CSVUtils/csv_cmp.pm

filter: Statementstmt 21.2% 20 / 94 Branchbran 0.0% 0 / 52 Conditioncond 0.0% 0 / 9 MC/DCmcdc 0.0% 0 / 6 Subroutinesub 50.0% 5 / 10 Podpod n/a totaltotal 14.6% 25 / 171 CC 32 SCAR 60.0
CC
32
Cov
13%
__ANON__
650.0
__ANON__
30.0
__ANON__
20.0
CRAP 402.2
↑ Summary CSVUtils/csv_find_values.pm »
1 package App::CSVUtils::csv_cmp;
2  
3 1 ▶use 5.010001;
Subroutine BEGIN: called CC=1 SCAR=0.0
4 1 ▶use strict;
Subroutine BEGIN: called CC=1 SCAR=0.0
5 1 ▶use warnings;
Subroutine BEGIN: called CC=1 SCAR=0.0
6 1 ▶use Log::ger;
Subroutine BEGIN: called CC=1 SCAR=0.0
7  
8 1 our $AUTHORITY = 'cpan:PERLANCAR'; # AUTHORITY
9 1 our $DATE = '2026-07-09'; # DATE
10 1 our $DIST = 'App-CSVUtils'; # DIST
11 1 our $VERSION = '1.038'; # VERSION
12  
13 1 ▶use App::CSVUtils qw(
Subroutine BEGIN: called CC=1 SCAR=0.0
14 gen_csv_util
15 1 );
16  
17 gen_csv_util(
18 name => 'csv_cmp',
19 summary => 'Compare two CSV files value by value',
20 description => <<'MARKDOWN',
21  
22 This utility is modelled after the Unix command `cmp`; it compares two CSV files
23 value by value and ignore quoting (and can be instructed to ignore whitespaces,
24 case difference).
25  
26 If all the values of two CSV files are identical, then utility will exit with
27 code 0. If a value differ, this utility will stop, print the difference and exit
28 with code 1.
29  
30 If `-l` (`--detail`) option is specified, all differences will be reported. Note
31 that in `cmp` Unix command, the `-l` option is called `--verbose`. The detailed
32 report is in the form of CSV:
33  
34 rownum,fieldnum,value1,value2
35  
36 where `rownum` begins at 1 (for header row), `fieldnum` begins at 1 (first
37 field), `value1` is the value in first CSV file, `value2` is the value in the
38 second CSV file.
39  
40 Other notes:
41  
42 * If none of the field selection options are used, it means all fields are
43 included (equivalent to `--include-all-fields`).
44  
45 * Field selection will be performed on the first CSV file, then the indexes will
46 be used for the second CSV file.
47  
48 MARKDOWN
49 add_args => {
50 %App::CSVUtils::argspecsopt_field_selection,
51 %App::CSVUtils::argspecsopt_show_selected_fields,
52  
53 detail => {
54 summary => 'Report all differences instead of just the first one',
55 schema => 'true*',
56 cmdline_aliases => {l=>{}},
57 },
58 quiet => {
59 summary => 'Do not report, just signal via exit code',
60 schema => 'true*',
61 cmdline_aliases => {q=>{}},
62 },
63 ignore_case => {
64 summary => 'Ignore case difference',
65 schema => 'bool*',
66 cmdline_aliases => {i=>{}},
67 },
68 ignore_leading_ws => {
69 summary => 'Ignore leading whitespaces',
70 schema => 'bool*',
71 },
72 ignore_trailing_ws => {
73 summary => 'Ignore trailing whitespaces',
74 schema => 'bool*',
75 },
76 ignore_ws => {
77 summary => 'Ignore leading & trailing whitespaces',
78 schema => 'bool*',
79 },
80 },
81 tags => [
82 'accepts-regex', # for selecting fields
83 'category:comparing',
84 ],
85  
86 examples => [
87 {
88 summary => 'Compare two identical files, will output nothing and exits 0',
89 argv => ['file.csv', 'file.csv'],
90 test => 0,
91 'x.doc.show_result' => 0,
92 },
93 {
94 summary => 'Compare two CSV files case-insensitively (-i), show detailed report (-l)',
95 argv => ['file1.csv', 'file2.csv', '-il'],
96 test => 0,
97 'x.doc.show_result' => 0,
98 },
99 ],
100  
101 reads_multiple_csv => 1,
102  
103 0 ▶ on_begin => sub {
Subroutine __ANON__: not called CC=4 SCAR=30.0
104 0 my $r = shift;
105  
106 0 ▶ unless ($r->{util_args}{input_filenames} && @{ $r->{util_args}{input_filenames} } == 2) {
Branch: 1 branch
BranchTrueFalseTotal
unless ($r->{"util_args"}{"input_filenames"} and @{$r->{"util_args"}{"input_filenames"};} == 2) 0 0 0
Condition0%0/3
$r->{"util_args"}{"input_filenames"} and @{$r->{"util_args"}{"input_filenames"};} == 2
!ll&&!rl&&r
000
MC/DC0%0/2
$r->{"util_args"}{"input_filenames"} and @{$r->{"util_args"}{"input_filenames"};} == 2
$r->{"util_args"}{"input_filenames"} @{$r->{"util_args"}{"input_filenames"};} == 2
Truth table0%0/3
A and B
A= $r->{"util_args"}{"input_filenames"}B= @{$r->{"util_args"}{"input_filenames"};} == 2
ABresult
0X0
100
111
107 0 die [400, "Please specify exactly two files to compare"];
108 }
109 # no point in generating detailed report if we're not going to show it
110 0 ▶ $r->{util_args}{detail} = 0 if $r->{util_args}{quiet};
Branch: 1 branch
BranchTrueFalseTotal
if $r->{"util_args"}{"quiet"} 0 0 0
111 },
112  
113 0 ▶ before_open_input_files => sub {
Subroutine __ANON__: not called CC=1 SCAR=6.9
114 0 my $r = shift;
115  
116 # we add the following keys to the stash
117 0 $r->{all_input_rows} = [[], []]; # including header row. format: [ [row1 of csv1, row2 of csv1...], [row1 of csv2, row2 of csv2, ...] ]
118 0 $r->{selected_fields_idx_array_sorted} = undef;
119 },
120  
121 0 ▶ on_input_header_row => sub {
Subroutine __ANON__: not called CC=5 SCAR=34.0
122 0 my $r = shift;
123  
124 0 push @{ $r->{all_input_rows}[ $r->{input_filenum}-1 ] }, $r->{input_fields};
125  
126 0 ▶ if ($r->{input_filenum} == 1) {
Branch: 1 branch
BranchTrueFalseTotal
if ($r->{"input_filenum"} == 1) 0 0 0
127 # set selected_fields_idx_array_sorted
128 0 my $res = App::CSVUtils::_select_fields($r->{input_fields}, $r->{input_fields_idx}, $r->{util_args}, 'all');
129 0 ▶ die $res unless $res->[0] == 100;
Branch: 1 branch
BranchTrueFalseTotal
unless $res->[0] == 100 0 0 0
130 0 my $selected_fields = $res->[2][0];
131 0 my $selected_fields_idx_array = $res->[2][1];
132 0 ▶ die [412, "At least one field must be selected"]
Branch: 1 branch
BranchTrueFalseTotal
unless @$selected_fields 0 0 0
133 unless @$selected_fields;
134 0 $r->{selected_fields_idx_array_sorted} = [sort { $b <=> $a } @$selected_fields_idx_array];
135  
136 0 ▶ if ($r->{util_args}{show_selected_fields}) {
Branch: 1 branch
BranchTrueFalseTotal
if ($r->{"util_args"}{"show_selected_fields"}) 0 0 0
137 0 $r->{wants_skip_files}++;
138 0 $r->{result} = [200, "OK", $selected_fields];
139 0 return;
140 }
141 }
142 },
143  
144 0 ▶ on_input_data_row => sub {
Subroutine __ANON__: not called CC=1 SCAR=6.9
145 0 my $r = shift;
146  
147 0 push @{ $r->{all_input_rows}[ $r->{input_filenum}-1 ] }, $r->{input_row};
148 },
149  
150 0 ▶ after_close_input_files => sub {
Subroutine __ANON__: not called CC=25 SCAR=64.8
151 0 my $r = shift;
152  
153 0 $r->{output_fields} = ["rownum","fieldnum","value1","value2"];
154  
155 0 my $exit_code = 0;
156 0 my $numrows1 = @{ $r->{all_input_rows}[0] };
157 0 my $numrows2 = @{ $r->{all_input_rows}[1] };
158 0 my $numfields1 = @{ $r->{all_input_rows}[0][0] };
159 0 my $numfields2 = @{ $r->{all_input_rows}[1][0] };
160  
161 0 ▶ if ($numfields1 > $numfields2) {
Branch: 2 branches
BranchTrueFalseTotal
if ($numfields1 > $numfields2) { } 0 0 0
elsif ($numfields1 < $numfields2) { } 0 0 0
162 warn "csv-cmp: second CSV only has $numfields2 field(s) (vs $numfields1)\n"
163 0 ▶ unless $r->{util_args}{quiet};
Branch: 1 branch
BranchTrueFalseTotal
unless $r->{"util_args"}{"quiet"} 0 0 0
164 0 $exit_code = 1;
165 0 ▶ goto DONE unless $r->{util_args}{detail};
Branch: 1 branch
BranchTrueFalseTotal
unless $r->{"util_args"}{"detail"} 0 0 0
166 } elsif ($numfields1 < $numfields2) {
167 warn "csv-cmp: first CSV only has $numfields1 field(s) (vs $numfields2)\n"
168 0 ▶ unless $r->{util_args}{quiet};
Branch: 1 branch
BranchTrueFalseTotal
unless $r->{"util_args"}{"quiet"} 0 0 0
169 0 $exit_code = 1;
170 0 ▶ goto DONE unless $r->{util_args}{detail};
Branch: 1 branch
BranchTrueFalseTotal
unless $r->{"util_args"}{"detail"} 0 0 0
171 }
172  
173 0 ▶ my $numrows_min = $numrows1 < $numrows2 ? $numrows1 : $numrows2;
Branch: 1 branch
BranchTrueFalseTotal
$numrows1 < $numrows2 ? : 0 0 0
174 0 for my $rownum (1 .. $numrows_min) {
175 0 for my $j (@{ $r->{selected_fields_idx_array_sorted} }) {
176 0 my $fieldnum = $j+1;
177 0 my $origvalue1 = my $value1 = $r->{all_input_rows}[0][ $rownum-1 ][ $fieldnum-1 ];
178 0 my $origvalue2 = my $value2 = $r->{all_input_rows}[1][ $rownum-1 ][ $fieldnum-1 ];
179  
180 0 ▶ if ($r->{util_args}{ignore_case}) {
Branch: 1 branch
BranchTrueFalseTotal
if ($r->{"util_args"}{"ignore_case"}) 0 0 0
181 0 $value1 = lc $value1;
182 0 $value2 = lc $value2;
183 }
184 0 ▶ if ($r->{util_args}{ignore_ws} || $r->{util_args}{ignore_leading_ws}) {
Branch: 1 branch
BranchTrueFalseTotal
if ($r->{"util_args"}{"ignore_ws"} or $r->{"util_args"}{"ignore_leading_ws"}) 0 0 0
Condition0%0/3
$r->{"util_args"}{"ignore_ws"} or $r->{"util_args"}{"ignore_leading_ws"}
l!l&&r!l&&!r
000
MC/DC0%0/2
$r->{"util_args"}{"ignore_ws"} or $r->{"util_args"}{"ignore_leading_ws"}
$r->{"util_args"}{"ignore_ws"} $r->{"util_args"}{"ignore_leading_ws"}
Truth table0%0/3
A or B
A= $r->{"util_args"}{"ignore_ws"}B= $r->{"util_args"}{"ignore_leading_ws"}
ABresult
1X1
011
000
185 0 $value1 =~ s/\A\s+//s;
186 0 $value2 =~ s/\A\s+//s;
187 }
188 0 ▶ if ($r->{util_args}{ignore_ws} || $r->{util_args}{ignore_trailing_ws}) {
Branch: 1 branch
BranchTrueFalseTotal
if ($r->{"util_args"}{"ignore_ws"} or $r->{"util_args"}{"ignore_trailing_ws"}) 0 0 0
Condition0%0/3
$r->{"util_args"}{"ignore_ws"} or $r->{"util_args"}{"ignore_trailing_ws"}
l!l&&r!l&&!r
000
MC/DC0%0/2
$r->{"util_args"}{"ignore_ws"} or $r->{"util_args"}{"ignore_trailing_ws"}
$r->{"util_args"}{"ignore_ws"} $r->{"util_args"}{"ignore_trailing_ws"}
Truth table0%0/3
A or B
A= $r->{"util_args"}{"ignore_ws"}B= $r->{"util_args"}{"ignore_trailing_ws"}
ABresult
1X1
011
000
189 0 $value1 =~ s/\s+\z//s;
190 0 $value2 =~ s/\s+\z//s;
191 }
192  
193 0 ▶ if ($value1 ne $value2) {
Branch: 1 branch
BranchTrueFalseTotal
if ($value1 ne $value2) 0 0 0
194 0 $exit_code = 1;
195 0 ▶ if ($r->{util_args}{detail}) {
Branch: 1 branch
BranchTrueFalseTotal
if ($r->{'util_args'}{'detail'}) { } 0 0 0
196 $r->{code_print_row}->([$rownum, $fieldnum, $origvalue1, $origvalue2])
197 0 ▶ unless $r->{util_args}{quiet};
Branch: 1 branch
BranchTrueFalseTotal
unless $r->{"util_args"}{"quiet"} 0 0 0
198 } else {
199 warn "csv-cmp: Value differ at rownum $rownum fieldnum $fieldnum: '$origvalue1' vs '$origvalue2'\n"
200 0 ▶ unless $r->{util_args}{quiet};
Branch: 1 branch
BranchTrueFalseTotal
unless $r->{"util_args"}{"quiet"} 0 0 0
201 0 goto DONE;
202 }
203 }
204 } # for field
205 } # for row
206  
207 0 ▶ if ($numrows1 > $numrows2) {
Branch: 2 branches
BranchTrueFalseTotal
if ($numrows1 > $numrows2) { } 0 0 0
elsif ($numrows1 < $numrows2) { } 0 0 0
208 warn "csv-cmp: EOF: second CSV only has $numrows2 row(s) (vs $numrows1)\n"
209 0 ▶ unless $r->{util_args}{quiet};
Branch: 1 branch
BranchTrueFalseTotal
unless $r->{"util_args"}{"quiet"} 0 0 0
210 0 $exit_code = 1;
211 0 ▶ goto DONE unless $r->{util_args}{detail};
Branch: 1 branch
BranchTrueFalseTotal
unless $r->{"util_args"}{"detail"} 0 0 0
212 } elsif ($numrows1 < $numrows2) {
213 warn "csv-cmp: EOF: first CSV only has $numrows1 row(s) (vs $numrows2)\n"
214 0 ▶ unless $r->{util_args}{quiet};
Branch: 1 branch
BranchTrueFalseTotal
unless $r->{"util_args"}{"quiet"} 0 0 0
215 0 $exit_code = 1;
216 0 ▶ goto DONE unless $r->{util_args}{detail};
Branch: 1 branch
BranchTrueFalseTotal
unless $r->{"util_args"}{"detail"} 0 0 0
217 }
218  
219 DONE:
220 0 $r->{result} = [200, "OK", "", {"cmdline.exit_code"=>$exit_code}];
221 },
222 1 );
223  
224 1 1;
225 # ABSTRACT: Compare two CSV files value by value
226  
227 __END__
↑ Summary CSVUtils/csv_find_values.pm »