but should it be auto-exported?

Put it in @EXPORT_OK not in @EXPORT. Don't encourage non-self-documenting imports. It isn't a burden to type use Text::CSV_XS 'csv2list'; and the clarity provided to the next person reading the code is well worth it.

It isn't a burden to type use Text::CSV_XS 'csv2list';

Except for when using one liners. This:

perl -MText::CSV -E"csv2list(...)"
[download]

Is nicer than:

perl -MText::CSV=cvs2list -E"csv2list(...)"
[download]

(But then, I'd name the function just csv(). )

---

With the rise and rise of 'Social' network sites: 'Computers are making people easier to use everyday'

Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.

"Science is about questioning the status quo. Questioning authority".

In the absence of evidence, opinion is indistinguishable from prejudice.

See App::CCSV for oneliners ... although it could use some help with interface (it should come with existing configs)

Some assorted, unorganized feedback follows:

I'm all for a versatile, easy-to-use function. But please don't call it csv2list when it doesn't actually return a list.

There's already enough array vs. list confusion without you contributing even more to it :-).

csv2array is only slightly longer.

It would be really awesome if there would be corresponding array2csv function that accepts mostly the same arguments, and round-trip the data back to original CSV (or as close as possible).

I would also love to have an option to have the first row treated as table headers (and thust column names for the hash variant), which would also imply that there must be some option to force the return format that isn't passing in a headers array reference.

I am already converted to csv. I like your quest for the reverse trip. I have to ponder a bit if it can use the same function or if there really need to be two.

headers => "auto" is exactly what you ask for in the last paragraph.


Enjoy, Have FUN! H.Merijn

Tie::CSV_File seems like a nice abstraction on CSV. The only thing lacking is it would be nice to be able to tell it that the top CSV row specifies column names, or to be able to supply a list of column names, and then get an AoH instead of the default AoA.

Tie::CSV_File fills a niche, but does not lower the threshold for beginners. I even think that using tie is way outside of the problem I am trying to understand: to lower the barrier to use CSV *safely* and reliable without people complaining the API is too difficult.

Using tie, I personally prefer to use Tie::Hash::DBD in combination with DBD::CSV, as I can then transparently switch to DBD::SQLite, DBD::Pg or any other relational database when CSV turns out to be too slow.


Enjoy, Have FUN! H.Merijn

My only question when reading on the CB what what happens with duplicate keys. Perhaps a non-default mechanism to allow for it to emit an HoAoH. LA's case assumes (or ensures, take your pick) no duplicates, so it is a non-issue for her. Update: Quite right Tux. I skipped over the map line and took it as what was returned.

My only other issue would be with the name. csv2list implies that it is returning a list. It is returning a hash (well, hashref), not a list.

No, it returns an arrayref: a reference to a list of either anonymous hashes OR a list of anonymous arrays.

And - I agree with tye and BrowserUK - if it not exported by default, the simple name csv would do quite well.

As it is a list, duplicate keys are not a problem, unless you want a list of hashrefs and the fields contain duplicate names. In that case, the function provides an attribute to pass the field names yourself.


Enjoy, Have FUN! H.Merijn

Absolutely correct. E_READ_ERR. My post updated.

--MidLifeXis

Thanks for all the feedback so far. What I have created so far is this:

I'd rather like a function like this:

my $CSV = openCsv (file => "file.csv", headers => "auto");
while (my $row = <$CSV>) {
  print "$row->{Foo} has $row->{Bar} bar(s)\n";
}
[download]

There is nothing preventing READLINE() of a tied handle from returning an array or hash reference.

I understand the request, but I see this syntax as too conflicting with the existing syntax.

IF a function like that is to return an object of any kind, it would be a Text::CSV_XS object, more like this:

my $csv = openCsv (file => "file.csv", headers => "auto");
while (my $row = $csv->getline) { # note the missing $io
    say $row->{Foo};
    }
[download]

This fits nicely in the rest of the code and documentation and I will consider something like this (no promises).


Enjoy, Have FUN! H.Merijn

I have just uploaded Text-CSV_XS-1.04, which has defined the new functionality like this:

csv

This function is not exported by default and should be explicitly requested:

This is the first draft. This function will stay, but the arguments might change based on user feedback: esp. the headers attribute is not complete. The basics will stay.

This is an high-level function that aims at simple interfaces. It can be used to read/parse a CSV file or stream (the default behavior) or to produce a file or write to a stream (define the out attribute). It returns an array reference on parsing (or undef on fail) or the numeric value of "error_diag" on writing. When this function fails you can get to the error using the class call to "error_diag"

 my $aoa = csv (in => "test.csv") or
     die Text::CSV_XS->error_diag;
[download]

This function takes the arguments as key-value pairs. It can be passed as a list or as an anonymous hash:

 my $aoa = csv (  in => "test.csv", sep_char => ";");
 my $aoh = csv ({ in => $fh, headers => "auto" });
[download]

The arguments passed consist of two parts: the arguments to "csv" itself and the optional attributes to the CSV object used inside the function as enumerated and explained in "new".

If not overridden, the default options used for CSV are

These options are always set and cannot be altered

in

Used to specify the source. in can be a file name (e.g. "file.csv"), which will be opened for reading and closed when finished, a file handle (e.g. $fh or FH), a reference to a glob (e.g. \*ARGV), or - when your version of perl is not archaic - the glob itself (e.g. *STDIN).

When used with "out", it should be a reference to a CSV structure (AoA or AoH).

 my $aoa = csv (in => "file.csv");

 open my $fh, "<", "file.csv";
 my $aoa = csv (in => $fh);

 my $csv = [ [qw( Foo Bar )], [ 1, 2 ], [ 2, 3 ]];
 my $err = csv (in => $csv, out => "file.csv");
[download]

out

In output mode, the default CSV options when producing CSV are

The "fragment" attribute is ignored in output mode.

out can be a file name (e.g. "file.csv"), which will be opened for writing and closed when finished, a file handle (e.g. $fh or FH), a reference to a glob (e.g. \*STDOUT), or - when your version of perl is not archaic - the glob itself (e.g. *STDOUT).

encoding

If passed, it should be an encoding accepted by the :encoding() option to open. There is no default value. This attribute does not work in perl 5.6.x.

headers

If this attribute is not given, the default behavior is to produce an array of arrays.

If headers is given, it should be either an anonymous list of column names or a flag: auto or skip. When skip is used, the header will not be included in the output.

If auto is used, the first line of the CSV source will be read as the list of field headers and used to produce an array of hashes.

If headers is an anonymous list, it will be used instead

 my $aoh = csv (in => $fh, headers => [qw( Foo Bar )]);
 csv (in => $aoa, out => $fh, headers => [qw( code description price }
+]);
[download]

fragment

Only output the fragment as defined in the "fragment" method. This attribute is ignored when generating CSV. See "out".

Combining all of them could give something like

 use Text::CSV_XS qw( csv );
 my $aoh = csv (
     in       => "test.txt",
     encoding => "utf-8",
     headers  => "auto",
     sep_char => "|",
     fragment => "row=3;6-9;15-*",
     );
 say $aoh->[15]{Foo};
[download]

Thanks for all the feedback I got!