| line |
stmt |
bran |
cond |
sub |
pod |
time |
code |
|
1
|
|
|
|
|
|
|
=pod |
|
2
|
|
|
|
|
|
|
|
|
3
|
|
|
|
|
|
|
=head1 NAME |
|
4
|
|
|
|
|
|
|
|
|
5
|
|
|
|
|
|
|
ETL::Pipeline::Input::File::Table - Sequential input in rows and columns |
|
6
|
|
|
|
|
|
|
|
|
7
|
|
|
|
|
|
|
=head1 SYNOPSIS |
|
8
|
|
|
|
|
|
|
|
|
9
|
|
|
|
|
|
|
# In the input source... |
|
10
|
|
|
|
|
|
|
use Moose; |
|
11
|
|
|
|
|
|
|
with 'ETL::Pipeline::Input'; |
|
12
|
|
|
|
|
|
|
with 'ETL::Pipeline::Input::File'; |
|
13
|
|
|
|
|
|
|
with 'ETL::Pipeline::Input::File::Table'; |
|
14
|
|
|
|
|
|
|
... |
|
15
|
|
|
|
|
|
|
|
|
16
|
|
|
|
|
|
|
=head1 DESCRIPTION |
|
17
|
|
|
|
|
|
|
|
|
18
|
|
|
|
|
|
|
CSV (comma separated values) or Excel spreadsheet files represent data in a |
|
19
|
|
|
|
|
|
|
table structure. Each row is a record. Each column an individual field. This |
|
20
|
|
|
|
|
|
|
role provides some attributes common for this type of data. That way you don't |
|
21
|
|
|
|
|
|
|
have to reinvent the wheel every time. |
|
22
|
|
|
|
|
|
|
|
|
23
|
|
|
|
|
|
|
=cut |
|
24
|
|
|
|
|
|
|
|
|
25
|
|
|
|
|
|
|
package ETL::Pipeline::Input::File::Table; |
|
26
|
|
|
|
|
|
|
|
|
27
|
3
|
|
|
3
|
|
3000
|
use 5.014000; |
|
|
3
|
|
|
|
|
13
|
|
|
28
|
|
|
|
|
|
|
|
|
29
|
3
|
|
|
3
|
|
23
|
use List::AllUtils qw/indexes/; |
|
|
3
|
|
|
|
|
7
|
|
|
|
3
|
|
|
|
|
237
|
|
|
30
|
3
|
|
|
3
|
|
19
|
use Moose::Role; |
|
|
3
|
|
|
|
|
9
|
|
|
|
3
|
|
|
|
|
35
|
|
|
31
|
3
|
|
|
3
|
|
18640
|
use String::Util qw/hascontent trim/; |
|
|
3
|
|
|
|
|
10
|
|
|
|
3
|
|
|
|
|
456
|
|
|
32
|
|
|
|
|
|
|
|
|
33
|
|
|
|
|
|
|
|
|
34
|
|
|
|
|
|
|
our $VERSION = '2.00'; |
|
35
|
|
|
|
|
|
|
|
|
36
|
|
|
|
|
|
|
|
|
37
|
|
|
|
|
|
|
=head1 METHODS & ATTRIBUTES |
|
38
|
|
|
|
|
|
|
|
|
39
|
|
|
|
|
|
|
=head2 Arguments for L<ETL::Pipeline/input> |
|
40
|
|
|
|
|
|
|
|
|
41
|
|
|
|
|
|
|
=head3 no_column_names |
|
42
|
|
|
|
|
|
|
|
|
43
|
|
|
|
|
|
|
Tabular data usually has field names in the very first row. This makes it |
|
44
|
|
|
|
|
|
|
easier for a human being to read. Sometimes, though, there are no field names. |
|
45
|
|
|
|
|
|
|
The data starts on the very first row. |
|
46
|
|
|
|
|
|
|
|
|
47
|
|
|
|
|
|
|
Set B<no_column_name> to B<true> for these cases. Otherwise, the input source |
|
48
|
|
|
|
|
|
|
will load your first row of data as field names. |
|
49
|
|
|
|
|
|
|
|
|
50
|
|
|
|
|
|
|
$etl->input( 'Excel', no_column_names => 1 ); |
|
51
|
|
|
|
|
|
|
|
|
52
|
|
|
|
|
|
|
=cut |
|
53
|
|
|
|
|
|
|
|
|
54
|
|
|
|
|
|
|
has 'no_column_names' => ( |
|
55
|
|
|
|
|
|
|
default => 0, |
|
56
|
|
|
|
|
|
|
is => 'ro', |
|
57
|
|
|
|
|
|
|
isa => 'Bool', |
|
58
|
|
|
|
|
|
|
); |
|
59
|
|
|
|
|
|
|
|
|
60
|
|
|
|
|
|
|
|
|
61
|
|
|
|
|
|
|
=head1 SEE ALSO |
|
62
|
|
|
|
|
|
|
|
|
63
|
|
|
|
|
|
|
L<ETL::Pipeline>, L<ETL::Pipeline::Input>, L<ETL::Pipeline::Input::File> |
|
64
|
|
|
|
|
|
|
|
|
65
|
|
|
|
|
|
|
=head1 AUTHOR |
|
66
|
|
|
|
|
|
|
|
|
67
|
|
|
|
|
|
|
Robert Wohlfarth <robert.j.wohlfarth@vumc.org> |
|
68
|
|
|
|
|
|
|
|
|
69
|
|
|
|
|
|
|
=head1 LICENSE |
|
70
|
|
|
|
|
|
|
|
|
71
|
|
|
|
|
|
|
Copyright 2021 (c) Vanderbilt University Medical Center |
|
72
|
|
|
|
|
|
|
|
|
73
|
|
|
|
|
|
|
This program is free software; you can redistribute it and/or modify it under |
|
74
|
|
|
|
|
|
|
the same terms as Perl itself. |
|
75
|
|
|
|
|
|
|
|
|
76
|
|
|
|
|
|
|
=cut |
|
77
|
|
|
|
|
|
|
|
|
78
|
3
|
|
|
3
|
|
23
|
no Moose; |
|
|
3
|
|
|
|
|
9
|
|
|
|
3
|
|
|
|
|
29
|
|
|
79
|
|
|
|
|
|
|
|
|
80
|
|
|
|
|
|
|
# Required by Perl to load the module. |
|
81
|
|
|
|
|
|
|
1; |