Skip to content

Tokenizer::scan performance improvements #6

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

Open
sam-osborne wants to merge 2 commits into
base: master
Choose a base branch
from
Open

Tokenizer::scan performance improvements #6

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
372ad50
Performance improvements to Tokenizer::scan
sam-osborne Jun 28, 2019
517f673
clean up
sam-osborne Jul 11, 2019
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
187 changes: 103 additions & 84 deletions src/Handlebars/Tokenizer.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -95,6 +95,7 @@ public function scan($text, $delimiters = null)
if ($text instanceof HandlebarsString) {
$text = $text->getString();
}

$this->reset();

if ($delimiters = trim($delimiters)) {
Expand All @@ -103,99 +104,116 @@ public function scan($text, $delimiters = null)
$this->ctag = $ctag;
}

$openingTagLength = strlen($this->otag);
$closingTagLength = strlen($this->ctag);
$firstOpeningTagCharacter = $this->otag[0];
$firstClosingTagCharacter = $this->ctag[0];

$len = strlen($text);

for ($i = 0; $i < $len; $i++) {

$character = $text[$i];

switch ($this->state) {
case self::IN_TEXT:
if ($this->tagChange($this->otag, $text, $i)) {
$i--;
$this->flushBuffer();
$this->state = self::IN_TAG_TYPE;
} else {
if ($text[$i] == "\n") {
$this->filterLine();

case self::IN_TEXT:
if ($character === $firstOpeningTagCharacter && $this->tagChange($this->otag, $text, $i, $openingTagLength)
) {
$i--;
$this->flushBuffer();
$this->state = self::IN_TAG_TYPE;
} else {
$this->buffer .= $text[$i];
if ($character == "\n") {
$this->filterLine();
} else {
$this->buffer .= $character;
}
}
}
break;
break;

case self::IN_TAG_TYPE:

$i += strlen($this->otag) - 1;
if (isset($this->tagTypes[$text[$i + 1]])) {
$tag = $text[$i + 1];
$this->tagType = $tag;
} else {
$tag = null;
$this->tagType = self::T_ESCAPED;
}
case self::IN_TAG_TYPE:

if ($this->tagType === self::T_DELIM_CHANGE) {
$i = $this->changeDelimiters($text, $i);
$this->state = self::IN_TEXT;
} else {
if ($tag !== null) {
$i++;
$i += $openingTagLength - 1;
if (isset($this->tagTypes[$text[$i + 1]])) {
$tag = $text[$i + 1];
$this->tagType = $tag;
} else {
$tag = null;
$this->tagType = self::T_ESCAPED;
}
$this->state = self::IN_TAG;
}
$this->seenTag = $i;
break;

default:
if ($this->tagChange($this->ctag, $text, $i)) {
// Sections (Helpers) can accept parameters
// Same thing for Partials (little known fact)
if (in_array($this->tagType, [
self::T_SECTION,
self::T_PARTIAL,
self::T_PARTIAL_2]
)) {
$newBuffer = explode(' ', trim($this->buffer), 2);
$args = '';
if (count($newBuffer) == 2) {
$args = $newBuffer[1];

if ($this->tagType === self::T_DELIM_CHANGE) {
$i = $this->changeDelimiters($text, $i);
$openingTagLength = strlen($this->otag);
$closingTagLength = strlen($this->ctag);
$firstOpeningTagCharacter = $this->otag[0];
$firstClosingTagCharacter = $this->ctag[0];

$this->state = self::IN_TEXT;
} else {
if ($tag !== null) {
$i++;
}
$this->buffer = $newBuffer[0];
}
$t = [
self::TYPE => $this->tagType,
self::NAME => trim($this->buffer),
self::OTAG => $this->otag,
self::CTAG => $this->ctag,
self::INDEX => ($this->tagType == self::T_END_SECTION) ?
$this->seenTag - strlen($this->otag) :
$i + strlen($this->ctag),
];
if (isset($args)) {
$t[self::ARGS] = $args;
$this->state = self::IN_TAG;
}
$this->tokens[] = $t;
unset($t);
unset($args);
$this->buffer = '';
$i += strlen($this->ctag) - 1;
$this->state = self::IN_TEXT;
if ($this->tagType == self::T_UNESCAPED) {
if ($this->ctag == '}}') {
$i++;
} else {
// Clean up `{{{ tripleStache }}}` style tokens.
$lastIndex = count($this->tokens) - 1;
$lastName = $this->tokens[$lastIndex][self::NAME];
if (substr($lastName, -1) === '}') {
$this->tokens[$lastIndex][self::NAME] = trim(
substr($lastName, 0, -1)
);
$this->seenTag = $i;
break;

default:
if ($character === $firstClosingTagCharacter && $this->tagChange($this->ctag, $text, $i, $closingTagLength)) {
// Sections (Helpers) can accept parameters
// Same thing for Partials (little known fact)
if (in_array($this->tagType, [
self::T_SECTION,
self::T_PARTIAL,
self::T_PARTIAL_2]
)) {
$newBuffer = explode(' ', trim($this->buffer), 2);
$args = '';
if (count($newBuffer) == 2) {
$args = $newBuffer[1];
}
$this->buffer = $newBuffer[0];
}
$t = [
self::TYPE => $this->tagType,
self::NAME => trim($this->buffer),
self::OTAG => $this->otag,
self::CTAG => $this->ctag,
self::INDEX => ($this->tagType == self::T_END_SECTION) ?
$this->seenTag - $openingTagLength :
$i + strlen($this->ctag),
];
if (isset($args)) {
$t[self::ARGS] = $args;
}
$this->tokens[] = $t;
unset($t);
unset($args);
$this->buffer = '';
$i += strlen($this->ctag) - 1;
$this->state = self::IN_TEXT;
if ($this->tagType == self::T_UNESCAPED) {
if ($this->ctag == '}}') {
$i++;
} else {
// Clean up `{{{ tripleStache }}}` style tokens.
$lastIndex = count($this->tokens) - 1;
$lastName = $this->tokens[$lastIndex][self::NAME];
if (substr($lastName, -1) === '}') {
$this->tokens[$lastIndex][self::NAME] = trim(
substr($lastName, 0, -1)
);
}
}
}
} else {
$this->buffer .= $character;
}
} else {
$this->buffer .= $text[$i];
}
break;
break;
}

}

$this->filterLine(true);
Expand Down Expand Up @@ -320,15 +338,16 @@ protected function changeDelimiters($text, $index)
/**
* Test whether it's time to change tags.
*
* @param string $tag Current tag name
* @param string $text Mustache template source
* @param int $index Current tokenizer index
* @param string $tag Current tag name
* @param string $text Mustache template source
* @param int $index Current tokenizer index
* @param int $tagLength Length of the opening/closing tag string
*
* @return boolean True if this is a closing section tag
*/
protected function tagChange($tag, $text, $index)
protected function tagChange($tag, $text, $index, $tagLength)
{
return substr($text, $index, strlen($tag)) === $tag;
return substr($text, $index, $tagLength) === $tag;
}

}