1 | # ------------------------------------------------------------------------------ |
---|
2 | # (C) Crown copyright Met Office. All rights reserved. |
---|
3 | # For further details please refer to the file COPYRIGHT.txt |
---|
4 | # which you should have received as part of this distribution. |
---|
5 | # ------------------------------------------------------------------------------ |
---|
6 | use strict; |
---|
7 | use warnings; |
---|
8 | |
---|
9 | # ------------------------------------------------------------------------------ |
---|
10 | package Fcm::Build::Fortran; |
---|
11 | |
---|
12 | use Text::Balanced qw{extract_bracketed extract_delimited}; |
---|
13 | |
---|
14 | # Actions of this class |
---|
15 | my %ACTION_OF = (extract_interface => \&_extract_interface); |
---|
16 | |
---|
17 | # Regular expressions |
---|
18 | # Matches a variable attribute |
---|
19 | my $RE_ATTR = qr{ |
---|
20 | allocatable|dimension|external|intent|optional|parameter|pointer|save|target |
---|
21 | }imsx; |
---|
22 | # Matches a name |
---|
23 | my $RE_NAME = qr{[A-Za-z]\w*}imsx; |
---|
24 | # Matches a specification type |
---|
25 | my $RE_SPEC = qr{ |
---|
26 | character|complex|double\s*precision|integer|logical|real|type |
---|
27 | }imsx; |
---|
28 | # Matches the identifier of a program unit that does not have arguments |
---|
29 | my $RE_UNIT_BASE = qr{block\s*data|module|program}imsx; |
---|
30 | # Matches the identifier of a program unit that has arguments |
---|
31 | my $RE_UNIT_CALL = qr{function|subroutine}imsx; |
---|
32 | # Matches the identifier of any program unit |
---|
33 | my $RE_UNIT = qr{$RE_UNIT_BASE|$RE_UNIT_CALL}msx; |
---|
34 | my %RE = ( |
---|
35 | # A comment line |
---|
36 | COMMENT => qr{\A\s*(?:!|\z)}msx, |
---|
37 | # A trailing comment, capture the expression before the comment |
---|
38 | COMMENT_END => qr{\A([^'"]*?)\s*!.*\z}msx, |
---|
39 | # A contination marker, capture the expression before the marker |
---|
40 | CONT => qr{\A(.*)&\s*\z}msx, |
---|
41 | # A contination marker at the beginning of a line, capture the marker and |
---|
42 | # the expression after the marker |
---|
43 | CONT_LEAD => qr{\A(\s*&)(.*)\z}msx, |
---|
44 | # Capture a variable identifier, removing any type component expression |
---|
45 | NAME_COMP => qr{\b($RE_NAME)(?:\s*\%\s*$RE_NAME)*\b}msx, |
---|
46 | # Matches the first identifier in a line |
---|
47 | NAME_LEAD => qr{\A\s*$RE_NAME\s*}msx, |
---|
48 | # Captures a name identifier after a comma, and the expression after |
---|
49 | NAME_LIST => qr{\A(?:.*?)\s*,\s*($RE_NAME)\b(.*)\z}msx, |
---|
50 | # Captures the next quote character |
---|
51 | QUOTE => qr{\A[^'"]*(['"])}msx, |
---|
52 | # Matches an attribute declaration |
---|
53 | TYPE_ATTR => qr{\A\s*($RE_ATTR)\b}msx, |
---|
54 | # Matches a type declaration |
---|
55 | TYPE_SPEC => qr{\A\s*($RE_SPEC)\b}msx, |
---|
56 | # Captures the expression after one or more program unit attributes |
---|
57 | UNIT_ATTR => qr{\A\s*(?:(?:elemental|recursive|pure)\s+)+(.*)\z}imsx, |
---|
58 | # Captures the identifier and the symbol of a program unit with no arguments |
---|
59 | UNIT_BASE => qr{\A\s*($RE_UNIT_BASE)\s+($RE_NAME)\s*\z}imsx, |
---|
60 | # Captures the identifier and the symbol of a program unit with arguments |
---|
61 | UNIT_CALL => qr{\A\s*($RE_UNIT_CALL)\s+($RE_NAME)\b}imsx, |
---|
62 | # Captures the end of a program unit, its identifier and its symbol |
---|
63 | UNIT_END => qr{\A\s*(end)(?:\s+($RE_NAME)(?:\s+($RE_NAME))?)?\s*\z}imsx, |
---|
64 | # Captures the expression after a program unit type specification |
---|
65 | UNIT_SPEC => qr{\A\s*$RE_SPEC\b(.*)\z}imsx, |
---|
66 | ); |
---|
67 | |
---|
68 | # Keywords in type declaration statements |
---|
69 | my %TYPE_DECL_KEYWORD_SET = map { ($_, 1) } qw{ |
---|
70 | allocatable |
---|
71 | dimension |
---|
72 | in |
---|
73 | inout |
---|
74 | intent |
---|
75 | kind |
---|
76 | len |
---|
77 | optional |
---|
78 | out |
---|
79 | parameter |
---|
80 | pointer |
---|
81 | save |
---|
82 | target |
---|
83 | }; |
---|
84 | |
---|
85 | # Creates and returns an instance of this class. |
---|
86 | sub new { |
---|
87 | my ($class) = @_; |
---|
88 | bless( |
---|
89 | sub { |
---|
90 | my $key = shift(); |
---|
91 | if (!exists($ACTION_OF{$key})) { |
---|
92 | return; |
---|
93 | } |
---|
94 | $ACTION_OF{$key}->(@_); |
---|
95 | }, |
---|
96 | $class, |
---|
97 | ); |
---|
98 | } |
---|
99 | |
---|
100 | # Methods. |
---|
101 | for my $key (keys(%ACTION_OF)) { |
---|
102 | no strict qw{refs}; |
---|
103 | *{$key} = sub { my $self = shift(); $self->($key, @_) }; |
---|
104 | } |
---|
105 | |
---|
106 | # Extracts the calling interfaces of top level subroutines and functions from |
---|
107 | # the $handle for reading Fortran sources. |
---|
108 | sub _extract_interface { |
---|
109 | my ($handle) = @_; |
---|
110 | map { _present_line($_) } @{_reduce_to_interface(_load($handle))}; |
---|
111 | } |
---|
112 | |
---|
113 | # Reads $handle for the next Fortran statement, handling continuations. |
---|
114 | sub _load { |
---|
115 | my ($handle) = @_; |
---|
116 | my $ctx = {signature_token_set_of => {}, statements => []}; |
---|
117 | my $state = { |
---|
118 | in_contains => undef, # in a "contains" section of a program unit |
---|
119 | in_interface => undef, # in an "interface" block |
---|
120 | in_quote => undef, # in a multi-line quote |
---|
121 | stack => [], # program unit stack |
---|
122 | }; |
---|
123 | my $NEW_STATEMENT = sub { |
---|
124 | { name => q{}, # statement name, e.g. function, integer, ... |
---|
125 | lines => [], # original lines in the statement |
---|
126 | line_number => 0, # line number (start) in the original source |
---|
127 | symbol => q{}, # name of a program unit (signature, end) |
---|
128 | type => q{}, # e.g. signature, use, type, attr, end |
---|
129 | value => q{}, # the actual value of the statement |
---|
130 | }; |
---|
131 | }; |
---|
132 | my $statement; |
---|
133 | LINE: |
---|
134 | while (my $line = readline($handle)) { |
---|
135 | if (!defined($statement)) { |
---|
136 | $statement = $NEW_STATEMENT->(); |
---|
137 | } |
---|
138 | my $value = $line; |
---|
139 | chomp($value); |
---|
140 | # Pre-processor directives and continuation |
---|
141 | if (!$statement->{line_number} && index($value, '#') == 0) { |
---|
142 | $statement->{line_number} = $.; |
---|
143 | $statement->{name} = 'cpp'; |
---|
144 | } |
---|
145 | if ($statement->{name} eq 'cpp') { |
---|
146 | push(@{$statement->{lines}}, $line); |
---|
147 | $statement->{value} .= $value; |
---|
148 | if (rindex($value, '\\') != length($value) - 1) { |
---|
149 | $statement = undef; |
---|
150 | } |
---|
151 | next LINE; |
---|
152 | } |
---|
153 | # Normal Fortran |
---|
154 | if ($value =~ $RE{COMMENT}) { |
---|
155 | next LINE; |
---|
156 | } |
---|
157 | if (!$statement->{line_number}) { |
---|
158 | $statement->{line_number} = $.; |
---|
159 | } |
---|
160 | my ($cont_head, $cont_tail); |
---|
161 | if ($statement->{line_number} != $.) { # is a continuation |
---|
162 | ($cont_head, $cont_tail) = $value =~ $RE{CONT_LEAD}; |
---|
163 | if ($cont_head) { |
---|
164 | $value = $cont_tail; |
---|
165 | } |
---|
166 | } |
---|
167 | # Correctly handle ! and & in quotes |
---|
168 | my ($head, $tail) = (q{}, $value); |
---|
169 | if ($state->{in_quote} && index($value, $state->{in_quote}) >= 0) { |
---|
170 | my $index = index($value, $state->{in_quote}); |
---|
171 | $head = substr($value, 0, $index + 1); |
---|
172 | $tail |
---|
173 | = length($value) > $index + 1 |
---|
174 | ? substr($value, $index + 2) |
---|
175 | : q{}; |
---|
176 | $state->{in_quote} = undef; |
---|
177 | } |
---|
178 | if (!$state->{in_quote}) { |
---|
179 | while ($tail) { |
---|
180 | if (index($tail, q{!}) >= 0) { |
---|
181 | if (!($tail =~ s/$RE{COMMENT_END}/$1/)) { |
---|
182 | ($head, $tail, $state->{in_quote}) |
---|
183 | = _load_extract_quote($head, $tail); |
---|
184 | } |
---|
185 | } |
---|
186 | else { |
---|
187 | while (index($tail, q{'}) > 0 |
---|
188 | || index($tail, q{"}) > 0) |
---|
189 | { |
---|
190 | ($head, $tail, $state->{in_quote}) |
---|
191 | = _load_extract_quote($head, $tail); |
---|
192 | } |
---|
193 | $head .= $tail; |
---|
194 | $tail = q{}; |
---|
195 | } |
---|
196 | } |
---|
197 | } |
---|
198 | $cont_head ||= q{}; |
---|
199 | push(@{$statement->{lines}}, $cont_head . $head . $tail . "\n"); |
---|
200 | $statement->{value} .= $head . $tail; |
---|
201 | # Process a statement only if it is marked with a continuation |
---|
202 | if (!($statement->{value} =~ s/$RE{CONT}/$1/)) { |
---|
203 | $statement->{value} =~ s{\s+\z}{}msx; |
---|
204 | if (_process($statement, $ctx, $state)) { |
---|
205 | push(@{$ctx->{statements}}, $statement); |
---|
206 | } |
---|
207 | $statement = undef; |
---|
208 | } |
---|
209 | } |
---|
210 | return $ctx; |
---|
211 | } |
---|
212 | |
---|
213 | # Helper, removes a quoted string from $tail. |
---|
214 | sub _load_extract_quote { |
---|
215 | my ($head, $tail) = @_; |
---|
216 | my ($extracted, $remainder, $prefix) |
---|
217 | = extract_delimited($tail, q{'"}, qr{[^'"]*}msx, q{}); |
---|
218 | if ($extracted) { |
---|
219 | return ($head . $prefix . $extracted, $remainder); |
---|
220 | } |
---|
221 | else { |
---|
222 | my ($quote) = $tail =~ $RE{QUOTE}; |
---|
223 | return ($head . $tail, q{}, $quote); |
---|
224 | } |
---|
225 | } |
---|
226 | |
---|
227 | # Study statements and put attributes into array $statements |
---|
228 | sub _process { |
---|
229 | my ($statement, $ctx, $state) = @_; |
---|
230 | my $name; |
---|
231 | |
---|
232 | # End Interface |
---|
233 | if ($state->{in_interface}) { |
---|
234 | if ($statement->{value} =~ qr{\A\s*end\s*interface\b}imsx) { |
---|
235 | $state->{in_interface} = 0; |
---|
236 | } |
---|
237 | return; |
---|
238 | } |
---|
239 | |
---|
240 | # End Program Unit |
---|
241 | if (@{$state->{stack}} && $statement->{value} =~ qr{\A\s*end\b}imsx) { |
---|
242 | my ($end, $type, $symbol) = lc($statement->{value}) =~ $RE{UNIT_END}; |
---|
243 | if (!$end) { |
---|
244 | return; |
---|
245 | } |
---|
246 | my ($top_type, $top_symbol) = @{$state->{stack}->[-1]}; |
---|
247 | if (!$type |
---|
248 | || $top_type eq $type && (!$symbol || $top_symbol eq $symbol)) |
---|
249 | { |
---|
250 | pop(@{$state->{stack}}); |
---|
251 | if ($state->{in_contains} && !@{$state->{stack}}) { |
---|
252 | $state->{in_contains} = 0; |
---|
253 | } |
---|
254 | if (!$state->{in_contains}) { |
---|
255 | $statement->{name} = $top_type; |
---|
256 | $statement->{symbol} = $top_symbol; |
---|
257 | $statement->{type} = 'end'; |
---|
258 | return $statement; |
---|
259 | } |
---|
260 | } |
---|
261 | return; |
---|
262 | } |
---|
263 | |
---|
264 | # Interface/Contains |
---|
265 | ($name) = $statement->{value} =~ qr{\A\s*(contains|interface)\b}imsx; |
---|
266 | if ($name) { |
---|
267 | $state->{'in_' . lc($name)} = 1; |
---|
268 | return; |
---|
269 | } |
---|
270 | |
---|
271 | # Program Unit |
---|
272 | my ($type, $symbol, @tokens) = _process_prog_unit($statement->{value}); |
---|
273 | if ($type) { |
---|
274 | push(@{$state->{stack}}, [$type, $symbol]); |
---|
275 | if ($state->{in_contains}) { |
---|
276 | return; |
---|
277 | } |
---|
278 | $statement->{name} = lc($type); |
---|
279 | $statement->{type} = 'signature'; |
---|
280 | $statement->{symbol} = lc($symbol); |
---|
281 | $ctx->{signature_token_set_of}{$symbol} |
---|
282 | = {map { (lc($_) => 1) } @tokens}; |
---|
283 | return $statement; |
---|
284 | } |
---|
285 | if ($state->{in_contains}) { |
---|
286 | return; |
---|
287 | } |
---|
288 | |
---|
289 | # Use |
---|
290 | if ($statement->{value} =~ qr{\A\s*(use)\b}imsx) { |
---|
291 | $statement->{name} = 'use'; |
---|
292 | $statement->{type} = 'use'; |
---|
293 | return $statement; |
---|
294 | } |
---|
295 | |
---|
296 | # Type Declarations |
---|
297 | ($name) = $statement->{value} =~ $RE{TYPE_SPEC}; |
---|
298 | if ($name) { |
---|
299 | $name =~ s{\s}{}gmsx; |
---|
300 | $statement->{name} = lc($name); |
---|
301 | $statement->{type} = 'type'; |
---|
302 | return $statement; |
---|
303 | } |
---|
304 | |
---|
305 | # Attribute Statements |
---|
306 | ($name) = $statement->{value} =~ $RE{TYPE_ATTR}; |
---|
307 | if ($name) { |
---|
308 | $statement->{name} = $name; |
---|
309 | $statement->{type} = 'attr'; |
---|
310 | return $statement; |
---|
311 | } |
---|
312 | } |
---|
313 | |
---|
314 | # Parse a statement for program unit header. Returns a list containing the type, |
---|
315 | # the symbol and the signature tokens of the program unit. |
---|
316 | sub _process_prog_unit { |
---|
317 | my ($string) = @_; |
---|
318 | my ($type, $symbol, @args) = (q{}, q{}); |
---|
319 | # Is it a blockdata, module or program? |
---|
320 | ($type, $symbol) = $string =~ $RE{UNIT_BASE}; |
---|
321 | if ($type) { |
---|
322 | $type = lc($type); |
---|
323 | $type =~ s{\s*}{}gmsx; |
---|
324 | return ($type, $symbol); |
---|
325 | } |
---|
326 | # Remove the attribute and type declaration of a procedure |
---|
327 | $string =~ s/$RE{UNIT_ATTR}/$1/; |
---|
328 | my ($match) = $string =~ $RE{UNIT_SPEC}; |
---|
329 | if ($match) { |
---|
330 | $string = $match; |
---|
331 | extract_bracketed($string); |
---|
332 | } |
---|
333 | # Is it a function or subroutine? |
---|
334 | ($type, $symbol) = lc($string) =~ $RE{UNIT_CALL}; |
---|
335 | if (!$type) { |
---|
336 | return; |
---|
337 | } |
---|
338 | my $extracted = extract_bracketed($string, q{()}, qr{[^(]*}msx); |
---|
339 | |
---|
340 | # Get signature tokens from SUBROUTINE/FUNCTION |
---|
341 | if ($extracted) { |
---|
342 | $extracted =~ s{\s}{}gmsx; |
---|
343 | @args = split(q{,}, substr($extracted, 1, length($extracted) - 2)); |
---|
344 | if ($type eq 'function') { |
---|
345 | my $result = extract_bracketed($string, q{()}, qr{[^(]*}msx); |
---|
346 | if ($result) { |
---|
347 | $result =~ s{\A\(\s*(.*?)\s*\)\z}{$1}msx; # remove braces |
---|
348 | push(@args, $result); |
---|
349 | } |
---|
350 | else { |
---|
351 | push(@args, $symbol); |
---|
352 | } |
---|
353 | } |
---|
354 | } |
---|
355 | return (lc($type), lc($symbol), map { lc($_) } @args); |
---|
356 | } |
---|
357 | |
---|
358 | # Reduces the list of statements to contain only the interface block. |
---|
359 | sub _reduce_to_interface { |
---|
360 | my ($ctx) = @_; |
---|
361 | my (%token_set, @interface_statements); |
---|
362 | STATEMENT: |
---|
363 | for my $statement (reverse(@{$ctx->{statements}})) { |
---|
364 | if ($statement->{type} eq 'end' |
---|
365 | && grep { $_ eq $statement->{name} } qw{subroutine function}) |
---|
366 | { |
---|
367 | push(@interface_statements, $statement); |
---|
368 | %token_set |
---|
369 | = %{$ctx->{signature_token_set_of}{$statement->{symbol}}}; |
---|
370 | next STATEMENT; |
---|
371 | } |
---|
372 | if ($statement->{type} eq 'signature' |
---|
373 | && grep { $_ eq $statement->{name} } qw{subroutine function}) |
---|
374 | { |
---|
375 | push(@interface_statements, $statement); |
---|
376 | %token_set = (); |
---|
377 | next STATEMENT; |
---|
378 | } |
---|
379 | if ($statement->{type} eq 'use') { |
---|
380 | my ($head, $tail) |
---|
381 | = split(qr{\s*:\s*}msx, lc($statement->{value}), 2); |
---|
382 | if ($tail) { |
---|
383 | my @imports = map { [split(qr{\s*=>\s*}msx, $_, 2)] } |
---|
384 | split(qr{\s*,\s*}msx, $tail); |
---|
385 | my @useful_imports |
---|
386 | = grep { exists($token_set{$_->[0]}) } @imports; |
---|
387 | if (!@useful_imports) { |
---|
388 | next STATEMENT; |
---|
389 | } |
---|
390 | if (@imports != @useful_imports) { |
---|
391 | my @token_strings |
---|
392 | = map { $_->[0] . ($_->[1] ? ' => ' . $_->[1] : q{}) } |
---|
393 | @useful_imports; |
---|
394 | my ($last, @rest) = reverse(@token_strings); |
---|
395 | my @token_lines |
---|
396 | = (reverse(map { $_ . q{,&} } @rest), $last); |
---|
397 | push( |
---|
398 | @interface_statements, |
---|
399 | { lines => [ |
---|
400 | sprintf("%s:&\n", $head), |
---|
401 | (map { sprintf(" & %s\n", $_) } @token_lines), |
---|
402 | ] |
---|
403 | }, |
---|
404 | ); |
---|
405 | next STATEMENT; |
---|
406 | } |
---|
407 | } |
---|
408 | push(@interface_statements, $statement); |
---|
409 | next STATEMENT; |
---|
410 | } |
---|
411 | if ($statement->{type} eq 'attr') { |
---|
412 | my ($spec, @tokens) = ($statement->{value} =~ /$RE{NAME_COMP}/g); |
---|
413 | if (grep { exists($token_set{$_}) } @tokens) { |
---|
414 | for my $token (@tokens) { |
---|
415 | $token_set{$token} = 1; |
---|
416 | } |
---|
417 | push(@interface_statements, $statement); |
---|
418 | next STATEMENT; |
---|
419 | } |
---|
420 | } |
---|
421 | if ($statement->{type} eq 'type') { |
---|
422 | my ($variable_string, $spec_string) |
---|
423 | = reverse(split('::', lc($statement->{value}), 2)); |
---|
424 | if ($spec_string) { |
---|
425 | $spec_string =~ s{$RE{NAME_LEAD}}{}msx; |
---|
426 | } |
---|
427 | else { |
---|
428 | # The first expression in the statement is the type + attrib |
---|
429 | $variable_string =~ s{$RE{NAME_LEAD}}{}msx; |
---|
430 | $spec_string = extract_bracketed($variable_string, '()', |
---|
431 | qr{[\s\*]*}msx); |
---|
432 | } |
---|
433 | # Useful tokens are those that comes after a comma |
---|
434 | my $tail = q{,} . lc($variable_string); |
---|
435 | my @tokens; |
---|
436 | while ($tail) { |
---|
437 | if ($tail =~ qr{\A\s*['"]}msx) { |
---|
438 | extract_delimited($tail, q{'"}, qr{\A[^'"]*}msx, q{}); |
---|
439 | } |
---|
440 | elsif ($tail =~ qr{\A\s*\(}msx) { |
---|
441 | extract_bracketed($tail, '()', qr{\A[^(]*}msx); |
---|
442 | } |
---|
443 | else { |
---|
444 | my $token; |
---|
445 | ($token, $tail) = $tail =~ $RE{NAME_LIST}; |
---|
446 | if ($token && $token_set{$token}) { |
---|
447 | @tokens = ($variable_string =~ /$RE{NAME_COMP}/g); |
---|
448 | $tail = q{}; |
---|
449 | } |
---|
450 | } |
---|
451 | } |
---|
452 | if (@tokens && $spec_string) { |
---|
453 | my @spec_tokens = (lc($spec_string) =~ /$RE{NAME_COMP}/g); |
---|
454 | push( |
---|
455 | @tokens, |
---|
456 | ( grep { !exists($TYPE_DECL_KEYWORD_SET{$_}) } |
---|
457 | @spec_tokens |
---|
458 | ), |
---|
459 | ); |
---|
460 | } |
---|
461 | if (grep { exists($token_set{$_}) } @tokens) { |
---|
462 | for my $token (@tokens) { |
---|
463 | $token_set{$token} = 1; |
---|
464 | } |
---|
465 | push(@interface_statements, $statement); |
---|
466 | next STATEMENT; |
---|
467 | } |
---|
468 | } |
---|
469 | } |
---|
470 | if (!@interface_statements) { |
---|
471 | return []; |
---|
472 | } |
---|
473 | [ {lines => ["interface\n"]}, |
---|
474 | reverse(@interface_statements), |
---|
475 | {lines => ["end interface\n"]}, |
---|
476 | ]; |
---|
477 | } |
---|
478 | |
---|
479 | # Processes and returns the line of the statement. |
---|
480 | sub _present_line { |
---|
481 | my ($statement) = @_; |
---|
482 | map { |
---|
483 | s{\s+}{ }gmsx; # collapse multiple spaces |
---|
484 | s{\s+\z}{\n}msx; # remove trailing spaces |
---|
485 | $_; |
---|
486 | } @{$statement->{lines}}; |
---|
487 | } |
---|
488 | |
---|
489 | # ------------------------------------------------------------------------------ |
---|
490 | 1; |
---|
491 | __END__ |
---|
492 | |
---|
493 | =head1 NAME |
---|
494 | |
---|
495 | Fcm::Build::Fortran |
---|
496 | |
---|
497 | =head1 SYNOPSIS |
---|
498 | |
---|
499 | use Fcm::Build::Fortran; |
---|
500 | my $fortran_util = Fcm::Build::Fortran->new(); |
---|
501 | open(my($handle), '<', $path_to_a_fortran_source_file); |
---|
502 | print($fortran_util->extract_interface($handle)); # prints interface |
---|
503 | close($handle); |
---|
504 | |
---|
505 | =head1 DESCRIPTION |
---|
506 | |
---|
507 | A class to analyse Fortran source. Currently, it has a single method to extract |
---|
508 | the calling interfaces of top level subroutines and functions in a Fortran |
---|
509 | source. |
---|
510 | |
---|
511 | =head1 METHODS |
---|
512 | |
---|
513 | =over 4 |
---|
514 | |
---|
515 | =item $class->new() |
---|
516 | |
---|
517 | Creates and returns an instance of this class. |
---|
518 | |
---|
519 | =item $instance->extract_interface($handle) |
---|
520 | |
---|
521 | Extracts the calling interfaces of top level subroutines and functions in a |
---|
522 | Fortran source that can be read from $handle. Returns an interface block as a |
---|
523 | list of lines. |
---|
524 | |
---|
525 | =back |
---|
526 | |
---|
527 | =head1 ACKNOWLEDGEMENT |
---|
528 | |
---|
529 | This module is inspired by the logic developed by the European Centre |
---|
530 | for Medium-Range Weather Forecasts (ECMWF). |
---|
531 | |
---|
532 | =head1 COPYRIGHT |
---|
533 | |
---|
534 | (C) Crown copyright Met Office. All rights reserved. |
---|
535 | |
---|
536 | =cut |
---|