vendor/wikimedia/purtle/src/N3Quoter.php

   1 <?php
   2
   3 namespace Wikimedia\Purtle;
   4
   5 /**
   6  * Helper class for quoting literals and URIs in N3 output.
   7  * Optionally supports shorthand and prefix resolution.
   8  *
   9  * @license GPL-2.0+
  10  * @author Daniel Kinzler
  11  */
  12 class N3Quoter {
  13
  14         /**
  15          * @var UnicodeEscaper
  16          */
  17         private $escaper = null;
  18
  19         /**
  20          * @param bool $escapeUnicode
  21          */
  22         public function setEscapeUnicode( $escapeUnicode ) {
  23                 $this->escaper = $escapeUnicode ? new UnicodeEscaper() : null;
  24         }
  25
  26         public function escapeIRI( $iri ) {
  27                 //FIXME: apply unicode escaping?!
  28                 return strtr( $iri, [
  29                                 ' ' => '%20',
  30                                 '"' => '%22',
  31                                 '<' => '%3C',
  32                                 '>' => '%3E',
  33                                 '\\' => '%5C',
  34                                 '`' => '%60',
  35                                 '^' => '%5E',
  36                                 '|' => '%7C',
  37                                 '{' => '%7B',
  38                                 '}' => '%7D',
  39                 ] );
  40         }
  41
  42         public function escapeLiteral( $s ) {
  43                 // String escapes. Note that the N3 spec is more restrictive than the Turtle and TR
  44                 // specifications, see <https://www.w3.org/TeamSubmission/n3/#escaping>
  45                 // and <https://www.w3.org/TR/turtle/#string>
  46                 // and <https://www.w3.org/TR/n-triples/#grammar-production-literal>.
  47                 // Allowed escapes according to the N3 spec are:
  48                 // ECHAR        ::=     '\' [tbnrf"'\]
  49                 // The single quote however does not require escaping when used in double quotes.
  50                 $escaped = strtr( $s, [
  51                         "\x00" => '\u0000',
  52                         "\x01" => '\u0001',
  53                         "\x02" => '\u0002',
  54                         "\x03" => '\u0003',
  55                         "\x04" => '\u0004',
  56                         "\x05" => '\u0005',
  57                         "\x06" => '\u0006',
  58                         "\x07" => '\u0007',
  59                         "\x08" => '\b',
  60                         "\x09" => '\t',
  61                         "\x0A" => '\n',
  62                         "\x0B" => '\u000B',
  63                         "\x0C" => '\f',
  64                         "\x0D" => '\r',
  65                         "\x0E" => '\u000E',
  66                         "\x0F" => '\u000F',
  67                         "\x10" => '\u0010',
  68                         "\x11" => '\u0011',
  69                         "\x12" => '\u0012',
  70                         "\x13" => '\u0013',
  71                         "\x14" => '\u0014',
  72                         "\x15" => '\u0015',
  73                         "\x16" => '\u0016',
  74                         "\x17" => '\u0017',
  75                         "\x18" => '\u0018',
  76                         "\x19" => '\u0019',
  77                         "\x1A" => '\u001A',
  78                         "\x1B" => '\u001B',
  79                         "\x1C" => '\u001C',
  80                         "\x1D" => '\u001D',
  81                         "\x1E" => '\u001E',
  82                         "\x1F" => '\u001F',
  83                         '"' => '\"',
  84                         '\\' => '\\\\',
  85                 ] );
  86
  87                 if ( $this->escaper !== null ) {
  88                         $escaped = $this->escaper->escapeString( $escaped );
  89                 }
  90
  91                 return $escaped;
  92         }
  93
  94 }