summaryrefslogtreecommitdiff
path: root/systems/texlive/tlnet/tlpkg/tlperl/lib/TAP/Parser/Scheduler.pm
blob: 7e3ddc2c086beb7c2c987933078d7af2211252ee (plain)
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
package TAP::Parser::Scheduler;

use strict;
use warnings;

use Carp;
use TAP::Parser::Scheduler::Job;
use TAP::Parser::Scheduler::Spinner;

=head1 NAME

TAP::Parser::Scheduler - Schedule tests during parallel testing

=head1 VERSION

Version 3.43

=cut

our $VERSION = '3.43';

=head1 SYNOPSIS

    use TAP::Parser::Scheduler;

=head1 DESCRIPTION

=head1 METHODS

=head2 Class Methods

=head3 C<new>

    my $sched = TAP::Parser::Scheduler->new(tests => \@tests);
    my $sched = TAP::Parser::Scheduler->new(
        tests => [ ['t/test_name.t','Test Description'], ... ],
        rules => \%rules,
    );

Given 'tests' and optional 'rules' as input, returns a new
C<TAP::Parser::Scheduler> object.  Each member of C<@tests> should be either a
a test file name, or a two element arrayref, where the first element is a test
file name, and the second element is a test description. By default, we'll use
the test name as the description.

The optional C<rules> attribute provides direction on which tests should be run
in parallel and which should be run sequentially. If no rule data structure is
provided, a default data structure is used which makes every test eligible to
be run in parallel:

    { par => '**' },

The rules data structure is documented more in the next section.

=head2 Rules data structure

The "C<rules>" data structure is the the heart of the scheduler. It allows you
to express simple rules like "run all tests in sequence" or "run all tests in
parallel except these five tests.". However, the rules structure also supports
glob-style pattern matching and recursive definitions, so you can also express
arbitarily complicated patterns.

The rule must only have one top level key: either 'par' for "parallel" or 'seq'
for "sequence".

Values must be either strings with possible glob-style matching, or arrayrefs
of strings or hashrefs which follow this pattern recursively.

Every element in an arrayref directly below a 'par' key is eligible to be run
in parallel, while vavalues directly below a 'seq' key must be run in sequence.

=head3 Rules examples

Here are some examples:

    # All tests be run in parallel (the default rule)
    { par => '**' },

    # Run all tests in sequence, except those starting with "p"
    { par => 't/p*.t' },

    # Run all tests in parallel, except those starting with "p"
    {
        seq => [
                  { seq => 't/p*.t' },
                  { par => '**'     },
               ],
    }

    # Run some  startup tests in sequence, then some parallel tests then some
    # teardown tests in sequence.
    {
        seq => [
            { seq => 't/startup/*.t' },
            { par => ['t/a/*.t','t/b/*.t','t/c/*.t'], }
            { seq => 't/shutdown/*.t' },
        ],
    },


=head3 Rules resolution

=over 4

=item * By default, all tests are eligible to be run in parallel. Specifying any of your own rules removes this one.

=item * "First match wins". The first rule that matches a test will be the one that applies.

=item * Any test which does not match a rule will be run in sequence at the end of the run.

=item * The existence of a rule does not imply selecting a test. You must still specify the tests to run.

=item * Specifying a rule to allow tests to run in parallel does not make the run in parallel. You still need specify the number of parallel C<jobs> in your Harness object.

=back

=head3 Glob-style pattern matching for rules

We implement our own glob-style pattern matching. Here are the patterns it supports:

    ** is any number of characters, including /, within a pathname
    * is zero or more characters within a filename/directory name
    ? is exactly one character within a filename/directory name
    {foo,bar,baz} is any of foo, bar or baz.
    \ is an escape character

=cut

sub new {
    my $class = shift;

    croak "Need a number of key, value pairs" if @_ % 2;

    my %args  = @_;
    my $tests = delete $args{tests} || croak "Need a 'tests' argument";
    my $rules = delete $args{rules} || { par => '**' };

    croak "Unknown arg(s): ", join ', ', sort keys %args
      if keys %args;

    # Turn any simple names into a name, description pair. TODO: Maybe
    # construct jobs here?
    my $self = bless {}, $class;

    $self->_set_rules( $rules, $tests );

    return $self;
}

# Build the scheduler data structure.
#
# SCHEDULER-DATA ::= JOB
#                ||  ARRAY OF ARRAY OF SCHEDULER-DATA
#
# The nested arrays are the key to scheduling. The outer array contains
# a list of things that may be executed in parallel. Whenever an
# eligible job is sought any element of the outer array that is ready to
# execute can be selected. The inner arrays represent sequential
# execution. They can only proceed when the first job is ready to run.

sub _set_rules {
    my ( $self, $rules, $tests ) = @_;

    # Convert all incoming tests to job objects. 
    # If no test description is provided use the file name as the description. 
    my @tests = map { TAP::Parser::Scheduler::Job->new(@$_) }
      map { 'ARRAY' eq ref $_ ? $_ : [ $_, $_ ] } @$tests;
    my $schedule = $self->_rule_clause( $rules, \@tests );

    # If any tests are left add them as a sequential block at the end of
    # the run.
    $schedule = [ [ $schedule, @tests ] ] if @tests;

    $self->{schedule} = $schedule;
}

sub _rule_clause {
    my ( $self, $rule, $tests ) = @_;
    croak 'Rule clause must be a hash'
      unless 'HASH' eq ref $rule;

    my @type = keys %$rule;
    croak 'Rule clause must have exactly one key'
      unless @type == 1;

    my %handlers = (
        par => sub {
            [ map { [$_] } @_ ];
        },
        seq => sub { [ [@_] ] },
    );

    my $handler = $handlers{ $type[0] }
      || croak 'Unknown scheduler type: ', $type[0];
    my $val = $rule->{ $type[0] };

    return $handler->(
        map {
            'HASH' eq ref $_
              ? $self->_rule_clause( $_, $tests )
              : $self->_expand( $_, $tests )
          } 'ARRAY' eq ref $val ? @$val : $val
    );
}

sub _glob_to_regexp {
    my ( $self, $glob ) = @_;
    my $nesting;
    my $pattern;

    while (1) {
        if ( $glob =~ /\G\*\*/gc ) {

            # ** is any number of characters, including /, within a pathname
            $pattern .= '.*?';
        }
        elsif ( $glob =~ /\G\*/gc ) {

            # * is zero or more characters within a filename/directory name
            $pattern .= '[^/]*';
        }
        elsif ( $glob =~ /\G\?/gc ) {

            # ? is exactly one character within a filename/directory name
            $pattern .= '[^/]';
        }
        elsif ( $glob =~ /\G\{/gc ) {

            # {foo,bar,baz} is any of foo, bar or baz.
            $pattern .= '(?:';
            ++$nesting;
        }
        elsif ( $nesting and $glob =~ /\G,/gc ) {

            # , is only special inside {}
            $pattern .= '|';
        }
        elsif ( $nesting and $glob =~ /\G\}/gc ) {

            # } that matches { is special. But unbalanced } are not.
            $pattern .= ')';
            --$nesting;
        }
        elsif ( $glob =~ /\G(\\.)/gc ) {

            # A quoted literal
            $pattern .= $1;
        }
        elsif ( $glob =~ /\G([\},])/gc ) {

            # Sometimes meta characters
            $pattern .= '\\' . $1;
        }
        else {

            # Eat everything that is not a meta character.
            $glob =~ /\G([^{?*\\\},]*)/gc;
            $pattern .= quotemeta $1;
        }
        return $pattern if pos $glob == length $glob;
    }
}

sub _expand {
    my ( $self, $name, $tests ) = @_;

    my $pattern = $self->_glob_to_regexp($name);
    $pattern = qr/^ $pattern $/x;
    my @match = ();

    for ( my $ti = 0; $ti < @$tests; $ti++ ) {
        if ( $tests->[$ti]->filename =~ $pattern ) {
            push @match, splice @$tests, $ti, 1;
            $ti--;
        }
    }

    return @match;
}

=head2 Instance Methods

=head3 C<get_all>

Get a list of all remaining tests.

=cut

sub get_all {
    my $self = shift;
    my @all  = $self->_gather( $self->{schedule} );
    $self->{count} = @all;
    @all;
}

sub _gather {
    my ( $self, $rule ) = @_;
    return unless defined $rule;
    return $rule unless 'ARRAY' eq ref $rule;
    return map { defined() ? $self->_gather($_) : () } map {@$_} @$rule;
}

=head3 C<get_job>

Return the next available job as L<TAP::Parser::Scheduler::Job> object or
C<undef> if none are available. Returns a L<TAP::Parser::Scheduler::Spinner> if
the scheduler still has pending jobs but none are available to run right now.

=cut

sub get_job {
    my $self = shift;
    $self->{count} ||= $self->get_all;
    my @jobs = $self->_find_next_job( $self->{schedule} );
    if (@jobs) {
        --$self->{count};
        return $jobs[0];
    }

    return TAP::Parser::Scheduler::Spinner->new
      if $self->{count};

    return;
}

sub _not_empty {
    my $ar = shift;
    return 1 unless 'ARRAY' eq ref $ar;
    for (@$ar) {
        return 1 if _not_empty($_);
    }
    return;
}

sub _is_empty { !_not_empty(@_) }

sub _find_next_job {
    my ( $self, $rule ) = @_;

    my @queue = ();
    my $index = 0;
    while ( $index < @$rule ) {
        my $seq = $rule->[$index];

        # Prune any exhausted items.
        shift @$seq while @$seq && _is_empty( $seq->[0] );
        if (@$seq) {
            if ( defined $seq->[0] ) {
                if ( 'ARRAY' eq ref $seq->[0] ) {
                    push @queue, $seq;
                }
                else {
                    my $job = splice @$seq, 0, 1, undef;
                    $job->on_finish( sub { shift @$seq } );
                    return $job;
                }
            }
            ++$index;
        }
        else {

            # Remove the empty sub-array from the array
            splice @$rule, $index, 1;
        }
    }

    for my $seq (@queue) {
        if ( my @jobs = $self->_find_next_job( $seq->[0] ) ) {
            return @jobs;
        }
    }

    return;
}

=head3 C<as_string>

Return a human readable representation of the scheduling tree.
For example:

    my @tests = (qw{
        t/startup/foo.t 
        t/shutdown/foo.t
    
        t/a/foo.t t/b/foo.t t/c/foo.t t/d/foo.t
    });
    my $sched = TAP::Parser::Scheduler->new(
        tests => \@tests,
        rules => {
            seq => [
                { seq => 't/startup/*.t' },
                { par => ['t/a/*.t','t/b/*.t','t/c/*.t'] },
                { seq => 't/shutdown/*.t' },
            ],
        },
    );

Produces:

    par:
      seq:
        par:
          seq:
            par:
              seq:
                't/startup/foo.t'
            par:
              seq:
                't/a/foo.t'
              seq:
                't/b/foo.t'
              seq:
                't/c/foo.t'
            par:
              seq:
                't/shutdown/foo.t'
        't/d/foo.t'


=cut


sub as_string {
    my $self = shift;
    return $self->_as_string( $self->{schedule} );
}

sub _as_string {
    my ( $self, $rule, $depth ) = ( shift, shift, shift || 0 );
    my $pad    = ' ' x 2;
    my $indent = $pad x $depth;
    if ( !defined $rule ) {
        return "$indent(undef)\n";
    }
    elsif ( 'ARRAY' eq ref $rule ) {
        return unless @$rule;
        my $type = ( 'par', 'seq' )[ $depth % 2 ];
        return join(
            '', "$indent$type:\n",
            map { $self->_as_string( $_, $depth + 1 ) } @$rule
        );
    }
    else {
        return "$indent'" . $rule->filename . "'\n";
    }
}

1;