Ray Smith has asked for the wisdom of the Perl Monks concerning the following question:

I have used Win32::OLE to do some parsing of MS Word files. I've been unsuccessful in my attempt to the paragraph numbers from the source file.

The following is a likely example in Visual Basic, but I've not been able to successfully translate it into Perl.

If Selection.Paragraphs(1).Range.ListParagraphs.Count = 1 Then
   MsgBox Selection.Paragraphs(1).Range.ListFormat.ListLevelNumber
Else
   MsgBox "Not a numbered paragraph"
End If

[download]

The following is a snipit of my working OLE MS Word access: 

sub doc2pt {
    my (
        $docfile,
        $docpt,     # Converted text file
        ) = @_;
    
    require Win32::OLE;
    require Win32::OLE::Enum;
                                    # NOTE: Win32::OLE appears to need
+ abs path
    my $document = Win32::OLE -> GetObject(abs_path($docfile));
    die "Can't GetObject($docfile) $!\n" if !defined($document);
    print "Extracting Text ...\n";
    open DOCPT, ">$docpt" or die "Can't open docpt file $docpt: $!";

    my $paragraphs = $document->Paragraphs();
    my $enumerate = new Win32::OLE::Enum($paragraphs);
    while (defined($paragraph = $enumerate->Next())) {
        my $style = $paragraph->{Style}->{NameLocal};
        print DOCPT "style==>$style\n";
###        my $range = $paragraph->{Range};
###        my $count = $range->ListParagraph();
###        if (defined($count)) {
###            my $paranum = $range->ListFormatListLevelNumber();
###            print DOCPT "paranum==>$paranum\n";
###        }

        my $text = $paragraph->{Range}->{Text};
        $text =~ s/[\n\r]//g;
        $text =~ s/\x0b/\n/g;
        print DOCPT "text==>$text\n";
    }
    close DOCPT;
}

[download]

I would greatly appreciate any suggestions on how to solve this problem.

Replies are listed 'Best First'.
Re: Using OLE to find MS Word paragraph numbers
by Util (Priest) on Oct 20, 2011 at 14:13 UTC

    1. Add my before $paragraph in while(defined($paragraph=$enumerate->Next())).
    2. Typo: Change ListParagraph to ListParagraphs.
    3. Typo: Change ListFormatListLevelNumber to ListFormat->ListLevelNumber.
    4. You might need ListFormat->ListValue instead of (or in addition to) ListFormat->ListLevelNumber.
    5. To match your VB code, check for $count->{Count} == 1 inside the if (defined $count) {...} block.

    Working, tested code:

    #!/usr/bin/perl
use strict;
use warnings;
use Cwd qw( abs_path );
use Data::Dumper;

doc2pt( "C:/a/perls/pm/PM_932635_data.doc" );

sub doc2pt {
    die if @_ != 1;
    my ( $doc_path ) = @_;
    
    ( my $out_path = $doc_path ) =~ s{\.doc$}{.out}
        or die "doc_path '$doc_path' does not end in '.doc'";

    require Win32::OLE;
    require Win32::OLE::Enum;

    # NOTE: Win32::OLE appears to need abs path
    my $abs_doc_path = abs_path($doc_path);

    my $document = Win32::OLE->GetObject($abs_doc_path)
        or die "Can't GetObject($abs_doc_path) $!\n";
    print "Extracting Text ...\n";
    open my $out_fh, '>', $out_path
        or die "Can't open output file '$out_path': $!";

    my $debug = sub {
        die if @_ != 2;
        my ( $name, $value ) = @_;
        local $Data::Dumper::Useqq = 1;
        local $Data::Dumper::Terse = 1;
        printf {$out_fh} "%-10s ==> %s", $name, Dumper $value;
    };

    my $paragraphs = $document->Paragraphs();
    my $enumerate = Win32::OLE::Enum->new($paragraphs);
    while ( my $paragraph = $enumerate->Next() ) {
        my $style = $paragraph->{Style}->{NameLocal};
        $debug->( style => $style );
        my $range = $paragraph->{Range};
        my $count = $range->ListParagraphs();
        if ( defined $count ) {
            my $real_count = $count->{Count};
            my $paranum = $range->ListFormat->ListLevelNumber();
            my $paraval = $range->ListFormat->ListValue();
            $debug->( real_count => $real_count );
            $debug->( paranum    => $paranum );
            $debug->( paraval    => $paraval );
        }
        my $text = $paragraph->{Range}->{Text};
        $text =~ tr{\n\r}{}d;
        $text =~ tr{\x0b}{\n};
        $debug->( text => $text );
    }
    close $out_fh;
}

    [download]

[reply]
[d/l]
[select]

Back to Seekers of Perl Wisdom