CAB/Format/Raw/Perl.pm

filter: Statementstmt 43.0% 28 / 65 Branchbran 0.0% 0 / 16 Conditioncond 0.0% 0 / 12 MC/DCmcdc 0.0% 0 / 7 Subroutinesub 50.0% 9 / 18 Podpod 100% 9 / 9 totaltotal 36.2% 46 / 127 CC 14 SCAR 41.2
CC
14
Cov
30%
parseRawString
156.0
new
6.0
fromString
6.0
CRAP 61.8
« CAB/Format.pm ↑ Summary CAB/Format/XmlTokWrapFast.pm »
1 ## -*- Mode: CPerl -*-
2 ##
3 ## File: DTA::CAB::Format::Raw::Perl.pm
4 ## Author: Bryan Jurish <moocow@cpan.org>
5 ## Description: Datum parser: raw untokenized text (pure-perl hack)
6  
7 package DTA::CAB::Format::Raw::Perl;
8 1 ▶use DTA::CAB::Format;
Subroutine BEGIN: called CC=1 SCAR=0.0
9 1 ▶use DTA::CAB::Format::Raw::Base;
Subroutine BEGIN: called CC=1 SCAR=0.0
10 1 ▶use DTA::CAB::Datum ':all';
Subroutine BEGIN: called CC=1 SCAR=0.0
11 1 ▶use IO::File;
Subroutine BEGIN: called CC=1 SCAR=0.0
12 1 ▶use Encode qw(encode decode);
Subroutine BEGIN: called CC=1 SCAR=0.0
13 1 ▶use Carp;
Subroutine BEGIN: called CC=1 SCAR=0.0
14 1 ▶use strict;
Subroutine BEGIN: called CC=1 SCAR=0.0
15  
16 ##==============================================================================
17 ## Globals
18 ##==============================================================================
19  
20 1 our @ISA = qw(DTA::CAB::Format::Raw::Base);
21  
22 1 ▶BEGIN {
Subroutine BEGIN: called CC=1 SCAR=0.0
23 1 DTA::CAB::Format->registerFormat(name=>__PACKAGE__, short=>'raw-perl', filenameRegex=>qr/\.(?i:raw-perl|txt-perl)$/);
24 }
25  
26 ## %ABBREVS
27 ## + default abbreviations
28 ## + set below
29 1 our (%ABBREVS);
30  
31 ##==============================================================================
32 ## Constructors etc.
33 ##==============================================================================
34  
35 ## $fmt = CLASS_OR_OBJ->new(%args)
36 ## + object structure: assumed HASH
37 ## {
38 ## ##-- Input
39 ## doc => $doc, ##-- buffered input document
40 ## abbrevs => \%abbrevs, ##-- hash of known abbrevs (default: \%ABBREVS)
41 ##
42 ## ##-- Output
43 ## outbuf => $stringBuffer, ##-- buffered output: DISABLED
44 ## #level => $formatLevel, ##-- n/a
45 ##
46 ## ##-- Common
47 ## utf8 => $bool,
48 0 ▶sub new {
Subroutine new: not called CC=2 SCAR=17.9
Pod documented
49 0 my $that = shift;
50 0 ▶ my $fmt = bless({
Condition0%0/3
ref $that || $that
l!l&&r!l&&!r
000
MC/DC0%0/2
ref $that || $that
ref $that $that
Truth table0%0/3
A || B
A= ref $thatB= $that
ABresult
1X1
011
000
51 ##-- common
52 utf8 => 1,
53  
54 ##-- input
55 doc => undef,
56 abbrevs => \%ABBREVS,
57  
58 ##-- output
59 #outbuf => '',
60  
61 ##-- user args
62 @_
63 }, ref($that)||$that);
64 0 return $fmt;
65 }
66  
67 ##==============================================================================
68 ## Methods: Persistence
69 ##==============================================================================
70  
71 ## @keys = $class_or_obj->noSaveKeys()
72 ## + returns list of keys not to be saved
73 ## + default just returns empty list
74 0 ▶sub noSaveKeys {
Subroutine noSaveKeys: not called CC=1 SCAR=6.9
Pod documented
75 0 return (shift->SUPER::noSaveKeys(), qw(doc outbuf));
76 }
77  
78 ##==============================================================================
79 ## Methods: Input
80 ##==============================================================================
81  
82 ##--------------------------------------------------------------
83 ## Methods: Input: Input selection
84  
85 ## $fmt = $fmt->close()
86 0 ▶sub close {
Subroutine close: not called CC=1 SCAR=6.9
Pod documented
87 0 delete($_[0]{doc});
88 0 return $_[0]->SUPER::close(@_[1..$#_]);
89 }
90  
91 ## $fmt = $fmt->fromString( $string)
92 ## $fmt = $fmt->fromString(\$string)
93 ## + select input from string $string
94 0 ▶sub fromString {
Subroutine fromString: not called CC=2 SCAR=17.9
Pod documented
95 0 my $fmt = shift;
96 0 $fmt->close();
97 0 ▶ return $fmt->parseRawString(ref($_[0]) ? $_[0] : \$_[0]);
Branch: 1 branch
BranchTrueFalseTotal
ref $_[0] ? : 0 0 0
98 }
99  
100 ## $fmt = $fmt->fromFh($fh)
101 ## + override calls $fmt->fromFh_str()
102 0 ▶sub fromFh {
Subroutine fromFh: not called CC=1 SCAR=6.9
Pod documented
103 0 return $_[0]->fromFh_str(@_[1..$#_]);
104 }
105  
106 ##--------------------------------------------------------------
107 ## Methods: Input: Local
108  
109 ## $fmt = $fmt->parseRawString(\$str)
110 ## + guts for fromString(): parse string $str into local document buffer.
111 0 ▶sub parseRawString {
Subroutine parseRawString: not called CC=12 SCAR=50.5
Pod documented
112 0 my ($fmt,$src) = @_;
113 0 ▶ utf8::decode($$src) if ($fmt->{utf8} && !utf8::is_utf8($$src));
Branch: 1 branch
BranchTrueFalseTotal
if $fmt->{"utf8"} and not utf8::is_utf8($$src) 0 0 0
Condition0%0/3
$fmt->{"utf8"} and not utf8::is_utf8($$src)
!ll&&!rl&&r
000
MC/DC0%0/2
$fmt->{"utf8"} and not utf8::is_utf8($$src)
$fmt->{"utf8"} not utf8::is_utf8($$src)
Truth table0%0/3
A and not(B)
A= $fmt->{"utf8"}B= not utf8::is_utf8($$src)
ABresult
0X0
100
111
114  
115 ##-- step 1: basic tokenization
116 0 my (@toks);
117 0 ▶ while ($$src =~ m/(
Branch: 1 branch
BranchTrueFalseTotal
if ($$src =~ m(( (?:([[:alpha:]_\-\255\#\@]+)[\-\254](?:\n\s*)([[:alpha:]_\-\255\#\@]+)) ##-- line-broken alphabetics | (?i:[IVXLCDM\#\@]+\.) ##-- dotted roman numerals (hack) | (?:[[:alpha:]\#\@]\.) ##-- dotted single-letter abbreviations | (?:[[:digit:]\#\@]+[[:alpha:]_\#\@]+) ##-- numbers with optional alphabetic suffixes | (?:[\-\+]?[[:digit:]_\#\@]*[[:digit:]_\,\.\#\@]+) ##-- comma- and/or dot-separated numbers | (?:\,\,|\`\`|\'\'|\-+|\.\.+|\[Formel\]) ##-- special punctuation sequences | (?:[[:alpha:]_\-\254\255\#\@]+) ##-- "normal" alphabetics (with "#", "@" ~= unknown) | (?:[[:punct:]]+) ##-- "normal" punctuation characters | (?:[^[:punct:][:digit:][:space:]]+) ##-- "strange" alaphbetic tokens | (?:\S+) ##-- HACK: anything else ) )gsx) 0 0 0
118 (?:([[:alpha:]_\-­\#\@]+)[\-¬](?:\n\s*)([[:alpha:]_\-­\#\@]+)) ##-- line-broken alphabetics
119 | (?i:[IVXLCDM\#\@]+\.) ##-- dotted roman numerals (hack)
120 | (?:[[:alpha:]\#\@]\.) ##-- dotted single-letter abbreviations
121 | (?:[[:digit:]\#\@]+[[:alpha:]_\#\@]+) ##-- numbers with optional alphabetic suffixes
122 | (?:[\-\+]?[[:digit:]_\#\@]*[[:digit:]_\,\.\#\@]+) ##-- comma- and\/or dot-separated numbers
123 | (?:\,\,|\`\`|\'\'|\-+|\.\.+|\[Formel\]) ##-- special punctuation sequences
124 | (?:[[:alpha:]_\-¬­\#\@]+) ##-- "normal" alphabetics (with "#", "@" ~= unknown)
125 | (?:[[:punct:]]+) ##-- "normal" punctuation characters
126 | (?:[^[:punct:][:digit:][:space:]]+) ##-- "strange" alaphbetic tokens
127 | (?:\S+) ##-- HACK: anything else
128 )
129 /xsg)
130 {
131 0 ▶ push(@toks, (defined($2) ? "$2$3" : $1));
Branch: 1 branch
BranchTrueFalseTotal
defined $2 ? : 0 0 0
132 }
133  
134 ##-- step 2: abbreviation & eos detection
135 0 my $abbrevs = $fmt->{abbrevs};
136 0 my $s = [];
137 0 my @sents = ($s);
138 0 my ($toki);
139 0 ▶ for ($toki=0; $toki <= $#toks; $toki++) {
Branch: 1 branch
BranchTrueFalseTotal
if ($toki <= $#toks) 0 0 0
140 0 ▶ if (exists($abbrevs->{$toks[$toki]}) && $toki < $#toks && $toks[$toki+1] eq '.') {
Branch: 2 branches
BranchTrueFalseTotal
if (exists $$abbrevs{$toks[$toki]} and $toki < $#toks and $toks[$toki + 1] eq '.') { } 0 0 0
elsif ($toks[$toki] =~ /^[\.\?\!]+$/) { } 0 0 0
Condition0%0/6
exists $$abbrevs{$toks[$toki]} and $toki < $#toks and $toks[$toki + 1] eq "."
!ll&&!rl&&r
000
exists $$abbrevs{$toks[$toki]} and $toki < $#toks
!ll&&!rl&&r
000
MC/DC0%0/3
exists $$abbrevs{$toks[$toki]} and $toki < $#toks and $toks[$toki + 1] eq "."
exists $$abbrevs{$toks[$toki]} $toki < $#toks $toks[$toki + 1] eq "."
Truth table0%0/4
A and B and C
A= exists $$abbrevs{$toks[$toki]}B= $toki < $#toksC= $toks[$toki + 1] eq "."
ABCresult
0XX0
10X0
1100
1111
141 ##-- abbreviation
142 0 push(@$s, "$toks[$toki].");
143 0 $toki++;
144 }
145 elsif ($toks[$toki] =~ /^[\.\?\!]+$/) {
146 ##-- sentence-final punctuation
147 0 push(@$s, $toks[$toki]);
148 0 push(@sents, $s=[]);
149 }
150 else {
151 ##-- normal token
152 0 push(@$s, $toks[$toki]);
153 }
154 }
155 0 ▶ pop(@sents) if (!@{$sents[$#sents]});
Branch: 1 branch
BranchTrueFalseTotal
unless @{$sents[$#sents];} 0 0 0
156  
157 ##-- step 3: build doc
158 0 foreach (@sents) {
159 0 @$_ = map { {text=>$_} } @$_;
160 }
161 0 $fmt->{doc} = bless({body=>[map { {tokens=>$_} } @sents]}, 'DTA::CAB::Document');
162 0 return $fmt;
163 }
164  
165  
166 ##--------------------------------------------------------------
167 ## Methods: Input: Generic API
168  
169 ## $doc = $fmt->parseDocument()
170 0 ▶sub parseDocument {
Subroutine parseDocument: not called CC=1 SCAR=6.9
Pod documented
171 0 return $_[0]{doc};
172 }
173  
174  
175 ##==============================================================================
176 ## Methods: Output
177 ## + output not supported
178 ##==============================================================================
179  
180 ##--------------------------------------------------------------
181 ## Methods: Output: Generic
182  
183 ## $type = $fmt->mimeType()
184 ## + default returns text/plain
185 0 ▶sub mimeType { return 'text/plain'; }
Subroutine mimeType: not called CC=1 SCAR=6.9
Pod documented
186  
187 ## $ext = $fmt->defaultExtension()
188 ## + returns default filename extension for this format
189 0 ▶sub defaultExtension { return '.raw'; }
Subroutine defaultExtension: not called CC=1 SCAR=6.9
Pod documented
190  
191  
192 ##==============================================================================
193 ## Initialization
194 1 ▶BEGIN {
Subroutine BEGIN: called CC=1 SCAR=0.0
195 ## @ABBREVS: most frequent 128 abbreviations from DTA (as of Wed, 01 Dec 2010 11:26:04 +0100)
196 ## + derived from file ../automata/eqlemma/dta-abbrevs.nontrivial.tfa
197 ## - itself derived by hand from ../automata/eqlemma/cab.toka.db, ../automata/words/dta-words.tf
198 ## using Lingua::TT hackery and Data::Dumper
199 1 my @ABBREVS =
200 (
201 'Abb',
202 "Ab\x{17f}",
203 'Anm',
204 'Ann',
205 'Arch',
206 'Art',
207 'Aufl',
208 'Bd',
209 'Bde',
210 'Br',
211 'Cap',
212 'Ch',
213 'Chr',
214 'Co',
215 'Cod',
216 'Ctr',
217 "Di\x{17f}\x{17f}",
218 'Dr',
219 'Ew',
220 'Fig',
221 'Fr',
222 "Ge\x{17f}",
223 "Ge\x{17f}ch",
224 'Gr',
225 'Hist',
226 'Hr',
227 'Hrn',
228 'Jahrh',
229 'Journ',
230 'Kap',
231 'Kilogr',
232 "K\x{f6}nigl",
233 'Lib',
234 'Lit',
235 'Matth',
236 'Mill',
237 'Mr',
238 'No',
239 'Nov',
240 'Nr',
241 'Num',
242 'Ol',
243 'Pal',
244 'Pf',
245 'Pfd',
246 'Pfr',
247 'Plut',
248 'Prof',
249 'Proz',
250 "Re\x{17f}cr",
251 'Sal',
252 'Sept',
253 'Sr',
254 'St',
255 'StGB',
256 'Staatsr',
257 'Str',
258 'Tab',
259 'Taf',
260 'Th',
261 'Thl',
262 'Thlr',
263 'Tit',
264 'Tom',
265 'Verf',
266 'Vergl',
267 'Vgl',
268 'Vol',
269 'Zeitschr',
270 'Ziff',
271 'acc',
272 'adj',
273 "ag\x{17f}",
274 'ahd',
275 'altn',
276 "angel\x{17f}",
277 'art',
278 'betr',
279 'cap',
280 'cit',
281 'dat',
282 'dergl',
283 'dgl',
284 'diss',
285 'ed',
286 'engl',
287 'eod',
288 'etc',
289 'fem',
290 'ff',
291 'fg',
292 'fig',
293 'fl',
294 'fr',
295 'geb',
296 'gl',
297 'goth',
298 'gr',
299 'griech',
300 'jun',
301 'ker',
302 'lat',
303 'lib',
304 'lit',
305 "ma\x{17f}c",
306 'med',
307 'mhd',
308 'min',
309 'nat',
310 'neutr',
311 'nhd',
312 'nom',
313 'pag',
314 'part',
315 'pl',
316 'pr',
317 'praes',
318 'praet',
319 "prae\x{17f}",
320 'resp',
321 'sing',
322 'tab',
323 'urspr',
324 'vergl',
325 'vgl',
326 "zu\x{17f}",
327 "\x{17f}t",
328 "\x{17f}ub\x{17f}t",
329  
330 "usw",
331 );
332  
333 1 %ABBREVS = map {($_=>undef)} @ABBREVS;
334 }
335  
336  
337  
338 1 1; ##-- be happy
339  
340 __END__
« CAB/Format.pm ↑ Summary CAB/Format/XmlTokWrapFast.pm »