version 2, including all changes.
.
Rev |
Author |
# |
Line |
1 |
perry |
1 |
PERLTIE |
|
|
2 |
!!!PERLTIE |
|
|
3 |
NAME |
|
|
4 |
SYNOPSIS |
|
|
5 |
DESCRIPTION |
|
|
6 |
SEE ALSO |
|
|
7 |
BUGS |
|
|
8 |
AUTHOR |
|
|
9 |
---- |
|
|
10 |
!!NAME |
|
|
11 |
|
|
|
12 |
|
|
|
13 |
perltie - how to hide an object class in a simple variable |
|
|
14 |
!!SYNOPSIS |
|
|
15 |
|
|
|
16 |
|
|
|
17 |
tie VARIABLE, CLASSNAME, LIST |
|
|
18 |
$object = tied VARIABLE |
|
|
19 |
untie VARIABLE |
|
|
20 |
!!DESCRIPTION |
|
|
21 |
|
|
|
22 |
|
|
|
23 |
Prior to release 5.0 of Perl, a programmer could use |
|
|
24 |
''dbmopen()'' to connect an on-disk database in the |
|
|
25 |
standard Unix dbm(3x) format magically to a |
|
|
26 |
%HASH in their program. However, their Perl was |
|
|
27 |
either built with one particular dbm library or another, but |
|
|
28 |
not both, and you couldn't extend this mechanism to other |
|
|
29 |
packages or types of variables. |
|
|
30 |
|
|
|
31 |
|
|
|
32 |
Now you can. |
|
|
33 |
|
|
|
34 |
|
|
|
35 |
The ''tie()'' function binds a variable to a class |
|
|
36 |
(package) that will provide the implementation for access |
|
|
37 |
methods for that variable. Once this magic has been |
|
|
38 |
performed, accessing a tied variable automatically triggers |
|
|
39 |
method calls in the proper class. The complexity of the |
|
|
40 |
class is hidden behind magic methods calls. The method names |
|
|
41 |
are in ALL CAPS , which is a convention that |
|
|
42 |
Perl uses to indicate that they're called implicitly rather |
|
|
43 |
than explicitly--just like the ''BEGIN |
|
|
44 |
()'' and ''END ()'' |
|
|
45 |
functions. |
|
|
46 |
|
|
|
47 |
|
|
|
48 |
In the ''tie()'' call, VARIABLE is the name of |
|
|
49 |
the variable to be enchanted. CLASSNAME is the name |
|
|
50 |
of a class implementing objects of the correct type. Any |
|
|
51 |
additional arguments in the LIST are passed to the |
|
|
52 |
appropriate constructor method for that class--meaning |
|
|
53 |
''TIESCALAR ()'', |
|
|
54 |
''TIEARRAY ()'', ''TIEHASH |
|
|
55 |
()'', or ''TIEHANDLE ()''. (Typically |
|
|
56 |
these are arguments such as might be passed to the |
|
|
57 |
''dbminit()'' function of C.) The object returned by the |
|
|
58 |
``new'' method is also returned by the ''tie()'' |
|
|
59 |
function, which would be useful if you wanted to access |
|
|
60 |
other methods in CLASSNAME. (You don't actually |
|
|
61 |
have to return a reference to a right ``type'' (e.g., |
|
|
62 |
HASH or CLASSNAME) so long as it's a |
|
|
63 |
properly blessed object.) You can also retrieve a reference |
|
|
64 |
to the underlying object using the ''tied()'' |
|
|
65 |
function. |
|
|
66 |
|
|
|
67 |
|
|
|
68 |
Unlike ''dbmopen()'', the ''tie()'' function will not |
|
|
69 |
use or require a module for you--you need |
|
|
70 |
to do that explicitly yourself. |
|
|
71 |
|
|
|
72 |
|
|
|
73 |
__Tying Scalars__ |
|
|
74 |
|
|
|
75 |
|
|
|
76 |
A class implementing a tied scalar should define the |
|
|
77 |
following methods: TIESCALAR , |
|
|
78 |
FETCH , STORE , and possibly |
|
|
79 |
UNTIE and/or DESTROY |
|
|
80 |
. |
|
|
81 |
|
|
|
82 |
|
|
|
83 |
Let's look at each in turn, using as an example a tie class |
|
|
84 |
for scalars that allows the user to do something |
|
|
85 |
like: |
|
|
86 |
|
|
|
87 |
|
|
|
88 |
tie $his_speed, 'Nice', getppid(); |
|
|
89 |
tie $my_speed, 'Nice', $$; |
|
|
90 |
And now whenever either of those variables is accessed, its current system priority is retrieved and returned. If those variables are set, then the process's priority is changed! |
|
|
91 |
|
|
|
92 |
|
|
|
93 |
We'll use Jarkko Hietaniemi jhi@iki.fi'' |
|
|
94 |
''PRIO_PROCESS , PRIO_MIN , and |
|
|
95 |
PRIO_MAX constants from your system, as well |
|
|
96 |
as the ''getpriority()'' and ''setpriority()'' system |
|
|
97 |
calls. Here's the preamble of the class. |
|
|
98 |
|
|
|
99 |
|
|
|
100 |
package Nice; |
|
|
101 |
use Carp; |
|
|
102 |
use BSD::Resource; |
|
|
103 |
use strict; |
|
|
104 |
$Nice::DEBUG = 0 unless defined $Nice::DEBUG; |
|
|
105 |
TIESCALAR classname, LIST |
|
|
106 |
|
|
|
107 |
|
|
|
108 |
This is the constructor for the class. That means it is |
|
|
109 |
expected to return a blessed reference to a new scalar |
|
|
110 |
(probably anonymous) that it's creating. For |
|
|
111 |
example: |
|
|
112 |
|
|
|
113 |
|
|
|
114 |
sub TIESCALAR { |
|
|
115 |
my $class = shift; |
|
|
116 |
my $pid = shift $$; # 0 means me |
|
|
117 |
if ($pid !~ /^d+$/) { |
|
|
118 |
carp |
|
|
119 |
unless (kill 0, $pid) { # EPERM or ERSCH, no doubt |
|
|
120 |
carp |
|
|
121 |
return bless $pid, $class; |
|
|
122 |
} |
|
|
123 |
This tie class has chosen to return an error rather than raising an exception if its constructor should fail. While this is how ''dbmopen()'' works, other classes may well not wish to be so forgiving. It checks the global variable $^W to see whether to emit a bit of noise anyway. |
|
|
124 |
|
|
|
125 |
|
|
|
126 |
FETCH this |
|
|
127 |
|
|
|
128 |
|
|
|
129 |
This method will be triggered every time the tied variable |
|
|
130 |
is accessed (read). It takes no arguments beyond its self |
|
|
131 |
reference, which is the object representing the scalar we're |
|
|
132 |
dealing with. Because in this case we're using just a |
|
|
133 |
SCALAR ref for the tied scalar object, a |
|
|
134 |
simple $$self allows the method to get at the real value |
|
|
135 |
stored there. In our example below, that real value is the |
|
|
136 |
process ID to which we've tied our |
|
|
137 |
variable. |
|
|
138 |
|
|
|
139 |
|
|
|
140 |
sub FETCH { |
|
|
141 |
my $self = shift; |
|
|
142 |
confess |
|
|
143 |
This time we've decided to blow up (raise an exception) if the renice fails--there's no place for us to return an error otherwise, and it's probably the right thing to do. |
|
|
144 |
|
|
|
145 |
|
|
|
146 |
STORE this, value |
|
|
147 |
|
|
|
148 |
|
|
|
149 |
This method will be triggered every time the tied variable |
|
|
150 |
is set (assigned). Beyond its self reference, it also |
|
|
151 |
expects one (and only one) argument--the new value the user |
|
|
152 |
is trying to assign. |
|
|
153 |
|
|
|
154 |
|
|
|
155 |
sub STORE { |
|
|
156 |
my $self = shift; |
|
|
157 |
confess |
|
|
158 |
if ($new_nicety |
|
|
159 |
if ($new_nicety |
|
|
160 |
unless (defined setpriority(PRIO_PROCESS, $$self, $new_nicety)) { |
|
|
161 |
confess |
|
|
162 |
|
|
|
163 |
|
|
|
164 |
UNTIE this |
|
|
165 |
|
|
|
166 |
|
|
|
167 |
This method will be triggered when the untie |
|
|
168 |
occurs. This can be useful if the class needs to know when |
|
|
169 |
no further calls will be made. (Except |
|
|
170 |
DESTROY of course.) See below for more |
|
|
171 |
details. |
|
|
172 |
|
|
|
173 |
|
|
|
174 |
DESTROY this |
|
|
175 |
|
|
|
176 |
|
|
|
177 |
This method will be triggered when the tied variable needs |
|
|
178 |
to be destructed. As with other object classes, such a |
|
|
179 |
method is seldom necessary, because Perl deallocates its |
|
|
180 |
moribund object's memory for you automatically--this isn't C |
|
|
181 |
++ , you know. We'll use a |
|
|
182 |
DESTROY method here for debugging purposes |
|
|
183 |
only. |
|
|
184 |
|
|
|
185 |
|
|
|
186 |
sub DESTROY { |
|
|
187 |
my $self = shift; |
|
|
188 |
confess |
|
|
189 |
|
|
|
190 |
|
|
|
191 |
That's about all there is to it. Actually, it's more than |
|
|
192 |
all there is to it, because we've done a few nice things |
|
|
193 |
here for the sake of completeness, robustness, and general |
|
|
194 |
aesthetics. Simpler TIESCALAR classes are |
|
|
195 |
certainly possible. |
|
|
196 |
|
|
|
197 |
|
|
|
198 |
__Tying Arrays__ |
|
|
199 |
|
|
|
200 |
|
|
|
201 |
A class implementing a tied ordinary array should define the |
|
|
202 |
following methods: TIEARRAY , |
|
|
203 |
FETCH , STORE , |
|
|
204 |
FETCHSIZE , STORESIZE and |
|
|
205 |
perhaps UNTIE and/or DESTROY |
|
|
206 |
. |
|
|
207 |
|
|
|
208 |
|
|
|
209 |
FETCHSIZE and STORESIZE are |
|
|
210 |
used to provide $#array and equivalent |
|
|
211 |
scalar(@array) access. |
|
|
212 |
|
|
|
213 |
|
|
|
214 |
The methods POP , PUSH , |
|
|
215 |
SHIFT , UNSHIFT , |
|
|
216 |
SPLICE , DELETE , and |
|
|
217 |
EXISTS are required if the perl operator with |
|
|
218 |
the corresponding (but lowercase) name is to operate on the |
|
|
219 |
tied array. The __Tie::Array__ class can be used as a |
|
|
220 |
base class to implement the first five of these in terms of |
|
|
221 |
the basic methods above. The default implementations of |
|
|
222 |
DELETE and EXISTS in |
|
|
223 |
__Tie::Array__ simply croak. |
|
|
224 |
|
|
|
225 |
|
|
|
226 |
In addition EXTEND will be called when perl |
|
|
227 |
would have pre-extended allocation in a real |
|
|
228 |
array. |
|
|
229 |
|
|
|
230 |
|
|
|
231 |
For this discussion, we'll implement an array whose elements |
|
|
232 |
are a fixed size at creation. If you try to create an |
|
|
233 |
element larger than the fixed size, you'll take an |
|
|
234 |
exception. For example: |
|
|
235 |
|
|
|
236 |
|
2 |
perry |
237 |
use !FixedElem_Array; |
|
|
238 |
tie @array, '!FixedElem_Array', 3; |
1 |
perry |
239 |
$array[[0] = 'cat'; # ok. |
|
|
240 |
$array[[1] = 'dogs'; # exception, length('dogs') |
|
|
241 |
The preamble code for the class is as follows: |
|
|
242 |
|
|
|
243 |
|
2 |
perry |
244 |
package !FixedElem_Array; |
1 |
perry |
245 |
use Carp; |
|
|
246 |
use strict; |
|
|
247 |
TIEARRAY classname, LIST |
|
|
248 |
|
|
|
249 |
|
|
|
250 |
This is the constructor for the class. That means it is |
|
|
251 |
expected to return a blessed reference through which the new |
|
|
252 |
array (probably an anonymous ARRAY ref) will |
|
|
253 |
be accessed. |
|
|
254 |
|
|
|
255 |
|
|
|
256 |
In our example, just to show you that you don't |
|
|
257 |
''really'' have to return an ARRAY |
|
|
258 |
reference, we'll choose a HASH reference to |
|
|
259 |
represent our object. A HASH works out well |
|
|
260 |
as a generic record type: the {ELEMSIZE} field will |
|
|
261 |
store the maximum element size allowed, and the |
|
|
262 |
{ARRAY} field will hold the true |
|
|
263 |
ARRAY ref. If someone outside the class tries |
|
|
264 |
to dereference the object returned (doubtless thinking it an |
|
|
265 |
ARRAY ref), they'll blow up. This just goes |
|
|
266 |
to show you that you should respect an object's |
|
|
267 |
privacy. |
|
|
268 |
|
|
|
269 |
|
|
|
270 |
sub TIEARRAY { |
|
|
271 |
my $class = shift; |
|
|
272 |
my $elemsize = shift; |
|
|
273 |
if ( @_ $elemsize =~ /D/ ) { |
|
|
274 |
croak |
|
|
275 |
|
|
|
276 |
|
|
|
277 |
FETCH this, index |
|
|
278 |
|
|
|
279 |
|
|
|
280 |
This method will be triggered every time an individual |
|
|
281 |
element the tied array is accessed (read). It takes one |
|
|
282 |
argument beyond its self reference: the index whose value |
|
|
283 |
we're trying to fetch. |
|
|
284 |
|
|
|
285 |
|
|
|
286 |
sub FETCH { |
|
|
287 |
my $self = shift; |
|
|
288 |
my $index = shift; |
|
|
289 |
return $self- |
|
|
290 |
If a negative array index is used to read from an array, the index will be translated to a positive one internally by calling FETCHSIZE before being passed to FETCH . |
|
|
291 |
|
|
|
292 |
|
|
|
293 |
As you may have noticed, the name of the |
|
|
294 |
FETCH method (et al.) is the same for all |
|
|
295 |
accesses, even though the constructors differ in names ( |
|
|
296 |
TIESCALAR vs TIEARRAY ). While |
|
|
297 |
in theory you could have the same class servicing several |
|
|
298 |
tied types, in practice this becomes cumbersome, and it's |
|
|
299 |
easiest to keep them at simply one tie type per |
|
|
300 |
class. |
|
|
301 |
|
|
|
302 |
|
|
|
303 |
STORE this, index, value |
|
|
304 |
|
|
|
305 |
|
|
|
306 |
This method will be triggered every time an element in the |
|
|
307 |
tied array is set (written). It takes two arguments beyond |
|
|
308 |
its self reference: the index at which we're trying to store |
|
|
309 |
something and the value we're trying to put |
|
|
310 |
there. |
|
|
311 |
|
|
|
312 |
|
|
|
313 |
In our example, undef is really |
|
|
314 |
$self- number of spaces so we have a |
|
|
315 |
little more work to do here: |
|
|
316 |
|
|
|
317 |
|
|
|
318 |
sub STORE { |
|
|
319 |
my $self = shift; |
|
|
320 |
my( $index, $value ) = @_; |
|
|
321 |
if ( length $value |
|
|
322 |
Negative indexes are treated the same as with FETCH . |
|
|
323 |
|
|
|
324 |
|
|
|
325 |
FETCHSIZE this |
|
|
326 |
|
|
|
327 |
|
|
|
328 |
Returns the total number of items in the tied array |
|
|
329 |
associated with object ''this''. (Equivalent to |
|
|
330 |
scalar(@array)). For example: |
|
|
331 |
|
|
|
332 |
|
|
|
333 |
sub FETCHSIZE { |
|
|
334 |
my $self = shift; |
|
|
335 |
return scalar @{$self- |
|
|
336 |
|
|
|
337 |
|
|
|
338 |
STORESIZE this, count |
|
|
339 |
|
|
|
340 |
|
|
|
341 |
Sets the total number of items in the tied array associated |
|
|
342 |
with object ''this'' to be ''count''. If this makes |
|
|
343 |
the array larger then class's mapping of undef |
|
|
344 |
should be returned for new positions. If the array becomes |
|
|
345 |
smaller then entries beyond count should be |
|
|
346 |
deleted. |
|
|
347 |
|
|
|
348 |
|
|
|
349 |
In our example, 'undef' is really an element containing |
|
|
350 |
$self- number of spaces. |
|
|
351 |
Observe: |
|
|
352 |
|
|
|
353 |
|
|
|
354 |
sub STORESIZE { |
|
|
355 |
my $self = shift; |
|
|
356 |
my $count = shift; |
|
|
357 |
if ( $count |
|
|
358 |
|
|
|
359 |
|
|
|
360 |
EXTEND this, count |
|
|
361 |
|
|
|
362 |
|
|
|
363 |
Informative call that array is likely to grow to have |
|
|
364 |
''count'' entries. Can be used to optimize allocation. |
|
|
365 |
This method need do nothing. |
|
|
366 |
|
|
|
367 |
|
|
|
368 |
In our example, we want to make sure there are no blank |
|
|
369 |
(undef) entries, so EXTEND will make use |
|
|
370 |
of STORESIZE to fill elements as |
|
|
371 |
needed: |
|
|
372 |
|
|
|
373 |
|
|
|
374 |
sub EXTEND { |
|
|
375 |
my $self = shift; |
|
|
376 |
my $count = shift; |
|
|
377 |
$self- |
|
|
378 |
|
|
|
379 |
|
|
|
380 |
EXISTS this, key |
|
|
381 |
|
|
|
382 |
|
|
|
383 |
Verify that the element at index ''key'' exists in the |
|
|
384 |
tied array ''this''. |
|
|
385 |
|
|
|
386 |
|
|
|
387 |
In our example, we will determine that if an element |
|
|
388 |
consists of $self- spaces only, it |
|
|
389 |
does not exist: |
|
|
390 |
|
|
|
391 |
|
|
|
392 |
sub EXISTS { |
|
|
393 |
my $self = shift; |
|
|
394 |
my $index = shift; |
|
|
395 |
return 0 if ! defined $self- |
|
|
396 |
|
|
|
397 |
|
|
|
398 |
DELETE this, key |
|
|
399 |
|
|
|
400 |
|
|
|
401 |
Delete the element at index ''key'' from the tied array |
|
|
402 |
''this''. |
|
|
403 |
|
|
|
404 |
|
|
|
405 |
In our example, a deleted item is $self-{ |
|
|
406 |
ELEMSIZE } |
|
|
407 |
|
|
|
408 |
|
|
|
409 |
sub DELETE { |
|
|
410 |
my $self = shift; |
|
|
411 |
my $index = shift; |
|
|
412 |
return $self- |
|
|
413 |
|
|
|
414 |
|
|
|
415 |
CLEAR this |
|
|
416 |
|
|
|
417 |
|
|
|
418 |
Clear (remove, delete, ...) all values from the tied array |
|
|
419 |
associated with object ''this''. For |
|
|
420 |
example: |
|
|
421 |
|
|
|
422 |
|
|
|
423 |
sub CLEAR { |
|
|
424 |
my $self = shift; |
|
|
425 |
return $self- |
|
|
426 |
|
|
|
427 |
|
|
|
428 |
PUSH this, LIST |
|
|
429 |
|
|
|
430 |
|
|
|
431 |
Append elements of ''LIST'' to the array. |
|
|
432 |
For example: |
|
|
433 |
|
|
|
434 |
|
|
|
435 |
sub PUSH { |
|
|
436 |
my $self = shift; |
|
|
437 |
my @list = @_; |
|
|
438 |
my $last = $self- |
|
|
439 |
|
|
|
440 |
|
|
|
441 |
POP this |
|
|
442 |
|
|
|
443 |
|
|
|
444 |
Remove last element of the array and return it. For |
|
|
445 |
example: |
|
|
446 |
|
|
|
447 |
|
|
|
448 |
sub POP { |
|
|
449 |
my $self = shift; |
|
|
450 |
return pop @{$self- |
|
|
451 |
|
|
|
452 |
|
|
|
453 |
SHIFT this |
|
|
454 |
|
|
|
455 |
|
|
|
456 |
Remove the first element of the array (shifting other |
|
|
457 |
elements down) and return it. For example: |
|
|
458 |
|
|
|
459 |
|
|
|
460 |
sub SHIFT { |
|
|
461 |
my $self = shift; |
|
|
462 |
return shift @{$self- |
|
|
463 |
|
|
|
464 |
|
|
|
465 |
UNSHIFT this, |
|
|
466 |
LIST |
|
|
467 |
|
|
|
468 |
|
|
|
469 |
Insert LIST elements at the beginning of the |
|
|
470 |
array, moving existing elements up to make room. For |
|
|
471 |
example: |
|
|
472 |
|
|
|
473 |
|
|
|
474 |
sub UNSHIFT { |
|
|
475 |
my $self = shift; |
|
|
476 |
my @list = @_; |
|
|
477 |
my $size = scalar( @list ); |
|
|
478 |
# make room for our list |
|
|
479 |
@{$self- |
|
|
480 |
|
|
|
481 |
|
|
|
482 |
SPLICE this, offset, length, |
|
|
483 |
LIST |
|
|
484 |
|
|
|
485 |
|
|
|
486 |
Perform the equivalent of splice on the |
|
|
487 |
array. |
|
|
488 |
|
|
|
489 |
|
|
|
490 |
''offset'' is optional and defaults to zero, negative |
|
|
491 |
values count back from the end of the array. |
|
|
492 |
|
|
|
493 |
|
|
|
494 |
''length'' is optional and defaults to rest of the |
|
|
495 |
array. |
|
|
496 |
|
|
|
497 |
|
|
|
498 |
''LIST'' may be empty. |
|
|
499 |
|
|
|
500 |
|
|
|
501 |
Returns a list of the original ''length'' elements at |
|
|
502 |
''offset''. |
|
|
503 |
|
|
|
504 |
|
|
|
505 |
In our example, we'll use a little shortcut if there is a |
|
|
506 |
''LIST'' : |
|
|
507 |
|
|
|
508 |
|
|
|
509 |
sub SPLICE { |
|
|
510 |
my $self = shift; |
|
|
511 |
my $offset = shift 0; |
|
|
512 |
my $length = shift $self- |
|
|
513 |
|
|
|
514 |
|
|
|
515 |
UNTIE this |
|
|
516 |
|
|
|
517 |
|
|
|
518 |
Will be called when untie happens. (See |
|
|
519 |
below.) |
|
|
520 |
|
|
|
521 |
|
|
|
522 |
DESTROY this |
|
|
523 |
|
|
|
524 |
|
|
|
525 |
This method will be triggered when the tied variable needs |
|
|
526 |
to be destructed. As with the scalar tie class, this is |
|
|
527 |
almost never needed in a language that does its own garbage |
|
|
528 |
collection, so this time we'll just leave it |
|
|
529 |
out. |
|
|
530 |
|
|
|
531 |
|
|
|
532 |
__Tying Hashes__ |
|
|
533 |
|
|
|
534 |
|
|
|
535 |
Hashes were the first Perl data type to be tied (see |
|
|
536 |
''dbmopen()''). A class implementing a tied hash should |
|
|
537 |
define the following methods: TIEHASH is the |
|
|
538 |
constructor. FETCH and STORE |
|
|
539 |
access the key and value pairs. EXISTS |
|
|
540 |
reports whether a key is present in the hash, and |
|
|
541 |
DELETE deletes one. CLEAR |
|
|
542 |
empties the hash by deleting all the key and value pairs. |
|
|
543 |
FIRSTKEY and NEXTKEY implement |
|
|
544 |
the ''keys()'' and ''each()'' functions to iterate |
|
|
545 |
over all the keys. UNTIE is called when |
|
|
546 |
untie happens, and DESTROY is called |
|
|
547 |
when the tied variable is garbage collected. |
|
|
548 |
|
|
|
549 |
|
|
|
550 |
If this seems like a lot, then feel free to inherit from |
|
|
551 |
merely the standard Tie::Hash module for most of your |
|
|
552 |
methods, redefining only the interesting ones. See Tie::Hash |
|
|
553 |
for details. |
|
|
554 |
|
|
|
555 |
|
|
|
556 |
Remember that Perl distinguishes between a key not existing |
|
|
557 |
in the hash, and the key existing in the hash but having a |
|
|
558 |
corresponding value of undef. The two possibilities |
|
|
559 |
can be tested with the exists() and |
|
|
560 |
defined() functions. |
|
|
561 |
|
|
|
562 |
|
|
|
563 |
Here's an example of a somewhat interesting tied hash class: |
|
|
564 |
it gives you a hash representing a particular user's dot |
|
|
565 |
files. You index into the hash with the name of the file |
|
|
566 |
(minus the dot) and you get back that dot file's contents. |
|
|
567 |
For example: |
|
|
568 |
|
|
|
569 |
|
2 |
perry |
570 |
use !DotFiles; |
|
|
571 |
tie %dot, '!DotFiles'; |
1 |
perry |
572 |
if ( $dot{profile} =~ /MANPATH/ |
|
|
573 |
$dot{login} =~ /MANPATH/ |
|
|
574 |
$dot{cshrc} =~ /MANPATH/ ) |
|
|
575 |
{ |
|
|
576 |
print |
|
|
577 |
Or here's another sample of using our tied class: |
|
|
578 |
|
|
|
579 |
|
2 |
perry |
580 |
tie %him, '!DotFiles', 'daemon'; |
1 |
perry |
581 |
foreach $f ( keys %him ) { |
|
|
582 |
printf |
2 |
perry |
583 |
In our tied hash !DotFiles example, we use a regular hash for the object containing several important fields, of which only the {LIST} field will be what the user thinks of as the real hash. |
1 |
perry |
584 |
|
|
|
585 |
|
|
|
586 |
USER |
|
|
587 |
|
|
|
588 |
|
|
|
589 |
whose dot files this object represents |
|
|
590 |
|
|
|
591 |
|
|
|
592 |
HOME |
|
|
593 |
|
|
|
594 |
|
|
|
595 |
where those dot files live |
|
|
596 |
|
|
|
597 |
|
|
|
598 |
CLOBBER |
|
|
599 |
|
|
|
600 |
|
|
|
601 |
whether we should try to change or remove those dot |
|
|
602 |
files |
|
|
603 |
|
|
|
604 |
|
|
|
605 |
LIST |
|
|
606 |
|
|
|
607 |
|
|
|
608 |
the hash of dot file names and content mappings |
|
|
609 |
|
|
|
610 |
|
|
|
611 |
Here's the start of ''Dotfiles.pm'': |
|
|
612 |
|
|
|
613 |
|
2 |
perry |
614 |
package !DotFiles; |
1 |
perry |
615 |
use Carp; |
|
|
616 |
sub whowasi { (caller(1))[[3] . '()' } |
|
|
617 |
my $DEBUG = 0; |
|
|
618 |
sub debug { $DEBUG = @_ ? shift : 1 } |
|
|
619 |
For our example, we want to be able to emit debugging info to help in tracing during development. We keep also one convenience function around internally to help print out warnings; ''whowasi()'' returns the function name that calls it. |
|
|
620 |
|
|
|
621 |
|
2 |
perry |
622 |
Here are the methods for the !DotFiles tied |
1 |
perry |
623 |
hash. |
|
|
624 |
|
|
|
625 |
|
|
|
626 |
TIEHASH classname, |
|
|
627 |
LIST |
|
|
628 |
|
|
|
629 |
|
|
|
630 |
This is the constructor for the class. That means it is |
|
|
631 |
expected to return a blessed reference through which the new |
|
|
632 |
object (probably but not necessarily an anonymous hash) will |
|
|
633 |
be accessed. |
|
|
634 |
|
|
|
635 |
|
|
|
636 |
Here's the constructor: |
|
|
637 |
|
|
|
638 |
|
|
|
639 |
sub TIEHASH { |
|
|
640 |
my $self = shift; |
|
|
641 |
my $user = shift $ |
|
|
642 |
my $node = { |
|
|
643 |
USER = |
|
|
644 |
opendir(DIR, $dir) |
|
|
645 |
croak |
|
|
646 |
It's probably worth mentioning that if you're going to filetest the return values out of a readdir, you'd better prepend the directory in question. Otherwise, because we didn't ''chdir()'' there, it would have been testing the wrong file. |
|
|
647 |
|
|
|
648 |
|
|
|
649 |
FETCH this, key |
|
|
650 |
|
|
|
651 |
|
|
|
652 |
This method will be triggered every time an element in the |
|
|
653 |
tied hash is accessed (read). It takes one argument beyond |
|
|
654 |
its self reference: the key whose value we're trying to |
|
|
655 |
fetch. |
|
|
656 |
|
|
|
657 |
|
2 |
perry |
658 |
Here's the fetch for our !DotFiles example. |
1 |
perry |
659 |
|
|
|
660 |
|
|
|
661 |
sub FETCH { |
|
|
662 |
carp |
|
|
663 |
unless (exists $self- |
|
|
664 |
if (defined $self- |
|
|
665 |
It was easy to write by having it call the Unix cat(1) command, but it would probably be more portable to open the file manually (and somewhat more efficient). Of course, because dot files are a Unixy concept, we're not that concerned. |
|
|
666 |
|
|
|
667 |
|
|
|
668 |
STORE this, key, value |
|
|
669 |
|
|
|
670 |
|
|
|
671 |
This method will be triggered every time an element in the |
|
|
672 |
tied hash is set (written). It takes two arguments beyond |
|
|
673 |
its self reference: the index at which we're trying to store |
|
|
674 |
something, and the value we're trying to put |
|
|
675 |
there. |
|
|
676 |
|
|
|
677 |
|
2 |
perry |
678 |
Here in our !DotFiles example, we'll be careful not to let |
1 |
perry |
679 |
them try to overwrite the file unless they've called the |
|
|
680 |
''clobber()'' method on the original object reference |
|
|
681 |
returned by ''tie()''. |
|
|
682 |
|
|
|
683 |
|
|
|
684 |
sub STORE { |
|
|
685 |
carp |
|
|
686 |
croak |
|
|
687 |
open(F, |
|
|
688 |
If they wanted to clobber something, they might say: |
|
|
689 |
|
|
|
690 |
|
|
|
691 |
$ob = tie %daemon_dots, 'daemon'; |
|
|
692 |
$ob- |
|
|
693 |
Another way to lay hands on a reference to the underlying object is to use the ''tied()'' function, so they might alternately have set clobber using: |
|
|
694 |
|
|
|
695 |
|
|
|
696 |
tie %daemon_dots, 'daemon'; |
|
|
697 |
tied(%daemon_dots)- |
|
|
698 |
The clobber method is simply: |
|
|
699 |
|
|
|
700 |
|
|
|
701 |
sub clobber { |
|
|
702 |
my $self = shift; |
|
|
703 |
$self- |
|
|
704 |
|
|
|
705 |
|
|
|
706 |
DELETE this, key |
|
|
707 |
|
|
|
708 |
|
|
|
709 |
This method is triggered when we remove an element from the |
|
|
710 |
hash, typically by using the ''delete()'' function. |
|
|
711 |
Again, we'll be careful to check whether they really want to |
|
|
712 |
clobber files. |
|
|
713 |
|
|
|
714 |
|
|
|
715 |
sub DELETE { |
|
|
716 |
carp |
|
|
717 |
my $self = shift; |
|
|
718 |
my $dot = shift; |
|
|
719 |
my $file = $self- |
|
|
720 |
The value returned by DELETE becomes the return value of the call to ''delete()''. If you want to emulate the normal behavior of ''delete()'', you should return whatever FETCH would have returned for this key. In this example, we have chosen instead to return a value which tells the caller whether the file was successfully deleted. |
|
|
721 |
|
|
|
722 |
|
|
|
723 |
CLEAR this |
|
|
724 |
|
|
|
725 |
|
|
|
726 |
This method is triggered when the whole hash is to be |
|
|
727 |
cleared, usually by assigning the empty list to |
|
|
728 |
it. |
|
|
729 |
|
|
|
730 |
|
|
|
731 |
In our example, that would remove all the user's dot files! |
|
|
732 |
It's such a dangerous thing that they'll have to set |
|
|
733 |
CLOBBER to something higher than 1 to make it |
|
|
734 |
happen. |
|
|
735 |
|
|
|
736 |
|
|
|
737 |
sub CLEAR { |
|
|
738 |
carp |
|
|
739 |
|
|
|
740 |
|
|
|
741 |
EXISTS this, key |
|
|
742 |
|
|
|
743 |
|
|
|
744 |
This method is triggered when the user uses the |
|
|
745 |
''exists()'' function on a particular hash. In our |
|
|
746 |
example, we'll look at the {LIST} hash element for |
|
|
747 |
this: |
|
|
748 |
|
|
|
749 |
|
|
|
750 |
sub EXISTS { |
|
|
751 |
carp |
|
|
752 |
|
|
|
753 |
|
|
|
754 |
FIRSTKEY this |
|
|
755 |
|
|
|
756 |
|
|
|
757 |
This method will be triggered when the user is going to |
|
|
758 |
iterate through the hash, such as via a ''keys()'' or |
|
|
759 |
''each()'' call. |
|
|
760 |
|
|
|
761 |
|
|
|
762 |
sub FIRSTKEY { |
|
|
763 |
carp |
|
|
764 |
|
|
|
765 |
|
|
|
766 |
NEXTKEY this, lastkey |
|
|
767 |
|
|
|
768 |
|
|
|
769 |
This method gets triggered during a ''keys()'' or |
|
|
770 |
''each()'' iteration. It has a second argument which is |
|
|
771 |
the last key that had been accessed. This is useful if |
|
|
772 |
you're carrying about ordering or calling the iterator from |
|
|
773 |
more than one sequence, or not really storing things in a |
|
|
774 |
hash anywhere. |
|
|
775 |
|
|
|
776 |
|
|
|
777 |
For our example, we're using a real hash so we'll do just |
|
|
778 |
the simple thing, but we'll have to go through the |
|
|
779 |
LIST field indirectly. |
|
|
780 |
|
|
|
781 |
|
|
|
782 |
sub NEXTKEY { |
|
|
783 |
carp |
|
|
784 |
|
|
|
785 |
|
|
|
786 |
UNTIE this |
|
|
787 |
|
|
|
788 |
|
|
|
789 |
This is called when untie occurs. |
|
|
790 |
|
|
|
791 |
|
|
|
792 |
DESTROY this |
|
|
793 |
|
|
|
794 |
|
|
|
795 |
This method is triggered when a tied hash is about to go out |
|
|
796 |
of scope. You don't really need it unless you're trying to |
|
|
797 |
add debugging or have auxiliary state to clean up. Here's a |
|
|
798 |
very simple function: |
|
|
799 |
|
|
|
800 |
|
|
|
801 |
sub DESTROY { |
|
|
802 |
carp |
|
|
803 |
|
|
|
804 |
|
|
|
805 |
Note that functions such as ''keys()'' and |
|
|
806 |
''values()'' may return huge lists when used on large |
|
|
807 |
objects, like DBM files. You may prefer to |
|
|
808 |
use the ''each()'' function to iterate over such. |
|
|
809 |
Example: |
|
|
810 |
|
|
|
811 |
|
|
|
812 |
# print out history file offsets |
|
|
813 |
use NDBM_File; |
|
|
814 |
tie(%HIST, 'NDBM_File', '/usr/lib/news/history', 1, 0); |
|
|
815 |
while (($key,$val) = each %HIST) { |
|
|
816 |
print $key, ' = ', unpack('L',$val), |
|
|
817 |
|
|
|
818 |
|
2 |
perry |
819 |
__Tying !FileHandles__ |
1 |
perry |
820 |
|
|
|
821 |
|
|
|
822 |
This is partially implemented now. |
|
|
823 |
|
|
|
824 |
|
|
|
825 |
A class implementing a tied filehandle should define the |
|
|
826 |
following methods: TIEHANDLE , at least one |
|
|
827 |
of PRINT , PRINTF , |
|
|
828 |
WRITE , READLINE , |
|
|
829 |
GETC , READ , and possibly |
|
|
830 |
CLOSE , UNTIE and |
|
|
831 |
DESTROY . The class can also provide: |
|
|
832 |
BINMODE , OPEN , |
|
|
833 |
EOF , FILENO , |
|
|
834 |
SEEK , TELL - if the |
|
|
835 |
corresponding perl operators are used on the |
|
|
836 |
handle. |
|
|
837 |
|
|
|
838 |
|
|
|
839 |
It is especially useful when perl is embedded in some other |
|
|
840 |
program, where output to STDOUT and |
|
|
841 |
STDERR may have to be redirected in some |
|
|
842 |
special way. See nvi and the Apache module for |
|
|
843 |
examples. |
|
|
844 |
|
|
|
845 |
|
|
|
846 |
In our example we're going to create a shouting |
|
|
847 |
handle. |
|
|
848 |
|
|
|
849 |
|
|
|
850 |
package Shout; |
|
|
851 |
TIEHANDLE classname, LIST |
|
|
852 |
|
|
|
853 |
|
|
|
854 |
This is the constructor for the class. That means it is |
|
|
855 |
expected to return a blessed reference of some sort. The |
|
|
856 |
reference can be used to hold some internal |
|
|
857 |
information. |
|
|
858 |
|
|
|
859 |
|
|
|
860 |
sub TIEHANDLE { print |
|
|
861 |
|
|
|
862 |
|
|
|
863 |
WRITE this, LIST |
|
|
864 |
|
|
|
865 |
|
|
|
866 |
This method will be called when the handle is written to via |
|
|
867 |
the syswrite function. |
|
|
868 |
|
|
|
869 |
|
|
|
870 |
sub WRITE { |
|
|
871 |
$r = shift; |
|
|
872 |
my($buf,$len,$offset) = @_; |
|
|
873 |
print |
|
|
874 |
|
|
|
875 |
|
|
|
876 |
PRINT this, LIST |
|
|
877 |
|
|
|
878 |
|
|
|
879 |
This method will be triggered every time the tied handle is |
|
|
880 |
printed to with the print() function. Beyond its |
|
|
881 |
self reference it also expects the list that was passed to |
|
|
882 |
the print function. |
|
|
883 |
|
|
|
884 |
|
|
|
885 |
sub PRINT { $r = shift; $$r++; print join($,,map(uc($_),@_)),$\ } |
|
|
886 |
|
|
|
887 |
|
|
|
888 |
PRINTF this, LIST |
|
|
889 |
|
|
|
890 |
|
|
|
891 |
This method will be triggered every time the tied handle is |
|
|
892 |
printed to with the printf() function. Beyond its |
|
|
893 |
self reference it also expects the format and list that was |
|
|
894 |
passed to the printf function. |
|
|
895 |
|
|
|
896 |
|
|
|
897 |
sub PRINTF { |
|
|
898 |
shift; |
|
|
899 |
my $fmt = shift; |
|
|
900 |
print sprintf($fmt, @_). |
|
|
901 |
|
|
|
902 |
|
|
|
903 |
READ this, LIST |
|
|
904 |
|
|
|
905 |
|
|
|
906 |
This method will be called when the handle is read from via |
|
|
907 |
the read or sysread |
|
|
908 |
functions. |
|
|
909 |
|
|
|
910 |
|
|
|
911 |
sub READ { |
|
|
912 |
my $self = shift; |
|
|
913 |
my $$bufref = $_[[0]; |
|
|
914 |
my(undef,$len,$offset) = @_; |
|
|
915 |
print |
|
|
916 |
|
|
|
917 |
|
|
|
918 |
READLINE this |
|
|
919 |
|
|
|
920 |
|
|
|
921 |
This method will be called when the handle is read from via |
|
|
922 |
HANDLE |
|
|
923 |
|
|
|
924 |
|
|
|
925 |
sub READLINE { $r = shift; |
|
|
926 |
|
|
|
927 |
|
|
|
928 |
GETC this |
|
|
929 |
|
|
|
930 |
|
|
|
931 |
This method will be called when the getc function |
|
|
932 |
is called. |
|
|
933 |
|
|
|
934 |
|
|
|
935 |
sub GETC { print |
|
|
936 |
|
|
|
937 |
|
|
|
938 |
CLOSE this |
|
|
939 |
|
|
|
940 |
|
|
|
941 |
This method will be called when the handle is closed via the |
|
|
942 |
close function. |
|
|
943 |
|
|
|
944 |
|
|
|
945 |
sub CLOSE { print |
|
|
946 |
|
|
|
947 |
|
|
|
948 |
UNTIE this |
|
|
949 |
|
|
|
950 |
|
|
|
951 |
As with the other types of ties, this method will be called |
|
|
952 |
when untie happens. It may be appropriate to ``auto |
|
|
953 |
CLOSE '' when this occurs. |
|
|
954 |
|
|
|
955 |
|
|
|
956 |
DESTROY this |
|
|
957 |
|
|
|
958 |
|
|
|
959 |
As with the other types of ties, this method will be called |
|
|
960 |
when the tied handle is about to be destroyed. This is |
|
|
961 |
useful for debugging and possibly cleaning up. |
|
|
962 |
|
|
|
963 |
|
|
|
964 |
sub DESTROY { print |
|
|
965 |
|
|
|
966 |
|
|
|
967 |
Here's how to use our little example: |
|
|
968 |
|
|
|
969 |
|
|
|
970 |
tie(*FOO,'Shout'); |
|
|
971 |
print FOO |
|
|
972 |
|
|
|
973 |
|
|
|
974 |
__UNTIE this__ |
|
|
975 |
|
|
|
976 |
|
|
|
977 |
You can define for all tie types an UNTIE |
|
|
978 |
method that will be called at ''untie()''. |
|
|
979 |
|
|
|
980 |
|
|
|
981 |
__The__ untie __Gotcha__ |
|
|
982 |
|
|
|
983 |
|
|
|
984 |
If you intend making use of the object returned from either |
|
|
985 |
''tie()'' or ''tied()'', and if the tie's target class |
|
|
986 |
defines a destructor, there is a subtle gotcha you |
|
|
987 |
''must'' guard against. |
|
|
988 |
|
|
|
989 |
|
|
|
990 |
As setup, consider this (admittedly rather contrived) |
|
|
991 |
example of a tie; all it does is use a file to keep a log of |
|
|
992 |
the values assigned to a scalar. |
|
|
993 |
|
|
|
994 |
|
|
|
995 |
package Remember; |
|
|
996 |
use strict; |
|
|
997 |
use warnings; |
|
|
998 |
use IO::File; |
|
|
999 |
sub TIESCALAR { |
|
|
1000 |
my $class = shift; |
|
|
1001 |
my $filename = shift; |
|
|
1002 |
my $handle = new IO::File |
|
|
1003 |
print $handle |
|
|
1004 |
sub FETCH { |
|
|
1005 |
my $self = shift; |
|
|
1006 |
return $self- |
|
|
1007 |
sub STORE { |
|
|
1008 |
my $self = shift; |
|
|
1009 |
my $value = shift; |
|
|
1010 |
my $handle = $self- |
|
|
1011 |
sub DESTROY { |
|
|
1012 |
my $self = shift; |
|
|
1013 |
my $handle = $self- |
|
|
1014 |
1; |
|
|
1015 |
Here is an example that makes use of this tie: |
|
|
1016 |
|
|
|
1017 |
|
|
|
1018 |
use strict; |
|
|
1019 |
use Remember; |
|
|
1020 |
my $fred; |
|
|
1021 |
tie $fred, 'Remember', 'myfile.txt'; |
|
|
1022 |
$fred = 1; |
|
|
1023 |
$fred = 4; |
|
|
1024 |
$fred = 5; |
|
|
1025 |
untie $fred; |
|
|
1026 |
system |
|
|
1027 |
This is the output when it is executed: |
|
|
1028 |
|
|
|
1029 |
|
|
|
1030 |
The Start |
|
|
1031 |
1 |
|
|
1032 |
4 |
|
|
1033 |
5 |
|
|
1034 |
The End |
|
|
1035 |
So far so good. Those of you who have been paying attention will have spotted that the tied object hasn't been used so far. So lets add an extra method to the Remember class to allow comments to be included in the file -- say, something like this: |
|
|
1036 |
|
|
|
1037 |
|
|
|
1038 |
sub comment { |
|
|
1039 |
my $self = shift; |
|
|
1040 |
my $text = shift; |
|
|
1041 |
my $handle = $self- |
|
|
1042 |
And here is the previous example modified to use the comment method (which requires the tied object): |
|
|
1043 |
|
|
|
1044 |
|
|
|
1045 |
use strict; |
|
|
1046 |
use Remember; |
|
|
1047 |
my ($fred, $x); |
|
|
1048 |
$x = tie $fred, 'Remember', 'myfile.txt'; |
|
|
1049 |
$fred = 1; |
|
|
1050 |
$fred = 4; |
|
|
1051 |
comment $x |
|
|
1052 |
When this code is executed there is no output. Here's why: |
|
|
1053 |
|
|
|
1054 |
|
|
|
1055 |
When a variable is tied, it is associated with the object |
|
|
1056 |
which is the return value of the TIESCALAR , |
|
|
1057 |
TIEARRAY , or TIEHASH |
|
|
1058 |
function. This object normally has only one reference, |
|
|
1059 |
namely, the implicit reference from the tied variable. When |
|
|
1060 |
''untie()'' is called, that reference is destroyed. Then, |
|
|
1061 |
as in the first example above, the object's destructor ( |
|
|
1062 |
DESTROY ) is called, which is normal for |
|
|
1063 |
objects that have no more valid references; and thus the |
|
|
1064 |
file is closed. |
|
|
1065 |
|
|
|
1066 |
|
|
|
1067 |
In the second example, however, we have stored another |
|
|
1068 |
reference to the tied object in $x. That means that |
|
|
1069 |
when ''untie()'' gets called there will still be a valid |
|
|
1070 |
reference to the object in existence, so the destructor is |
|
|
1071 |
not called at that time, and thus the file is not closed. |
|
|
1072 |
The reason there is no output is because the file buffers |
|
|
1073 |
have not been flushed to disk. |
|
|
1074 |
|
|
|
1075 |
|
|
|
1076 |
Now that you know what the problem is, what can you do to |
|
|
1077 |
avoid it? Prior to the introduction of the optional |
|
|
1078 |
UNTIE method the only way was the good old |
|
|
1079 |
-w flag. Which will spot any instances where you |
|
|
1080 |
call ''untie()'' and there are still valid references to |
|
|
1081 |
the tied object. If the second script above this near the |
|
|
1082 |
top use warnings 'untie' or was run with the |
|
|
1083 |
-w flag, Perl prints this warning |
|
|
1084 |
message: |
|
|
1085 |
|
|
|
1086 |
|
|
|
1087 |
untie attempted while 1 inner references still exist |
|
|
1088 |
To get the script to work properly and silence the warning make sure there are no valid references to the tied object ''before untie()'' is called: |
|
|
1089 |
|
|
|
1090 |
|
|
|
1091 |
undef $x; |
|
|
1092 |
untie $fred; |
|
|
1093 |
Now that UNTIE exists the class designer can decide which parts of the class functionality are really associated with untie and which with the object being destroyed. What makes sense for a given class depends on whether the inner references are being kept so that non-tie-related methods can be called on the object. But in most cases it probably makes sense to move the functionality that would have been in DESTROY to the UNTIE method. |
|
|
1094 |
|
|
|
1095 |
|
|
|
1096 |
If the UNTIE method exists then the warning |
|
|
1097 |
above does not occur. Instead the UNTIE |
|
|
1098 |
method is passed the count of ``extra'' references and can |
|
|
1099 |
issue its own warning if appropriate. e.g. to replicate the |
|
|
1100 |
no UNTIE case this method can be |
|
|
1101 |
used: |
|
|
1102 |
|
|
|
1103 |
|
|
|
1104 |
sub UNTIE |
|
|
1105 |
{ |
|
|
1106 |
my ($obj,$count) = @_; |
|
|
1107 |
carp |
|
|
1108 |
!!SEE ALSO |
|
|
1109 |
|
|
|
1110 |
|
|
|
1111 |
See DB_File or Config for some interesting ''tie()'' |
|
|
1112 |
implementations. A good starting point for many ''tie()'' |
|
|
1113 |
implementations is with one of the modules Tie::Scalar, |
|
|
1114 |
Tie::Array, Tie::Hash, or Tie::Handle. |
|
|
1115 |
!!BUGS |
|
|
1116 |
|
|
|
1117 |
|
|
|
1118 |
You cannot easily tie a multilevel data structure (such as a |
|
|
1119 |
hash of hashes) to a dbm file. The first problem is that all |
|
|
1120 |
but GDBM and Berkeley DB have |
|
|
1121 |
size limitations, but beyond that, you also have problems |
|
|
1122 |
with how references are to be represented on disk. One |
|
|
1123 |
experimental module that does attempt to address this need |
|
|
1124 |
partially is the MLDBM module. Check your |
|
|
1125 |
nearest CPAN site as described in perlmodlib |
|
|
1126 |
for source code to MLDBM . |
|
|
1127 |
|
|
|
1128 |
|
|
|
1129 |
Tied filehandles are still incomplete. ''sysopen()'', |
|
|
1130 |
''truncate()'', ''flock()'', ''fcntl()'', |
|
|
1131 |
''stat()'' and -X can't currently be |
|
|
1132 |
trapped. |
|
|
1133 |
!!AUTHOR |
|
|
1134 |
|
|
|
1135 |
|
|
|
1136 |
Tom Christiansen |
|
|
1137 |
|
|
|
1138 |
|
|
|
1139 |
TIEHANDLE by Sven Verdoolaege |
|
|
1140 |
skimo@dns.ufsia.ac.be'' |
|
|
1141 |
''dougm@osf.org'''' |
|
|
1142 |
|
|
|
1143 |
|
|
|
1144 |
UNTIE by Nick Ing-Simmons |
|
|
1145 |
nick@ing-simmons.net'''' |
|
|
1146 |
|
|
|
1147 |
|
|
|
1148 |
Tying Arrays by Casey Tweten |
|
|
1149 |
crt@kiski.net'''' |
|
|
1150 |
---- |