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
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
|
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8"/>
<head>
<title>Penlight Documentation</title>
<link rel="stylesheet" href="../ldoc.css" type="text/css" />
</head>
<body>
<div id="container">
<div id="product">
<div id="product_logo"></div>
<div id="product_name"><big><b></b></big></div>
<div id="product_description"></div>
</div> <!-- id="product" -->
<div id="main">
<!-- Menu -->
<div id="navigation">
<br/>
<h1>Penlight</h1>
<ul>
<li><a href="../index.html">Index</a></li>
</ul>
<h2>Contents</h2>
<ul>
<li><a href="#Reading_Data_Files">Reading Data Files</a></li>
<li><a href="#Reading_Unstructured_Text_Data">Reading Unstructured Text Data</a></li>
<li><a href="#Reading_Columnar_Data">Reading Columnar Data</a></li>
<li><a href="#Reading_Configuration_Files">Reading Configuration Files</a></li>
<li><a href="#Lexical_Scanning">Lexical Scanning</a></li>
<li><a href="#XML">XML</a></li>
</ul>
<h2>Topics</h2>
<ul>
<li><a href="../topics/01-introduction.md.html">01-introduction.md</a></li>
<li><a href="../topics/02-arrays.md.html">02-arrays.md</a></li>
<li><a href="../topics/03-strings.md.html">03-strings.md</a></li>
<li><a href="../topics/04-paths.md.html">04-paths.md</a></li>
<li><a href="../topics/05-dates.md.html">05-dates.md</a></li>
<li><strong>06-data.md</strong></li>
<li><a href="../topics/07-functional.md.html">07-functional.md</a></li>
<li><a href="../topics/08-additional.md.html">08-additional.md</a></li>
<li><a href="../topics/09-discussion.md.html">09-discussion.md</a></li>
</ul>
<h2>Modules</h2>
<ul>
<li><a href="../modules/pl.html">pl</a></li>
<li><a href="../modules/pl.Date.html">pl.Date</a></li>
<li><a href="../modules/pl.List.html">pl.List</a></li>
<li><a href="../modules/pl.Map.html">pl.Map</a></li>
<li><a href="../modules/pl.MultiMap.html">pl.MultiMap</a></li>
<li><a href="../modules/pl.OrderedMap.html">pl.OrderedMap</a></li>
<li><a href="../modules/pl.Set.html">pl.Set</a></li>
<li><a href="../modules/pl.app.html">pl.app</a></li>
<li><a href="../modules/pl.array2d.html">pl.array2d</a></li>
<li><a href="../modules/pl.class.html">pl.class</a></li>
<li><a href="../modules/pl.comprehension.html">pl.comprehension</a></li>
<li><a href="../modules/pl.config.html">pl.config</a></li>
<li><a href="../modules/pl.data.html">pl.data</a></li>
<li><a href="../modules/pl.dir.html">pl.dir</a></li>
<li><a href="../modules/pl.file.html">pl.file</a></li>
<li><a href="../modules/pl.func.html">pl.func</a></li>
<li><a href="../modules/pl.input.html">pl.input</a></li>
<li><a href="../modules/pl.lapp.html">pl.lapp</a></li>
<li><a href="../modules/pl.lexer.html">pl.lexer</a></li>
<li><a href="../modules/pl.luabalanced.html">pl.luabalanced</a></li>
<li><a href="../modules/pl.operator.html">pl.operator</a></li>
<li><a href="../modules/pl.path.html">pl.path</a></li>
<li><a href="../modules/pl.permute.html">pl.permute</a></li>
<li><a href="../modules/pl.pretty.html">pl.pretty</a></li>
<li><a href="../modules/pl.seq.html">pl.seq</a></li>
<li><a href="../modules/pl.sip.html">pl.sip</a></li>
<li><a href="../modules/pl.strict.html">pl.strict</a></li>
<li><a href="../modules/pl.stringio.html">pl.stringio</a></li>
<li><a href="../modules/pl.stringx.html">pl.stringx</a></li>
<li><a href="../modules/pl.tablex.html">pl.tablex</a></li>
<li><a href="../modules/pl.template.html">pl.template</a></li>
<li><a href="../modules/pl.test.html">pl.test</a></li>
<li><a href="../modules/pl.text.html">pl.text</a></li>
<li><a href="../modules/pl.utils.html">pl.utils</a></li>
<li><a href="../modules/pl.xml.html">pl.xml</a></li>
</ul>
<h2>Examples</h2>
<ul>
<li><a href="../examples/seesubst.lua.html">seesubst.lua</a></li>
<li><a href="../examples/sipscan.lua.html">sipscan.lua</a></li>
<li><a href="../examples/symbols.lua.html">symbols.lua</a></li>
<li><a href="../examples/test-cmp.lua.html">test-cmp.lua</a></li>
<li><a href="../examples/test-data.lua.html">test-data.lua</a></li>
<li><a href="../examples/test-listcallbacks.lua.html">test-listcallbacks.lua</a></li>
<li><a href="../examples/test-pretty.lua.html">test-pretty.lua</a></li>
<li><a href="../examples/test-symbols.lua.html">test-symbols.lua</a></li>
<li><a href="../examples/testapp.lua.html">testapp.lua</a></li>
<li><a href="../examples/testclone.lua.html">testclone.lua</a></li>
<li><a href="../examples/testconfig.lua.html">testconfig.lua</a></li>
<li><a href="../examples/testglobal.lua.html">testglobal.lua</a></li>
<li><a href="../examples/testinputfields.lua.html">testinputfields.lua</a></li>
<li><a href="../examples/testinputfields2.lua.html">testinputfields2.lua</a></li>
<li><a href="../examples/testxml.lua.html">testxml.lua</a></li>
<li><a href="../examples/which.lua.html">which.lua</a></li>
</ul>
</div>
<div id="content">
<h1>Topic <code>06-data.md</code></h1>
<h2>Data</h2>
<p><a name="Reading_Data_Files"></a></p>
<h3>Reading Data Files</h3>
<p>The first thing to consider is this: do you actually need to write a custom file reader? And if the answer is yes, the next question is: can you write the reader in as clear a way as possible? Correctness, Robustness, and Speed; pick the first two and the third can be sorted out later, <em>if necessary</em>.</p>
<p>A common sort of data file is the configuration file format commonly used on Unix systems. This format is often called a <em>property</em> file in the Java world.</p>
<pre>
# Read timeout <span class="keyword">in</span> seconds
read.timeout=<span class="number">10</span>
# Write timeout <span class="keyword">in</span> seconds
write.timeout=<span class="number">10</span>
</pre>
<p>Here is a simple Lua implementation:</p>
<pre>
<span class="comment">-- property file parsing with Lua string patterns
</span> props = []
<span class="keyword">for</span> line <span class="keyword">in</span> <span class="global">io</span>.lines() <span class="keyword">do</span>
<span class="keyword">if</span> line:find(<span class="string">'#,1,true) ~= 1 and not line:find('</span>^%s*$<span class="string">') then
local var,value = line:match('</span>([^=]+)=(.*)')
props[var] = value
<span class="keyword">end</span>
<span class="keyword">end</span>
</pre>
<p>Very compact, but it suffers from a similar disease in equivalent Perl programs; it uses odd string patterns which are ‘lexically noisy’. Noisy code like this slows the casual reader down. (For an even more direct way of doing this, see the next section, ‘Reading Configuration Files’)</p>
<p>Another implementation, using the Penlight libraries:</p>
<pre>
<span class="comment">-- property file parsing with extended string functions
</span> <span class="global">require</span> <span class="string">'pl'</span>
stringx.import()
props = []
<span class="keyword">for</span> line <span class="keyword">in</span> <span class="global">io</span>.lines() <span class="keyword">do</span>
<span class="keyword">if</span> <span class="keyword">not</span> line:startswith(<span class="string">'#'</span>) <span class="keyword">and</span> <span class="keyword">not</span> line:isspace() <span class="keyword">then</span>
<span class="keyword">local</span> var,value = line:splitv(<span class="string">'='</span>)
props[var] = value
<span class="keyword">end</span>
<span class="keyword">end</span>
</pre>
<p>This is more self-documenting; it is generally better to make the code express the <em>intention</em>, rather than having to scatter comments everywhere – comments are necessary, of course, but mostly to give the higher view of your intention that cannot be expressed in code. It is slightly slower, true, but in practice the speed of this script is determined by I/O, so further optimization is unnecessary.</p>
<p><a name="Reading_Unstructured_Text_Data"></a></p>
<h3>Reading Unstructured Text Data</h3>
<p>Text data is sometimes unstructured, for example a file containing words. The <a href="../modules/pl.input.html#">pl.input</a> module has a number of functions which makes processing such files easier. For example, a script to count the number of words in standard input using <code>import.words</code>:</p>
<pre>
<span class="comment">-- countwords.lua
</span> <span class="global">require</span> <span class="string">'pl'</span>
<span class="keyword">local</span> k = <span class="number">1</span>
<span class="keyword">for</span> w <span class="keyword">in</span> input.words(<span class="global">io</span>.stdin) <span class="keyword">do</span>
k = k + <span class="number">1</span>
<span class="keyword">end</span>
<span class="global">print</span>(<span class="string">'count'</span>,k)
</pre>
<p>Or this script to calculate the average of a set of numbers using <a href="../modules/pl.input.html#numbers">input.numbers</a> :</p>
<pre>
<span class="comment">-- average.lua
</span> <span class="global">require</span> <span class="string">'pl'</span>
<span class="keyword">local</span> k = <span class="number">1</span>
<span class="keyword">local</span> sum = <span class="number">0</span>
<span class="keyword">for</span> n <span class="keyword">in</span> input.numbers(<span class="global">io</span>.stdin) <span class="keyword">do</span>
sum = sum + n
k = k + <span class="number">1</span>
<span class="keyword">end</span>
<span class="global">print</span>(<span class="string">'average'</span>,sum/k)
</pre>
<p>These scripts can be improved further by <em>eliminating loops</em> In the last case, there is a perfectly good function <a href="../modules/pl.seq.html#sum">seq.sum</a> which can already take a sequence of numbers and calculate these numbers for us:</p>
<pre>
<span class="comment">-- average2.lua
</span> <span class="global">require</span> <span class="string">'pl'</span>
<span class="keyword">local</span> total,n = seq.sum(input.numbers())
<span class="global">print</span>(<span class="string">'average'</span>,total/n)
</pre>
<p>A further simplification here is that if <code>numbers</code> or <code>words</code> are not passed an argument, they will grab their input from standard input. The first script can be rewritten:</p>
<pre>
<span class="comment">-- countwords2.lua
</span> <span class="global">require</span> <span class="string">'pl'</span>
<span class="global">print</span>(<span class="string">'count'</span>,seq.count(input.words()))
</pre>
<p>A useful feature of a sequence generator like <code>numbers</code> is that it can read from a string source. Here is a script to calculate the sums of the numbers on each line in a file:</p>
<pre>
<span class="comment">-- sums.lua
</span> <span class="keyword">for</span> line <span class="keyword">in</span> <span class="global">io</span>.lines() <span class="keyword">do</span>
<span class="global">print</span>(seq.sum(input.numbers(line))
<span class="keyword">end</span>
</pre>
<p><a name="Reading_Columnar_Data"></a></p>
<h3>Reading Columnar Data</h3>
<p>It is very common to find data in columnar form, either space or comma-separated, perhaps with an initial set of column headers. Here is a typical example:</p>
<pre>
EventID Magnitude LocationX LocationY LocationZ
<span class="number">981124001</span> <span class="number">2.0</span> <span class="number">18988.4</span> <span class="number">10047.1</span> <span class="number">4149.7</span>
<span class="number">981125001</span> <span class="number">0.8</span> <span class="number">19104.0</span> <span class="number">9970.4</span> <span class="number">5088.7</span>
<span class="number">981127003</span> <span class="number">0.5</span> <span class="number">19012.5</span> <span class="number">9946.9</span> <span class="number">3831.2</span>
...
</pre>
<p><a href="../modules/pl.input.html#fields">input.fields</a> is designed to extract several columns, given some delimiter (default to whitespace). Here is a script to calculate the average X location of all the events:</p>
<pre>
<span class="comment">-- avg-x.lua
</span> <span class="global">require</span> <span class="string">'pl'</span>
<span class="global">io</span>.read() <span class="comment">-- skip the header line
</span> <span class="keyword">local</span> sum,count = seq.sum(input.fields {<span class="number">3</span>})
<span class="global">print</span>(sum/count)
</pre>
<p><a href="../modules/pl.input.html#fields">input.fields</a> is passed either a field count, or a list of column indices, starting at one as usual. So in this case we're only interested in column 3. If you pass it a field count, then you get every field up to that count:</p>
<pre>
<span class="keyword">for</span> id,mag,locX,locY,locZ <span class="keyword">in</span> input.fields (<span class="number">5</span>) <span class="keyword">do</span>
....
<span class="keyword">end</span>
</pre>
<p><a href="../modules/pl.input.html#fields">input.fields</a> by default tries to convert each field to a number. It will skip lines which clearly don’t match the pattern, but will abort the script if there are any fields which cannot be converted to numbers.</p>
<p>The second parameter is a delimiter, by default spaces. ‘ ’ is understood to mean ‘any number of spaces’, i.e. ‘%s+’. Any Lua string pattern can be used.</p>
<p>The third parameter is a <em>data source</em>, by default standard input (defined by <a href="../modules/pl.input.html#create_getter">input.create_getter</a> .) It assumes that the data source has a <code>read</code> method which brings in the next line, i.e. it is a ‘file-like’ object. As a special case, a string will be split into its lines:</p>
<pre>
> <span class="keyword">for</span> x,y <span class="keyword">in</span> input.fields(<span class="number">2</span>,<span class="string">' '</span>,<span class="string">'10 20\n30 40\n'</span>) <span class="keyword">do</span> <span class="global">print</span>(x,y) <span class="keyword">end</span>
<span class="number">10</span> <span class="number">20</span>
<span class="number">30</span> <span class="number">40</span>
</pre>
<p>Note the default behaviour for bad fields, which is to show the offending line number:</p>
<pre>
> <span class="keyword">for</span> x,y <span class="keyword">in</span> input.fields(<span class="number">2</span>,<span class="string">' '</span>,<span class="string">'10 20\n30 40x\n'</span>) <span class="keyword">do</span> <span class="global">print</span>(x,y) <span class="keyword">end</span>
<span class="number">10</span> <span class="number">20</span>
line <span class="number">2</span>: cannot convert <span class="string">'40x'</span> to number
</pre>
<p>This behaviour of <a href="../modules/pl.input.html#fields">input.fields</a> is appropriate for a script which you want to fail immediately with an appropriate <em>user</em> error message if conversion fails. The fourth optional parameter is an options table: <code>{no_fail=true}</code> means that conversion is attempted but if it fails it just returns the string, rather as AWK would operate. You are then responsible for checking the type of the returned field. <code>{no_convert=true}</code> switches off conversion altogether and all fields are returned as strings.</p>
<p>Sometimes it is useful to bring a whole dataset into memory, for operations such as extracting columns. Penlight provides a flexible reader specifically for reading this kind of data, using the <a href="../modules/pl.data.html#">data</a> module. Given a file looking like this:</p>
<pre>
x,y
<span class="number">10</span>,<span class="number">20</span>
<span class="number">2</span>,<span class="number">5</span>
<span class="number">40</span>,<span class="number">50</span>
</pre>
<p>Then <a href="../modules/pl.data.html#read">data.read</a> will create a table like this, with each row represented by a sublist:</p>
<pre>
> t = data.read <span class="string">'test.txt'</span>
> pretty.dump(t)
{{<span class="number">10</span>,<span class="number">20</span>},{<span class="number">2</span>,<span class="number">5</span>},{<span class="number">40</span>,<span class="number">50</span>},fieldnames={<span class="string">'x'</span>,<span class="string">'y'</span>},delim=<span class="string">','</span>}
</pre>
<p>You can now analyze this returned table using the supplied methods. For instance, the method <code>column_by_name</code> returns a table of all the values of that column.</p>
<pre>
<span class="comment">-- testdata.lua
</span> <span class="global">require</span> <span class="string">'pl'</span>
d = data.read(<span class="string">'fev.txt'</span>)
<span class="keyword">for</span> _,name <span class="keyword">in</span> <span class="global">ipairs</span>(d.fieldnames) <span class="keyword">do</span>
<span class="keyword">local</span> col = d:column_by_name(name)
<span class="keyword">if</span> <span class="global">type</span>(col[<span class="number">1</span>]) == <span class="string">'number'</span> <span class="keyword">then</span>
<span class="keyword">local</span> total,n = seq.sum(col)
utils.printf(<span class="string">"Average for %s is %f\n"</span>,name,total/n)
<span class="keyword">end</span>
<span class="keyword">end</span>
</pre>
<p><a href="../modules/pl.data.html#read">data.read</a> tries to be clever when given data; by default it expects a first line of column names, unless any of them are numbers. It tries to deduce the column delimiter by looking at the first line. Sometimes it guesses wrong; these things can be specified explicitly. The second optional parameter is an options table: can override <code>delim</code> (a string pattern), <code>fieldnames</code> (a list or comma-separated string), specify <code>no_convert</code> (default is to convert), numfields (indices of columns known to be numbers, as a list) and <code>thousands_dot</code> (when the thousands separator in Excel CSV is ‘.’)</p>
<p>A very powerful feature is a way to execute SQL-like queries on such data:</p>
<pre>
<span class="comment">-- queries on tabular data
</span> <span class="global">require</span> <span class="string">'pl'</span>
<span class="keyword">local</span> d = data.read(<span class="string">'xyz.txt'</span>)
<span class="keyword">local</span> q = d:<span class="global">select</span>(<span class="string">'x,y,z where x > 3 and z < 2 sort by y'</span>)
<span class="keyword">for</span> x,y,z <span class="keyword">in</span> q <span class="keyword">do</span>
<span class="global">print</span>(x,y,z)
<span class="keyword">end</span>
</pre>
<p>Please note that the format of queries is restricted to the following syntax:</p>
<pre>
FIELDLIST [ <span class="string">'where'</span> CONDITION ] [ <span class="string">'sort by'</span> FIELD [asc|desc]]
</pre>
<p>Any valid Lua code can appear in <code>CONDITION</code>; remember it is <em>not</em> SQL and you have to use <code>==</code> (this warning comes from experience.)</p>
<p>For this to work, <em>field names must be Lua identifiers</em>. So <a href="../modules/pl.data.html#read">read</a> will massage fieldnames so that all non-alphanumeric chars are replaced with underscores.</p>
<p><a href="../modules/pl.data.html#read">read</a> can handle standard CSV files fine, although doesn’t try to be a full-blown CSV parser. Spreadsheet programs are not always the best tool to process such data, strange as this might seem to some people. This is a toy CSV file; to appreciate the problem, imagine thousands of rows and dozens of columns like this:</p>
<pre>
Department Name,Employee ID,Project,Hours Booked
sales,<span class="number">1231</span>,overhead,<span class="number">4</span>
sales,<span class="number">1255</span>,overhead,<span class="number">3</span>
engineering,<span class="number">1501</span>,development,<span class="number">5</span>
engineering,<span class="number">1501</span>,maintenance,<span class="number">3</span>
engineering,<span class="number">1433</span>,maintenance,<span class="number">10</span>
</pre>
<p>The task is to reduce the dataset to a relevant set of rows and columns, perhaps do some processing on row data, and write the result out to a new CSV file. The <code>write_row</code> method uses the delimiter to write the row to a file; <code>Data.select_row</code> is like <code>Data.select</code>, except it iterates over <em>rows</em>, not fields; this is necessary if we are dealing with a lot of columns!</p>
<pre>
names = {[<span class="number">1501</span>]=<span class="string">'don'</span>,[<span class="number">1433</span>]=<span class="string">'dilbert'</span>}
keepcols = {<span class="string">'Employee_ID'</span>,<span class="string">'Hours_Booked'</span>}
t:write_row (outf,{<span class="string">'Employee'</span>,<span class="string">'Hours_Booked'</span>})
q = t:select_row {
fields=keepcols,
where=<span class="keyword">function</span>(row) <span class="keyword">return</span> row[<span class="number">1</span>]==<span class="string">'engineering'</span> <span class="keyword">end</span>
}
<span class="keyword">for</span> row <span class="keyword">in</span> q <span class="keyword">do</span>
row[<span class="number">1</span>] = names[row[<span class="number">1</span>]]
t:write_row(outf,row)
<span class="keyword">end</span>
</pre>
<p><code>Data.select_row</code> and <code>Data.select</code> can be passed a table specifying the query; a list of field names, a function defining the condition and an optional parameter <code>sort_by</code>. It isn’t really necessary here, but if we had a more complicated row condition (such as belonging to a specified set) then it is not generally possible to express such a condition as a query string, without resorting to hackery such as global variables.</p>
<p>Data does not have to come from files, nor does it necessarily come from the lab or the accounts department. On Linux, <code>ps aux</code> gives you a full listing of all processes running on your machine. It is straightforward to feed the output of this command into <a href="../modules/pl.data.html#read">data.read</a> and perform useful queries on it. Notice that non-identifier characters like ‘%’ get converted into underscores:</p>
<pre>
<span class="global">require</span> <span class="string">'pl'</span>
f = <span class="global">io</span>.popen <span class="string">'ps aux'</span>
s = data.read (f,{last_field_collect=<span class="keyword">true</span>})
f:close()
<span class="global">print</span>(s.fieldnames)
<span class="global">print</span>(s:column_by_name <span class="string">'USER'</span>)
qs = <span class="string">'COMMAND,_MEM where _MEM > 5 and USER=="steve"'</span>
<span class="keyword">for</span> name,mem <span class="keyword">in</span> s:<span class="global">select</span>(qs) <span class="keyword">do</span>
<span class="global">print</span>(mem,name)
<span class="keyword">end</span>
</pre>
<p>I've always been an admirer of the AWK programming language; with <a href="../modules/pl.data.html#filter">filter</a> you can get Lua programs which are just as compact:</p>
<pre>
<span class="comment">-- printxy.lua
</span> <span class="global">require</span> <span class="string">'pl'</span>
data.filter <span class="string">'x,y where x > 3'</span>
</pre>
<p>It is common enough to have data files without headers of field names. <a href="../modules/pl.data.html#read">data.read</a> makes a special exception for such files if all fields are numeric. Since there are no column names to use in query expressions, you can use AWK-like column indexes, e.g. ‘$1,$2 where $1 > 3’. I have a little executable script on my system called <code>lf</code> which looks like this:</p>
<pre>
#!/usr/bin/env lua
<span class="global">require</span> <span class="string">'pl.data'</span>.filter(arg[<span class="number">1</span>])
</pre>
<p>And it can be used generally as a filter command to extract columns from data. (The column specifications may be expressions or even constants.)</p>
<pre>
$ lf <span class="string">'$1,$5/10'</span> < test.dat
</pre>
<p>(As with AWK, please note the single-quotes used in this command; this prevents the shell trying to expand the column indexes. If you are on Windows, then you are fine, but it is still necessary to quote the expression in double-quotes so it is passed as one argument to your batch file.)</p>
<p>As a tutorial resource, have a look at <a href="../examples/test-data.lua.html#">test-data.lua</a> in the PL tests directory for other examples of use, plus comments.</p>
<p>The data returned by <a href="../modules/pl.data.html#read">read</a> or constructed by <code>Data.copy_select</code> from a query is basically just an array of rows: <code>{{1,2},{3,4}}</code>. So you may use <a href="../modules/pl.data.html#read">read</a> to pull in any array-like dataset, and process with any function that expects such a implementation. In particular, the functions in <a href="../modules/pl.array2d.html#">array2d</a> will work fine with this data. In fact, these functions are available as methods; e.g. <a href="../modules/pl.array2d.html#flatten">array2d.flatten</a> can be called directly like so to give us a one-dimensional list:</p>
<pre>
v = data.read(<span class="string">'dat.txt'</span>):flatten()
</pre>
<p>The data is also in exactly the right shape to be treated as matrices by <a href="http://lua-users.org/wiki/LuaMatrix">LuaMatrix</a>:</p>
<pre>
> matrix = <span class="global">require</span> <span class="string">'matrix'</span>
> m = matrix(data.read <span class="string">'mat.txt'</span>)
> = m
<span class="number">1</span> <span class="number">0.2</span> <span class="number">0.3</span>
<span class="number">0.2</span> <span class="number">1</span> <span class="number">0.1</span>
<span class="number">0.1</span> <span class="number">0.2</span> <span class="number">1</span>
> = m^<span class="number">2</span> <span class="comment">-- same as m*m
</span> <span class="number">1.07</span> <span class="number">0.46</span> <span class="number">0.62</span>
<span class="number">0.41</span> <span class="number">1.06</span> <span class="number">0.26</span>
<span class="number">0.24</span> <span class="number">0.42</span> <span class="number">1.05</span>
</pre>
<p><a href="../modules/pl.data.html#write">write</a> will write matrices back to files for you.</p>
<p>Finally, for the curious, the global variable <code>_DEBUG</code> can be used to print out the actual iterator function which a query generates and dynamically compiles. By using code generation, we can get pretty much optimal performance out of arbitrary queries.</p>
<pre>
> lua -lpl -e <span class="string">"_DEBUG=true"</span> -e <span class="string">"data.filter 'x,y where x > 4 sort by x'"</span> < test.txt
<span class="keyword">return</span> <span class="keyword">function</span> (t)
<span class="keyword">local</span> i = <span class="number">0</span>
<span class="keyword">local</span> v
<span class="keyword">local</span> ls = {}
<span class="keyword">for</span> i,v <span class="keyword">in</span> <span class="global">ipairs</span>(t) <span class="keyword">do</span>
<span class="keyword">if</span> v[<span class="number">1</span>] > <span class="number">4</span> <span class="keyword">then</span>
ls[#ls+<span class="number">1</span>] = v
<span class="keyword">end</span>
<span class="keyword">end</span>
<span class="global">table</span>.sort(ls,<span class="keyword">function</span>(v1,v2)
<span class="keyword">return</span> v1[<span class="number">1</span>] < v2[<span class="number">1</span>]
<span class="keyword">end</span>)
<span class="keyword">local</span> n = #ls
<span class="keyword">return</span> <span class="keyword">function</span>()
i = i + <span class="number">1</span>
v = ls[i]
<span class="keyword">if</span> i > n <span class="keyword">then</span> <span class="keyword">return</span> <span class="keyword">end</span>
<span class="keyword">return</span> v[<span class="number">1</span>],v[<span class="number">2</span>]
<span class="keyword">end</span>
<span class="keyword">end</span>
<span class="number">10</span>,<span class="number">20</span>
<span class="number">40</span>,<span class="number">50</span>
</pre>
<p><a name="Reading_Configuration_Files"></a></p>
<h3>Reading Configuration Files</h3>
<p>The <a href="../modules/pl.config.html#">config</a> module provides a simple way to convert several kinds of configuration files into a Lua table. Consider the simple example:</p>
<pre>
# test.config
# Read timeout <span class="keyword">in</span> seconds
read.timeout=<span class="number">10</span>
# Write timeout <span class="keyword">in</span> seconds
write.timeout=<span class="number">5</span>
#acceptable ports
ports = <span class="number">1002</span>,<span class="number">1003</span>,<span class="number">1004</span>
</pre>
<p>This can be easily brought in using <a href="../modules/pl.config.html#read">config.read</a> and the result shown using <a href="../modules/pl.pretty.html#write">pretty.write</a> :</p>
<pre>
<span class="comment">-- readconfig.lua
</span> <span class="keyword">local</span> config = <span class="global">require</span> <span class="string">'pl.config'</span>
<span class="keyword">local</span> pretty= <span class="global">require</span> <span class="string">'pl.pretty'</span>
<span class="keyword">local</span> t = config.read(arg[<span class="number">1</span>])
<span class="global">print</span>(pretty.write(t))
</pre>
<p>and the output of <code>lua readconfig.lua test.config</code> is:</p>
<pre>
{
ports = {
<span class="number">1002</span>,
<span class="number">1003</span>,
<span class="number">1004</span>
},
write_timeout = <span class="number">5</span>,
read_timeout = <span class="number">10</span>
}
</pre>
<p>That is, <a href="../modules/pl.config.html#read">config.read</a> will bring in all key/value pairs, ignore # comments, and ensure that the key names are proper Lua identifiers by replacing non-identifier characters with ‘_’. If the values are numbers, then they will be converted. (So the value of <code>t.write_timeout</code> is the number 5). In addition, any values which are separated by commas will be converted likewise into an array.</p>
<p>Any line can be continued with a backslash. So this will all be considered one line:</p>
<pre>
names=one,two,three, \
four,five,six,seven, \
eight,nine,ten
</pre>
<p>Windows-style INI files are also supported. The section structure of INI files translates naturally to nested tables in Lua:</p>
<pre>
; test.ini
[timeouts]
read=<span class="number">10</span> ; Read timeout <span class="keyword">in</span> seconds
write=<span class="number">5</span> ; Write timeout <span class="keyword">in</span> seconds
[portinfo]
ports = <span class="number">1002</span>,<span class="number">1003</span>,<span class="number">1004</span>
</pre>
<p> The output is:</p>
<pre>
{
portinfo = {
ports = {
<span class="number">1002</span>,
<span class="number">1003</span>,
<span class="number">1004</span>
}
},
timeouts = {
write = <span class="number">5</span>,
read = <span class="number">10</span>
}
}
</pre>
<p>You can now refer to the write timeout as <code>t.timeouts.write</code>.</p>
<p>As a final example of the flexibility of <a href="../modules/pl.config.html#read">config.read</a> , if passed this simple comma-delimited file</p>
<pre>
one,two,three
<span class="number">10</span>,<span class="number">20</span>,<span class="number">30</span>
<span class="number">40</span>,<span class="number">50</span>,<span class="number">60</span>
<span class="number">1</span>,<span class="number">2</span>,<span class="number">3</span>
</pre>
<p>it will produce the following table:</p>
<pre>
{
{ <span class="string">"one"</span>, <span class="string">"two"</span>, <span class="string">"three"</span> },
{ <span class="number">10</span>, <span class="number">20</span>, <span class="number">30</span> },
{ <span class="number">40</span>, <span class="number">50</span>, <span class="number">60</span> },
{ <span class="number">1</span>, <span class="number">2</span>, <span class="number">3</span> }
}
</pre>
<p><a href="../modules/pl.config.html#read">config.read</a> isn’t designed to read all CSV files in general, but intended to support some Unix configuration files not structured as key-value pairs, such as ‘/etc/passwd’.</p>
<p>This function is intended to be a Swiss Army Knife of configuration readers, but it does have to make assumptions, and you may not like them. So there is an optional extra parameter which allows some control, which is table that may have the following fields:</p>
<pre>
{
variablilize = <span class="keyword">true</span>,
convert_numbers = <span class="keyword">true</span>,
trim_space = <span class="keyword">true</span>,
list_delim = <span class="string">','</span>
}
</pre>
<p><code>variablilize</code> is the option that converted <code>write.timeout</code> in the first example to the valid Lua identifier <code>write_timeout</code>. If <code>convert_numbers</code> is true, then an attempt is made to convert any string that starts like a number. <code>trim_space</code> ensures that there is no starting or trailing whitespace with values, and <code>list_delim</code> is the character that will be used to decide whether to split a value up into a list (it may be a Lua string pattern such as ‘%s+’.)</p>
<p>For instance, the password file in Unix is colon-delimited:</p>
<pre>
t = config.read(<span class="string">'/etc/passwd'</span>,{list_delim=<span class="string">':'</span>})
</pre>
<p>This produces the following output on my system (only last two lines shown):</p>
<pre>
{
...
{
<span class="string">"user"</span>,
<span class="string">"x"</span>,
<span class="string">"1000"</span>,
<span class="string">"1000"</span>,
<span class="string">"user,,,"</span>,
<span class="string">"/home/user"</span>,
<span class="string">"/bin/bash"</span>
},
{
<span class="string">"sdonovan"</span>,
<span class="string">"x"</span>,
<span class="string">"1001"</span>,
<span class="string">"1001"</span>,
<span class="string">"steve donovan,28,,"</span>,
<span class="string">"/home/sdonovan"</span>,
<span class="string">"/bin/bash"</span>
}
}
</pre>
<p>You can get this into a more sensible format, where the usernames are the keys, with:</p>
<pre>
t = tablex.pairmap(<span class="keyword">function</span>(k,v) <span class="keyword">return</span> v,v[<span class="number">1</span>] <span class="keyword">end</span>,t)
</pre>
<p>and you get:</p>
<pre>
{ ...
sdonovan = {
<span class="string">"sdonovan"</span>,
<span class="string">"x"</span>,
<span class="string">"1001"</span>,
<span class="string">"1001"</span>,
<span class="string">"steve donovan,28,,"</span>,
<span class="string">"/home/sdonovan"</span>,
<span class="string">"/bin/bash"</span>
}
...
}
</pre>
<p><a id="lexer"/></p>
<p><a name="Lexical_Scanning"></a></p>
<h3>Lexical Scanning</h3>
<p>Although Lua’s string pattern matching is very powerful, there are times when something more powerful is needed. <a href="../modules/pl.lexer.html#scan">pl.lexer.scan</a> provides lexical scanners which <em>tokenizes</em> a string, classifying tokens into numbers, strings, etc.</p>
<pre>
> lua -lpl
Lua <span class="number">5.1</span>.<span class="number">4</span> Copyright (C) <span class="number">1994</span>-<span class="number">2008</span> Lua.org, PUC-Rio
> tok = lexer.scan <span class="string">'alpha = sin(1.5)'</span>
> = tok()
iden alpha
> = tok()
= =
> = tok()
iden sin
> = tok()
( (
> = tok()
number <span class="number">1.5</span>
> = tok()
) )
> = tok()
(<span class="keyword">nil</span>)
</pre>
<p>The scanner is a function, which is repeatedly called and returns the <em>type</em> and <em>value</em> of the token. Recognized basic types are ‘iden’,‘string’,‘number’, and ‘space’. and everything else is represented by itself. Note that by default the scanner will skip any ‘space’ tokens.</p>
<p>‘comment’ and ‘keyword’ aren’t applicable to the plain scanner, which is not language-specific, but a scanner which understands Lua is available. It recognizes the Lua keywords, and understands both short and long comments and strings.</p>
<pre>
> <span class="keyword">for</span> t,v <span class="keyword">in</span> lexer.lua <span class="string">'for i=1,n do'</span> <span class="keyword">do</span> <span class="global">print</span>(t,v) <span class="keyword">end</span>
keyword <span class="keyword">for</span>
iden i
= =
number <span class="number">1</span>
, ,
iden n
keyword <span class="keyword">do</span>
</pre>
<p>A lexical scanner is useful where you have highly-structured data which is not nicely delimited by newlines. For example, here is a snippet of a in-house file format which it was my task to maintain:</p>
<pre>
points (<span class="number">818344.1</span>,-<span class="number">20389.7</span>,-<span class="number">0.1</span>),(<span class="number">818337.9</span>,-<span class="number">20389.3</span>,-<span class="number">0.1</span>),(<span class="number">818332.5</span>,-<span class="number">20387.8</span>,-<span class="number">0.1</span>)
,(<span class="number">818327.4</span>,-<span class="number">20388</span>,-<span class="number">0.1</span>),(<span class="number">818322</span>,-<span class="number">20387.7</span>,-<span class="number">0.1</span>),(<span class="number">818316.3</span>,-<span class="number">20388.6</span>,-<span class="number">0.1</span>)
,(<span class="number">818309.7</span>,-<span class="number">20389.4</span>,-<span class="number">0.1</span>),(<span class="number">818303.5</span>,-<span class="number">20390.6</span>,-<span class="number">0.1</span>),(<span class="number">818295.8</span>,-<span class="number">20388.3</span>,-<span class="number">0.1</span>)
,(<span class="number">818290.5</span>,-<span class="number">20386.9</span>,-<span class="number">0.1</span>),(<span class="number">818285.2</span>,-<span class="number">20386.1</span>,-<span class="number">0.1</span>),(<span class="number">818279.3</span>,-<span class="number">20383.6</span>,-<span class="number">0.1</span>)
,(<span class="number">818274</span>,-<span class="number">20381.2</span>,-<span class="number">0.1</span>),(<span class="number">818274</span>,-<span class="number">20380.7</span>,-<span class="number">0.1</span>);
</pre>
<p>Here is code to extract the points using <a href="../modules/pl.lexer.html#">pl.lexer</a> :</p>
<pre>
<span class="comment">-- assume 's' contains the text above...
</span> <span class="keyword">local</span> lexer = <span class="global">require</span> <span class="string">'pl.lexer'</span>
<span class="keyword">local</span> expecting = lexer.expecting
<span class="keyword">local</span> append = <span class="global">table</span>.insert
<span class="keyword">local</span> tok = lexer.scan(s)
<span class="keyword">local</span> points = {}
<span class="keyword">local</span> t,v = tok() <span class="comment">-- should be 'iden','points'
</span>
<span class="keyword">while</span> t ~= <span class="string">';'</span> <span class="keyword">do</span>
c = {}
expecting(tok,<span class="string">'('</span>)
c.x = expecting(tok,<span class="string">'number'</span>)
expecting(tok,<span class="string">','</span>)
c.y = expecting(tok,<span class="string">'number'</span>)
expecting(tok,<span class="string">','</span>)
c.z = expecting(tok,<span class="string">'number'</span>)
expecting(tok,<span class="string">')'</span>)
t,v = tok() <span class="comment">-- either ',' or ';'
</span> append(points,c)
<span class="keyword">end</span>
</pre>
<p>The <code>expecting</code> function grabs the next token and if the type doesn’t match, it throws an error. (<a href="../modules/pl.lexer.html#">pl.lexer</a> , unlike other PL libraries, raises errors if something goes wrong, so you should wrap your code in <a href="http://www.lua.org/manual/5.1/manual.html#pdf-pcall">pcall</a> to catch the error gracefully.)</p>
<p>The scanners all have a second optional argument, which is a table which controls whether you want to exclude spaces and/or comments. The default for <a href="../modules/pl.lexer.html#lua">lexer.lua</a> is <code>{space=true,comments=true}</code>. There is a third optional argument which determines how string and number tokens are to be processsed.</p>
<p>The ultimate highly-structured data is of course, program source. Here is a snippet from ‘text-lexer.lua’:</p>
<pre>
<span class="global">require</span> <span class="string">'pl'</span>
lines = <span class="string">[[
for k,v in pairs(t) do
if type(k) == 'number' then
print(v) -- array-like case
else
print(k,v)
end
end
]]</span>
ls = List()
<span class="keyword">for</span> tp,val <span class="keyword">in</span> lexer.lua(lines,{space=<span class="keyword">true</span>,comments=<span class="keyword">true</span>}) <span class="keyword">do</span>
<span class="global">assert</span>(tp ~= <span class="string">'space'</span> <span class="keyword">and</span> tp ~= <span class="string">'comment'</span>)
<span class="keyword">if</span> tp == <span class="string">'keyword'</span> <span class="keyword">then</span> ls:append(val) <span class="keyword">end</span>
<span class="keyword">end</span>
test.asserteq(ls,List{<span class="string">'for'</span>,<span class="string">'in'</span>,<span class="string">'do'</span>,<span class="string">'if'</span>,<span class="string">'then'</span>,<span class="string">'else'</span>,<span class="string">'end'</span>,<span class="string">'end'</span>})
</pre>
<p>Here is a useful little utility that identifies all common global variables found in a lua module:</p>
<pre>
<span class="comment">-- testglobal.lua
</span> <span class="global">require</span> <span class="string">'pl'</span>
<span class="keyword">local</span> txt,err = utils.readfile(arg[<span class="number">1</span>])
<span class="keyword">if</span> <span class="keyword">not</span> txt <span class="keyword">then</span> <span class="keyword">return</span> <span class="global">print</span>(err) <span class="keyword">end</span>
<span class="keyword">local</span> globals = List()
<span class="keyword">for</span> t,v <span class="keyword">in</span> lexer.lua(txt) <span class="keyword">do</span>
<span class="keyword">if</span> t == <span class="string">'iden'</span> <span class="keyword">and</span> _G[v] <span class="keyword">then</span>
globals:append(v)
<span class="keyword">end</span>
<span class="keyword">end</span>
pretty.dump(seq.count_map(globals))
</pre>
<p>Rather then dumping the whole list, with its duplicates, we pass it through <a href="../modules/pl.seq.html#count_map">seq.count_map</a> which turns the list into a table where the keys are the values, and the associated values are the number of times those values occur in the sequence. Typical output looks like this:</p>
<pre>
{
<span class="global">type</span> = <span class="number">2</span>,
<span class="global">pairs</span> = <span class="number">2</span>,
<span class="global">table</span> = <span class="number">2</span>,
<span class="global">print</span> = <span class="number">3</span>,
<span class="global">tostring</span> = <span class="number">2</span>,
<span class="global">require</span> = <span class="number">1</span>,
<span class="global">ipairs</span> = <span class="number">4</span>
}
</pre>
<p>You could further pass this through <a href="../modules/pl.tablex.html#keys">tablex.keys</a> to get a unique list of symbols. This can be useful when writing ‘strict’ Lua modules, where all global symbols must be defined as locals at the top of the file.</p>
<p>For a more detailed use of <a href="../modules/pl.lexer.html#scan">lexer.scan</a> , please look at <a href="../examples/testxml.lua.html#">testxml.lua</a> in the examples directory.</p>
<p><a name="XML"></a></p>
<h3>XML</h3>
<p>New in the 0.9.7 release is some support for XML. This is a large topic, and Penlight does not provide a full XML stack, which is properly the task of a more specialized library.</p>
<h4>Parsing and Pretty-Printing</h4>
<p>The semi-standard XML parser in the Lua universe is <a href="">lua-expat</a>. In particular, it has a function called <code>lxp.lom.parse</code> which will parse XML into the Lua Object Model (LOM) format. However, it does not provide a way to convert this data back into XML text. <a href="../modules/pl.xml.html#parse">xml.parse</a> will use this function, <em>if</em> <code>lua-expat</code> is available, and otherwise switches back to a pure Lua parser originally written by Roberto Ierusalimschy.</p>
<p>The resulting document object knows how to render itself as a string, which is useful for debugging:</p>
<pre>
> d = xml.parse <span class="string">"<nodes><node id='1'>alice</node></nodes>"</span>
> = d
<nodes><node id=<span class="string">'1'</span>>alice</node></nodes>
> pretty.dump (d)
{
{
<span class="string">"alice"</span>,
attr = {
<span class="string">"id"</span>,
id = <span class="string">"1"</span>
},
tag = <span class="string">"node"</span>
},
attr = {
},
tag = <span class="string">"nodes"</span>
}
</pre>
<p>Looking at the actual shape of the data reveals the structure of LOM:</p>
<ul>
<li>every element has a <code>tag</code> field with its name</li>
<li>plus a <code>attr</code> field which is a table containing the attributes as fields, and also as an array. It is always present.</li>
<li>the children of the element are the array part of the element, so <code>d[1]</code> is the first child of <code>d</code>, etc.</li>
</ul>
<p>It could be argued that having attributes also as the array part of <code>attr</code> is not essential (you generally cannot depend on attribute order in XML) but that’s how it goes with this standard.</p>
<p><code>lua-expat</code> is another <em>soft dependency</em> of Penlight; generally, the fallback parser is good enough for straightforward XML as is commonly found in configuration files, etc. <code>doc.basic_parse</code> is not intended to be a proper conforming parser (it’s only sixty lines) but it handles simple kinds of documents that do not have comments or DTD directives. It is intelligent enough to ignore the <code><?xml</code> directive and that is about it.</p>
<p>You can get pretty-printing by explicitly calling <a href="../modules/pl.xml.html#tostring">xml.tostring</a> and passing it the initial indent and the per-element indent:</p>
<pre>
> = xml.<span class="global">tostring</span>(d,<span class="string">''</span>,<span class="string">' '</span>)
<nodes>
<node id=<span class="string">'1'</span>>alice</node>
</nodes>
</pre>
<p>There is a fourth argument which is the <em>attribute indent</em>:</p>
<pre>
> a = xml.parse <span class="string">"<frodo name='baggins' age='50' type='hobbit'/>"</span>
> = xml.<span class="global">tostring</span>(a,<span class="string">''</span>,<span class="string">' '</span>,<span class="string">' '</span>)
<frodo
<span class="global">type</span>=<span class="string">'hobbit'</span>
name=<span class="string">'baggins'</span>
age=<span class="string">'50'</span>
/>
</pre>
<h4>Parsing and Working with Configuration Files</h4>
<p>It’s common to find configurations expressed with XML these days. It’s straightforward to ‘walk’ the LOM data and extract the data in the form you want:</p>
<pre>
<span class="global">require</span> <span class="string">'pl'</span>
<span class="keyword">local</span> config = <span class="string">[[
<config>
<alpha>1.3</alpha>
<beta>10</beta>
<name>bozo</name>
</config>
]]</span>
<span class="keyword">local</span> d,err = xml.parse(config)
<span class="keyword">local</span> t = {}
<span class="keyword">for</span> item <span class="keyword">in</span> d:childtags() <span class="keyword">do</span>
t[item.tag] = item[<span class="number">1</span>]
<span class="keyword">end</span>
pretty.dump(t)
<span class="comment">--->
</span> {
beta = <span class="string">"10"</span>,
alpha = <span class="string">"1.3"</span>,
name = <span class="string">"bozo"</span>
}
</pre>
<p>The only gotcha is that here we must use the <code>Doc:childtags</code> method, which will skip over any text elements.</p>
<p>A more involved example is this excerpt from <code>serviceproviders.xml</code>, which is usually found at <code>/usr/share/mobile-broadband-provider-info/serviceproviders.xml</code> on Debian/Ubuntu Linux systems.</p>
<pre>
d = xml.parse <span class="string">[[
<serviceproviders format="2.0">
<country code="za">
<provider>
<name>Cell-c</name>
<gsm>
<network-id mcc="655" mnc="07"/>
<apn value="internet">
<username>Cellcis</username>
<dns>196.7.0.138</dns>
<dns>196.7.142.132</dns>
</apn>
</gsm>
</provider>
<provider>
<name>MTN</name>
<gsm>
<network-id mcc="655" mnc="10"/>
<apn value="internet">
<dns>196.11.240.241</dns>
<dns>209.212.97.1</dns>
</apn>
</gsm>
</provider>
<provider>
<name>Vodacom</name>
<gsm>
<network-id mcc="655" mnc="01"/>
<apn value="internet">
<dns>196.207.40.165</dns>
<dns>196.43.46.190</dns>
</apn>
<apn value="unrestricted">
<name>Unrestricted</name>
<dns>196.207.32.69</dns>
<dns>196.43.45.190</dns>
</apn>
</gsm>
</provider>
<provider>
<name>Virgin Mobile</name>
<gsm>
<apn value="vdata">
<dns>196.7.0.138</dns>
<dns>196.7.142.132</dns>
</apn>
</gsm>
</provider>
</country>
</serviceproviders>
]]</span>
</pre>
<p>Getting the names of the providers per-country is straightforward:</p>
<pre>
<span class="keyword">local</span> t = {}
<span class="keyword">for</span> country <span class="keyword">in</span> d:childtags() <span class="keyword">do</span>
<span class="keyword">local</span> providers = {}
t[country.tag] = providers
<span class="keyword">for</span> provider <span class="keyword">in</span> country:childtags() <span class="keyword">do</span>
<span class="global">table</span>.insert(providers,provider:child_with_name(<span class="string">'name'</span>):get_text())
<span class="keyword">end</span>
<span class="keyword">end</span>
pretty.dump(t)
<span class="comment">-->
</span> {
country = {
<span class="string">"Cell-c"</span>,
<span class="string">"MTN"</span>,
<span class="string">"Vodacom"</span>,
<span class="string">"Virgin Mobile"</span>
}
}
</pre>
<h4>Generating XML with ‘xmlification’</h4>
<p>This feature is inspired by the <code>htmlify</code> function used by <a href="http://keplerproject.github.com/orbit/">Orbit</a> to simplify HTML generation, except that no function environment magic is used; the <code>tags</code> function returns a set of <em>constructors</em> for elements of the given tag names.</p>
<pre>
> nodes, node = xml.tags <span class="string">'nodes, node'</span>
> = node <span class="string">'alice'</span>
<node>alice</node>
> = nodes { node {id=<span class="string">'1'</span>,<span class="string">'alice'</span>}}
<nodes><node id=<span class="string">'1'</span>>alice</node></nodes>
</pre>
<p>The flexibility of Lua tables is very useful here, since both the attributes and the children of an element can be encoded naturally. The argument to these tag constructors is either a single value (like a string) or a table where the attributes are the named keys and the children are the array values.</p>
<h4>Generating XML using Templates</h4>
<p>A template is a little XML document which contains dollar-variables. The <code>subst</code> method on a document is fed an array of tables containing values for these variables. Note how the parent tag name is specified:</p>
<pre>
> templ = xml.parse <span class="string">"<node id='$id'>$name</node>"</span>
> = templ:subst {tag=<span class="string">'nodes'</span>, {id=<span class="number">1</span>,name=<span class="string">'alice'</span>},{id=<span class="number">2</span>,name=<span class="string">'john'</span>}}
<nodes><node id=<span class="string">'1'</span>>alice</node><node id=<span class="string">'2'</span>>john</node></nodes>
</pre>
<h4>Extracting Data using Templates</h4>
<p>Matching goes in the opposite direction. We have a document, and would like to extract values from it using a pattern.</p>
<p>A common use of this is parsing the XML result of API queries. The <a href="http://blog.programmableweb.com/2010/02/08/googles-secret-weather-api/">(undocumented) Google Weather API</a> is a good example. Grabbing the result of <code>http://www.google.com/ig/api?weather=Johannesburg,ZA" we get something like this, after pretty-printing:</code></p>
<pre>
<xml_api_reply version=<span class="string">'1'</span>>
<weather module_id=<span class="string">'0'</span> tab_id=<span class="string">'0'</span> mobile_zipped=<span class="string">'1'</span> section=<span class="string">'0'</span> row=<span class="string">'0'</span> mobile_row=<span class="string">'0'</span>>
<forecast_information>
<city data=<span class="string">'Johannesburg, Gauteng'</span>/>
<postal_code data=<span class="string">'Johannesburg,ZA'</span>/>
<latitude_e6 data=<span class="string">''</span>/>
<longitude_e6 data=<span class="string">''</span>/>
<forecast_date data=<span class="string">'2010-10-02'</span>/>
<current_date_time data=<span class="string">'2010-10-02 18:30:00 +0000'</span>/>
<unit_system data=<span class="string">'US'</span>/>
</forecast_information>
<current_conditions>
<condition data=<span class="string">'Clear'</span>/>
<temp_f data=<span class="string">'75'</span>/>
<temp_c data=<span class="string">'24'</span>/>
<humidity data=<span class="string">'Humidity: 19%'</span>/>
<icon data=<span class="string">'/ig/images/weather/sunny.gif'</span>/>
<wind_condition data=<span class="string">'Wind: NW at 7 mph'</span>/>
</current_conditions>
<forecast_conditions>
<day_of_week data=<span class="string">'Sat'</span>/>
<low data=<span class="string">'60'</span>/>
<high data=<span class="string">'89'</span>/>
<icon data=<span class="string">'/ig/images/weather/sunny.gif'</span>/>
<condition data=<span class="string">'Clear'</span>/>
</forecast_conditions>
....
</weather>
</xml_api_reply>
</pre>
<p>Assume that the above XML has been read into <code>google</code>. The idea is to write a pattern looking like a template, and use it to extract some values of interest:</p>
<pre>
t = <span class="string">[[
<weather>
<current_conditions>
<condition data='$condition'/>
<temp_c data='$temp'/>
</current_conditions>
</weather>
]]</span>
<span class="keyword">local</span> res, ret = google:match(t)
pretty.dump(res)
</pre>
<p>And the output is:</p>
<pre>
{
condition = <span class="string">"Clear"</span>,
temp = <span class="string">"24"</span>
}
</pre>
<p>The <code>match</code> method can be passed a LOM document or some text, which will be parsed first. Note that <code>$NUMBER</code> is treated specially as a numerical index, so that <code>$1</code> is the first element of the resulting array, etc.</p>
</div> <!-- id="content" -->
</div> <!-- id="main" -->
<div id="about">
<i>generated by <a href="http://github.com/stevedonovan/LDoc">LDoc 1.2</a></i>
</div> <!-- id="about" -->
</div> <!-- id="container" -->
</body>
</html>
|
