Perl Testing: A Developer's Notebook
by Ian Langworth, chromatic

The unconfirmed error reports are from readers.  They have not yet been 
approved or disproved by the author or editor and represent solely the
opinion of the reader.

Here's a key to the markup:
        [page-number]: serious technical mistake
        {page-number}: minor technical mistake
        <page-number>: important language/formatting problem
        (page-number): language change or minor formatting problem
        ?page-number?: reader question or request for clarification

This page was updated March 4, 2008.


UNCONFIRMED  errors and comments from readers: 

(3) 3rd line;
On page 1, 3rd last line, you say 'if you have a C compiler', and then on page 3, 3rd line you say 'that has an appropriate compiler'. I think the text on page 3 was meant to read 'that does not have.'


[14]  2nd "What About" Answer;
"and add the following lines to the end of analyze_sentence.t:". Calculating the
number of words in $sentence here, at the end of the script, will result in a failure
because we are testing for 18 words. However, a few lines above, we changed $sentence
to "Rampaging ideas flutter greedily.", which is four words.



{14}  second A: on page;
as written test 6 in analyze_sentence.t will fail because @word will have 4 elements,
not 18. To fix the example I made an amateurish change to this:

#!perl

use strict;
use warnings;

use Test::More tests =>6 ;

my @subs = qw( words count_words );

#use_ok( 'AnalyzeSentence', @subs   );
#BEGIN { use_ok( 'AnalyzeSentence', @subs, '$WORD_SEPARATOR' ) or exit;} #as example
for testing variable $WORD_SEPARATOR

BEGIN
        {
            my @subs = qw( words count_words );
            use_ok( 'AnalyzeSentence', @subs, '$WORD_SEPARATOR' ) or exit;
        }

can_ok( __PACKAGE__, 'words'       );
can_ok( __PACKAGE__, 'count_words' );

my $sentence1 =
    'Queen Esther, ruler of the Frog-Human Alliance, briskly devours a monumental ice
cream sundae in her honor.';

my @words = words( $sentence1 );
ok( @words == 17, 'words(  ) should return all words in sentence' );

my $sentence = 'Rampaging ideas flutter greedily.';
my $count = count_words( $sentence );

ok( $count == 4, 'count_words(  ) should handle simple sentences' );

$WORD_SEPARATOR = qr/(?:\s|-)+/;
@words = words( $sentence1 );
ok( @words == 18, '...respecting $WORD_SEPARATOR, if set' );


{15} Around the 3th line;
The BEGIN statement errata is already posted, but it will still fail because the last defining of $sentence only has 4 words. In order to get the test right, the last three lines of analyze_sentence.t should be as shown below. Note that you need to type in once again the "Queen Esther" sentence because we want the word count to be 18 due to "Frog-Human" now being counted as two words.

$sentence =
'Queen Esther, ruler of the Frog-Human Alliance, briskly devours a
monumental ice cream sundae in her honor.';

$WORD_SEPARATOR = qr/(?:\s|-)+/;
@words = words( $sentence );
ok( @words == 18, '... respecting $WORD_SEPARATOR, if set' );


[51]  example - between.t;
There is still an error using test_out() instead of test_pass():

> perl between.t
1..3
not ok 1 - simple alphabetical comparison
#   Failed test 'simple alphabetical comparison'
#   in between.t at line 14.
# STDOUT is:
# ok 1 - simple alphabetical comparison
#
# not:
# simple alphabetical comparison
#
# as expected
not ok 2 - simple numeric comparison
#   Failed test 'simple numeric comparison'
#   in between.t at line 19.
# STDOUT is:
# ok 1 - simple numeric comparison
#
# not:
# simple numeric comparison
#
# as expected
ok 3 - failed comparison with diagnostics



(51)  example - between.t;
Further to my previous report regarding test_out(), the solution is to change the
first two instances of test_out() to:

test_out("ok 1 - $desc");



(54)  1st code example new_harness.pl;
The following line in new_harness.pl reads...
    my %results = $strap->analyze_file( $file );
It should be...
    my $results = $strap->analyze_file( $file );
----------------------------------------------------------------
The following line in new_harness.pl reads...
    printf <<END_REPORT, $file, @results{qw( max seen ok skip todo bonus )};
It should be...
    printf <<END_REPORT, $file, @$results{qw( max seen ok skip todo bonus )};



{54}  2nd paragraph;
Page 53 says that the new harness script needs to be stored as 'new_harness.pl'
somewhere in the $PATH. On page 54 this script is executed as 'new_harness
t/strap*t'.

Either the script needs to be stored as 'new_harness' inside $PATH or it needs to be
executed as 'new_harness.pl t/strap*t.



{157}  output from `prove -v queue.t';
The output from `prove' on this page is the same as the output from pages 153-154.
That is, it's the output of `prove' when run against the original Queue::Test module
described on page 152, which doesn't have any fixtures.

The updated Queue::Test module described on pages 156-157 uses a fixture to create
two Queue objects before running each of the other test methods. The output from
`prove' when run against this updated module should be something like the text below,
which matches the description in the second paragraph on page 157 (under the heading
"What just happened?"):

  "...There are a total of six isa checks..."

t/queue....1..12
ok 1 - The object isa Queue
ok 2 - The object isa Queue
ok 3 - empty queue
ok 4 - first item
ok 5 - second item
ok 6 - The object isa Queue
ok 7 - The object isa Queue
ok 8 - queue is now larger
ok 9 - The object isa Queue
ok 10 - The object isa Queue
ok 11 - an empty queue
ok 12 - a queue with some elements
ok
All tests successful.
Files=1, Tests=12,  0 wallclock secs ( 0.08 cusr +  0.00 csys =  0.08 CPU)