no local | |
3923 variable with the given index, | |
3924 and raises an error when called with a <code>level</code> out of range. | |
3925 (You can call <code>getinfo</code> to check whether the level is valid.) | |
3926 Otherwise, it returns the name of the local variable. | |
3927 | |
3928 | |
3929 <p> | |
3930 See <a href="#pdf-debug.getlocal"><code>debug.getlocal</code></a> for more information about | |
3931 variable indices and names. | |
3932 | |
3933 | |
3934 | |
3935 | |
3936 <p> | |
3937 <hr><h3><a name="pdf-debug.setmetatable"><code>debug.setmetatable (value, table)</code></a></h3> | |
3938 | |
3939 | |
3940 <p> | |
3941 Sets the metatable for the given <code>value</code> to the given <code>table</code> | |
3942 (which can be <b>nil</b>). | |
3943 Returns <code>value</code>. | |
3944 | |
3945 | |
3946 | |
3947 | |
3948 <p> | |
3949 <hr><h3><a name="pdf-debug.setupvalue"><code>debug.setupvalue (f, up, value)</code></a></h3> | |
3950 | |
3951 | |
3952 <p> | |
3953 This function assigns the value <code>value</code> to the upvalue | |
3954 with index <code>up</code> of the function <code>f</code>. | |
3955 The function returns <b>nil</b> if there is no upvalue | |
3956 with the given index. | |
3957 Otherwise, it returns the name of the upvalue. | |
3958 | |
3959 | |
3960 | |
3961 | |
3962 <p> | |
3963 <hr><h3><a name="pdf-debug.setuservalue"><code>debug.setuservalue (udata, value)</code></a></h3> | |
3964 | |
3965 | |
3966 <p> | |
3967 Sets the given <code>value</code> as | |
3968 the Lua value associated to the given <code>udata</code>. | |
3969 <code>udata</code> must be a full userdata. | |
3970 | |
3971 | |
3972 <p> | |
3973 Returns <code>udata</code>. | |
3974 | |
3975 | |
3976 | |
3977 | |
3978 <p> | |
3979 <hr><h3><a name="pdf-debug.traceback"><code>debug.traceback ([thread,] [message [, level]])</code></a></h3> | |
3980 | |
3981 | |
3982 <p> | |
3983 If <code>message</code> is present but is neither a string nor <b>nil</b>, | |
3984 this function returns <code>message</code> without further processing. | |
3985 Otherwise, | |
3986 it returns a string with a traceback of the call stack. | |
3987 The optional <code>message</code> string is appended | |
3988 at the beginning of the traceback. | |
3989 An optional <code>level</code> number tells at which level | |
3990 to start the traceback | |
3991 (default is 1, the function calling <code>traceback</code>). | |
3992 | |
3993 | |
3994 | |
3995 | |
3996 <p> | |
3997 <hr><h3><a name="pdf-debug.upvalueid"><code>debug.upvalueid (f, n)</code></a></h3> | |
3998 | |
3999 | |
4000 <p> | |
4001 Returns a unique identifier (as a light userdata) | |
4002 for the upvalue numbered <code>n</code> | |
4003 from the given function. | |
4004 | |
4005 | |
4006 <p> | |
4007 These unique identifiers allow a program to check whether different | |
4008 closures share upvalues. | |
4009 Lua closures that share an upvalue | |
4010 (that is, that access a same external local variable) | |
4011 will return identical ids for those upvalue indices. | |
4012 | |
4013 | |
4014 | |
4015 | |
4016 <p> | |
4017 <hr><h3><a name="pdf-debug.upvaluejoin"><code>debug.upvaluejoin (f1, n1, f2, n2)</code></a></h3> | |
4018 | |
4019 | |
4020 <p> | |
4021 Make the <code>n1</code>-th upvalue of the Lua closure <code>f1</code> | |
4022 refer to the <code>n2</code>-th upvalue of the Lua closure <code>f2</code>. | |
4023 | |
4024 | |
4025 | |
4026 | |
4027 | |
4028 | |
4029 | |
4030 <h1>7 – <a name="7">Lua Standalone</a></h1> | |
4031 | |
4032 <p> | |
4033 Although Lua has been designed as an extension language, | |
4034 to be embedded in a host C program, | |
4035 it is also frequently used as a standalone language. | |
4036 An interpreter for Lua as a standalone language, | |
4037 called simply <code>lua</code>, | |
4038 is provided with the standard distribution. | |
4039 The standalone interpreter includes | |
4040 all standard libraries, including the debug library. | |
4041 Its usage is: | |
4042 | |
4043 <pre> | |
4044 lua [options] [script [args]] | |
4045 </pre><p> | |
4046 The options are: | |
4047 | |
4048 <ul> | |
4049 <li><b><code>-e <em>stat</em></code>: </b> executes string <em>stat</em>;</li> | |
4050 <li><b><code>-l <em>mod</em></code>: </b> "requires" <em>mod</em>;</li> | |
4051 <li><b><code>-i</code>: </b> enters interactive mode after running <em>script</em>;</li> | |
4052 <li><b><code>-v</code>: </b> prints version information;</li> | |
4053 <li><b><code>-E</code>: </b> ignores environment variables;</li> | |
4054 <li><b><code>--</code>: </b> stops handling options;</li> | |
4055 <li><b><code>-</code>: </b> executes <code>stdin</code> as a file and stops handling options.</li> | |
4056 </ul><p> | |
4057 After handling its options, <code>lua</code> runs the given <em>script</em>. | |
4058 When called without arguments, | |
4059 <code>lua</code> behaves as <code>lua -v -i</code> | |
4060 when the standard input (<code>stdin</code>) is a terminal, | |
4061 and as <code>lua -</code> otherwise. | |
4062 | |
4063 | |
4064 <p> | |
4065 When called without option <code>-E</code>, | |
4066 the interpreter checks for an environment variable <a name="pdf-LUA_INIT_5_3"><code>LUA_INIT_5_3</code></a> | |
4067 (or <a name="pdf-LUA_INIT"><code>LUA_INIT</code></a> if the versioned name is not defined) | |
4068 before running any argument. | |
4069 If the variable content has the format <code>@<em>filename</em></code>, | |
4070 then <code>lua</code> executes the file. | |
4071 Otherwise, <code>lua</code> executes the string itself. | |
4072 | |
4073 | |
4074 <p> | |
4075 When called with option <code>-E</code>, | |
4076 besides ignoring <code>LUA_INIT</code>, | |
4077 Lua also ignores | |
4078 the values of <code>LUA_PATH</code> and <code>LUA_CPATH</code>, | |
4079 setting the values of | |
4080 <a href="#pdf-package.path"><code>package.path</code></a> and <a href="#pdf-package.cpath"><code>package.cpath</code></a> | |
4081 with the default paths defined in <code>luaconf.h</code>. | |
4082 | |
4083 | |
4084 <p> | |
4085 All options are handled in order, except <code>-i</code> and <code>-E</code>. | |
4086 For instance, an invocation like | |
4087 | |
4088 <pre> | |
4089 $ lua -e'a=1' -e 'print(a)' script.lua | |
4090 </pre><p> | |
4091 will first set <code>a</code> to 1, then print the value of <code>a</code>, | |
4092 and finally run the file <code>script.lua</code> with no arguments. | |
4093 (Here <code>$</code> is the shell prompt. Your prompt may be different.) | |
4094 | |
4095 | |
4096 <p> | |
4097 Before running any code, | |
4098 <code>lua</code> collects all command-line arguments | |
4099 in a global table called <code>arg</code>. | |
4100 The script name goes to index 0, | |
4101 the first argument after the script name goes to index 1, | |
4102 and so on. | |
4103 Any arguments before the script name | |
4104 (that is, the interpreter name plus its options) | |
4105 go to negative indices. | |
4106 For instance, in the call | |
4107 | |
4108 <pre> | |
4109 $ lua -la b.lua t1 t2 | |
4110 </pre><p> | |
4111 the table is like this: | |
4112 | |
4113 <pre> | |
4114 arg = { [-2] = "lua", [-1] = "-la", | |
4115 [0] = "b.lua", | |
4116 [1] = "t1", [2] = "t2" } | |
4117 </pre><p> | |
4118 If there is no script in the call, | |
4119 the interpreter name goes to index 0, | |
4120 followed by the other arguments. | |
4121 For instance, the call | |
4122 | |
4123 <pre> | |
4124 $ lua -e "print(arg[1])" | |
4125 </pre><p> | |
4126 will print "<code>-e</code>". | |
4127 If there is a script, | |
4128 the script is called with parameters | |
4129 <code>arg[1]</code>, ···, <code>arg[#arg]</code>. | |
4130 (Like all chunks in Lua, | |
4131 the script is compiled as a vararg function.) | |
4132 | |
4133 | |
4134 <p> | |
4135 In interactive mode, | |
4136 Lua repeatedly prompts and waits for a line. | |
4137 After reading a line, | |
4138 Lua first try to interpret the line as an expression. | |
4139 If it succeeds, it prints its value. | |
4140 Otherwise, it interprets the line as a statement. | |
4141 If you write an incomplete statement, | |
4142 the interpreter waits for its completion | |
4143 by issuing a different prompt. | |
4144 | |
4145 | |
4146 <p> | |
4147 In case of unprotected errors in the script, | |
4148 the interpreter reports the error to the standard error stream. | |
4149 If the error object is not a string but | |
4150 has a metamethod <code>__to_string</code>, | |
4151 the interpreter calls this metamethod to produce the final message. | |
4152 Otherwise, the interpreter converts the error object to a string | |
4153 and adds a stack traceback to it. | |
4154 | |
4155 | |
4156 <p> | |
4157 When finishing normally, | |
4158 the interpreter closes its main Lua state | |
4159 (see <a href="#lua_close"><code>lua_close</code></a>). | |
4160 The script can avoid this step by | |
4161 calling <a href="#pdf-os.exit"><code>os.exit</code></a> to terminate. | |
4162 | |
4163 | |
4164 <p> | |
4165 To allow the use of Lua as a | |
4166 script interpreter in Unix systems, | |
4167 the standalone interpreter skips | |
4168 the first line of a chunk if it starts with <code>#</code>. | |
4169 Therefore, Lua scripts can be made into executable programs | |
4170 by using <code>chmod +x</code> and the <code>#!</code> form, | |
4171 as in | |
4172 | |
4173 <pre> | |
4174 #!/usr/local/bin/lua | |
4175 </pre><p> | |
4176 (Of course, | |
4177 the location of the Lua interpreter may be different in your machine. | |
4178 If <code>lua</code> is in your <code>PATH</code>, | |
4179 then | |
4180 | |
4181 <pre> | |
4182 #!/usr/bin/env lua | |
4183 </pre><p> | |
4184 is a more portable solution.) | |
4185 | |
4186 | |
4187 | |
4188 <h1>8 – <a name="8">Incompatibilities with the Previous Version</a></h1> | |
4189 | |
4190 <p> | |
4191 Here we list the incompatibilities that you may find when moving a program | |
4192 from Lua 5.2 to Lua 5.3. | |
4193 You can avoid some incompatibilities by compiling Lua with | |
4194 appropriate options (see file <code>luaconf.h</code>). | |
4195 However, | |
4196 all these compatibility options will be removed in the future. | |
4197 | |
4198 | |
4199 <p> | |
4200 Lua versions can always change the C API in ways that | |
4201 do not imply source-code changes in a program, | |
4202 such as the numeric values for constants | |
4203 or the implementation of functions as macros. | |
4204 Therefore, | |
4205 you should not assume that binaries are compatible between | |
4206 different Lua versions. | |
4207 Always recompile clients of the Lua API when | |
4208 using a new version. | |
4209 | |
4210 | |
4211 <p> | |
4212 Similarly, Lua versions can always change the internal representation | |
4213 of precompiled chunks; | |
4214 precompiled chunks are not compatible between different Lua versions. | |
4215 | |
4216 | |
4217 <p> | |
4218 The standard paths in the official distribution may | |
4219 change between versions. | |
4220 | |
4221 | |
4222 | |
4223 <h2>8.1 – <a name="8.1">Changes in the Language</a></h2> | |
4224 <ul> | |
4225 | |
4226 <li> | |
4227 The main difference between Lua 5.2 and Lua 5.3 is the | |
4228 introduction of an integer subtype for numbers. | |
4229 Although this change should not affect "normal" computations, | |
4230 some computations | |
4231 (mainly those that involve some kind of overflow) | |
4232 can give different results. | |
4233 | |
4234 | |
4235 <p> | |
4236 You can fix these differences by forcing a number to be a float | |
4237 (in Lua 5.2 all numbers were float), | |
4238 in particular writing constants with an ending <code>.0</code> | |
4239 or using <code>x = x + 0.0</code> to convert a variable. | |
4240 (This recommendation is only for a quick fix | |
4241 for an occasional incompatibility; | |
4242 it is not a general guideline for good programming. | |
4243 For good programming, | |
4244 use floats where you need floats | |
4245 and integers where you need integers.) | |
4246 </li> | |
4247 | |
4248 <li> | |
4249 The conversion of a float to a string now adds a <code>.0</code> suffix | |
4250 to the result if it looks like an integer. | |
4251 (For instance, the float 2.0 will be printed as <code>2.0</code>, | |
4252 not as <code>2</code>.) | |
4253 You should always use an explicit format | |
4254 when you need a specific format for numbers. | |
4255 | |
4256 | |
4257 <p> | |
4258 (Formally this is not an incompatibility, | |
4259 because Lua does not specify how numbers are formatted as strings, | |
4260 but some programs assumed a specific format.) | |
4261 </li> | |
4262 | |
4263 <li> | |
4264 The generational mode for the garbage collector was removed. | |
4265 (It was an experimental feature in Lua 5.2.) | |
4266 </li> | |
4267 | |
4268 </ul> | |
4269 | |
4270 | |
4271 | |
4272 | |
4273 <h2>8.2 – <a name="8.2">Changes in the Libraries</a></h2> | |
4274 <ul> | |
4275 | |
4276 <li> | |
4277 The <code>bit32</code> library has been deprecated. | |
4278 It is easy to require a compatible external library or, | |
4279 better yet, to replace its functions with appropriate bitwise operations. | |
4280 (Keep in mind that <code>bit32</code> operates on 32-bit integers, | |
4281 while the bitwise operators in standard Lua operate on 64-bit integers.) | |
4282 </li> | |
4283 | |
4284 <li> | |
4285 The Table library now respects metamethods | |
4286 for setting and getting elements. | |
4287 </li> | |
4288 | |
4289 <li> | |
4290 The <a href="#pdf-ipairs"><code>ipairs</code></a> iterator now respects metamethods and | |
4291 its <code>__ipairs</code> metamethod has been deprecated. | |
4292 </li> | |
4293 | |
4294 <li> | |
4295 Option names in <a href="#pdf-io.read"><code>io.read</code></a> do not have a starting '<code>*</code>' anymore. | |
4296 For compatibility, Lua will continue to ignore this character. | |
4297 </li> | |
4298 | |
4299 <li> | |
4300 The following functions were deprecated in the mathematical library: | |
4301 <code>atan2</code>, <code>cosh</code>, <code>sinh</code>, <code>tanh</code>, <code>pow</code>, | |
4302 <code>frexp</code>, and <code>ldexp</code>. | |
4303 You can replace <code>math.pow(x,y)</code> with <code>x^y</code>; | |
4304 you can replace <code>math.atan2</code> with <code>math.atan</code>, | |
4305 which now accepts one or two parameters; | |
4306 you can replace <code>math.ldexp(x,exp)</code> with <code>x * 2.0^exp</code>. | |
4307 For the other operations, | |
4308 you can either use an external library or | |
4309 implement them in Lua. | |
4310 </li> | |
4311 | |
4312 <li> | |
4313 The searcher for C loaders used by <a href="#pdf-require"><code>require</code></a> | |
4314 changed the way it handles versioned names. | |
4315 Now, the version should come after the module name | |
4316 (as is usual in most other tools). | |
4317 For compatibility, that searcher still tries the old format | |
4318 if it cannot find an open function according to the new style. | |
4319 (Lua 5.2 already worked that way, | |
4320 but it did not document the change.) | |
4321 </li> | |
4322 | |
4323 </ul> | |
4324 | |
4325 | |
4326 | |
4327 | |
4328 <h2>8.3 – <a name="8.3">Changes in the API</a></h2> | |
4329 | |
4330 | |
4331 <ul> | |
4332 | |
4333 <li> | |
4334 Continuation functions now receive as parameters what they needed | |
4335 to get through <code>lua_getctx</code>, | |
4336 so <code>lua_getctx</code> has been removed. | |
4337 Adapt your code accordingly. | |
4338 </li> | |
4339 | |
4340 <li> | |
4341 Function <a href="#lua_dump"><code>lua_dump</code></a> has an extra parameter, <code>strip</code>. | |
4342 Use 0 as the value of this parameter to get the old behavior. | |
4343 </li> | |
4344 | |
4345 <li> | |
4346 Functions to inject/project unsigned integers | |
4347 (<code>lua_pushunsigned</code>, <code>lua_tounsigned</code>, <code>lua_tounsignedx</code>, | |
4348 <code>luaL_checkunsigned</code>, <code>luaL_optunsigned</code>) | |
4349 were deprecated. | |
4350 Use their signed equivalents with a type cast. | |
4351 </li> | |
4352 | |
4353 <li> | |
4354 Macros to project non-default integer types | |
4355 (<code>luaL_checkint</code>, <code>luaL_optint</code>, <code>luaL_checklong</code>, <code>luaL_optlong</code>) | |
4356 were deprecated. | |
4357 Use their equivalent over <a href="#lua_Integer"><code>lua_Integer</code></a> with a type cast | |
4358 (or, when possible, use <a href="#lua_Integer"><code>lua_Integer</code></a> in your code). | |
4359 </li> | |
4360 | |
4361 </ul> | |
4362 | |
4363 | |
4364 | |
4365 | |
4366 <h1>9 – <a name="9">The Complete Syntax of Lua</a></h1> | |
4367 | |
4368 <p> | |
4369 Here is the complete syntax of Lua in extended BNF. | |
4370 As usual in extended BNF, | |
4371 {A} means 0 or more As, | |
4372 and [A] means an optional A. | |
4373 (For operator precedences, see <a href="#3.4.8">§3.4.8</a>; | |
4374 for a description of the terminals | |
4375 Name, Numeral, | |
4376 and LiteralString, see <a href="#3.1">§3.1</a>.) | |
4377 | |
4378 | |
4379 | |
4380 | |
4381 <pre> | |
4382 | |
4383 chunk ::= block | |
4384 | |
4385 block ::= {stat} [retstat] | |
4386 | |
4387 stat ::= ‘<b>;</b>’ | | |
4388 varlist ‘<b>=</b>’ explist | | |
4389 functioncall | | |
4390 label | | |
4391 <b>break</b> | | |
4392 <b>goto</b> Name | | |
4393 <b>do</b> block <b>end</b> | | |
4394 <b>while</b> exp <b>do</b> block <b>end</b> | | |
4395 <b>repeat</b> block <b>until</b> exp | | |
4396 <b>if</b> exp <b>then</b> block {<b>elseif</b> exp <b>then</b> block} [<b>else</b> block] <b>end</b> | | |
4397 <b>for</b> Name ‘<b>=</b>’ exp ‘<b>,</b>’ exp [‘<b>,</b>’ exp] <b>do</b> block <b>end</b> | | |
4398 <b>for</b> namelist <b>in</b> explist <b>do</b> block <b>end</b> | | |
4399 <b>function</b> funcname funcbody | | |
4400 <b>local</b> <b>function</b> Name funcbody | | |
4401 <b>local</b> namelist [‘<b>=</b>’ explist] | |
4402 | |
4403 retstat ::= <b>return</b> [explist] [‘<b>;</b>’] | |
4404 | |
4405 label ::= ‘<b>::</b>’ Name ‘<b>::</b>’ | |
4406 | |
4407 funcname ::= Name {‘<b>.</b>’ Name} [‘<b>:</b>’ Name] | |
4408 | |
4409 varlist ::= var {‘<b>,</b>’ var} | |
4410 | |
4411 var ::= Name | prefixexp ‘<b>[</b>’ exp ‘<b>]</b>’ | prefixexp ‘<b>.</b>’ Name | |
4412 | |
4413 namelist ::= Name {‘<b>,</b>’ Name} | |
4414 | |
4415 explist ::= exp {‘<b>,</b>’ exp} | |
4416 | |
4417 exp ::= <b>nil</b> | <b>false</b> | <b>true</b> | Numeral | LiteralString | ‘<b>...</b>’ | functiondef | | |
4418 prefixexp | tableconstructor | exp binop exp | unop exp | |
4419 | |
4420 prefixexp ::= var | functioncall | ‘<b>(</b>’ exp ‘<b>)</b>’ | |
4421 | |
4422 functioncall ::= prefixexp args | prefixexp ‘<b>:</b>’ Name args | |
4423 | |
4424 args ::= ‘<b>(</b>’ [explist] ‘<b>)</b>’ | tableconstructor | LiteralString | |
4425 | |
4426 functiondef ::= <b>function</b> funcbody | |
4427 | |
4428 funcbody ::= ‘<b>(</b>’ [parlist] ‘<b>)</b>’ block <b>end</b> | |
4429 | |
4430 parlist ::= namelist [‘<b>,</b>’ ‘<b>...</b>’] | ‘<b>...</b>’ | |
4431 | |
4432 tableconstructor ::= ‘<b>{</b>’ [fieldlist] ‘<b>}</b>’ | |
4433 | |
4434 fieldlist ::= field {fieldsep field} [fieldsep] | |
4435 | |
4436 field ::= ‘<b>[</b>’ exp ‘<b>]</b>’ ‘<b>=</b>’ exp | Name ‘<b>=</b>’ exp | exp | |
4437 | |
4438 fieldsep ::= ‘<b>,</b>’ | ‘<b>;</b>’ | |
4439 | |
4440 binop ::= ‘<b>+</b>’ | ‘<b>-</b>’ | ‘<b>*</b>’ | ‘<b>/</b>’ | ‘<b>//</b>’ | ‘<b>^</b>’ | ‘<b>%</b>’ | | |
4441 ‘<b>&</b>’ | ‘<b>~</b>’ | ‘<b>|</b>’ | ‘<b>>></b>’ | ‘<b><<</b>’ | ‘<b>..</b>’ | | |
4442 ‘<b><</b>’ | ‘<b><=</b>’ | ‘<b>></b>’ | ‘<b>>=</b>’ | ‘<b>==</b>’ | ‘<b>~=</b>’ | | |
4443 <b>and</b> | <b>or</b> | |
4444 | |
4445 unop ::= ‘<b>-</b>’ | <b>not</b> | ‘<b>#</b>’ | ‘<b>~</b>’ | |
4446 | |
4447 </pre> | |
4448 | |
4449 <p> | |
4450 | |
4451 | |
4452 | |
4453 | |
4454 | |
4455 | |
4456 | |
4457 | |
4458 <HR> | |
4459 <SMALL CLASS="footer"> | |
4460 Last update: | |
4461 Fri Jan 16 00:58:20 BRST 2015 | |
4462 </SMALL> | |
4463 <!-- | |
4464 Last change: minor edit | |
4465 --> | |
4466 | |
4467 </div> | |
4468 </body> | |
4469 </html> | |
4470 <% | |
4471 end |