Tokenizer::__construct() - Code Metrics - Inspection of "Optimize the token comparison" - Masterminds/html5-php - Measure and Improve Code Quality continuously with Scrutinizer

Completed
Pull Request — master (#154)

by Christophe
created 2018-11-24 12:11 UTC
Tokenizer::__construct() A

↳ Parent: Tokenizer
Complexity

Conditions	1
Paths	1
Size

Total Lines
Duplication

Lines	0
Ratio	0 %
Code Coverage

Tests	5
CRAP Score	1
Importance

Changes
Metric	Value
dl	0
loc	6
ccs	5
cts	5
cp	1
rs	10
c	0
b	0
f	0
cc	1
nc	1
nop	3
crap	1
<?php
namespace Masterminds\HTML5\Parser;

use Masterminds\HTML5\Elements;

/**
 * The HTML5 tokenizer.
 *
 * The tokenizer's role is reading data from the scanner and gathering it into
 * semantic units. From the tokenizer, data is emitted to an event handler,
 * which may (for example) create a DOM tree.
 *
 * The HTML5 specification has a detailed explanation of tokenizing HTML5. We
 * follow that specification to the maximum extent that we can. If you find
 * a discrepancy that is not documented, please file a bug and/or submit a
 * patch.
 *
 * This tokenizer is implemented as a recursive descent parser.
 *
 * Within the API documentation, you may see references to the specific section
 * of the HTML5 spec that the code attempts to reproduce. Example: 8.2.4.1.
 * This refers to section 8.2.4.1 of the HTML5 CR specification.
 *
 * @see http://www.w3.org/TR/2012/CR-html5-20121217/
 */
class Tokenizer
{

    protected $scanner;

    protected $events;

    protected $tok;

    /**
     * Buffer for text.
     */
    protected $text = '';

    // When this goes to false, the parser stops.
    protected $carryOn = true;

    protected $textMode = 0; // TEXTMODE_NORMAL;
    protected $untilTag = null;

    const CONFORMANT_XML = 'xml';
    const CONFORMANT_HTML = 'html';
    protected $mode = self::CONFORMANT_HTML;

    /**
     * Create a new tokenizer.
     *
     * Typically, parsing a document involves creating a new tokenizer, giving
     * it a scanner (input) and an event handler (output), and then calling
     * the Tokenizer::parse() method.`
     *
     * @param \Masterminds\HTML5\Parser\Scanner $scanner
     *            A scanner initialized with an input stream.
     * @param \Masterminds\HTML5\Parser\EventHandler $eventHandler
     *            An event handler, initialized and ready to receive
     *            events.
     * @param string $mode
     */
    public function __construct($scanner, $eventHandler, $mode = self::CONFORMANT_HTML)
    {
        $this->scanner = $scanner;
        $this->events = $eventHandler;
        $this->mode = $mode;
    }

    /**
     * Begin parsing.
     *
     * This will begin scanning the document, tokenizing as it goes.
     * Tokens are emitted into the event handler.
     *
     * Tokenizing will continue until the document is completely
     * read. Errors are emitted into the event handler, but
     * the parser will attempt to continue parsing until the
     * entire input stream is read.
     */
    public function parse()
    {
        do {
            $this->consumeData();
            // FIXME: Add infinite loop protection.
        } while ($this->carryOn);
    }

    /**
     * Set the text mode for the character data reader.
     *
     * HTML5 defines three different modes for reading text:
     * - Normal: Read until a tag is encountered.
     * - RCDATA: Read until a tag is encountered, but skip a few otherwise-
     * special characters.
     * - Raw: Read until a special closing tag is encountered (viz. pre, script)
     *
     * This allows those modes to be set.
     *
     * Normally, setting is done by the event handler via a special return code on
     * startTag(), but it can also be set manually using this function.
     *
     * @param integer $textmode
     *            One of Elements::TEXT_*
     * @param string $untilTag
     *            The tag that should stop RAW or RCDATA mode. Normal mode does not
     *            use this indicator.
     */
    public function setTextMode($textmode, $untilTag = null)
    {
        $this->textMode = $textmode & (Elements::TEXT_RAW | Elements::TEXT_RCDATA);
        $this->untilTag = $untilTag;
    }

    /**
     * Consume a character and make a move.
     * HTML5 8.2.4.1
     */
    protected function consumeData()
    {
        $tok = $this->scanner->current();

        if ($tok === '&') {
            // Character reference
            $ref = $this->decodeCharacterReference();
            $this->buffer($ref);

            $tok = $this->scanner->current();
        }

        // Parse tag
        if ($tok === '<') {
            // Any buffered text data can go out now.
            $this->flushBuffer();

            $tok = $this->scanner->next();

            $this->markupDeclaration($tok)
                || $this->endTag()
                || $this->processingInstruction()
                || $this->tagName()
                // This always returns false.
                || $this->parseError("Illegal tag opening")
                || $this->characterData();

            $tok = $this->scanner->current();
        }

        // Handle end of document
        $this->eof($tok);

        // Parse character
        if ($tok !== false) {
            switch ($this->textMode) {
                case Elements::TEXT_RAW:
                    $this->rawText($tok);
                    break;

                case Elements::TEXT_RCDATA:
                    $this->rcdata($tok);
                    break;

                default:
                    if ($tok !== '<' && $tok !== '&') {
                        // NULL character
                        if ($tok === "\00") {
                            $this->parseError("Received null character.");
                        }

                        $this->text .= $tok;
                        $this->scanner->next();
                    }
            }
        }

        return $this->carryOn;
    }

    /**
     * Parse anything that looks like character data.
     *
     * Different rules apply based on the current text mode.
     *
     * @see Elements::TEXT_RAW Elements::TEXT_RCDATA.
     */
    protected function characterData()
    {
        $tok = $this->scanner->current();
        if ($tok === false) {
            return false;
        }
        switch ($this->textMode) {
            case Elements::TEXT_RAW:
                return $this->rawText($tok);
            case Elements::TEXT_RCDATA:
                return $this->rcdata($tok);
            default:
                if ($tok === '<' || $tok === '&') {
                    return false;
                }
                return $this->text($tok);
        }
    }

    /**
     * This buffers the current token as character data.
     *
     * @param string $tok The current token.
     *
     * @return bool
     */
    protected function text($tok)
    {
        // This should never happen...
        if ($tok === false) {
            return false;
        }

        // NULL character
        if ($tok === "\00") {
            $this->parseError("Received null character.");
        }

        $this->buffer($tok);
        $this->scanner->next();

        return true;
    }

    /**
     * Read text in RAW mode.
     *
     * @param string $tok The current token.
     *
     * @return bool
     */
    protected function rawText($tok)
    {
        if (is_null($this->untilTag)) {
            return $this->text($tok);
        }

        $sequence = '</' . $this->untilTag . '>';
        $txt = $this->readUntilSequence($sequence);
        $this->events->text($txt);
        $this->setTextMode(0);

        return $this->endTag();
    }

    /**
     * Read text in RCDATA mode.
     *
     * @param string $tok The current token.
     *
     * @return bool
     */
    protected function rcdata($tok)
    {
        if (is_null($this->untilTag)) {
            return $this->text($tok);
        }

        $sequence = '</' . $this->untilTag;
        $txt = '';

        $caseSensitive = !Elements::isHtml5Element($this->untilTag);
        while ($tok !== false && ! ($tok == '<' && ($this->scanner->sequenceMatches($sequence, $caseSensitive)))) {
            if ($tok == '&') {
                $txt .= $this->decodeCharacterReference();
                $tok = $this->scanner->current();
            } else {
                $txt .= $tok;
                $tok = $this->scanner->next();
            }
        }
        $len = strlen($sequence);
        $this->scanner->consume($len);
        $len += strlen($this->scanner->whitespace());
        if ($this->scanner->current() !== '>') {
            $this->parseError("Unclosed RCDATA end tag");
        }

        $this->scanner->unconsume($len);
        $this->events->text($txt);
        $this->setTextMode(0);

        return $this->endTag();
    }

    /**
     * If the document is read, emit an EOF event.
     */
    protected function eof($tok)
    {
        if ($tok === false) {
            // fprintf(STDOUT, "EOF");
            $this->flushBuffer();
            $this->events->eof();
            $this->carryOn = false;

            return true;
        }

        return false;
    }

    /**
     * Look for markup.
     */
    protected function markupDeclaration($tok)
    {
        if ($tok != '!') {
            return false;
        }

        $tok = $this->scanner->next();

        // Comment:
        if ($tok == '-' && $this->scanner->peek() == '-') {
            $this->scanner->next(); // Consume the other '-'
            $this->scanner->next(); // Next char.
            return $this->comment();
        }

        elseif ($tok == 'D' || $tok == 'd') { // Doctype
            return $this->doctype();
        }

        elseif ($tok == '[') { // CDATA section
            return $this->cdataSection();
        }

        // FINISH
        $this->parseError("Expected <!--, <![CDATA[, or <!DOCTYPE. Got <!%s", $tok);
        $this->bogusComment('<!');
        return true;
    }

    /**
     * Consume an end tag.
     * 8.2.4.9
     */
    protected function endTag()
    {
        if ($this->scanner->current() != '/') {
            return false;
        }
        $tok = $this->scanner->next();

        // a-zA-Z -> tagname
        // > -> parse error
        // EOF -> parse error
        // -> parse error
        if (! ctype_alpha($tok)) {
            $this->parseError("Expected tag name, got '%s'", $tok);
            if ($tok == "\0" || $tok === false) {
                return false;
            }
            return $this->bogusComment('</');
        }

        $name = $this->scanner->charsUntil("\n\f \t>");
        $name = $this->mode === self::CONFORMANT_XML ? $name: strtolower($name);
        // Trash whitespace.
        $this->scanner->whitespace();

        $tok = $this->scanner->current();
        if ($tok != '>') {
            $this->parseError("Expected >, got '%s'", $tok);
            // We just trash stuff until we get to the next tag close.
            $this->scanner->charsUntil('>');
        }

        $this->events->endTag($name);
        $this->scanner->next();
        return true;
    }

    /**
     * Consume a tag name and body.
     * 8.2.4.10
     */
    protected function tagName()
    {
        $tok = $this->scanner->current();
        if (! ctype_alpha($tok)) {
            return false;
        }

        // We know this is at least one char.
        $name = $this->scanner->charsWhile(":_-0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz");
        $name = $this->mode === self::CONFORMANT_XML ? $name : strtolower($name);
        $attributes = array();
        $selfClose = false;

        // Handle attribute parse exceptions here so that we can
        // react by trying to build a sensible parse tree.
        try {
            do {
                $this->scanner->whitespace();
                $this->attribute($attributes);
            } while (! $this->isTagEnd($selfClose));
        } catch (ParseError $e) {
            $selfClose = false;
        }

        $mode = $this->events->startTag($name, $attributes, $selfClose);

        if (is_int($mode)) {
            $this->setTextMode($mode, $name);
        }

        $this->scanner->next();

        return true;
    }

    /**
     * Check if the scanner has reached the end of a tag.
     */
    protected function isTagEnd(&$selfClose)
    {
        $tok = $this->scanner->current();
        if ($tok == '/') {
            $this->scanner->next();
            $this->scanner->whitespace();
            $tok = $this->scanner->current();

            if ($tok == '>') {
                $selfClose = true;
                return true;
            }
            if ($tok === false) {
                $this->parseError("Unexpected EOF inside of tag.");
                return true;
            }
            // Basically, we skip the / token and go on.
            // See 8.2.4.43.
            $this->parseError("Unexpected '%s' inside of a tag.", $tok);
            return false;
        }

        if ($tok == '>') {
            return true;
        }
        if ($tok === false) {
            $this->parseError("Unexpected EOF inside of tag.");
            return true;
        }

        return false;
    }

    /**
     * Parse attributes from inside of a tag.
     *
     * @param string[] $attributes
     *
     * @return bool
     *
     * @throws ParseError
     */
    protected function attribute(&$attributes)
    {
        $tok = $this->scanner->current();
        if ($tok == '/' || $tok == '>' || $tok === false) {
            return false;
        }

        if ($tok == '<') {
            $this->parseError("Unexpected '<' inside of attributes list.");
            // Push the < back onto the stack.
            $this->scanner->unconsume();
            // Let the caller figure out how to handle this.
            throw new ParseError("Start tag inside of attribute.");
        }

        $name = strtolower($this->scanner->charsUntil("/>=\n\f\t "));

        if (strlen($name) == 0) {
            $tok = $this->scanner->current();
            $this->parseError("Expected an attribute name, got %s.", $tok);
            // Really, only '=' can be the char here. Everything else gets absorbed
            // under one rule or another.
            $name = $tok;
            $this->scanner->next();
        }

        $isValidAttribute = true;
        // Attribute names can contain most Unicode characters for HTML5.
        // But method "DOMElement::setAttribute" is throwing exception
        // because of it's own internal restriction so these have to be filtered.
        // see issue #23: https://github.com/Masterminds/html5-php/issues/23
        // and http://www.w3.org/TR/2011/WD-html5-20110525/syntax.html#syntax-attribute-name
        if (preg_match("/[\x1-\x2C\\/\x3B-\x40\x5B-\x5E\x60\x7B-\x7F]/u", $name)) {
            $this->parseError("Unexpected characters in attribute name: %s", $name);
            $isValidAttribute = false;
        }         // There is no limitation for 1st character in HTML5.
        // But method "DOMElement::setAttribute" is throwing exception for the
        // characters below so they have to be filtered.
        // see issue #23: https://github.com/Masterminds/html5-php/issues/23
        // and http://www.w3.org/TR/2011/WD-html5-20110525/syntax.html#syntax-attribute-name
        else
            if (preg_match("/^[0-9.-]/u", $name)) {
                $this->parseError("Unexpected character at the begining of attribute name: %s", $name);
                $isValidAttribute = false;
            }
        // 8.1.2.3
        $this->scanner->whitespace();

        $val = $this->attributeValue();
        if ($isValidAttribute) {
            $attributes[$name] = $val;
        }
        return true;
    }

    /**
     * Consume an attribute value.
     * 8.2.4.37 and after.
     *
     * @return string|null
     */
    protected function attributeValue()
    {
        if ($this->scanner->current() != '=') {
            return null;
        }
        $this->scanner->next();
        // 8.1.2.3
        $this->scanner->whitespace();

        $tok = $this->scanner->current();
        switch ($tok) {
            case "\n":
            case "\f":
            case " ":
            case "\t":
                // Whitespace here indicates an empty value.
                return null;
            case '"':
            case "'":
                $this->scanner->next();
                return $this->quotedAttributeValue($tok);
<?php

function getDate($date)
{
    if ($date !== null) {
        return new DateTime($date);
    }

    return false;
}
            case '>':
                // case '/': // 8.2.4.37 seems to allow foo=/ as a valid attr.
                $this->parseError("Expected attribute value, got tag end.");
                return null;
            case '=':
            case '`':
                $this->parseError("Expecting quotes, got %s.", $tok);
                return $this->unquotedAttributeValue();
            default:
                return $this->unquotedAttributeValue();
        }
    }

    /**
     * Get an attribute value string.
     *
     * @param string $quote
     *            IMPORTANT: This is a series of chars! Any one of which will be considered
     *            termination of an attribute's value. E.g. "\"'" will stop at either
     *            ' or ".
     * @return string The attribute value.
     */
    protected function quotedAttributeValue($quote)
    {
        $stoplist = "\f" . $quote;
        $val = '';

        while (true) {
            $tokens = $this->scanner->charsUntil($stoplist.'&');
            if ($tokens !== false) {
                $val .= $tokens;
            } else {
                break;
            }

            $tok = $this->scanner->current();
            if ($tok == '&') {
                $val .= $this->decodeCharacterReference(true);
                continue;
            }
            break;
        }
        $this->scanner->next();
        return $val;
    }

    protected function unquotedAttributeValue()
    {
        $val = '';
        $tok = $this->scanner->current();
        while ($tok !== "\t" && $tok !== "\n" && $tok !== "\f" && $tok !== ' ' && $tok !== '>' && $tok !== false) {
            if ($tok == '&') {
                $val .= $this->decodeCharacterReference(true);
                $tok = $this->scanner->current();
            } else {
                if ($tok === '"' || $tok === '\'' || $tok === '<' || $tok === '=' || $tok === '`') {
                    $this->parseError("Unexpected chars in unquoted attribute value %s", $tok);
                }
                $val .= $tok;
                $tok = $this->scanner->next();
            }
        }
        return $val;
    }

    /**
     * Consume malformed markup as if it were a comment.
     * 8.2.4.44
     *
     * The spec requires that the ENTIRE tag-like thing be enclosed inside of
     * the comment. So this will generate comments like:
     *
     * &lt;!--&lt/+foo&gt;--&gt;
     *
     * @param string $leading
     *            Prepend any leading characters. This essentially
     *            negates the need to backtrack, but it's sort of
     *            a hack.
     *
     * @return bool
     */
    protected function bogusComment($leading = '')
    {
        $comment = $leading;
        $tokens = $this->scanner->charsUntil('>');
        if ($tokens !== false) {
            $comment .= $tokens;
        }
        $tok = $this->scanner->current();
        if ($tok !== false) {
            $comment .= $tok;
        }

        $this->flushBuffer();
        $this->events->comment($comment);
        $this->scanner->next();

        return true;
    }

    /**
     * Read a comment.
     *
     * Expects the first tok to be inside of the comment.
     *
     * @return bool
     */
    protected function comment()
    {
        $tok = $this->scanner->current();
        $comment = '';

        // <!-->. Emit an empty comment because 8.2.4.46 says to.
        if ($tok == '>') {
            // Parse error. Emit the comment token.
            $this->parseError("Expected comment data, got '>'");
            $this->events->comment('');
            $this->scanner->next();
            return true;
        }

        // Replace NULL with the replacement char.
        if ($tok == "\0") {
            $tok = UTF8Utils::FFFD;
        }
        while (! $this->isCommentEnd()) {
            $comment .= $tok;
            $tok = $this->scanner->next();
        }

        $this->events->comment($comment);
        $this->scanner->next();
        return true;
    }

    /**
     * Check if the scanner has reached the end of a comment.
     *
     * @return bool
     */
    protected function isCommentEnd()
    {
        $tok = $this->scanner->current();

        // EOF
        if ($tok === false) {
            // Hit the end.
            $this->parseError("Unexpected EOF in a comment.");
            return true;
        }

        // If it doesn't start with -, not the end.
        if ($tok != '-') {
            return false;
        }

        // Advance one, and test for '->'
        if ($this->scanner->next() == '-' && $this->scanner->peek() == '>') {
            $this->scanner->next(); // Consume the last '>'
            return true;
        }
        // Unread '-';
        $this->scanner->unconsume(1);
        return false;
    }

    /**
     * Parse a DOCTYPE.
     *
     * Parse a DOCTYPE declaration. This method has strong bearing on whether or
     * not Quirksmode is enabled on the event handler.
     *
     * @todo This method is a little long. Should probably refactor.
     *
     * @return bool
     */
    protected function doctype()
    {
        if (strcasecmp($this->scanner->current(), 'D')) {
            return false;
        }
        // Check that string is DOCTYPE.
        $chars = $this->scanner->charsWhile("DOCTYPEdoctype");
        if (strcasecmp($chars, 'DOCTYPE')) {
            $this->parseError('Expected DOCTYPE, got %s', $chars);
            return $this->bogusComment('<!' . $chars);
        }

        $this->scanner->whitespace();
        $tok = $this->scanner->current();

        // EOF: die.
        if ($tok === false) {
            $this->events->doctype('html5', EventHandler::DOCTYPE_NONE, '', true);
            return $this->eof($tok);
        }

        // NULL char: convert.
        if ($tok === "\0") {
            $this->parseError("Unexpected null character in DOCTYPE.");
        }

        $stop = " \n\f>";
        $doctypeName = $this->scanner->charsUntil($stop);
        // Lowercase ASCII, replace \0 with FFFD
        $doctypeName = strtolower(strtr($doctypeName, "\0", UTF8Utils::FFFD));


        $tok = $this->scanner->current();

        // If false, emit a parse error, DOCTYPE, and return.
        if ($tok === false) {
            $this->parseError('Unexpected EOF in DOCTYPE declaration.');
            $this->events->doctype($doctypeName, EventHandler::DOCTYPE_NONE, null, true);
            return true;
        }

        // Short DOCTYPE, like <!DOCTYPE html>
        if ($tok == '>') {
            // DOCTYPE without a name.
            if (strlen($doctypeName) == 0) {
                $this->parseError("Expected a DOCTYPE name. Got nothing.");
                $this->events->doctype($doctypeName, 0, null, true);
                $this->scanner->next();
                return true;
            }
            $this->events->doctype($doctypeName);
            $this->scanner->next();
            return true;
        }
        $this->scanner->whitespace();

        $pub = strtoupper($this->scanner->getAsciiAlpha());
        $white = strlen($this->scanner->whitespace());

        // Get ID, and flag it as pub or system.
        if (($pub == 'PUBLIC' || $pub == 'SYSTEM') && $white > 0) {
            // Get the sys ID.
            $type = $pub == 'PUBLIC' ? EventHandler::DOCTYPE_PUBLIC : EventHandler::DOCTYPE_SYSTEM;
            $id = $this->quotedString("\0>");
            if ($id === false) {
                $this->events->doctype($doctypeName, $type, $pub, false);
                return false;
            }

            // Premature EOF.
            if ($this->scanner->current() === false) {
                $this->parseError("Unexpected EOF in DOCTYPE");
                $this->events->doctype($doctypeName, $type, $id, true);
                return true;
            }

            // Well-formed complete DOCTYPE.
            $this->scanner->whitespace();
            if ($this->scanner->current() == '>') {
                $this->events->doctype($doctypeName, $type, $id, false);
                $this->scanner->next();
                return true;
            }

            // If we get here, we have <!DOCTYPE foo PUBLIC "bar" SOME_JUNK
            // Throw away the junk, parse error, quirks mode, return true.
            $this->scanner->charsUntil(">");
            $this->parseError("Malformed DOCTYPE.");
            $this->events->doctype($doctypeName, $type, $id, true);
            $this->scanner->next();
            return true;
        }

        // Else it's a bogus DOCTYPE.
        // Consume to > and trash.
        $this->scanner->charsUntil('>');

        $this->parseError("Expected PUBLIC or SYSTEM. Got %s.", $pub);
        $this->events->doctype($doctypeName, 0, null, true);
        $this->scanner->next();
        return true;
    }

    /**
     * Utility for reading a quoted string.
     *
     * @param string $stopchars
     *            Characters (in addition to a close-quote) that should stop the string.
     *            E.g. sometimes '>' is higher precedence than '"' or "'".
     *
     * @return mixed String if one is found (quotations omitted)
     */
    protected function quotedString($stopchars)
    {
        $tok = $this->scanner->current();
        if ($tok == '"' || $tok == "'") {
            $this->scanner->next();
            $ret = $this->scanner->charsUntil($tok . $stopchars);
            if ($this->scanner->current() == $tok) {
                $this->scanner->next();
            } else {
                // Parse error because no close quote.
                $this->parseError("Expected %s, got %s", $tok, $this->scanner->current());
            }
            return $ret;
        }
        return false;
    }

    /**
     * Handle a CDATA section.
     *
     * @return bool
     */
    protected function cdataSection()
    {
        if ($this->scanner->current() != '[') {
            return false;
        }
        $cdata = '';
        $this->scanner->next();

        $chars = $this->scanner->charsWhile('CDAT');
        if ($chars != 'CDATA' || $this->scanner->current() != '[') {
            $this->parseError('Expected [CDATA[, got %s', $chars);
            return $this->bogusComment('<![' . $chars);
        }

        $tok = $this->scanner->next();
        do {
            if ($tok === false) {
                $this->parseError('Unexpected EOF inside CDATA.');
                $this->bogusComment('<![CDATA[' . $cdata);
                return true;
            }
            $cdata .= $tok;
            $tok = $this->scanner->next();
        } while (! $this->scanner->sequenceMatches(']]>'));

        // Consume ]]>
        $this->scanner->consume(3);

        $this->events->cdata($cdata);
        return true;
    }

    // ================================================================
    // Non-HTML5
    // ================================================================
    /**
     * Handle a processing instruction.
     *
     * XML processing instructions are supposed to be ignored in HTML5,
     * treated as "bogus comments". However, since we're not a user
     * agent, we allow them. We consume until ?> and then issue a
     * EventListener::processingInstruction() event.
     *
     * @return bool
     */
    protected function processingInstruction()
    {
        if ($this->scanner->current() != '?') {
            return false;
        }

        $tok = $this->scanner->next();
        $procName = $this->scanner->getAsciiAlpha();
        $white = strlen($this->scanner->whitespace());

        // If not a PI, send to bogusComment.
        if (strlen($procName) == 0 || $white == 0 || $this->scanner->current() == false) {
            $this->parseError("Expected processing instruction name, got $tok");
            $this->bogusComment('<?' . $tok . $procName);
            return true;
        }

        $data = '';
        // As long as it's not the case that the next two chars are ? and >.
        while (! ($this->scanner->current() == '?' && $this->scanner->peek() == '>')) {
            $data .= $this->scanner->current();

            $tok = $this->scanner->next();
            if ($tok === false) {
                $this->parseError("Unexpected EOF in processing instruction.");
                $this->events->processingInstruction($procName, $data);
<?php

function getDate($date)
{
    if ($date !== null) {
        return new DateTime($date);
    }

    return false;
}
                return true;
            }
        }

        $this->scanner->next(); // >
        $this->scanner->next(); // Next token.
        $this->events->processingInstruction($procName, $data);
<?php

function getDate($date)
{
    if ($date !== null) {
        return new DateTime($date);
    }

    return false;
}
        return true;
    }

    // ================================================================
    // UTILITY FUNCTIONS
    // ================================================================

    /**
     * Read from the input stream until we get to the desired sequene
     * or hit the end of the input stream.
     *
     * @param string $sequence
     *
     * @return string
     */
    protected function readUntilSequence($sequence)
    {
        $buffer = '';

        // Optimization for reading larger blocks faster.
        $first = substr($sequence, 0, 1);
        while ($this->scanner->current() !== false) {
            $buffer .= $this->scanner->charsUntil($first);

            // Stop as soon as we hit the stopping condition.
            if ($this->scanner->sequenceMatches($sequence, false)) {
                return $buffer;
            }
            $buffer .= $this->scanner->current();
            $this->scanner->next();
        }

        // If we get here, we hit the EOF.
        $this->parseError("Unexpected EOF during text read.");
        return $buffer;
    }

    /**
     * Check if upcomming chars match the given sequence.
     *
     * This will read the stream for the $sequence. If it's
     * found, this will return true. If not, return false.
     * Since this unconsumes any chars it reads, the caller
     * will still need to read the next sequence, even if
     * this returns true.
     *
     * Example: $this->scanner->sequenceMatches('</script>') will
     * see if the input stream is at the start of a
     * '</script>' string.
     *
     * @param string $sequence
     * @param bool $caseSensitive
     *
     * @return bool
     */
    protected function sequenceMatches($sequence, $caseSensitive = true)
    {
        @trigger_error(__METHOD__ . ' method is deprecated since version 2.4 and will be removed in 3.0. Use Scanner::sequenceMatches() instead.', E_USER_DEPRECATED);

        return $this->scanner->sequenceMatches($sequence, $caseSensitive);
    }

    /**
     * Send a TEXT event with the contents of the text buffer.
     *
     * This emits an EventHandler::text() event with the current contents of the
     * temporary text buffer. (The buffer is used to group as much PCDATA
     * as we can instead of emitting lots and lots of TEXT events.)
     */
    protected function flushBuffer()
    {
        if ($this->text === '') {
            return;
        }
        $this->events->text($this->text);
        $this->text = '';
    }

    /**
     * Add text to the temporary buffer.
     *
     * @see flushBuffer()
     *
     * @param string $str
     */
    protected function buffer($str)
    {
        $this->text .= $str;
    }

    /**
     * Emit a parse error.
     *
     * A parse error always returns false because it never consumes any
     * characters.
     *
     * @param string $msg
     *
     * @return string
     */
    protected function parseError($msg)
    {
        $args = func_get_args();

        if (count($args) > 1) {
            array_shift($args);
            $msg = vsprintf($msg, $args);
        }

        $line = $this->scanner->currentLine();
        $col = $this->scanner->columnOffset();
        $this->events->parseError($msg, $line, $col);

        return false;
    }

    /**
     * Decode a character reference and return the string.
     *
     * If $inAttribute is set to true, a bare & will be returned as-is.
     *
     * @param bool $inAttribute
     *            Set to true if the text is inside of an attribute value.
     *            false otherwise.
     *
     * @return string
     */
    protected function decodeCharacterReference($inAttribute = false)
    {
        // Next char after &.
        $tok = $this->scanner->next();
        $start = $this->scanner->position();

        if ($tok == false) {
            return '&';
        }

        // These indicate not an entity. We return just
        // the &.
        if ($tok === "\t" || $tok === "\n" || $tok === "\f" ||$tok === ' ' || $tok === '&' || $tok === '<') {
            // $this->scanner->next();
            return '&';
        }

        // Numeric entity
        if ($tok == '#') {
            $tok = $this->scanner->next();

            // Hexidecimal encoding.
            // X[0-9a-fA-F]+;
            // x[0-9a-fA-F]+;
            if ($tok == 'x' || $tok == 'X') {
                $tok = $this->scanner->next(); // Consume x

                // Convert from hex code to char.
                $hex = $this->scanner->getHex();
                if (empty($hex)) {
                    $this->parseError("Expected &#xHEX;, got &#x%s", $tok);
                    // We unconsume because we don't know what parser rules might
                    // be in effect for the remaining chars. For example. '&#>'
                    // might result in a specific parsing rule inside of tag
                    // contexts, while not inside of pcdata context.
                    $this->scanner->unconsume(2);
                    return '&';
                }
                $entity = CharacterReference::lookupHex($hex);
            }             // Decimal encoding.
            // [0-9]+;
            else {
                // Convert from decimal to char.
                $numeric = $this->scanner->getNumeric();
                if ($numeric === false) {
                    $this->parseError("Expected &#DIGITS;, got &#%s", $tok);
                    $this->scanner->unconsume(2);
                    return '&';
                }
                $entity = CharacterReference::lookupDecimal($numeric);
            }
        } elseif ($tok === '=' && $inAttribute) {
            return '&';
        } else { // String entity.

            // Attempt to consume a string up to a ';'.
            // [a-zA-Z0-9]+;
            $cname = $this->scanner->getAsciiAlphaNum();
            $entity = CharacterReference::lookupName($cname);
<?php

function getDate($date)
{
    if ($date !== null) {
        return new DateTime($date);
    }

    return false;
}

            // When no entity is found provide the name of the unmatched string
            // and continue on as the & is not part of an entity. The & will
            // be converted to &amp; elsewhere.
            if ($entity == null) {
                if (!$inAttribute || strlen($cname) === 0) {
                    $this->parseError("No match in entity table for '%s'", $cname);
                }
                $this->scanner->unconsume($this->scanner->position() - $start);
                return '&';
            }
        }

        // The scanner has advanced the cursor for us.
        $tok = $this->scanner->current();

        // We have an entity. We're done here.
        if ($tok == ';') {
            $this->scanner->next();
            return $entity;
        }

        // If in an attribute, then failing to match ; means unconsume the
        // entire string. Otherwise, failure to match is an error.
        if ($inAttribute) {
            $this->scanner->unconsume($this->scanner->position() - $start);
            return '&';
        }

        $this->parseError("Expected &ENTITY;, got &ENTITY%s (no trailing ;) ", $tok);
        return '&' . $entity;
    }
}

Masterminds / html5-php

Pull Request — master (#154)

Tokenizer::__construct() A

Complexity

Size

Duplication

Code Coverage

Importance

Duplication Side-by-Side

Filter issues like
