source: gsdl/trunk/perllib/plugins/TEXTPlug.pm@ 15119

###########################################################################
#
# TEXTPlug.pm -- simple text plugin
# A component of the Greenstone digital library software
# from the New Zealand Digital Library Project at the 
# University of Waikato, New Zealand.
#
# Copyright (C) 1999 New Zealand Digital Library Project
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
#
###########################################################################

# creates simple single-level document. Adds Title metadata 
# of first line of text (up to 100 characters long).

# 12/05/02 Added usage datastructure - John Thompson

package TEXTPlug;

use BasPlug;

use strict;
no strict 'refs'; # allow filehandles to be variables and viceversa

sub BEGIN {
    @TEXTPlug::ISA = ('BasPlug');
}

my $arguments =
    [ { 'name' => "process_exp",
    'desc' => "{BasPlug.process_exp}",
    'type' => "regexp",
    'deft' => &get_default_process_exp(),
    'reqd' => "no" } ,
      { 'name' => "title_sub",
    'desc' => "{TEXTPlug.title_sub}",
    'type' => "regexp",
    'deft' => "",
    'reqd' => "no" } ];

my $options = { 'name'     => "TEXTPlug",
        'desc'     => "{TEXTPlug.desc}",
        'abstract' => "no",
        'inherits' => "yes",
        'srcreplaceable' => "yes", # Source docs in regular txt format can be replaced with GS-generated html
        'args'     => $arguments };


sub new {
    my ($class) = shift (@_);
    my ($pluginlist,$inputargs,$hashArgOptLists) = @_;
    push(@$pluginlist, $class);

    if(defined $arguments){ push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});}
    if(defined $options) { push(@{$hashArgOptLists->{"OptList"}},$options)};

    my $self = new BasPlug($pluginlist, $inputargs, $hashArgOptLists);

    return bless $self, $class;
}

sub get_default_process_exp {
    my $self = shift (@_);

    return q^(?i)\.te?xt$^;
}

# do plugin specific processing of doc_obj
sub process {
    my $self = shift (@_);
    my ($textref, $pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli) = @_;
    my $outhandle = $self->{'outhandle'};

    print STDERR "<Processing n='$file' p='TEXTPlug'>\n" if ($gli);
    print $outhandle "TEXTPlug: processing $file\n" 
    if $self->{'verbosity'} > 1;
    
    my $cursection = $doc_obj->get_top_section();
    # get title metadata
    # (don't need to get title if it has been passed
    # in from another plugin)
    if (!defined $metadata->{'Title'}) {
    my ($title) = $$textref;
    $title =~ /^\s+/s;
    if (defined $self->{'title_sub'} &&
           $self->{'title_sub'}) {$title =~ s/$self->{'title_sub'}//;}
    $title =~ /^\s*([^\n]*)/s; $title=$1;
    if (length($title) > 100) {
        $title = substr ($title, 0, 100) . "...";
    }
        $title =~ s/\[/&#91;/g;
        $title =~ s/\[/&#93;/g;
        $title =~ s/\</&#60;/g;
        $title =~ s/\>/&#62;/g;
    $doc_obj->add_utf8_metadata ($cursection, "Title", $title);
    }
    # Add FileFormat metadata
    $doc_obj->add_metadata($cursection, "FileFormat", "TEXT");
    

    # we need to escape the escape character, or else mg will convert into
    # eg literal newlines, instead of leaving the text as '\n'
    $$textref =~ s/\\/\\\\/g; # macro language
    $$textref =~ s/_/\\_/g; # macro language
    $$textref =~ s/</&lt;/g;
    $$textref =~ s/>/&gt;/g;

    # insert preformat tags and add text to document object
    $doc_obj->add_utf8_text($cursection, "<pre>\n$$textref\n</pre>");

    return 1;
}

# replace_srcdoc_with_html.pl requires all subroutines that support src_replaceable
# to contain a method called tmp_area_convert_file - this is indeed the case with all
# Perl modules that are subclasses of ConvertToPlug.pm, but as we want TEXTPlug to also
# be srcreplaceable and because TEXTPlug does not inherit from ConvertToPlug.pm, we have
# this ugly solution: same subroutine name.
# Despite the subroutine name, this method does not in fact create the output html file in
# some tmp folder. Instead, it creates the html file in the same folder as the input_filename
# and writes the contents as html paragraphs nested inside <html><head></head><body></body></html>
# It also sets the encoding of the html document created to UTF-8 in the head's meta tag.
# Note: doesn't seem to be able to cope with <br /> and <meta ... /> -> slashes are a problem.
# As a consequence, we resorted to making it not proper xhtml but just regular html.\
# The output file's name will be utf8 AND might not be the same as the input file's name
# (for instance, the output filename may have a number appended to it if there is already an html
# file in the input folder with the same name).  
sub tmp_area_convert_file {
    my $self = shift (@_);
    my ($output_ext, $input_filename) = @_;
    
    #my $outhandle = $self->{'outhandle'};
    #my $failhandle = $self->{'failhandle'};
    #my $convert_to_ext = $self->{'convert_to_ext'};
    
    # derive output filename from input filename
    my ($tailname, $dirname, $suffix)
    = &File::Basename::fileparse($input_filename, "\\.[^\\.]+\$");

    # convert to utf-8 otherwise we have problems with the doc.xml file
    # later on
    &unicode::ensure_utf8(\$tailname); # TO DO: does this change the filename or not?

    #my $output_filename = $tailname$output_ext;#output_ext has to be html!
    my $output_filename = &util::filename_cat($dirname, $tailname.".html");
    
    # read contents of text file line by line into an array
    # create an HTML file from the text file
    # Recreate the original file for writing the updated contents
    unless(open(TEXT, "<$input_filename")) {  # open it as a new file for writing
    print STDERR "TEXTPlug.pm: Unable to open and read from $input_filename for converting to html...ERROR\n";
    return 0;
    }
       
    my @lines = ();
    my $line;
    my $newpara = 1; # true whenever we're going to start a new para
    
    while ($line=<TEXT>) { 
    # replace < and > with their character encodings
    $line =~ s/</&lt;/g;
    $line =~ s/>/.*&gt;/g;
    
    if ($line =~ /^\s*$/) { # line is empty
        if(!$newpara) {
        push(@lines, "</p>".$line); # end of a paragraph, leave empty line in there
        $newpara = 1;
        } # If it is a new paragraph, we do nothing
    } else { # a line with text
        if($newpara) {
        push(@lines, "<p>".$line); # start a new paragraph
        $newpara = 0;
        } else { # text-line is not a new paragraph, but just a new line
        push(@lines, "<br>\n".$line); # put a break. It doesn't seem to accept <br />
        }
    }
    }
    close TEXT; 

    # we've come to the last line of input file, make sure that the text ends on </p>
    # get the last line and check it has a </p> at the end:
    $line = pop(@lines);
    my $endpara = "</p>";
    unless($line =~ m/$endpara$/ ) { # if no </p> at end of last para, then append it
    $line = $line.$endpara;
    } 
    push(@lines, $line); # either way, put the last line back

    # write everything into the html file along with html start and end parts:
    my $count = "1";
    # create a sensible html file, don't overwrite pre-existing files of the same name
    # just concatenate a number instead
    while(-e "$output_filename") {
    $output_filename = &util::filename_cat($dirname, $tailname.$count.".html");
    $count++;
    }
    # try creating this new file writing and try opening it for writing, else exit with error value
    unless(open(HTML, ">$output_filename")) {  # open the new html file for writing
    print STDERR "TEXTPlug.pm: Unable to create $output_filename for writing $tailname$suffix txt converted to html...ERROR\n";
    return 0;
    }

    # append html start and end sections to the txt contents of the input file
    print HTML "<html>\n<head>\n<title>$tailname</title>\n";
    # we'll set the content to UTF-8 encoding
    print HTML "<meta http-equiv=\"Content-Type\" content=\"text/html; charset=UTF-8\">\n"; # /> not accepted
    print HTML "</head>\n<body>\n";
    # go through each line and write it to the file:
    foreach $line (@lines) {
    print HTML $line;
    }
    print HTML "\n</body></html>\n"; 
    close HTML;
 
    return $output_filename;
}

1;