line |
stmt |
bran |
cond |
sub |
pod |
time |
code |
1
|
|
|
|
|
|
|
package Treex::Block::Read::BaseTextReader; |
2
|
|
|
|
|
|
|
$Treex::Block::Read::BaseTextReader::VERSION = '2.20150928'; |
3
|
2
|
|
|
2
|
|
1573
|
use Moose; |
|
2
|
|
|
|
|
4
|
|
|
2
|
|
|
|
|
19
|
|
4
|
2
|
|
|
2
|
|
11563
|
use Treex::Core::Common; |
|
2
|
|
|
|
|
4
|
|
|
2
|
|
|
|
|
18
|
|
5
|
|
|
|
|
|
|
extends 'Treex::Block::Read::BaseReader'; |
6
|
|
|
|
|
|
|
#use File::Slurp 9999; |
7
|
2
|
|
|
2
|
|
9808
|
use PerlIO::via::gzip; |
|
2
|
|
|
|
|
4
|
|
|
2
|
|
|
|
|
761
|
|
8
|
|
|
|
|
|
|
|
9
|
|
|
|
|
|
|
# By default read from STDIN |
10
|
|
|
|
|
|
|
has '+from' => ( |
11
|
|
|
|
|
|
|
default => '-', |
12
|
|
|
|
|
|
|
handles => [qw(current_filename current_filehandle file_number _set_file_number next_filehandle)], |
13
|
|
|
|
|
|
|
); |
14
|
|
|
|
|
|
|
|
15
|
|
|
|
|
|
|
has lines_per_doc => ( isa => 'Int', is => 'ro', default => 0 ); |
16
|
|
|
|
|
|
|
has merge_files => ( isa => 'Bool', is => 'ro', default => 0 ); |
17
|
|
|
|
|
|
|
has encoding => ( isa => 'Str', is => 'ro', default => 'utf8' ); |
18
|
|
|
|
|
|
|
|
19
|
|
|
|
|
|
|
sub BUILD { |
20
|
0
|
|
|
0
|
|
|
my ($self) = @_; |
21
|
0
|
0
|
|
|
|
|
if ( $self->lines_per_doc ) { |
22
|
0
|
|
|
|
|
|
$self->set_is_one_doc_per_file(0); |
23
|
|
|
|
|
|
|
} |
24
|
0
|
|
|
|
|
|
return; |
25
|
|
|
|
|
|
|
} |
26
|
|
|
|
|
|
|
|
27
|
|
|
|
|
|
|
sub next_document_text { |
28
|
0
|
|
|
0
|
|
|
my ($self) = @_; |
29
|
0
|
0
|
|
|
|
|
if ( $self->is_one_doc_per_file ) { |
30
|
0
|
|
|
|
|
|
return $self->from->next_file_text(); |
31
|
|
|
|
|
|
|
} |
32
|
|
|
|
|
|
|
|
33
|
0
|
|
|
|
|
|
my $text = ''; |
34
|
|
|
|
|
|
|
LINE: |
35
|
0
|
|
|
|
|
|
for my $line ( 1 .. $self->lines_per_doc ) { |
36
|
0
|
|
|
|
|
|
$line = $self->from->next_line(); |
37
|
0
|
0
|
|
|
|
|
if (!defined $line){ |
38
|
0
|
0
|
0
|
|
|
|
return if $text eq '' && !$self->from->has_next_file(); |
39
|
0
|
|
|
|
|
|
last LINE; |
40
|
|
|
|
|
|
|
} |
41
|
|
|
|
|
|
|
|
42
|
0
|
|
|
|
|
|
$text .= $line; |
43
|
|
|
|
|
|
|
} |
44
|
0
|
|
|
|
|
|
return $text; |
45
|
|
|
|
|
|
|
} |
46
|
|
|
|
|
|
|
|
47
|
|
|
|
|
|
|
1; |
48
|
|
|
|
|
|
|
|
49
|
|
|
|
|
|
|
__END__ |
50
|
|
|
|
|
|
|
|
51
|
|
|
|
|
|
|
=pod |
52
|
|
|
|
|
|
|
|
53
|
|
|
|
|
|
|
=encoding utf-8 |
54
|
|
|
|
|
|
|
|
55
|
|
|
|
|
|
|
=for Pod::Coverage BUILD |
56
|
|
|
|
|
|
|
|
57
|
|
|
|
|
|
|
=head1 NAME |
58
|
|
|
|
|
|
|
|
59
|
|
|
|
|
|
|
Treex::Block::Read::BaseTextReader - abstract ancestor for document readers |
60
|
|
|
|
|
|
|
|
61
|
|
|
|
|
|
|
=head1 VERSION |
62
|
|
|
|
|
|
|
|
63
|
|
|
|
|
|
|
version 2.20150928 |
64
|
|
|
|
|
|
|
|
65
|
|
|
|
|
|
|
=head1 DESCRIPTION |
66
|
|
|
|
|
|
|
|
67
|
|
|
|
|
|
|
This class serves as an common ancestor for document readers, |
68
|
|
|
|
|
|
|
that have parameter C<from> with a space or comma separated list of filenames |
69
|
|
|
|
|
|
|
to be loaded and load the documents from plain text files. |
70
|
|
|
|
|
|
|
It is designed to implement the L<Treex::Core::DocumentReader> interface. |
71
|
|
|
|
|
|
|
|
72
|
|
|
|
|
|
|
In derived classes you need to define the C<next_document> method, |
73
|
|
|
|
|
|
|
and you can use C<next_document_text> and C<new_document> methods. |
74
|
|
|
|
|
|
|
|
75
|
|
|
|
|
|
|
=head1 ATTRIBUTES |
76
|
|
|
|
|
|
|
|
77
|
|
|
|
|
|
|
=over |
78
|
|
|
|
|
|
|
|
79
|
|
|
|
|
|
|
=item language (required) |
80
|
|
|
|
|
|
|
|
81
|
|
|
|
|
|
|
=item lines_per_doc |
82
|
|
|
|
|
|
|
|
83
|
|
|
|
|
|
|
If you want to split one file to more documents. |
84
|
|
|
|
|
|
|
The default is 0 which means, don't split. |
85
|
|
|
|
|
|
|
|
86
|
|
|
|
|
|
|
=item merge_files |
87
|
|
|
|
|
|
|
|
88
|
|
|
|
|
|
|
Merge the content of all files (specified in C<from> attribute) into one stream. |
89
|
|
|
|
|
|
|
Useful in combination with C<lines_per_doc> to get equally-sized documents |
90
|
|
|
|
|
|
|
even from non-equally-sized files. |
91
|
|
|
|
|
|
|
|
92
|
|
|
|
|
|
|
=item encoding |
93
|
|
|
|
|
|
|
|
94
|
|
|
|
|
|
|
What is the encoding of the input files. E.g. C<utf8> (the default), C<cp1250> etc. |
95
|
|
|
|
|
|
|
|
96
|
|
|
|
|
|
|
=back |
97
|
|
|
|
|
|
|
|
98
|
|
|
|
|
|
|
=head1 METHODS |
99
|
|
|
|
|
|
|
|
100
|
|
|
|
|
|
|
=over |
101
|
|
|
|
|
|
|
|
102
|
|
|
|
|
|
|
=item next_document_text |
103
|
|
|
|
|
|
|
|
104
|
|
|
|
|
|
|
Returns a content of each file (specified in C<from> attribute) as a text string. |
105
|
|
|
|
|
|
|
|
106
|
|
|
|
|
|
|
=item next_filehandle |
107
|
|
|
|
|
|
|
|
108
|
|
|
|
|
|
|
Helper method - you can use this instead of C<next_document_text> |
109
|
|
|
|
|
|
|
if you don't want to load the whole text into memory |
110
|
|
|
|
|
|
|
(but do e.g. SAX-like parsing). |
111
|
|
|
|
|
|
|
|
112
|
|
|
|
|
|
|
=back |
113
|
|
|
|
|
|
|
|
114
|
|
|
|
|
|
|
=head1 SEE |
115
|
|
|
|
|
|
|
|
116
|
|
|
|
|
|
|
L<Treex::Block::Read::BaseReader> |
117
|
|
|
|
|
|
|
L<Treex::Block::Read::Text> |
118
|
|
|
|
|
|
|
|
119
|
|
|
|
|
|
|
=head1 AUTHOR |
120
|
|
|
|
|
|
|
|
121
|
|
|
|
|
|
|
Martin Popel |
122
|
|
|
|
|
|
|
|
123
|
|
|
|
|
|
|
=head1 COPYRIGHT AND LICENSE |
124
|
|
|
|
|
|
|
|
125
|
|
|
|
|
|
|
Copyright © 2011-2012 by Institute of Formal and Applied Linguistics, Charles University in Prague |
126
|
|
|
|
|
|
|
|
127
|
|
|
|
|
|
|
This module is free software; you can redistribute it and/or modify it under the same terms as Perl itself. |