Skip to content

Optimize dictionary parsing #508

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
PrinsFrank merged 5 commits into from
Aug 21, 2026
Merged

Optimize dictionary parsing #508

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
44dd84c
Keep track of current Context seperately instead of constantly re-ret…
PrinsFrank Aug 21, 2026
150ab4c
Replace in_array calls with strict checks
PrinsFrank Aug 21, 2026
b2113aa
Replace match with two ifs to speed up DictionaryParser
PrinsFrank Aug 21, 2026
975e168
Remove nested key/value buffer calls
PrinsFrank Aug 21, 2026
d61dce2
Remove usage of InfiniteBuffer in CompressedObjectByteOffsetParser
PrinsFrank Aug 21, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 0 additions & 12 deletions src/Document/Dictionary/DictionaryParseContext/NestingContext.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -55,12 +55,6 @@ public function addToKeyBuffer(string $char): self {
return $this;
}

public function removeFromKeyBuffer(int $nChars = 1): self {
$this->getKeyBuffer()->removeChar($nChars);

return $this;
}

public function getValueBuffer(): InfiniteBuffer {
return $this->valueBuffer[$this->currentLevel] ??= new InfiniteBuffer();
}
Expand All @@ -71,12 +65,6 @@ public function addToValueBuffer(string $char): self {
return $this;
}

public function removeFromValueBuffer(int $nChars = 1): self {
$this->getValueBuffer()->removeChar($nChars);

return $this;
}

/** @return list<string> */
public function getKeysFromRoot(): array {
$keysFromRoot = [];
Expand Down
84 changes: 52 additions & 32 deletions src/Document/Dictionary/DictionaryParser.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,87 +23,107 @@ class DictionaryParser {
public static function parse(?EncryptionContext $encryptionContext, Stream $stream, int $startPos, int $nrOfBytes): Dictionary {
$dictionaryArray = [];
$nestingContext = (new NestingContext())->setContext(DictionaryParseContext::ROOT);
$keyBuffer = $nestingContext->getKeyBuffer();
$valueBuffer = $nestingContext->getValueBuffer();
$arrayNestingLevel = 0;
$previousChar = $secondToLastChar = null;
$contextBeforeComment = $previousIndexLevelDecrease = $previousIndexLevelIncrease = null;
$previousChar = $secondToLastChar = $currentContext = $contextBeforeComment = $previousIndexLevelDecrease = $previousIndexLevelIncrease = null;
foreach ($stream->chars($startPos, $nrOfBytes) as $index => $char) {
if ($char === DelimiterCharacter::LESS_THAN_SIGN->value
&& $previousChar === DelimiterCharacter::LESS_THAN_SIGN->value
&& $secondToLastChar !== LiteralStringEscapeCharacter::REVERSE_SOLIDUS->value
&& $nestingContext->getContext() !== DictionaryParseContext::VALUE_IN_SQUARE_BRACKETS
&& $currentContext !== DictionaryParseContext::VALUE_IN_SQUARE_BRACKETS
&& $previousIndexLevelIncrease !== $index - 1) {
if ($nestingContext->getContext() === DictionaryParseContext::KEY) {
$nestingContext->removeFromKeyBuffer();
if ($currentContext === DictionaryParseContext::KEY) {
$keyBuffer->removeChar();
}

$previousIndexLevelIncrease = $index;
$nestingContext->setContext(DictionaryParseContext::DICTIONARY)->incrementNesting()->setContext(DictionaryParseContext::DICTIONARY);
} elseif ($char === DelimiterCharacter::LESS_THAN_SIGN->value && $nestingContext->getContext() === DictionaryParseContext::KEY) {
$currentContext = DictionaryParseContext::DICTIONARY;
$keyBuffer = $nestingContext->getKeyBuffer();
$valueBuffer = $nestingContext->getValueBuffer();
} elseif ($char === DelimiterCharacter::LESS_THAN_SIGN->value && $currentContext === DictionaryParseContext::KEY) {
$nestingContext->setContext(DictionaryParseContext::VALUE);
$currentContext = DictionaryParseContext::VALUE;
} elseif ($char === DelimiterCharacter::GREATER_THAN_SIGN->value
&& $previousChar === DelimiterCharacter::GREATER_THAN_SIGN->value
&& $secondToLastChar !== LiteralStringEscapeCharacter::REVERSE_SOLIDUS->value
&& $nestingContext->getContext() !== DictionaryParseContext::VALUE_IN_SQUARE_BRACKETS
&& $currentContext !== DictionaryParseContext::VALUE_IN_SQUARE_BRACKETS
&& $previousIndexLevelDecrease !== $index - 1) {
$nestingContext->removeFromValueBuffer();
$valueBuffer->removeChar();
self::flush($dictionaryArray, $nestingContext);
$previousIndexLevelDecrease = $index;
$nestingContext->decrementNesting()->flush();
$currentContext = $nestingContext->getContext();
$keyBuffer = $nestingContext->getKeyBuffer();
$valueBuffer = $nestingContext->getValueBuffer();
} elseif ($char === DelimiterCharacter::SOLIDUS->value
&& $previousChar !== LiteralStringEscapeCharacter::REVERSE_SOLIDUS->value
&& $nestingContext->getContext() !== DictionaryParseContext::VALUE_IN_SQUARE_BRACKETS) {
if ($nestingContext->getContext() === DictionaryParseContext::DICTIONARY) {
&& $currentContext !== DictionaryParseContext::VALUE_IN_SQUARE_BRACKETS) {
if ($currentContext === DictionaryParseContext::DICTIONARY) {
$nestingContext->setContext(DictionaryParseContext::KEY);
} elseif ($nestingContext->getContext() === DictionaryParseContext::VALUE) {
$currentContext = DictionaryParseContext::KEY;
} elseif ($currentContext === DictionaryParseContext::VALUE) {
self::flush($dictionaryArray, $nestingContext);
$nestingContext->setContext(DictionaryParseContext::KEY);
} elseif ($nestingContext->getContext() === DictionaryParseContext::KEY || $nestingContext->getContext() === DictionaryParseContext::KEY_VALUE_SEPARATOR) {
$currentContext = DictionaryParseContext::KEY;
} elseif ($currentContext === DictionaryParseContext::KEY || $currentContext === DictionaryParseContext::KEY_VALUE_SEPARATOR) {
$nestingContext->setContext(DictionaryParseContext::VALUE);
$currentContext = DictionaryParseContext::VALUE;
}
} elseif ($char === WhitespaceCharacter::LINE_FEED->value && $nestingContext->getContext() !== DictionaryParseContext::VALUE_IN_SQUARE_BRACKETS) {
if ($nestingContext->getContext() === DictionaryParseContext::KEY) {
} elseif ($char === WhitespaceCharacter::LINE_FEED->value && $currentContext !== DictionaryParseContext::VALUE_IN_SQUARE_BRACKETS) {
if ($currentContext === DictionaryParseContext::KEY) {
$nestingContext->setContext(DictionaryParseContext::KEY_VALUE_SEPARATOR);
} elseif ($nestingContext->getContext() === DictionaryParseContext::VALUE) {
$currentContext = DictionaryParseContext::KEY_VALUE_SEPARATOR;
} elseif ($currentContext === DictionaryParseContext::VALUE) {
self::flush($dictionaryArray, $nestingContext);
} elseif ($nestingContext->getContext() === DictionaryParseContext::COMMENT) {
} elseif ($currentContext === DictionaryParseContext::COMMENT) {
$nestingContext->setContext($contextBeforeComment ?? DictionaryParseContext::DICTIONARY);
$currentContext = $contextBeforeComment ?? DictionaryParseContext::DICTIONARY;
$contextBeforeComment = null;
}
} elseif ($char === DelimiterCharacter::PERCENT_SIGN->value && $previousChar !== LiteralStringEscapeCharacter::REVERSE_SOLIDUS->value && $nestingContext->getContext() !== DictionaryParseContext::VALUE_IN_PARENTHESES) {
if ($nestingContext->getContext() === DictionaryParseContext::VALUE) {
} elseif ($char === DelimiterCharacter::PERCENT_SIGN->value && $previousChar !== LiteralStringEscapeCharacter::REVERSE_SOLIDUS->value && $currentContext !== DictionaryParseContext::VALUE_IN_PARENTHESES) {
if ($currentContext === DictionaryParseContext::VALUE) {
self::flush($dictionaryArray, $nestingContext);
$contextBeforeComment = DictionaryParseContext::DICTIONARY;
} else {
$contextBeforeComment = $nestingContext->getContext();
$contextBeforeComment = $currentContext;
}
$nestingContext->setContext(DictionaryParseContext::COMMENT);
} elseif (WhitespaceCharacter::tryFrom($char) !== null && $nestingContext->getContext() === DictionaryParseContext::KEY) {
$currentContext = DictionaryParseContext::COMMENT;
} elseif (WhitespaceCharacter::tryFrom($char) !== null && $currentContext === DictionaryParseContext::KEY) {
$nestingContext->setContext(DictionaryParseContext::KEY_VALUE_SEPARATOR);
} elseif ($char === DelimiterCharacter::LEFT_PARENTHESIS->value && (in_array($nestingContext->getContext(), [DictionaryParseContext::KEY, DictionaryParseContext::KEY_VALUE_SEPARATOR, DictionaryParseContext::VALUE], true))) {
$currentContext = DictionaryParseContext::KEY_VALUE_SEPARATOR;
} elseif ($char === DelimiterCharacter::LEFT_PARENTHESIS->value && ($currentContext === DictionaryParseContext::KEY || $currentContext === DictionaryParseContext::KEY_VALUE_SEPARATOR || $currentContext === DictionaryParseContext::VALUE)) {
$nestingContext->setContext(DictionaryParseContext::VALUE_IN_PARENTHESES);
} elseif ($char === DelimiterCharacter::RIGHT_PARENTHESIS->value && $previousChar !== LiteralStringEscapeCharacter::REVERSE_SOLIDUS->value && $nestingContext->getContext() === DictionaryParseContext::VALUE_IN_PARENTHESES) {
$currentContext = DictionaryParseContext::VALUE_IN_PARENTHESES;
} elseif ($char === DelimiterCharacter::RIGHT_PARENTHESIS->value && $previousChar !== LiteralStringEscapeCharacter::REVERSE_SOLIDUS->value && $currentContext === DictionaryParseContext::VALUE_IN_PARENTHESES) {
$nestingContext->setContext(DictionaryParseContext::VALUE);
} elseif ($char === DelimiterCharacter::LEFT_SQUARE_BRACKET->value && (in_array($nestingContext->getContext(), [DictionaryParseContext::KEY, DictionaryParseContext::KEY_VALUE_SEPARATOR, DictionaryParseContext::VALUE, DictionaryParseContext::VALUE_IN_SQUARE_BRACKETS], true))) {
$currentContext = DictionaryParseContext::VALUE;
} elseif ($char === DelimiterCharacter::LEFT_SQUARE_BRACKET->value && ($currentContext === DictionaryParseContext::KEY || $currentContext === DictionaryParseContext::KEY_VALUE_SEPARATOR || $currentContext === DictionaryParseContext::VALUE || $currentContext === DictionaryParseContext::VALUE_IN_SQUARE_BRACKETS)) {
$nestingContext->setContext(DictionaryParseContext::VALUE_IN_SQUARE_BRACKETS);
$currentContext = DictionaryParseContext::VALUE_IN_SQUARE_BRACKETS;
$arrayNestingLevel++;
} elseif ($char === DelimiterCharacter::RIGHT_SQUARE_BRACKET->value && $nestingContext->getContext() === DictionaryParseContext::VALUE_IN_SQUARE_BRACKETS) {
} elseif ($char === DelimiterCharacter::RIGHT_SQUARE_BRACKET->value && $currentContext === DictionaryParseContext::VALUE_IN_SQUARE_BRACKETS) {
$arrayNestingLevel--;
if ($arrayNestingLevel === 0) {
$nestingContext->setContext(DictionaryParseContext::VALUE);
$currentContext = DictionaryParseContext::VALUE;
}
} elseif (trim($char) !== '' && $nestingContext->getContext() === DictionaryParseContext::KEY_VALUE_SEPARATOR) {
} elseif (trim($char) !== '' && $currentContext === DictionaryParseContext::KEY_VALUE_SEPARATOR) {
$nestingContext->setContext(DictionaryParseContext::VALUE);
$currentContext = DictionaryParseContext::VALUE;
}

$secondToLastChar = $previousChar;
$previousChar = $char;
match ($nestingContext->getContext()) {
DictionaryParseContext::KEY => $nestingContext->addToKeyBuffer($char),
DictionaryParseContext::VALUE_IN_PARENTHESES,
DictionaryParseContext::VALUE_IN_SQUARE_BRACKETS,
DictionaryParseContext::VALUE => $nestingContext->addToValueBuffer($char),
default => null,
};
if ($currentContext === DictionaryParseContext::KEY) {
$keyBuffer->addChar($char);
} elseif ($currentContext === DictionaryParseContext::VALUE_IN_PARENTHESES
|| $currentContext === DictionaryParseContext::VALUE_IN_SQUARE_BRACKETS
|| $currentContext === DictionaryParseContext::VALUE) {
$valueBuffer->addChar($char);
}
}

return DictionaryFactory::fromArray($encryptionContext, $dictionaryArray);
Expand Down
4 changes: 2 additions & 2 deletions src/Document/Generic/Parsing/InfiniteBuffer.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,9 +38,9 @@ public function setValue(string $buffer): self {
return $this;
}

public function removeChar(int $nChars): self {
public function removeChar(): self {
if ($this->buffer !== '') {
$this->buffer = substr($this->buffer, 0, -$nChars);
$this->buffer = substr($this->buffer, 0, -1);
}

return $this;
Expand Down
16 changes: 6 additions & 10 deletions src/Document/Object/Item/CompressedObject/CompressedObjectByteOffsetParser.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,6 @@
use PrinsFrank\PdfParser\Document\Dictionary\DictionaryValue\Integer\IntegerValue;
use PrinsFrank\PdfParser\Document\Generic\Character\WhitespaceCharacter;
use PrinsFrank\PdfParser\Document\Generic\Marker;
use PrinsFrank\PdfParser\Document\Generic\Parsing\InfiniteBuffer;
use PrinsFrank\PdfParser\Document\Object\Item\CompressedObject\CompressedObjectContent\CompressedObjectContentParser;
use PrinsFrank\PdfParser\Exception\ParseFailureException;
use PrinsFrank\PdfParser\Exception\PdfParserException;
Expand All @@ -33,35 +32,32 @@ public static function parse(Stream $stream, int $startOffsetObject, int $endOff
$content = bin2hex(CompressedObjectContentParser::parseBinary(null, $stream, $startStreamPos, $length, $dictionary)->toString());
$first = $dictionary->getValueForKey(null, DictionaryKey::FIRST, IntegerValue::class)
?? throw new RuntimeException('Expected a dictionary entry for "First", none found');
$buffer = new InfiniteBuffer();
$buffer = '';
$previousObjectNumber = null;
$byteOffsets = [];
foreach (str_split(substr($content, 0, $first->value * 2), 2) as $char) {
$decodedChar = mb_chr((int) hexdec($char));
if (WhitespaceCharacter::tryFrom($decodedChar) !== null) {
$numberInBuffer = $buffer->__toString();
if (trim($numberInBuffer) === '') {
$buffer->flush();
if (trim($buffer) === '') {
continue;
}

if ($numberInBuffer !== (string) (int) $numberInBuffer) {
throw new ParseFailureException(sprintf('Number "%s" in buffer is not a valid number', $numberInBuffer));
if ($buffer !== (string) $numberInBuffer = (int) $buffer) {
throw new ParseFailureException(sprintf('Number "%s" in buffer is not a valid number', $buffer));
}

$numberInBuffer = (int) $numberInBuffer;
if ($previousObjectNumber !== null) {
$byteOffsets[$previousObjectNumber] = $numberInBuffer;
$previousObjectNumber = null;
} else {
$previousObjectNumber = $numberInBuffer;
}

$buffer->flush();
$buffer = '';
continue;
}

$buffer->addChar($decodedChar);
$buffer .= $decodedChar;
}

return new CompressedObjectByteOffsets($byteOffsets);
Expand Down
Loading