use JSON;
use autodie;
use Time::HiRes qw(time);
+use Encode;
sub make_basedir {
my $path = shift;
<title>%s</title>
<link rel="stylesheet" type="text/css" href="/BookReader/BookReader.css"/>
- <script type="text/javascript" src="http://www.archive.org/includes/jquery-1.4.2.min.js"></script>
- <script type="text/javascript" src="http://www.archive.org/bookreader/jquery-ui-1.8.5.custom.min.js"></script>
+ <script type="text/javascript" src="http://archive.org/includes/jquery-1.4.2.min.js"></script>
+ <script type="text/javascript" src="http://archive.org/bookreader/jquery-ui-1.8.5.custom.min.js"></script>
- <script type="text/javascript" src="http://www.archive.org/bookreader/dragscrollable.js"></script>
- <script type="text/javascript" src="http://www.archive.org/bookreader/jquery.colorbox-min.js"></script>
- <script type="text/javascript" src="http://www.archive.org/bookreader/jquery.ui.ipad.js"></script>
- <script type="text/javascript" src="http://www.archive.org/bookreader/jquery.bt.min.js"></script>
+ <script type="text/javascript" src="http://archive.org/bookreader/dragscrollable.js"></script>
+ <script type="text/javascript" src="http://archive.org/bookreader/jquery.colorbox-min.js"></script>
+ <script type="text/javascript" src="http://archive.org/bookreader/jquery.ui.ipad.js"></script>
+ <script type="text/javascript" src="http://archive.org/bookreader/jquery.bt.min.js"></script>
<script type="text/javascript" src="/BookReader/BookReader.js"></script>
// Return the width of a given page. Here we assume all images are 800 pixels wide
br.getPageWidth = function(index) {
if ( ! pages[index] ) return;
- return parseInt( pages[index][1] );
+ return parseInt( pages[index][1] );
}
// Return the height of a given page. Here we assume all images are 1200 pixels high
br.getPageHeight = function(index) {
if ( ! pages[index] ) return;
- return parseInt( pages[index][2] );
+ return parseInt( pages[index][2] );
}
// We load the images from archive.org -- you can modify this function to retrieve images
// using a different URL structure
br.getPageURI = function(index, reduce, rotate) {
if ( ! pages[index] ) return;
- // reduce and rotate are ignored in this simple implementation, but we
- // could e.g. look at reduce and load images from a different directory
- // or pass the information to an image server
- var url = pages[index][0] + '?reduce='+reduce;
- console.debug('getPageURI', index, reduce, rotate, url);
- return url;
+ // reduce and rotate are ignored in this simple implementation, but we
+ // could e.g. look at reduce and load images from a different directory
+ // or pass the information to an image server
+ var r = 1 << ( Math.ceil(reduce).toString(2).length - 1 ); // reduce to nearest higher pow 2
+ var url = pages[index][0] + '?reduce='+r;
+ console.debug('getPageURI', index, reduce, r, rotate, url);
+ return url;
}
// Return which side, left or right, that a given page should be displayed on
br.getPageSide = function(index) {
- if (0 == (index & 0x1)) {
- return 'R';
- } else {
- return 'L';
- }
+ if (0 == (index & 0x1)) {
+ return 'R';
+ } else {
+ return 'L';
+ }
}
// This function returns the left and right indices for the user-visible
// spread that contains the given index. The return values may be
// null if there is no facing page or the index is invalid.
br.getSpreadIndices = function(pindex) {
- var spreadIndices = [null, null];
- if ('rl' == this.pageProgression) {
- // Right to Left
- if (this.getPageSide(pindex) == 'R') {
- spreadIndices[1] = pindex;
- spreadIndices[0] = pindex + 1;
- } else {
- // Given index was LHS
- spreadIndices[0] = pindex;
- spreadIndices[1] = pindex - 1;
- }
- } else {
- // Left to right
- if (this.getPageSide(pindex) == 'L') {
- spreadIndices[0] = pindex;
- spreadIndices[1] = pindex + 1;
- } else {
- // Given index was RHS
- spreadIndices[1] = pindex;
- spreadIndices[0] = pindex - 1;
- }
- }
-
- return spreadIndices;
+ var spreadIndices = [null, null];
+ if ('rl' == this.pageProgression) {
+ // Right to Left
+ if (this.getPageSide(pindex) == 'R') {
+ spreadIndices[1] = pindex;
+ spreadIndices[0] = pindex + 1;
+ } else {
+ // Given index was LHS
+ spreadIndices[0] = pindex;
+ spreadIndices[1] = pindex - 1;
+ }
+ } else {
+ // Left to right
+ if (this.getPageSide(pindex) == 'L') {
+ spreadIndices[0] = pindex;
+ spreadIndices[1] = pindex + 1;
+ } else {
+ // Given index was RHS
+ spreadIndices[1] = pindex;
+ spreadIndices[0] = pindex - 1;
+ }
+ }
+
+ return spreadIndices;
}
// For a given "accessible page index" return the page number in the book.
// For example, index 5 might correspond to "Page 1" if there is front matter such
// as a title page and table of contents.
br.getPageNum = function(index) {
- return index+1;
+ return index+1;
}
// Total number of leafs
br.imagesBaseURL = '/BookReader/images/';
br.getEmbedCode = function(frameWidth, frameHeight, viewParams) {
- return "Embed code not supported in bookreader demo.";
+ return "Embed code not supported in bookreader demo.";
}
// Let's go!
PAGE
sub should_handle {
- my($self, $file) = @_;
- return -d $file || -f $file;
+ my($self, $file) = @_;
+ return -d $file || -f $file;
}
sub return_dir_redirect {
- my ($self, $env) = @_;
- my $uri = Plack::Request->new($env)->uri;
- return [ 301,
- [
- 'Location' => $uri . '/',
- 'Content-Type' => 'text/plain',
- 'Content-Length' => 8,
- ],
- [ 'Redirect' ],
- ];
+ my ($self, $env) = @_;
+ my $uri = Plack::Request->new($env)->uri;
+ return [ 301,
+ [
+ 'Location' => $uri . '/',
+ 'Content-Type' => 'text/plain',
+ 'Content-Length' => 8,
+ ],
+ [ 'Redirect' ],
+ ];
}
sub convert { gm('convert',@_) }
return $image;
}
+sub render_pdf_page {
+ my ( $pdf, $page, $path ) = @_;
+ my $t = time();
+
+ warn "# pdftocairo $pdf\n";
+ system('pdftocairo', '-jpeg', '-f', $page, '-l', $page, $pdf, $path);
+
+ my $image = sprintf( '%s-%03d.jpg', $path, $page );
+
+ die "can't find $image: $!" unless -r $image;
+
+ $t = time() - $t;
+ warn sprintf("## page: %d in %.2f s for %s\n", $page, $t, $image);
+ return $image;
+}
+
sub serve_path {
- my($self, $env, $path, $fullpath) = @_;
+ my($self, $env, $path, $fullpath) = @_;
my $req = Plack::Request->new($env);
- my $dir_url = $env->{SCRIPT_NAME} . $env->{PATH_INFO};
+ my $dir_url = $env->{SCRIPT_NAME} . $env->{PATH_INFO};
my @files = ();
my @page_files;
if ( -f $path && $path =~ s{/([^/]+\.pdf)$}{} ) {
push @page_files, $1;
warn "# single pdf: $path / $1\n";
- } elsif (-f $path ) {
+ } elsif ( -f $path ) {
if ( my $reduce = $req->param('reduce') ) {
$reduce = int($reduce); # BookReader javascript somethimes returns float
convert( '-scale', ( 100 / $reduce ) .'%', $path => $cache_path );
}
- return $self->SUPER::serve_path($env, $cache_path, $fullpath);
+ return $self->SUPER::serve_path($env, $cache_path, $fullpath);
}
- return $self->SUPER::serve_path($env, $path, $fullpath);
- } elsif ( -d $path ) {
+ return $self->SUPER::serve_path($env, $path, $fullpath);
+ } elsif ( -d $path ) {
if ($dir_url !~ m{/$}) {
return $self->return_dir_redirect($env);
warn "# page_files = ",dump( @page_files );
}
- my $dir = Plack::Util::encode_html( $env->{PATH_INFO} );
+ my $dir = Plack::Util::encode_html( $env->{PATH_INFO} );
my $page = 'empty';
if ( $req->param('bookreader') ) {
- my $pages; # []
- my $pages_path = "cache/$dir_url/bookreader.json";
+ my $pages; # []:
+ my $pages_path = "meta/$dir_url/bookreader.json";
if ( -e $pages_path ) {
$pages = decode_json read_file $pages_path;
} else {
if ( $page =~ m/\.pdf$/ ) {
die "$path/$page: $!" unless -r "$path/$page";
- my $info = `pdfinfo $path/$page`;
+ my $info = `pdfinfo "$path/$page"`;
warn "# pdfinfo $path/$page\n$info\n";
my $pdf_pages = $1 if ( $info =~ m/Pages:\s*(\d+)/s );
die "can't find number of pages for $path/$page in:\n$pdf_pages\n" unless $pdf_pages;
+ my $cache_path = "cache/$dir_url/$page";
+ my $txt = "$cache_path.txt";
+ make_basedir $txt;
+ system('pdftotext', "$path/$page", $txt);
+ warn "# pdftotext $txt ", -s $txt, " bytes\n";
+
+ my $is_bitmap = -s $txt == $pdf_pages;
+
+ $pdf_pages = $ENV{MAX_PAGES} if defined $ENV{MAX_PAGES} && $pdf_pages > $ENV{MAX_PAGES}; # FIXME
- $pdf_pages = $ENV{MAX_PAGES} if $pdf_pages > $ENV{MAX_PAGES}; # FIXME
+ warn "DIAG: bitmap:$is_bitmap pdf_pages:$pdf_pages\n";
foreach my $nr ( 1 .. $pdf_pages ) {
- my $cache_path = "cache/$dir_url/$page";
- my $page_url = convert_pdf_page( "$path/$page", $nr, $cache_path . '.' . $nr );
+ my $page_url = $is_bitmap
+ ? convert_pdf_page( "$path/$page", $nr, "$cache_path.$nr" )
+ : render_pdf_page( "$path/$page", $nr, "$cache_path" )
+ ;
warn "## ping $page_url\n";
my ( $w, $h, $size, $format ) = $image->ping($page_url);
warn "## image size $w*$h $size $format $page_url\n";
- push @$pages, [ "/$page_url", $w, $h ] if $w && $h;
+ my $url = decode('utf-8',"/$page_url");
+ push @$pages, [ $url, $w, $h ] if $w && $h;
}
} else {
die "$path/$page: $!" unless -r "$path/$page";
my ( $w, $h, $size, $format ) = $image->ping("$path/$page");
warn "# image size $w*$h $size $format $path/$page\n";
- push @$pages, [ "$dir_url/$page", $w, $h ] if $w && $h;
+ my $url = decode('utf-8',"$dir_url/$page");
+ push @$pages, [ $url, $w, $h ] if $w && $h;
}
}
make_basedir $pages_path;
- write_file $pages_path => encode_json( $pages );
+ write_file $pages_path, => encode_json( $pages );
warn "# created $pages_path ", -s $pages_path, " bytes\n";
}
warn "# pages = ",dump($pages);
- $page = sprintf $reader_page, $dir, encode_json( $pages ), $dir, '..';
+ $page = sprintf $reader_page, $dir, encode_json( $pages ), $dir, $dir =~ m/\/$/ ? '..' : '.';
} else {
}
- return [ 200, ['Content-Type' => 'text/html; charset=utf-8'], [ $page ] ];
+ return [ 200, ['Content-Type' => 'text/html; charset=utf-8'], [ $page ] ];
}
1;
projects / bookreader.git / blobdiff