Databases
StackOverflow2013
Postsdbo
POFastest CSV Parser in Perl
 

Note that there are some explanatory texts on larger screens.

plurals
  1. POFastest CSV Parser in Perl
    primarykey
    data
    text
    <p>I am creating a subroutine that:</p> <p>(1) Parses a CSV file;</p> <p>(2) And checks if all the rows in that file have the expected number of columns. It croaks if the number of columns is invalid.</p> <p>When the number of rows is ranging from thousands to millions, <strong>what do you think is the <em>most efficient</em> way to do it?</strong></p> <p>Right now, I'm trying out these implementations.</p> <p><strong>(1) Basic file parser</strong></p> <pre><code>open my $in_fh, '&lt;', $file or croak "Cannot open '$file': $OS_ERROR"; my $row_no = 0; while ( my $row = &lt;$in_fh&gt; ) { my @values = split (q{,}, $row); ++$row_no; if ( scalar @values &lt; $min_cols_no ) { croak "Invalid file format. File '$file' does not have '$min_cols_no' columns in line '$row_no'."; } } close $in_fh or croak "Cannot close '$file': $OS_ERROR"; </code></pre> <p><strong>(2) Using Text::CSV_XS (bind_columns and csv->getline)</strong></p> <pre><code>my $csv = Text::CSV_XS-&gt;new () or croak "Cannot use CSV: " . Text::CSV_XS-&gt;error_diag(); open my $in_fh, '&lt;', $file or croak "Cannot open '$file': $OS_ERROR"; my $row_no = 1; my @cols = @{$csv-&gt;getline($in_fh)}; my $row = {}; $csv-&gt;bind_columns (\@{$row}{@cols}); while ($csv-&gt;getline ($in_fh)) { ++$row_no; if ( scalar keys %$row &lt; $min_cols_no ) { croak "Invalid file format. File '$file' does not have '$min_cols_no' columns in line '$row_no'."; } } $csv-&gt;eof or $csv-&gt;error_diag(); close $in_fh or croak "Cannot close '$file': $OS_ERROR"; </code></pre> <p><strong>(3) Using Text::CSV_XS (csv->parse)</strong></p> <pre><code>my $csv = Text::CSV_XS-&gt;new() or croak "Cannot use CSV: " . Text::CSV_XS-&gt;error_diag(); open my $in_fh, '&lt;', $file or croak "Cannot open '$file': $OS_ERROR"; my $row_no = 0; while ( &lt;$in_fh&gt; ) { $csv-&gt;parse($_); ++$row_no; if ( scalar $csv-&gt;fields &lt; $min_cols_no ) { croak "Invalid file format. File '$file' does not have '$min_cols_no' columns in line '$row_no'."; } } $csv-&gt;eof or $csv-&gt;error_diag(); close $in_fh or croak "Cannot close '$file': $OS_ERROR"; </code></pre> <p><strong>(4) Using Parse::CSV</strong></p> <pre><code>use Parse::CSV; my $simple = Parse::CSV-&gt;new( file =&gt; $file ); my $row_no = 0; while ( my $array_ref = $simple-&gt;fetch ) { ++$row_no; if ( scalar @$array_ref &lt; $min_cols_no ) { croak "Invalid file format. File '$file' does not have '$min_cols_no' columns in line '$row_no'."; } } </code></pre> <p>I benchmark-ed them using the Benchmark module.</p> <pre><code>use Benchmark qw(timeit timestr timediff :hireswallclock); </code></pre> <p>And these are the <strong>numbers (in seconds)</strong> that I got:</p> <p><strong>1,000 lines of file:</strong></p> <blockquote> <p>Implementation 1: 0.0016</p> <p>Implementation 2: 0.0025</p> <p>Implementation 3: 0.0050</p> <p>Implementation 4: 0.0097</p> </blockquote> <p><strong>10,000 lines of file:</strong></p> <blockquote> <p>Implementation 1: 0.0204</p> <p>Implementation 2: 0.0244</p> <p>Implementation 3: 0.0523</p> <p>Implementation 4: 0.1050</p> </blockquote> <p><strong>1,500,000 lines of file:</strong></p> <blockquote> <p>Implementation 1: 1.8697</p> <p>Implementation 2: 3.1913</p> <p>Implementation 3: 7.8475</p> <p>Implementation 4: 15.6274</p> </blockquote> <p>Given these numbers, I would conclude that the simple parser is the fastest but from what I have read from different sources, Text::CSV_XS should be the fastest. </p> <p>Will someone enlighten me on this? Is there something wrong with how I used the modules? Thanks a lot for your help!</p>
    singulars
    1. PO
      singulars
      1. PTAnswer
    1. This table or related slice is empty.
    1. PTQuestion
    1. This table or related slice is empty.
    1. USCarlisle18
    plurals
    1. This table or related slice is empty.
    1. PL
      singulars
      1. LTLinked
    1. This table or related slice is empty.
    1. PO
      singulars
      1. PTAnswer
    2. PO
      singulars
      1. PTAnswer
    3. PO
      singulars
      1. PTAnswer
    1. VO
      singulars
      1. POFastest CSV Parser in Perl
      1. This table or related slice is empty.
      1. VTUpMod
    2. VO
      singulars
      1. POFastest CSV Parser in Perl
      1. This table or related slice is empty.
      1. VTUpMod
    3. VO
      singulars
      1. POFastest CSV Parser in Perl
      1. This table or related slice is empty.
      1. VTUpMod
    1. COWhy is speed so important to you? Are you sure that it is the slow spot of your code? Is your program speed even a problem? Are you sure you're not prematurely optimizing? If, for example, your program spends 90% of the time writing to the database, and 10% reading from the CSV file, then finding the fastest CSV parser is not the best way to spend your time.
      singulars
      1. POFastest CSV Parser in Perl
      1. USAndy Lester
    2. CO`'"That's the thing about strings", he said, "you can include commas, periods, and all other kinds of stuff in quotes!"'`
      singulars
      1. POFastest CSV Parser in Perl
      1. This table or related slice is empty.
    3. CO@AndyLester, speed is so important since this is an added functionality or step to the current workflow of almost all the systems that we have, and we don't want to add any additional time if possible. In this case, I would really like to learn from the experts if I was doing it the right way or not. I would also agree with you. It really makes a lot more sense to optimize the writing to the database since most of the time, the bottleneck happens there. FYI, we're using freebcp for inserting. Thanks again, Andy!
      singulars
      1. POFastest CSV Parser in Perl
      1. USCarlisle18
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload