Addressed UTF-8 encoding issues.

Neither iconv nor mb seem to be able to convert UTF-8 surrogates
into UTF-8. As I understand it, this is an extreme edge case.
Still, the behavior in both cases is that the surrogates are
stripped from the string. We test for that condition, now.

1 files changed, 12 insertions, 120 deletions

diff --git a/src/HTML5/Parser/StringInputStream.php b/src/HTML5/Parser/StringInputStream.php
index 5f41c21..9aa0b73 100644
--- a/src/HTML5/Parser/StringInputStream.php
+++ b/src/HTML5/Parser/StringInputStream.php
@@ -62,17 +62,18 @@ class StringInputStream implements InputStream {
    *
    * @param $data Data to parse
    */
-  public function __construct($data, $encoding = 'UTF-8') {
+  public function __construct($data, $encoding = 'UTF-8', $debug = '') {
 
-    $data = $this->convertToUTF8($data, $encoding);
+    $data = UTF8Utils::convertToUTF8($data, $encoding);
+    if ($debug) fprintf(STDOUT, $debug, $data, strlen($data));
 
     // There is good reason to question whether it makes sense to
     // do this here, since most of these checks are done during
     // parsing, and since this check doesn't actually *do* anything.
-    $e = UTF8Utils::checkForIllegalCodepoints($data);
-    if (!empty($e)) {
-      throw new ParseError("UTF-8 encoding issues: " . implode(', ', $e));
-    }
+    $this->errors = UTF8Utils::checkForIllegalCodepoints($data);
+    //if (!empty($e)) {
+    //  throw new ParseError("UTF-8 encoding issues: " . implode(', ', $e));
+    //}
 
     $data = $this->replaceLinefeeds($data);
 
@@ -82,63 +83,6 @@ class StringInputStream implements InputStream {
   }
 
   /**
-   * Convert data from the given encoding to UTF-8.
-   *
-   * This has not yet been tested with charactersets other than UTF-8. 
-   * It should work with ISO-8859-1/-13 and standard Latin Win charsets.
-   *
-   * @param string $data
-   *   The data to convert.
-   * @param string $encoding
-   *   A valid encoding. Examples: http://www.php.net/manual/en/mbstring.supported-encodings.php
-   */
-  protected function convertToUTF8($data, $encoding = 'UTF-8') {
-    /* Given an encoding, the bytes in the input stream must be
-    converted to Unicode characters for the tokeniser, as
-    described by the rules for that encoding, except that the
-    leading U+FEFF BYTE ORDER MARK character, if any, must not
-    be stripped by the encoding layer (it is stripped by the rule below).
-
-    Bytes or sequences of bytes in the original byte stream that
-    could not be converted to Unicode characters must be converted
-    to U+FFFD REPLACEMENT CHARACTER code points. */
-
-    // XXX currently assuming input data is UTF-8; once we
-    // build encoding detection this will no longer be the case
-    //
-    // We previously had an mbstring implementation here, but that
-    // implementation is heavily non-conforming, so it's been
-    // omitted.
-    if (function_exists('iconv') && $encoding != 'auto') {
-      // iconv has the following behaviors:
-      // - Overlong representations are ignored.
-      // - Beyond Plane 16 is replaced with a lower char.
-      // - Incomplete sequences generate a warning.
-      $data = @iconv($encoding, 'UTF-8//IGNORE', $data);
-    }
-    // MPB: Testing the newer mb_convert_encoding(). This might need
-    // to be removed again.
-    elseif (function_exists('mb_convert_encoding')) {
-      // mb library has the following behaviors:
-      // - UTF-16 surrogates result in FALSE.
-      // - Overlongs and outside Plane 16 result in empty strings.
-      $data = mb_convert_encoding($data, 'UTF-8', $encoding);
-    }
-    else {
-      // we can make a conforming native implementation
-      throw new Exception('Not implemented, please install mbstring or iconv');
-    }
-
-    /* One leading U+FEFF BYTE ORDER MARK character must be
-    ignored if any are present. */
-    if (substr($data, 0, 3) === "\xEF\xBB\xBF") {
-      $data = substr($data, 3);
-    }
-
-    return $data;
-  }
-
-  /**
    * Replace linefeed characters according to the spec.
    */
   protected function replaceLinefeeds($data) {
@@ -158,62 +102,6 @@ class StringInputStream implements InputStream {
   }
 
   /**
-   * Checks for Unicode code points that are not valid in a document.
-   *
-   * This stores a parse error for each error that is found.
-   */
-  protected function checkForIllegalCodepoints($data) {
-    if (!function_exists('preg_match_all')) {
-      throw \Exception('The PCRE library is not loaded or is not available.');
-    }
-
-    /* All U+0000 NULL characters in the input must be replaced
-    by U+FFFD REPLACEMENT CHARACTERs. Any occurrences of such
-    characters is a parse error. */
-    for ($i = 0, $count = substr_count($data, "\0"); $i < $count; $i++) {
-      $this->errors[] = array(
-        'type' => Tokenizer::PARSEERROR,
-        'data' => 'null-character'
-      );
-    }
-
-    /* Any occurrences of any characters in the ranges U+0001 to
-    U+0008, U+000B,  U+000E to U+001F,  U+007F  to U+009F,
-    U+D800 to U+DFFF , U+FDD0 to U+FDEF, and
-    characters U+FFFE, U+FFFF, U+1FFFE, U+1FFFF, U+2FFFE, U+2FFFF,
-    U+3FFFE, U+3FFFF, U+4FFFE, U+4FFFF, U+5FFFE, U+5FFFF, U+6FFFE,
-    U+6FFFF, U+7FFFE, U+7FFFF, U+8FFFE, U+8FFFF, U+9FFFE, U+9FFFF,
-    U+AFFFE, U+AFFFF, U+BFFFE, U+BFFFF, U+CFFFE, U+CFFFF, U+DFFFE,
-    U+DFFFF, U+EFFFE, U+EFFFF, U+FFFFE, U+FFFFF, U+10FFFE, and
-    U+10FFFF are parse errors. (These are all control characters
-    or permanently undefined Unicode characters.) */
-    // Check PCRE is loaded.
-    $count = preg_match_all(
-      '/(?:
-        [\x01-\x08\x0B\x0E-\x1F\x7F] # U+0001 to U+0008, U+000B,  U+000E to U+001F and U+007F
-      |
-        \xC2[\x80-\x9F] # U+0080 to U+009F
-      |
-        \xED(?:\xA0[\x80-\xFF]|[\xA1-\xBE][\x00-\xFF]|\xBF[\x00-\xBF]) # U+D800 to U+DFFFF
-      |
-        \xEF\xB7[\x90-\xAF] # U+FDD0 to U+FDEF
-      |
-        \xEF\xBF[\xBE\xBF] # U+FFFE and U+FFFF
-      |
-        [\xF0-\xF4][\x8F-\xBF]\xBF[\xBE\xBF] # U+nFFFE and U+nFFFF (1 <= n <= 10_{16})
-      )/x',
-      $data,
-      $matches
-    );
-    for ($i = 0; $i < $count; $i++) {
-      $this->errors[] = array(
-        'type' => Tokenizer::PARSEERROR,
-        'data' => 'invalid-codepoint'
-      );
-    }
-  }
-
-  /**
    * Returns the current line that the tokenizer is at.
    */
   public function currentLine() {
@@ -319,6 +207,10 @@ class StringInputStream implements InputStream {
    * end of the file.
    *
    * @note This performs bounds checking
+   *
+   * @return string
+   *   Returns the remaining text. If called when the InputStream is
+   *   already exhausted, it returns an empty string.
    */
   public function remainingChars() {
     if ($this->char < $this->EOF) {
@@ -326,7 +218,7 @@ class StringInputStream implements InputStream {
       $this->char = $this->EOF;
       return $data;
     }
-    return FALSE;
+    return '';//FALSE;
   }
 
   /**