1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
|
<html>
<head>
<title>grammar</title>
<link rel="stylesheet" href="doc.css" type="text/css"/>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8"/>
</head>
<body>
<div id="container">
<div id="product">
<p/>
<div id="product_logo">
<a href="http://www.lua.org"> <img alt="Lua logo" src="lua.gif"/> </a>
</div>
<div id="product_name"><big><strong>grammar</strong></big></div>
<div id="product_description"> LPeg grammar manipulation </div>
<br/>
<div id="product_description"> <small><em> <b>Version:</b> 0.1.2 </em></small></div>
<div id="product_description"> <small><em><b>Generated:</b> November 26, 2007 </em></small> </div>
</div> <!-- id="product" -->
<div id="main">
<div id="navigation">
<h1><a href="index.html">Modules</a></h1>
<ul>
<li><a href="grammar.html">grammar</a></li>
<li><a href="parser.html">parser</a></li>
<li><a href="scanner.html">scanner</a></li>
</ul>
<br/>
<h1>grammar</h1>
<ul>
<!--
<li><a href="#section_Description"> Description </a>
</li>
-->
<li><a href="#section_Description"> Description </a>
</li>
<!--
<li><a href="#section_Dependencies"> Dependencies </a>
</li>
-->
<li><a href="#section_Dependencies"> Dependencies </a>
</li>
<!--
<li><a href="#section_Operations"> Operations </a>
<ul>
<li><a href="#section_Operations_Piping"> Piping </a></li>
<li><a href="#section_Operations_Completing"> Completing </a></li>
<li><a href="#section_Operations_Applying"> Applying </a></li>
<li><a href="#section_Operations_Example"> Example </a></li>
</ul>
</li>
-->
<li><a href="#section_Operations"> Operations </a>
<ul>
<li><a href="#section_Piping"> Piping </a></li>
<li><a href="#section_Completing"> Completing </a></li>
<li><a href="#section_Applying"> Applying </a></li>
<li><a href="#section_Example"> Example </a></li>
</ul>
</li>
<li><a href="#functions"> Functions </a>
<ul>
<li><code><a href="#function_anyOf"> anyOf </a></code></li>
<li><code><a href="#function_apply"> apply </a></code></li>
<li><code><a href="#function_C"> C </a></code></li>
<li><code><a href="#function_complete"> complete </a></code></li>
<li><code><a href="#function_copy"> copy </a></code></li>
<li><code><a href="#function_Ct"> Ct </a></code></li>
<li><code><a href="#function_listOf"> listOf </a></code></li>
<li><code><a href="#function_pipe"> pipe </a></code></li>
</ul>
</li>
</ul>
</div> <!-- id="navigation" -->
<div id="content">
<p/>
<p/>
<h1><a name="section_Description"></a>Description</h1>
<p/>
This module defines a handful of operations which can be applied to
<a href="http://www.inf.puc-rio.br/~roberto/lpeg.html">LPeg</a> patterns and grammars in
general.
<p/>
<h1><a name="section_Dependencies"></a>Dependencies</h1>
<ul>
<li><a href="http://www.inf.puc-rio.br/~roberto/lpeg.html">LPeg</a>.</li></ul>
<p/>
<h1><a name="section_Operations"></a>Operations</h1>
<p/>
<h2><a name="section_Piping"></a>Piping</h2>
<p/>
Pattern matching dissociates the notion of <em>matching</em> from the notion of
<em>capturing</em>: matching checks if a given string follows a certain pattern,
and capturing generates values according to the match made. This division
allows interesting possibilities:
<ul>
<li>different problems can be solved by applying different captures to the same grammar;</li>
<li>captures may be defined separately;</li>
<li>captures may be done on top of other captures.</li></ul>
<p/>
Accounting for the first and second bullets, the grammar given in
<a href="parser.html">parser</a> has no captures, enabling the user to reuse it to solve any
problems that require a Lua grammar. One good example is documentation
generation, described in a little more detail <a href="#section_Example">below</a>.
<p/>
The third bullet depicts a more interesting idea: a capture might take the
result of another capture as input, doing a further transformation of the
original data. This capture chaining, with the latter ones using the former's
output as its input, is very similar to <a href="http://en.wikipedia.org/wiki/Pipeline_%28Unix%29">Unix pipelines</a>,
so this mechanism was named <strong>piping</strong>.
<p/>
<h2><a name="section_Completing"></a>Completing</h2>
<p/>
With piping, several levels of captures can be chained together up to the
most appropriate for the task at hand. Yet some levels might require extra rules, and modifications to existing ones, to ensure proper matching.
<p/>
To avoid manual copying, the new grammar should redefine only the necessary
rules, copying the rest from the older grammar. This action is dubbed
<strong>completing</strong>.
<p/>
<h2><a name="section_Applying"></a>Applying</h2>
<p/>
Once a new rule set is created and <a href="#section_Completing">completed</a>, and
all captures are correctly <a href="#section_Piping">piped</a>, all that's left is
to put them together, a process called <strong>applying</strong>. The result is a grammar ready for <a href="http://www.inf.puc-rio.br/~roberto/lpeg.html#lpeg">lpeg.P</a>
consumption, whose pattern will return the intended result when a match is made.
<p/>
<h2><a name="section_Example"></a>Example</h2>
<p/>
Let's consider the problem of documenting a Lua module. In this case, comments
must be captured before every function declaration when in the outermost scope:
<p/>
<pre class="example">
<font color="#808080"> -- the code to parse</font>
subject = <font color="#008000">[[
-- Calculates the sum a+b.
-- An extra line.
function sum (a, b)
-- code
end
-- f1: assume a variable assignment is not a proper declaration for an
-- exported function
f1 = function ()
-- code
end
while true do
-- this function is not in the outermost scope
function aux() end
end
function something:other(a, ...)
-- a global function without comments
end
]]</font>
</pre>
<p/>
In the code above only <code>sum</code> and <code>something:other</code> should be documented, as <code>f1</code> isn't properly (by our standards) declared and <code>aux</code> is not in the outermost scope.
<p/>
By combining <a href="http://www.inf.puc-rio.br/~roberto/lpeg.html">LPeg</a> and the modules <a href="scanner.html">scanner</a>, <a href="parser.html">parser</a> and <a href="grammar.html">grammar</a>, this specific problem can be solved as follows:
<p/>
<pre class="example">
<font color="#808080"> -- change only the initial rule and make no captures</font>
patt = <b>grammar.apply</b>(parser.rules, scanner.COMMENT^-<font color="#B00000"><i>1</i></font> * <b>lpeg.V</b><font color="#008000">'GlobalFunction'</font>, <font color="#0000FF"><b>nil</b></font>)
<font color="#808080"> -- transform the new grammar into a LPeg pattern</font>
patt = <b>lpeg.P</b>(patt)
<font color="#808080"> -- making a pattern that matches any Lua statement, also without captures</font>
Stat = <b>lpeg.P</b>( <b>grammar.apply</b>(parser.rules, <b>lpeg.V</b><font color="#008000">'Stat'</font>, <font color="#0000FF"><b>nil</b></font>) )
<font color="#808080"> -- a pattern which matches function declarations and skips statements in</font>
<font color="#808080"> -- inner scopes or undesired tokens</font>
patt = (patt + Stat + scanner.ANY)^<font color="#B00000"><i>0</i></font>
<font color="#808080"> -- matching a string</font>
<b>patt:match</b>(subject)
</pre>
<p/>
These are the relevant rules in <a href="parser.html#section_The_Grammar">the grammar</a>:
<p/>
<pre class="example">
GlobalFunction = <font color="#008000">'function'</font> * FuncName * FuncBody
FuncName = ID * (<font color="#008000">'.'</font> * ID)^<font color="#B00000"><i>0</i></font> * (<font color="#008000">':'</font> * ID)^-<font color="#B00000"><i>1</i></font>
FuncBody = <font color="#008000">'('</font> * (ParList + EPSILON) * <font color="#008000">')'</font> * Block * <font color="#008000">'end'</font>
ParList = NameList * (<font color="#008000">','</font> * <font color="#008000">'...'</font>)^-<font color="#B00000"><i>1</i></font>
NameList = ID * (<font color="#008000">','</font> * ID)^<font color="#B00000"><i>0</i></font>
ID = scanner.ID
EPSILON = <b>lpeg.P</b>(<font color="#0000FF"><b>true</b></font>)
</pre>
<p/>
It may seem that <code>ParList + EPSILON</code> could be substituted for <code>ParList^-1</code> (optionally match <code>ParList</code>), but then no captures would be made for empty parameter lists, and <code>GlobalFunction</code> would get all strings matched by <code>FuncBody</code>. The <code>EPSILON</code> rule acts in this manner as a placeholder in the argument list, avoiding any argument list processing in the capture function.
<p/>
Since no captures are being made, <a href="http://www.inf.puc-rio.br/~roberto/lpeg.html#basic">lpeg.match</a> doesn't return anything interesting. Here are some possible captures:
<p/>
<pre class="example">
<font color="#808080"> -- some interesting captures bundled up in a table. Note that the table keys</font>
<font color="#808080"> -- match the grammar rules we want to add captures to. Whatever rules aren't in</font>
<font color="#808080"> -- the rules table below will come from parser.rules .</font>
captures = {
[<font color="#B00000"><i>1</i></font>] = <font color="#0000FF"><b>function</b></font> (...) <font color="#808080"> -- the initial rule</font>
<font color="#0000FF"><b>return</b></font> <font color="#008000">'<function>'</font>..<font color = "#800040"><b>table.concat</b></font>{...}..<font color="#008000">'</function>'</font>
<font color="#0000FF"><b>end</b></font>,
GlobalFunction = <font color="#0000FF"><b>function</b></font> (name, parlist)
<font color="#0000FF"><b>return</b></font> <font color="#008000">'<name>'</font>..name..<font color="#008000">'</name><parlist>'</font>..(parlist <font color="#0000FF"><b>or</b></font> <font color="#008000">''</font>)..<font color="#008000">'</parlist>'</font>
<font color="#0000FF"><b>end</b></font>,
FuncName = grammar.C, <font color="#808080"> -- capture the raw text</font>
ParList = grammar.C, <font color="#808080"> -- capture the raw text</font>
COMMENT = scanner.comment2text, <font color="#808080"> -- remove the comment trappings</font>
}
<font color="#808080"> -- spacing rule</font>
<font color="#0000FF"><b>local</b></font> S = scanner.SPACE ^ <font color="#B00000"><i>0</i></font>
<font color="#808080"> -- rules table</font>
rules = {
[<font color="#B00000"><i>1</i></font>] = ((<b>lpeg.V</b><font color="#008000">'COMMENT'</font> *S) ^ <font color="#B00000"><i>0</i></font>) *S* <b>lpeg.V</b><font color="#008000">'GlobalFunction'</font>,
COMMENT = scanner.COMMENT,
}
<font color="#808080"> -- building the new grammar and adding the captures</font>
patt = <b>lpeg.P</b>( <b>grammar.apply</b>(parser.rules, rules, captures) )
<font color="#808080"> -- a pattern that matches a sequence of patts and concatenates the results</font>
patt = (patt + Stat + scanner.ANY)^<font color="#B00000"><i>0</i></font> / <font color="#0000FF"><b>function</b></font>(...)
<font color="#0000FF"><b>return</b></font> <font color = "#800040"><b>table.concat</b></font>({...}, <font color="#008000">'\n\n'</font>) <font color="#808080"> -- some line breaks for easier reading</font>
<font color="#0000FF"><b>end</b></font>
<font color="#808080"> -- finally, matching a string</font>
<b><font color ="#800040">print</font></b>(<b>patt:match</b>(subject))
</pre>
<p/>
<code>FuncBody</code> needs no captures, as <code>Block</code> and all its non-terminals have none; it
just needs to pass along any captures made by <code>ParList</code>. <code>NameList</code> and <code>ID</code> also have no captures, and the whole subject string is passed further.
<p/>
The printed result is:
<pre class="example">
<function>Calculates the sum a+b. An extra line.<name>sum</name><parlist>a, b</parlist></function>
<br/>
<function><name>something:other</name><parlist>a, ...</parlist></function>
</pre>
<p/>
<p/>
<p/>
<hr/>
<h1><a name="functions"></a> Functions </h1>
<table border="0" width="95%">
<tr>
<!-- <td> <pre class="example"><big><strong><a href="#functions_anyOf">anyOf</a></strong></big> (list)</pre> </td> -->
<td> <code><big><strong> <a href="#function_anyOf">anyOf</a></strong></big> (list) </code> </td>
<td> Returns a pattern which matches any of the patterns received. </td>
</tr>
<tr>
<!-- <td> <pre class="example"><big><strong><a href="#functions_apply">apply</a></strong></big> (grammar, rules, captures)</pre> </td> -->
<td> <code><big><strong> <a href="#function_apply">apply</a></strong></big> (grammar, rules, captures) </code> </td>
<td> <a href="#section_Completing">Completes</a> <code>rules</code> with <code>grammar</code> and then <a href="#Applying">applies</a> <code>captures</code>. </td>
</tr>
<tr>
<!-- <td> <pre class="example"><big><strong><a href="#functions_C">C</a></strong></big> ()</pre> </td> -->
<td> <code><big><strong> <a href="#function_C">C</a></strong></big> () </code> </td>
<td> A capture function, made so that <code>patt / C</code> is equivalent to <code>m.C(patt)</code>. It's intended to be used in capture tables, such as those required by <a href="#function_pipe">pipe</a> and <a href="#function_apply">apply</a>. </td>
</tr>
<tr>
<!-- <td> <pre class="example"><big><strong><a href="#functions_complete">complete</a></strong></big> (dest, orig)</pre> </td> -->
<td> <code><big><strong> <a href="#function_complete">complete</a></strong></big> (dest, orig) </code> </td>
<td> <a href="#section_Completing">Completes</a> <code>dest</code> with <code>orig</code>. </td>
</tr>
<tr>
<!-- <td> <pre class="example"><big><strong><a href="#functions_copy">copy</a></strong></big> (grammar)</pre> </td> -->
<td> <code><big><strong> <a href="#function_copy">copy</a></strong></big> (grammar) </code> </td>
<td> Creates a shallow copy of <code>grammar</code>. </td>
</tr>
<tr>
<!-- <td> <pre class="example"><big><strong><a href="#functions_Ct">Ct</a></strong></big> ()</pre> </td> -->
<td> <code><big><strong> <a href="#function_Ct">Ct</a></strong></big> () </code> </td>
<td> A capture function, made so that <code>patt / Ct</code> is equivalent to <code>m.Ct(patt)</code>. It's intended to be used in capture tables, such as those required by <a href="#function_pipe">pipe</a> and <a href="#function_apply">apply</a>. </td>
</tr>
<tr>
<!-- <td> <pre class="example"><big><strong><a href="#functions_listOf">listOf</a></strong></big> (patt, sep)</pre> </td> -->
<td> <code><big><strong> <a href="#function_listOf">listOf</a></strong></big> (patt, sep) </code> </td>
<td> Returns a pattern which matches a list of <code>patt</code>s, separated by <code>sep</code>. </td>
</tr>
<tr>
<!-- <td> <pre class="example"><big><strong><a href="#functions_pipe">pipe</a></strong></big> (dest, orig)</pre> </td> -->
<td> <code><big><strong> <a href="#function_pipe">pipe</a></strong></big> (dest, orig) </code> </td>
<td> <a href="#section_Piping">Pipes</a> the captures in <code>orig</code> to the ones in <code>dest</code>. </td>
</tr>
</table>
<p/><a name="function_anyOf"></a>
<hr/><code><big>anyOf (list)</big></code>
<ul>Returns a pattern which matches any of the patterns received.
<p/>
<strong>Example:</strong>
<pre class="example">
<font color="#0000FF"><b>local</b></font> g, s, m = <b><font color ="#800040">require</font></b> <font color="#008000">'leg.grammar'</font>, <b><font color ="#800040">require</font></b> <font color="#008000">'leg.scanner'</font>, <b><font color ="#800040">require</font></b> <font color="#008000">'lpeg'</font>
<font color="#808080"> -- match numbers or operators, capture the numbers</font>
<b><font color ="#800040">print</font></b>( (<b>g.anyOf</b> { <font color="#008000">'+'</font>, <font color="#008000">'-'</font>, <font color="#008000">'*'</font>, <font color="#008000">'/'</font>, <b>m.C</b>(s.NUMBER) }):match <font color="#008000">'34.5@23 * 56 / 45 - 45'</font> )
<font color="#808080"> --> prints 34.5</font>
</pre>
<p/>
<strong>Parameters:</strong><ul>
<li><code>list</code>: a list of zero or more LPeg patterns or values which can be fed to <a href="http://www.inf.puc-rio.br/~roberto/lpeg.html#lpeg">lpeg.P</a>.</li></ul>
<p/>
<strong>Returns:</strong><ul>
<li>a pattern which matches any of the patterns received.</li></ul></ul>
<p/><a name="function_apply"></a>
<hr/><code><big>apply (grammar, rules, captures)</big></code>
<ul><a href="#section_Completing">Completes</a> <code>rules</code> with <code>grammar</code> and then <a href="#Applying">applies</a> <code>captures</code>.
<p/>
<code>rules</code> can either be:<ul>
<li>a single pattern, which is taken to be the new initial rule, </li>
<li>a possibly incomplete LPeg grammar, as per <a href="#function_complete">complete</a>, or </li>
<li><code>nil</code>, which means no new rules are added.</li></ul>
<p/>
<code>captures</code> can either be:<ul>
<li>a capture table, as per <a href="#function_pipe">pipe</a>, or</li>
<li><code>nil</code>, which means no captures are applied.</li></ul>
<p/>
<strong>Parameters:</strong><ul>
<li><code>grammar</code>: the old grammar. It stays unmodified.</li>
<li><code>rules</code>: optional, the new rules. </li>
<li><code>captures</code>: optional, the final capture table.</li></ul>
<p/>
<strong>Returns:</strong><ul>
<li><code>rules</code>, suitably augmented by <code>grammar</code> and <code>captures</code>.</li></ul></ul>
<p/><a name="function_C"></a>
<hr/><code><big>C ()</big></code>
<ul>A capture function, made so that <code>patt / C</code> is equivalent to <code>m.C(patt)</code>. It's intended to be used in capture tables, such as those required by <a href="#function_pipe">pipe</a> and <a href="#function_apply">apply</a>.</ul>
<p/><a name="function_complete"></a>
<hr/><code><big>complete (dest, orig)</big></code>
<ul><a href="#section_Completing">Completes</a> <code>dest</code> with <code>orig</code>.
<p/>
<strong>Parameters:</strong><ul>
<li><code>dest</code>: the new grammar. Must be a table.</li>
<li><code>orig</code>: the original grammar. Must be a table.</li></ul>
<p/>
<strong>Returns:</strong><ul>
<li><code>dest</code>, with new rules inherited from <code>orig</code>.</li></ul></ul>
<p/><a name="function_copy"></a>
<hr/><code><big>copy (grammar)</big></code>
<ul>Creates a shallow copy of <code>grammar</code>.
<p/>
<strong>Parameters:</strong><ul>
<li><code>grammar</code>: a regular table.</li></ul>
<p/>
<strong>Returns:</strong><ul>
<li>a newly created table, with <code>grammar</code>'s keys and values.</li></ul></ul>
<p/><a name="function_Ct"></a>
<hr/><code><big>Ct ()</big></code>
<ul>A capture function, made so that <code>patt / Ct</code> is equivalent to <code>m.Ct(patt)</code>. It's intended to be used in capture tables, such as those required by <a href="#function_pipe">pipe</a> and <a href="#function_apply">apply</a>.</ul>
<p/><a name="function_listOf"></a>
<hr/><code><big>listOf (patt, sep)</big></code>
<ul>Returns a pattern which matches a list of <code>patt</code>s, separated by <code>sep</code>.
<p/>
<strong>Example:</strong> matching comma-separated values:
<pre class="example">
<font color="#0000FF"><b>local</b></font> g, m = <b><font color ="#800040">require</font></b> <font color="#008000">'leg.grammar'</font>, <b><font color ="#800040">require</font></b> <font color="#008000">'lpeg'</font>
<font color="#808080"> -- separator</font>
<font color="#0000FF"><b>local</b></font> sep = <b>m.P</b><font color="#008000">','</font> + <b>m.P</b><font color="#008000">'\n'</font>
<font color="#808080"> -- element: anything but sep, capture it</font>
<font color="#0000FF"><b>local</b></font> elem = <b>m.C</b>((<font color="#B00000"><i>1</i></font> - sep)^<font color="#B00000"><i>0</i></font>)
<font color="#808080"> -- pattern</font>
<font color="#0000FF"><b>local</b></font> patt = <b>g.listOf</b>(elem, sep)
<font color="#808080"> -- matching</font>
<b><font color ="#800040">print</font></b>( <b>patt:match</b> <font color="#008000">[[a, b, 'christmas eve'
d, evening; mate!
f]]</font>)
<font color="#808080"> --> prints out "a b 'christmas eve' d evening; mate! f"</font>
</pre>
<p/>
<strong>Parameters:</strong><ul>
<li><code>patt</code>: a LPeg pattern.</li>
<li><code>sep</code>: a LPeg pattern.</li></ul>
<p/>
<strong>Returns:</strong><ul>
<li>the following pattern: <pre class="example">patt * (sep * patt)^<font color="#B00000"><i>0</i></font></pre></li></ul></ul>
<p/><a name="function_pipe"></a>
<hr/><code><big>pipe (dest, orig)</big></code>
<ul><a href="#section_Piping">Pipes</a> the captures in <code>orig</code> to the ones in <code>dest</code>.
<p/>
<code>dest</code> and <code>orig</code> should be tables, with each key storing a capture function. Each capture in <code>dest</code> will be altered to use the results for the matching one in <code>orig</code> as input, using function composition. Should <code>orig</code> possess keys not in <code>dest</code>, <code>dest</code> will copy them.
<p/>
<strong>Parameters:</strong><ul>
<li><code>dest</code>: a capture table.</li>
<li><code>orig</code>: a capture table.</li></ul>
<p/>
<strong>Returns:</strong><ul>
<li><code>dest</code>, suitably modified.</li></ul></ul>
<hr/>
</div> <!-- id="content" -->
</div> <!-- id="main" -->
</div> <!-- id="container" -->
</body>
</html>
|
