Commit | Line | Data |
1a24938d |
1 | package SQL::Translator::Utils; |
2 | |
1a24938d |
3 | use strict; |
f27f9229 |
4 | use warnings; |
c092c5b3 |
5 | use Digest::SHA qw( sha1_hex ); |
bdf60588 |
6 | use File::Spec; |
45287c81 |
7 | use Scalar::Util qw(blessed); |
8 | use Try::Tiny; |
f8783818 |
9 | use Carp qw(carp); |
1a24938d |
10 | |
0c04c5a2 |
11 | our $VERSION = '1.59'; |
12 | our $DEFAULT_COMMENT = '-- '; |
bdf60588 |
13 | |
14 | use base qw(Exporter); |
0c04c5a2 |
15 | our @EXPORT_OK = qw( |
7b4b17aa |
16 | debug normalize_name header_comment parse_list_arg truncate_id_uniquely |
17 | $DEFAULT_COMMENT parse_mysql_version parse_dbms_version |
bdf60588 |
18 | ddl_parser_instance |
f8783818 |
19 | throw ex2err carp_ro |
118bb73f |
20 | ); |
11ad2df9 |
21 | use constant COLLISION_TAG_LENGTH => 8; |
1a24938d |
22 | |
1a24938d |
23 | sub debug { |
a2ba36ba |
24 | my ($pkg, $file, $line, $sub) = caller(0); |
1a24938d |
25 | { |
26 | no strict qw(refs); |
27 | return unless ${"$pkg\::DEBUG"}; |
28 | } |
29 | |
30 | $sub =~ s/^$pkg\:://; |
31 | |
32 | while (@_) { |
33 | my $x = shift; |
34 | chomp $x; |
35 | $x =~ s/\bPKG\b/$pkg/g; |
36 | $x =~ s/\bLINE\b/$line/g; |
37 | $x =~ s/\bSUB\b/$sub/g; |
38 | #warn '[' . $x . "]\n"; |
39 | print STDERR '[' . $x . "]\n"; |
40 | } |
41 | } |
42 | |
93d12e9c |
43 | sub normalize_name { |
ae48473b |
44 | my $name = shift or return ''; |
93d12e9c |
45 | |
46 | # The name can only begin with a-zA-Z_; if there's anything |
47 | # else, prefix with _ |
48 | $name =~ s/^([^a-zA-Z_])/_$1/; |
49 | |
50 | # anything other than a-zA-Z0-9_ in the non-first position |
51 | # needs to be turned into _ |
52 | $name =~ tr/[a-zA-Z0-9_]/_/c; |
53 | |
54 | # All duplicated _ need to be squashed into one. |
55 | $name =~ tr/_/_/s; |
56 | |
57 | # Trim a trailing _ |
58 | $name =~ s/_$//; |
59 | |
60 | return $name; |
61 | } |
62 | |
a2ba36ba |
63 | sub header_comment { |
64 | my $producer = shift || caller; |
65 | my $comment_char = shift; |
66 | my $now = scalar localtime; |
67 | |
68 | $comment_char = $DEFAULT_COMMENT |
69 | unless defined $comment_char; |
70 | |
71 | my $header_comment =<<"HEADER_COMMENT"; |
72 | ${comment_char} |
73 | ${comment_char}Created by $producer |
74 | ${comment_char}Created on $now |
75 | ${comment_char} |
76 | HEADER_COMMENT |
77 | |
78 | # Any additional stuff passed in |
79 | for my $additional_comment (@_) { |
80 | $header_comment .= "${comment_char}${additional_comment}\n"; |
81 | } |
82 | |
83 | return $header_comment; |
84 | } |
85 | |
e545d971 |
86 | sub parse_list_arg { |
87 | my $list = UNIVERSAL::isa( $_[0], 'ARRAY' ) ? shift : [ @_ ]; |
88 | |
51bb6fe0 |
89 | # |
90 | # This protects stringification of references. |
91 | # |
92 | if ( @$list && ref $list->[0] ) { |
93 | return $list; |
94 | } |
95 | # |
96 | # This processes string-like arguments. |
97 | # |
98 | else { |
ea93df61 |
99 | return [ |
51bb6fe0 |
100 | map { s/^\s+|\s+$//g; $_ } |
101 | map { split /,/ } |
102 | grep { defined && length } @$list |
103 | ]; |
104 | } |
118bb73f |
105 | } |
106 | |
f5405d47 |
107 | sub truncate_id_uniquely { |
108 | my ( $desired_name, $max_symbol_length ) = @_; |
109 | |
16fa91c0 |
110 | return $desired_name |
111 | unless defined $desired_name && length $desired_name > $max_symbol_length; |
f5405d47 |
112 | |
16fa91c0 |
113 | my $truncated_name = substr $desired_name, 0, |
11ad2df9 |
114 | $max_symbol_length - COLLISION_TAG_LENGTH - 1; |
f5405d47 |
115 | |
116 | # Hex isn't the most space-efficient, but it skirts around allowed |
117 | # charset issues |
118 | my $digest = sha1_hex($desired_name); |
11ad2df9 |
119 | my $collision_tag = substr $digest, 0, COLLISION_TAG_LENGTH; |
f5405d47 |
120 | |
121 | return $truncated_name |
122 | . '_' |
123 | . $collision_tag; |
124 | } |
125 | |
5d666b31 |
126 | |
5d666b31 |
127 | sub parse_mysql_version { |
128 | my ($v, $target) = @_; |
129 | |
130 | return undef unless $v; |
131 | |
132 | $target ||= 'perl'; |
133 | |
134 | my @vers; |
135 | |
ea93df61 |
136 | # X.Y.Z style |
5d666b31 |
137 | if ( $v =~ / ^ (\d+) \. (\d{1,3}) (?: \. (\d{1,3}) )? $ /x ) { |
138 | push @vers, $1, $2, $3; |
139 | } |
140 | |
ea93df61 |
141 | # XYYZZ (mysql) style |
5d666b31 |
142 | elsif ( $v =~ / ^ (\d) (\d{2}) (\d{2}) $ /x ) { |
143 | push @vers, $1, $2, $3; |
144 | } |
145 | |
ea93df61 |
146 | # XX.YYYZZZ (perl) style or simply X |
5d666b31 |
147 | elsif ( $v =~ / ^ (\d+) (?: \. (\d{3}) (\d{3}) )? $ /x ) { |
148 | push @vers, $1, $2, $3; |
149 | } |
150 | else { |
151 | #how do I croak sanely here? |
152 | die "Unparseable MySQL version '$v'"; |
153 | } |
154 | |
155 | if ($target eq 'perl') { |
156 | return sprintf ('%d.%03d%03d', map { $_ || 0 } (@vers) ); |
157 | } |
158 | elsif ($target eq 'mysql') { |
159 | return sprintf ('%d%02d%02d', map { $_ || 0 } (@vers) ); |
160 | } |
161 | else { |
162 | #how do I croak sanely here? |
163 | die "Unknown version target '$target'"; |
164 | } |
165 | } |
166 | |
7b4b17aa |
167 | sub parse_dbms_version { |
168 | my ($v, $target) = @_; |
169 | |
170 | return undef unless $v; |
171 | |
172 | my @vers; |
173 | |
ea93df61 |
174 | # X.Y.Z style |
7b4b17aa |
175 | if ( $v =~ / ^ (\d+) \. (\d{1,3}) (?: \. (\d{1,3}) )? $ /x ) { |
176 | push @vers, $1, $2, $3; |
177 | } |
178 | |
ea93df61 |
179 | # XX.YYYZZZ (perl) style or simply X |
7b4b17aa |
180 | elsif ( $v =~ / ^ (\d+) (?: \. (\d{3}) (\d{3}) )? $ /x ) { |
181 | push @vers, $1, $2, $3; |
182 | } |
183 | else { |
184 | #how do I croak sanely here? |
185 | die "Unparseable database server version '$v'"; |
186 | } |
187 | |
188 | if ($target eq 'perl') { |
189 | return sprintf ('%d.%03d%03d', map { $_ || 0 } (@vers) ); |
190 | } |
191 | elsif ($target eq 'native') { |
e0d18105 |
192 | return join '.' => grep defined, @vers; |
7b4b17aa |
193 | } |
194 | else { |
195 | #how do I croak sanely here? |
196 | die "Unknown version target '$target'"; |
197 | } |
198 | } |
5d666b31 |
199 | |
0eb3b94a |
200 | #my ($parsers_libdir, $checkout_dir); |
bdf60588 |
201 | sub ddl_parser_instance { |
0eb3b94a |
202 | |
bdf60588 |
203 | my $type = shift; |
204 | |
205 | # it may differ from our caller, even though currently this is not the case |
206 | eval "require SQL::Translator::Parser::$type" |
207 | or die "Unable to load grammar-spec container SQL::Translator::Parser::$type:\n$@"; |
208 | |
7e666ece |
209 | # handle DB2 in a special way, since the grammar source was lost :( |
210 | if ($type eq 'DB2') { |
211 | require SQL::Translator::Parser::DB2::Grammar; |
212 | return SQL::Translator::Parser::DB2::Grammar->new; |
213 | } |
214 | |
0eb3b94a |
215 | require Parse::RecDescent; |
216 | return Parse::RecDescent->new(do { |
217 | no strict 'refs'; |
218 | ${"SQL::Translator::Parser::${type}::GRAMMAR"} |
219 | || die "No \$SQL::Translator::Parser::${type}::GRAMMAR defined, unable to instantiate PRD parser\n" |
220 | }); |
221 | |
222 | # this is disabled until RT#74593 is resolved |
223 | =begin for general sadness |
224 | |
bdf60588 |
225 | unless ($parsers_libdir) { |
226 | |
227 | # are we in a checkout? |
228 | if ($checkout_dir = _find_co_root()) { |
229 | $parsers_libdir = File::Spec->catdir($checkout_dir, 'share', 'PrecompiledParsers'); |
230 | } |
231 | else { |
232 | require File::ShareDir; |
233 | $parsers_libdir = File::Spec->catdir( |
234 | File::ShareDir::dist_dir('SQL-Translator'), |
235 | 'PrecompiledParsers' |
236 | ); |
237 | } |
238 | |
239 | unshift @INC, $parsers_libdir; |
240 | } |
241 | |
242 | my $precompiled_mod = "Parse::RecDescent::DDL::SQLT::$type"; |
243 | |
244 | # FIXME FIXME FIXME |
245 | # Parse::RecDescent has horrible architecture where each precompiled parser |
246 | # instance shares global state with all its siblings |
247 | # What we do here is gross, but scarily efficient - the parser compilation |
248 | # is much much slower than an unload/reload cycle |
f2ab5843 |
249 | require Class::Unload; |
bdf60588 |
250 | Class::Unload->unload($precompiled_mod); |
251 | |
252 | # There is also a sub-namespace that P::RD uses, but simply unsetting |
253 | # $^W to stop redefine warnings seems to be enough |
254 | #Class::Unload->unload("Parse::RecDescent::$precompiled_mod"); |
255 | |
256 | eval "local \$^W; require $precompiled_mod" or do { |
257 | if ($checkout_dir) { |
258 | die "Unable to find precompiled grammar for $type - run Makefile.PL to generate it\n"; |
259 | } |
260 | else { |
261 | die "Unable to load precompiled grammar for $type... this is not supposed to happen if you are not in a checkout, please file a bugreport:\n$@" |
262 | } |
263 | }; |
264 | |
265 | my $grammar_spec_fn = $INC{"SQL/Translator/Parser/$type.pm"}; |
266 | my $precompiled_fn = $INC{"Parse/RecDescent/DDL/SQLT/$type.pm"}; |
267 | |
268 | if ( |
269 | (stat($grammar_spec_fn))[9] |
270 | > |
271 | (stat($precompiled_fn))[9] |
272 | ) { |
273 | die ( |
274 | "Grammar spec '$grammar_spec_fn' is newer than precompiled parser '$precompiled_fn'" |
275 | . ($checkout_dir |
276 | ? " - run Makefile.PL to regenerate stale versions\n" |
277 | : "... this is not supposed to happen if you are not in a checkout, please file a bugreport\n" |
278 | ) |
279 | ); |
280 | } |
281 | |
282 | return $precompiled_mod->new; |
0eb3b94a |
283 | =cut |
284 | |
bdf60588 |
285 | } |
286 | |
287 | # Try to determine the root of a checkout/untar if possible |
288 | # or return undef |
289 | sub _find_co_root { |
290 | |
291 | my @mod_parts = split /::/, (__PACKAGE__ . '.pm'); |
292 | my $rel_path = join ('/', @mod_parts); # %INC stores paths with / regardless of OS |
293 | |
294 | return undef unless ($INC{$rel_path}); |
295 | |
296 | # a bit convoluted, but what we do here essentially is: |
297 | # - get the file name of this particular module |
298 | # - do 'cd ..' as many times as necessary to get to lib/SQL/Translator/../../.. |
299 | |
300 | my $root = (File::Spec::Unix->splitpath($INC{$rel_path}))[1]; |
301 | for (1 .. @mod_parts) { |
302 | $root = File::Spec->catdir($root, File::Spec->updir); |
303 | } |
304 | |
305 | return ( -f File::Spec->catfile($root, 'Makefile.PL') ) |
306 | ? $root |
307 | : undef |
308 | ; |
309 | } |
310 | |
45287c81 |
311 | { |
312 | package SQL::Translator::Utils::Error; |
313 | |
314 | use overload |
315 | '""' => sub { ${$_[0]} }, |
316 | fallback => 1; |
317 | |
318 | sub new { |
319 | my ($class, $msg) = @_; |
320 | bless \$msg, $class; |
321 | } |
322 | } |
323 | |
324 | sub throw { |
325 | die SQL::Translator::Utils::Error->new($_[0]); |
326 | } |
327 | |
328 | sub ex2err { |
329 | my ($orig, $self, @args) = @_; |
330 | return try { |
331 | $self->$orig(@args); |
332 | } catch { |
333 | die $_ unless blessed($_) && $_->isa("SQL::Translator::Utils::Error"); |
334 | $self->error("$_"); |
335 | }; |
336 | } |
337 | |
f8783818 |
338 | sub carp_ro { |
339 | my ($name) = @_; |
340 | return sub { |
341 | my ($orig, $self) = (shift, shift); |
342 | carp "'$name' is a read-only accessor" if @_; |
343 | return $self->$orig; |
344 | }; |
345 | } |
346 | |
1a24938d |
347 | 1; |
348 | |
118bb73f |
349 | =pod |
1a24938d |
350 | |
351 | =head1 NAME |
352 | |
353 | SQL::Translator::Utils - SQL::Translator Utility functions |
354 | |
355 | =head1 SYNOPSIS |
356 | |
357 | use SQL::Translator::Utils qw(debug); |
358 | debug("PKG: Bad things happened"); |
359 | |
360 | =head1 DESCSIPTION |
361 | |
362 | C<SQL::Translator::Utils> contains utility functions designed to be |
363 | used from the other modules within the C<SQL::Translator> modules. |
364 | |
a2ba36ba |
365 | Nothing is exported by default. |
1a24938d |
366 | |
a2ba36ba |
367 | =head1 EXPORTED FUNCTIONS AND CONSTANTS |
1a24938d |
368 | |
369 | =head2 debug |
370 | |
371 | C<debug> takes 0 or more messages, which will be sent to STDERR using |
372 | C<warn>. Occurances of the strings I<PKG>, I<SUB>, and I<LINE> |
373 | will be replaced by the calling package, subroutine, and line number, |
e545d971 |
374 | respectively, as reported by C<caller(1)>. |
1a24938d |
375 | |
376 | For example, from within C<foo> in F<SQL/Translator.pm>, at line 666: |
377 | |
378 | debug("PKG: Error reading file at SUB/LINE"); |
379 | |
380 | Will warn |
381 | |
382 | [SQL::Translator: Error reading file at foo/666] |
383 | |
384 | The entire message is enclosed within C<[> and C<]> for visual clarity |
385 | when STDERR is intermixed with STDOUT. |
93d12e9c |
386 | |
387 | =head2 normalize_name |
388 | |
389 | C<normalize_name> takes a string and ensures that it is suitable for |
390 | use as an identifier. This means: ensure that it starts with a letter |
391 | or underscore, and that the rest of the string consists of only |
392 | letters, numbers, and underscores. A string that begins with |
393 | something other than [a-zA-Z] will be prefixer with an underscore, and |
394 | all other characters in the string will be replaced with underscores. |
395 | Finally, a trailing underscore will be removed, because that's ugly. |
396 | |
397 | normalize_name("Hello, world"); |
398 | |
399 | Produces: |
400 | |
401 | Hello_world |
402 | |
403 | A more useful example, from the C<SQL::Translator::Parser::Excel> test |
404 | suite: |
405 | |
406 | normalize_name("silly field (with random characters)"); |
407 | |
408 | returns: |
409 | |
410 | silly_field_with_random_characters |
411 | |
a2ba36ba |
412 | =head2 header_comment |
413 | |
414 | Create the header comment. Takes 1 mandatory argument (the producer |
415 | classname), an optional comment character (defaults to $DEFAULT_COMMENT), |
416 | and 0 or more additional comments, which will be appended to the header, |
417 | prefixed with the comment character. If additional comments are provided, |
418 | then a comment string must be provided ($DEFAULT_COMMENT is exported for |
419 | this use). For example, this: |
420 | |
421 | package My::Producer; |
422 | |
423 | use SQL::Translator::Utils qw(header_comment $DEFAULT_COMMENT); |
424 | |
425 | print header_comment(__PACKAGE__, |
e545d971 |
426 | $DEFAULT_COMMENT, |
a2ba36ba |
427 | "Hi mom!"); |
428 | |
429 | produces: |
430 | |
e545d971 |
431 | -- |
a2ba36ba |
432 | -- Created by My::Prodcuer |
433 | -- Created on Fri Apr 25 06:56:02 2003 |
e545d971 |
434 | -- |
a2ba36ba |
435 | -- Hi mom! |
e545d971 |
436 | -- |
a2ba36ba |
437 | |
438 | Note the gratuitous spacing. |
439 | |
118bb73f |
440 | =head2 parse_list_arg |
441 | |
442 | Takes a string, list or arrayref (all of which could contain |
443 | comma-separated values) and returns an array reference of the values. |
444 | All of the following will return equivalent values: |
445 | |
446 | parse_list_arg('id'); |
447 | parse_list_arg('id', 'name'); |
448 | parse_list_arg( 'id, name' ); |
449 | parse_list_arg( [ 'id', 'name' ] ); |
450 | parse_list_arg( qw[ id name ] ); |
451 | |
f5405d47 |
452 | =head2 truncate_id_uniquely |
453 | |
454 | Takes a string ($desired_name) and int ($max_symbol_length). Truncates |
455 | $desired_name to $max_symbol_length by including part of the hash of |
456 | the full name at the end of the truncated name, giving a high |
457 | probability that the symbol will be unique. For example, |
458 | |
459 | truncate_id_uniquely( 'a' x 100, 64 ) |
460 | truncate_id_uniquely( 'a' x 99 . 'b', 64 ); |
461 | truncate_id_uniquely( 'a' x 99, 64 ) |
462 | |
463 | Will give three different results; specifically: |
464 | |
465 | aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa_7f900025 |
466 | aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa_6191e39a |
467 | aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa_8cd96af2 |
468 | |
a2ba36ba |
469 | =head2 $DEFAULT_COMMENT |
470 | |
471 | This is the default comment string, '-- ' by default. Useful for |
472 | C<header_comment>. |
473 | |
5d666b31 |
474 | =head2 parse_mysql_version |
475 | |
ea93df61 |
476 | Used by both L<Parser::MySQL|SQL::Translator::Parser::MySQL> and |
5d666b31 |
477 | L<Producer::MySQL|SQL::Translator::Producer::MySQL> in order to provide a |
478 | consistent format for both C<< parser_args->{mysql_parser_version} >> and |
479 | C<< producer_args->{mysql_version} >> respectively. Takes any of the following |
480 | version specifications: |
481 | |
482 | 5.0.3 |
483 | 4.1 |
484 | 3.23.2 |
485 | 5 |
486 | 5.001005 (perl style) |
487 | 30201 (mysql style) |
488 | |
282bf498 |
489 | =head2 parse_dbms_version |
490 | |
491 | Takes a version string (X.Y.Z) or perl style (XX.YYYZZZ) and a target ('perl' |
492 | or 'native') transforms the string to the given target style. |
493 | to |
494 | |
4e43db0d |
495 | =head2 throw |
496 | |
497 | Throws the provided string as an object that will stringify back to the |
498 | original string. This stops it from being mangled by L<Moo>'s C<isa> |
499 | code. |
500 | |
501 | =head2 ex2err |
502 | |
503 | Wraps an attribute accessor to catch any exception raised using |
504 | L</throw> and store them in C<< $self->error() >>, finally returning |
505 | undef. A reference to this function can be passed directly to |
506 | L<Moo/around>. |
507 | |
508 | around foo => \&ex2err; |
509 | |
510 | around bar => sub { |
511 | my ($orig, $self) = (shift, shift); |
512 | return ex2err($orig, $self, @_) if @_; |
513 | ... |
514 | }; |
515 | |
f8783818 |
516 | =head2 carp_ro |
517 | |
518 | Takes a field name and returns a reference to a function can be used |
519 | L<around|Moo/around> a read-only accessor to make it L<carp|Carp/carp> |
520 | instead of die when passed an argument. |
521 | |
118bb73f |
522 | =head1 AUTHORS |
523 | |
524 | Darren Chamberlain E<lt>darren@cpan.orgE<gt>, |
11ad2df9 |
525 | Ken Y. Clark E<lt>kclark@cpan.orgE<gt>. |
118bb73f |
526 | |
527 | =cut |