Skip to content

Commit

Permalink
. Optimized regular expressions that remove useless Adobe Postscript …
Browse files Browse the repository at this point in the history 
…instructions and added new ones

Rewrote the CodePointToUtf8() method

. Handled a new way to specify font aliases : /TTx
  • Loading branch information
@christian-vigh
christian-vigh committed Aug 5, 2016
1 parent dfc137b commit d662ada
Showing 1 changed file with 34 additions and 35 deletions.
69 changes: 34 additions & 35 deletions PdfToText.phpclass
View file
Original file line number Diff line number Diff line change
Expand Up @@ -584,6 +584,11 @@
. Character translation results are now buffered, to avoid unnecessary calls to the MapCharacter()
method of the PdfTexterFontTable class.

[Version : 1.2.32] [Date : 2016/08/05] [Author : CV]
. Optimized regular expressions that remove useless Adobe Postscript instructions and added new ones.
. Rewrote the CodePointToUtf8() method.
. Handled a new way to specify font aliases : /TTx.

**************************************************************************************************************/


Expand Down Expand Up @@ -723,7 +728,7 @@ abstract class PdfObjectBase // extends Object
*-------------------------------------------------------------------------------------------------------------*/
protected function IsFontMap ( $object_data )
{
if ( preg_match ( '#<< \s* ( (/F \d+) | (/f-\d+-\d+) | (/[CT]\d+_\d+) ) \s+ .* >>#msx', $object_data ) )
if ( preg_match ( '#<< \s* ( (/F \d+) | (/f-\d+-\d+) | (/[CT]\d+_\d+) | (/TT \d+) ) \s+ .* >>#msx', $object_data ) )
return ( true ) ;
else
return ( false ) ;
Expand Down Expand Up @@ -989,21 +994,16 @@ abstract class PdfObjectBase // extends Object
*-------------------------------------------------------------------------------------------------------------*/
protected function CodePointToUtf8 ( $code )
{
// For simple case ($code <= 0xFFFF), use the mb_convert_encoding function which works well
if ( $code <= 0xFFFF )
if ( $code )
{
$entity = '&#' . sprintf ( "%d", $code ) . ';' ;
$result = '' ;

return ( mb_convert_encoding ( $entity, 'UTF-8', 'HTML-ENTITIES' ) ) ;
}
// For more complex cases, use a homemade solution
// (source : http://stackoverflow.com/questions/1805802/php-convert-unicode-codepoint-to-utf-8)
else if ( $code < 0x1FFFFF )
{
$result = chr ( ( $code >> 18 ) + 240 ) .
chr ( ( ( $code >> 12 ) & 63 ) + 128 ) .
chr ( ( ( $code >> 6 ) & 63 ) + 128 ) .
chr ( ( ( $code & 63 ) + 128 ) ) ;
while ( $code )
{
$entity = '&#x' . sprintf ( '%x', ( $code & 0xFFFF ) ) . ';' ;
$result = mb_convert_encoding ( $entity, 'UTF-8', 'HTML-ENTITIES' ) . $result ;
$code >>= 16 ;
}

return ( $result ) ;
}
Expand All @@ -1030,7 +1030,7 @@ abstract class PdfObjectBase // extends Object
class PdfToText extends PdfObjectBase
{
// Current version of the class
const VERSION = "1.2.29" ;
const VERSION = "1.2.32" ;

// Pdf processing options
const PDFOPT_NONE = 0x0000 ; // No extra option
Expand Down Expand Up @@ -1097,22 +1097,15 @@ class PdfToText extends PdfObjectBase
// %n - Will be replaced with a regex matching a decimal number.
private static $IgnoredInstructionsTemplates =
[
'%n{6} c \s+',
'%n{6} cm \s+',
'%n{4} re \s+',
'%n{2} m \s+',
'%n{2} l \s+',
'%n{4} y \s+',
'%n{4} v \s+',
'%n{3} scn \s+',
'%n{3} SCN \s+',
'%n w \s+',
'%n M \s+',
'\b BT \s+',
'\b ET \s+',
'\b EMC \s+',
'\b \/CS \d+ \s+ \w+',
'\b \/GS \d+ \s+ \w+',
'%n{6} ( (c) | (cm) ) \s+',
'%n{4} ( (re) | (y) | (v) ) \s+',
'%n{3} ( (scn) | (SCN) | (r) | (rg) ) \s+',
'%n{2} ( (m) | (l) ) \s+',
'%n ( (w) | (M) | (g) | (G) | (J) ) \s+',
'\b ( (BDC) | (BT) | (ET) | (EMC) ) \s+',
'\/( (CS \d+) | (GS \d+) | (Fm \d+) | (Im \d+) | (PlacedGraphic) ) \s+ \w+ \s*',
'\s+ [fhnqQSW] \s+',
'\/Span \s* << .*? >> [ \t\r\n>]*'
] ;
// Replacement regular expressions for %something constructs specified in the $IgnoredInstructions array
private static $ReplacementConstructs =
Expand Down Expand Up @@ -2162,7 +2155,7 @@ class PdfToText extends PdfObjectBase
{
echo "\n----------------------------------- TEXT #$object_id (size = " . strlen ( $data ) . " bytes, new size = " . strlen ( $new_data ) . " bytes)\n" ;
echo $data ;
echo "\n----------------------------------- NEWTEXT #$object_id\n" ;
echo "\n----------------------------------- OPTIMIZED TEXT #$object_id\n" ;
echo $new_data ;
}

Expand Down Expand Up @@ -2631,7 +2624,13 @@ class PdfToText extends PdfObjectBase
// "/F" : Specifies the font to be used for the next text output(s), using an indirect id
else if ( ! strncasecmp ( $token, '/F', 2 ) || ! strncasecmp ( $token, '/C', 2 ) || ! strncasecmp ( $token, '/T', 2 ) )
{
$resource = substr ( $token, 2 ) ;
// Special case for /TTx font specifications
if ( $token [2] == 'T' )
$length = 3 ;
else
$length = 2 ;

$resource = substr ( $token, $length ) ;

if ( $resource [0] == '-' )
$resource = substr ( $resource, 1 ) ;
Expand Down Expand Up @@ -3170,7 +3169,7 @@ class PdfTexterFontTable extends PdfObjectBase
// <</C0_0 5 0 R
public function AddFontMap ( $object_id, $object_data )
{
if ( preg_match_all ( '#/F(?P<font> \d+) \s+ (?P<object> \d+)#x', $object_data, $matches ) )
if ( preg_match_all ( '#/ ( (F) || (TT) ) (?P<font> \d+) \s+ (?P<object> \d+)#x', $object_data, $matches ) )
{
for ( $i = 0, $count = count ( $matches [ 'font' ] ) ; $i < $count ; $i ++ )
$this -> FontMap [ $matches [ 'font' ] [$i] ] = $matches [ 'object' ] [$i] ;
Expand All @@ -3180,7 +3179,7 @@ class PdfTexterFontTable extends PdfObjectBase
for ( $i = 0, $count = count ( $matches [ 'font' ] ) ; $i < $count ; $i ++ )
$this -> FontMap [ $matches [ 'font' ] [$i] ] = $matches [ 'object' ] [$i] ;
}
else if ( preg_match_all ( '#/C(?P<font> \d+ _ \d+) \s+ (?P<object> \d+)#x', $object_data, $matches ) )
else if ( preg_match_all ( '#/[CT] (?P<font> \d+ _ \d+) \s+ (?P<object> \d+)#x', $object_data, $matches ) )
{
for ( $i = 0, $count = count ( $matches [ 'font' ] ) ; $i < $count ; $i ++ )
$this -> FontMap [ $matches [ 'font' ] [$i] ] = $matches [ 'object' ] [$i] ;
Expand Down

0 comments on commit d662ada

Please sign in to comment.