TT/Sentence.pm

filter: Statementstmt 11.2% 7 / 62 Branchbran 0.0% 0 / 10 Conditioncond 0.0% 0 / 12 MC/DCmcdc 0.0% 0 / 8 Subroutinesub 15.3% 2 / 13 Podpod 0.0% 0 / 11 totaltotal 7.7% 9 / 116 CC 13 SCAR 26.6
CC
13
Cov
8%
guessRawString
42.0
rawString
12.0
new
6.0
CRAP 14.3
« TT/Enum.pm ↑ Summary TT/Packed/a.pm »
1 ## -*- Mode: CPerl -*-
2 ## File: Lingua::TT::Sentence.pm
3 ## Author: Bryan Jurish <TT/IO.pm>
4 ## Descript: TT I/O: sentences
5  
6  
7 package Lingua::TT::Sentence;
8 1 ▶use Lingua::TT::Token;
Subroutine BEGIN: called CC=1 SCAR=0.0
9 1 ▶use strict;
Subroutine BEGIN: called CC=1 SCAR=0.0
10  
11 ##==============================================================================
12 ## Globals & Constants
13  
14 ##==============================================================================
15 ## Constructors etc.
16  
17 ## $sent = CLASS_OR_OBJECT->new(@tokens)
18 ## + $sent: ARRAY-ref
19 ## [$tok1, $tok2, ..., $tokN]
20 0 ▶sub new {
Subroutine new: not called CC=2 SCAR=17.9
Pod undocumented
21 0 my $that = shift;
22 0 ▶ return bless([@_], ref($that)||$that);
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
23 }
24  
25 ## $sent = CLASS_OR_OBJECT->newFromString($str)
26 0 ▶sub newFromString {
Subroutine newFromString: not called CC=1 SCAR=6.9
Pod undocumented
27 0 return $_[0]->new()->fromString($_[1]);
28 }
29  
30 ## $sent2 = $sent->copy($depth)
31 ## + creates a copy of $sent
32 ## + if $deep is 0, only a shallow copy is created (tokens are shared)
33 ## + if $deep is >=1 (or <0), sentences are copied as well (tokens are copied)
34 0 ▶sub copy {
Subroutine copy: not called CC=2 SCAR=17.9
Pod undocumented
35 0 my ($sent,$deep) = @_;
36 0 my $sent2 = bless([],ref($sent));
37 0 ▶ @$sent2 = $deep ? (map {bless([@$_],ref($_))} @$sent) : @$sent;
Branch: 1 branch
BranchTrueFalseTotal
$deep ? : 0 0 0
38 0 return $sent2;
39 }
40  
41 ##==============================================================================
42 ## Methods: Access
43  
44 ## $bool = $sent->isEmpty()
45 ## + true iff $sent has no non-empty tokens
46 0 ▶sub isEmpty {
Subroutine isEmpty: not called CC=1 SCAR=6.9
Pod undocumented
47 0 return !grep {!$_->isEmpty} @{$_[0]};
48 }
49  
50 ## $sent = $sent->rmEmptyTokens()
51 ## + removes empty & undefined tokens from @$sent
52 0 ▶sub rmEmptyTokens {
Subroutine rmEmptyTokens: not called CC=2 SCAR=17.9
Pod undocumented
53 0 ▶ @{$_[0]} = grep {defined($_) && !$_->isEmpty} @{$_[0]};
Condition0%0/3
defined $_ && !$_->isEmpty
!ll&&!rl&&r
000
MC/DC0%0/2
defined $_ && !$_->isEmpty
defined $_ !$_->isEmpty
Truth table0%0/3
A && not(B)
A= defined $_B= !$_->isEmpty
ABresult
0X0
100
111
54 0 return $_[0];
55 }
56  
57 ## $sent = $sent->rmComments()
58 ## + removes comment pseudo-tokens from @$sent
59 0 ▶sub rmComments {
Subroutine rmComments: not called CC=2 SCAR=17.9
Pod undocumented
60 0 ▶ @{$_[0]} = grep {!defined($_) || !$_->isComment} @{$_[0]};
Condition0%0/3
defined $_ && $_->isComment
!ll&&!rl&&r
000
MC/DC0%0/2
defined $_ && $_->isComment
defined $_ $_->isComment
Truth table0%0/3
A && B
A= defined $_B= $_->isComment
ABresult
0X0
100
111
61 0 return $_[0];
62 }
63  
64 ## $sent = $sent->rmNonVanilla()
65 ## + removes non-vanilla tokens from @$sent
66 0 ▶sub rmNonVanilla {
Subroutine rmNonVanilla: not called CC=2 SCAR=17.9
Pod undocumented
67 0 ▶ @{$_[0]} = grep {defined($_) && $_->isVanilla} @{$_[0]};
Condition0%0/3
defined $_ && $_->isVanilla
!ll&&!rl&&r
000
MC/DC0%0/2
defined $_ && $_->isVanilla
defined $_ $_->isVanilla
Truth table0%0/3
A && B
A= defined $_B= $_->isVanilla
ABresult
0X0
100
111
68 0 return $_[0];
69 }
70  
71  
72 ##==============================================================================
73 ## Methods: I/O
74  
75 ## $str = $sent->toString()
76 ## + returns string representing $sent, but without terminating newline
77 0 ▶sub toString {
Subroutine toString: not called CC=1 SCAR=6.9
Pod undocumented
78 0 return join("\n", map {$_->toString} @{$_[0]})."\n";
79 }
80  
81 ## $sent = $sent->fromString($str)
82 ## + parses $sent from string $str
83 0 ▶sub fromString {
Subroutine fromString: not called CC=1 SCAR=6.9
Pod undocumented
84 #my ($sent,$str) = @_;
85 0 @{$_[0]} = map {Lingua::TT::Token->newFromString($_)} split(/[\r\n]+/,$_[1]);
86 0 return $_[0];
87 }
88  
89 ##==============================================================================
90 ## Methods: Raw Text (heuristic)
91  
92 ## $str = $sent->rawString()
93 ## + get raw text for sentence
94 ## + returns TEXT from first comment-line '%% $stxt=TEXT' or '%% Sentence ID\t=TEXT' if available
95 ## + otherwise, heuristically generates raw-text string from sentence tokens using $sent->guessRawString()
96 0 ▶sub rawString {
Subroutine rawString: not called CC=3 SCAR=24.8
Pod undocumented
97 0 foreach (@{$_[0]}) {
98 0 ▶ return $1 if ($_->[0] =~ /^%%(?:\s*\$stxt=| Sentence\b[^\t]*\t=)(.*)/);
Branch: 1 branch
BranchTrueFalseTotal
if $_->[0] =~ /^%%(?:\s*\$stxt=| Sentence\b[^\t]*\t=)(.*)/ 0 0 0
99 }
100 0 return $_[0]->guessRawString();
101 }
102  
103 ## $str = $sent->guessdRawString()
104 ## + guess raw text for sentence
105 0 ▶sub guessRawString {
Subroutine guessRawString: not called CC=6 SCAR=37.4
Pod undocumented
106 0 my $sent = shift; ##-- ( \@tok1, \@tok2, ..., \@tokN )
107 0 my @spaces = qw(); ##-- ( $space_before_tok1, ..., $space_before_tokN )
108 0 my @toks = grep {$_->[0] !~ /^\%\%/} @$sent; ##-- remove comments
109 0 my @words = map {$_->[0]} @toks;
110  
111 0 foreach (@words) {
112 0 ▶ $_ =~ s/_/ /g if ($_ =~ /^[0-9]{1,3}(?:_[0-9]{3})+(?:,[0-9]+)?$/); ##-- map underscore to space in separated numerals
Branch: 1 branch
BranchTrueFalseTotal
if $_ =~ /^[0-9]{1,3}(?:_[0-9]{3})+(?:,[0-9]+)?$/ 0 0 0
113 }
114  
115 ##-- insert boundary space
116 0 @spaces = map {''} @words;
117 0 my ($i,$w1,$w2);
118 0 foreach $i (1..$#words) {
119 0 ($w1,$w2) = @words[($i-1),$i];
120 0 ▶ next if ($w2 =~ /^(?:[\]\)\%\.\,\:\;\!\?]|\'+|\'[[:alpha:]]+)$/); ##-- no token-boundary space BEFORE these text types
Branch: 1 branch
BranchTrueFalseTotal
if $w2 =~ /^(?:[\]\)\%\.\,\:\;\!\?]|\'+|\'[[:alpha:]]+)$/ 0 0 0
121 0 ▶ next if ($w1 =~ /^(?:[\[\(]|\`+)$/); ##-- no token-boundary space AFTER these text types
Branch: 1 branch
BranchTrueFalseTotal
if $w1 =~ /^(?:[\[\(]|\`+)$/ 0 0 0
122 0 $spaces[$i] = ' '; ##-- default: add token-boundary space
123 }
124  
125 ##-- dump raw text
126 0 return join('', map {($spaces[$_],$words[$_])} (0..$#words));
127 }
128  
129 ##==============================================================================
130 ## Footer
131 1 1;
132  
133 __END__
« TT/Enum.pm ↑ Summary TT/Packed/a.pm »