1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
|
package SQL::Translator::Parser::xSV;
=head1 NAME
SQL::Translator::Parser::xSV - parser for arbitrarily delimited text files
=head1 SYNOPSIS
use SQL::Translator;
use SQL::Translator::Parser::xSV;
my $translator = SQL::Translator->new(
parser => 'xSV',
parser_args => { field_separator => "\t" },
);
=head1 DESCRIPTION
Parses arbitrarily delimited text files. See the
Text::RecordParser manpage for arguments on how to parse the file
(e.g., C<field_separator>, C<record_separator>). Other arguments
include:
=head1 OPTIONS
=over
=item * scan_fields
Indicates that the columns should be scanned to determine data types
and field sizes. True by default.
=item * trim_fields
A shortcut to sending filters to Text::RecordParser, will create
callbacks that trim leading and trailing spaces from fields and headers.
True by default.
=back
Field names will automatically be normalized by
C<SQL::Translator::Utils::normalize_name>.
=cut
use strict;
use warnings;
our @EXPORT;
our $VERSION = '1.59';
use Exporter;
use Text::ParseWords qw(quotewords);
use Text::RecordParser;
use SQL::Translator::Utils qw(debug normalize_name);
use base qw(Exporter);
@EXPORT = qw(parse);
#
# Passed a SQL::Translator instance and a string containing the data
#
sub parse {
my ( $tr, $data ) = @_;
my $args = $tr->parser_args;
my $parser = Text::RecordParser->new(
field_separator => $args->{'field_separator'} || ',',
record_separator => $args->{'record_separator'} || "\n",
data => $data,
header_filter => \&normalize_name,
);
$parser->field_filter( sub { $_ = shift || ''; s/^\s+|\s+$//g; $_ } )
unless defined $args->{'trim_fields'} && $args->{'trim_fields'} == 0;
my $schema = $tr->schema;
my $table = $schema->add_table( name => 'table1' );
#
# Get the field names from the first row.
#
$parser->bind_header;
my @field_names = $parser->field_list;
for ( my $i = 0; $i < @field_names; $i++ ) {
my $field = $table->add_field(
name => $field_names[$i],
data_type => 'char',
default_value => '',
size => 255,
is_nullable => 1,
is_auto_increment => undef,
) or die $table->error;
if ( $i == 0 ) {
$table->primary_key( $field->name );
$field->is_primary_key(1);
}
}
#
# If directed, look at every field's values to guess size and type.
#
unless (
defined $args->{'scan_fields'} &&
$args->{'scan_fields'} == 0
) {
my %field_info = map { $_, {} } @field_names;
while ( my $rec = $parser->fetchrow_hashref ) {
for my $field ( @field_names ) {
my $data = defined $rec->{ $field } ? $rec->{ $field } : '';
my $size = [ length $data ];
my $type;
if ( $data =~ /^-?\d+$/ ) {
$type = 'integer';
}
elsif (
$data =~ /^-?[,\d]+\.[\d+]?$/
||
$data =~ /^-?[,\d]+?\.\d+$/
||
$data =~ /^-?\.\d+$/
) {
$type = 'float';
my ( $w, $d ) =
map { s/,//g; length $_ || 1 } split( /\./, $data );
$size = [ $w + $d, $d ];
}
else {
$type = 'char';
}
for my $i ( 0, 1 ) {
next unless defined $size->[ $i ];
my $fsize = $field_info{ $field }{'size'}[ $i ] || 0;
if ( $size->[ $i ] > $fsize ) {
$field_info{ $field }{'size'}[ $i ] = $size->[ $i ];
}
}
$field_info{ $field }{ $type }++;
}
}
for my $field ( keys %field_info ) {
my $size = $field_info{ $field }{'size'} || [ 1 ];
my $data_type =
$field_info{ $field }{'char'} ? 'char' :
$field_info{ $field }{'float'} ? 'float' :
$field_info{ $field }{'integer'} ? 'integer' : 'char';
if ( $data_type eq 'char' && scalar @$size == 2 ) {
$size = [ $size->[0] + $size->[1] ];
}
my $field = $table->get_field( $field );
$field->size( $size );
$field->data_type( $data_type );
}
}
return 1;
}
1;
=pod
=head1 AUTHORS
Darren Chamberlain E<lt>darren@cpan.orgE<gt>,
Ken Y. Clark E<lt>kclark@cpan.orgE<gt>.
=head1 SEE ALSO
Text::RecordParser, SQL::Translator.
=cut
|
