/**
* Parses YAML into a PHP array.
*
* The parse method, when supplied with a YAML stream (string or file),
* will do its best to convert YAML in a file into a PHP array.
*
* Usage:
* <code>
* $array = Yaml::parse('config.yml');
* print_r($array);
* </code>
*
* As this method accepts both plain strings and file names as an input,
* you must validate the input before calling this method. Passing a file
* as an input is a deprecated feature and will be removed in 3.0.
*
* @param string $input Path to a YAML file or a string containing YAML
* @param Boolean $exceptionOnInvalidType True if an exception must be thrown on invalid types false otherwise
* @param Boolean $objectSupport True if object support is enabled, false otherwise
*
* @return array The YAML converted to a PHP array
*
* @throws ParseException If the YAML is not valid
*
* @api
*/
public static function parse($input, $exceptionOnInvalidType = false, $objectSupport = false)
{
// if input is a file, process it
$file = '';
if (strpos($input, "\n") === false && is_file($input)) {
if (false === is_readable($input)) {
throw new ParseException(sprintf('Unable to parse "%s" as the file is not readable.', $input));
}
$file = $input;
$input = file_get_contents($file);
}
$yaml = new Parser();
try {
return $yaml->parse($input, $exceptionOnInvalidType, $objectSupport);
} catch (ParseException $e) {
if ($file) {
$e->setParsedFile($file);
}
throw $e;
}
}
/**
* Parses a YAML string to a PHP value.
*
* @param string $value A YAML string
* @param Boolean $exceptionOnInvalidType true if an exception must be thrown on invalid types (a PHP resource or object), false otherwise
* @param Boolean $objectSupport true if object support is enabled, false otherwise
*
* @return mixed A PHP value
*
* @throws ParseException If the YAML is not valid
*/
public function parse($value, $exceptionOnInvalidType = false, $objectSupport = false)
{
$this->currentLineNb = -1;
$this->currentLine = '';
$this->lines = explode("\n", $this->cleanup($value));
if (function_exists('mb_detect_encoding') && false === mb_detect_encoding($value, 'UTF-8', true)) {
throw new ParseException('The YAML value does not appear to be valid UTF-8.');
}
if (function_exists('mb_internal_encoding') && (int) ini_get('mbstring.func_overload') & 2) {
$mbEncoding = mb_internal_encoding();
mb_internal_encoding('UTF-8');
}
$data = array();
$context = null;
while ($this->moveToNextLine()) {
if ($this->isCurrentLineEmpty()) {
continue;
}
// tab?
if ("\t" === $this->currentLine[0]) {
throw new ParseException('A YAML file cannot contain tabs as indentation.', $this->getRealCurrentLineNb() + 1, $this->currentLine);
}
$isRef = $isInPlace = $isProcessed = false;
if (preg_match('#^\\-((?P<leadspaces>\\s+)(?P<value>.+?))?\\s*$#u', $this->currentLine, $values)) {
if ($context && 'mapping' == $context) {
throw new ParseException('You cannot define a sequence item when in a mapping');
}
$context = 'sequence';
if (isset($values['value']) && preg_match('#^&(?P<ref>[^ ]+) *(?P<value>.*)#u', $values['value'], $matches)) {
$isRef = $matches['ref'];
$values['value'] = $matches['value'];
}
// array
if (!isset($values['value']) || '' == trim($values['value'], ' ') || 0 === strpos(ltrim($values['value'], ' '), '#')) {
$c = $this->getRealCurrentLineNb() + 1;
$parser = new Parser($c);
$parser->refs =& $this->refs;
$data[] = $parser->parse($this->getNextEmbedBlock(), $exceptionOnInvalidType, $objectSupport);
} else {
if (isset($values['leadspaces']) && ' ' == $values['leadspaces'] && preg_match('#^(?P<key>' . Inline::REGEX_QUOTED_STRING . '|[^ \'"\\{\\[].*?) *\\:(\\s+(?P<value>.+?))?\\s*$#u', $values['value'], $matches)) {
// this is a compact notation element, add to next block and parse
$c = $this->getRealCurrentLineNb();
$parser = new Parser($c);
$parser->refs =& $this->refs;
$block = $values['value'];
if ($this->isNextLineIndented()) {
$block .= "\n" . $this->getNextEmbedBlock($this->getCurrentLineIndentation() + 2);
}
$data[] = $parser->parse($block, $exceptionOnInvalidType, $objectSupport);
} else {
$data[] = $this->parseValue($values['value'], $exceptionOnInvalidType, $objectSupport);
}
}
} elseif (preg_match('#^(?P<key>' . Inline::REGEX_QUOTED_STRING . '|[^ \'"\\[\\{].*?) *\\:(\\s+(?P<value>.+?))?\\s*$#u', $this->currentLine, $values) && false === strpos($values['key'], ' #')) {
if ($context && 'sequence' == $context) {
throw new ParseException('You cannot define a mapping item when in a sequence');
}
$context = 'mapping';
// force correct settings
Inline::parse(null, $exceptionOnInvalidType, $objectSupport);
try {
$key = Inline::parseScalar($values['key']);
} catch (ParseException $e) {
$e->setParsedLine($this->getRealCurrentLineNb() + 1);
$e->setSnippet($this->currentLine);
throw $e;
}
if ('<<' === $key) {
if (isset($values['value']) && 0 === strpos($values['value'], '*')) {
$isInPlace = substr($values['value'], 1);
if (!array_key_exists($isInPlace, $this->refs)) {
throw new ParseException(sprintf('Reference "%s" does not exist.', $isInPlace), $this->getRealCurrentLineNb() + 1, $this->currentLine);
}
} else {
if (isset($values['value']) && $values['value'] !== '') {
$value = $values['value'];
} else {
$value = $this->getNextEmbedBlock();
}
$c = $this->getRealCurrentLineNb() + 1;
$parser = new Parser($c);
$parser->refs =& $this->refs;
$parsed = $parser->parse($value, $exceptionOnInvalidType, $objectSupport);
$merged = array();
if (!is_array($parsed)) {
throw new ParseException('YAML merge keys used with a scalar value instead of an array.', $this->getRealCurrentLineNb() + 1, $this->currentLine);
} elseif (isset($parsed[0])) {
// Numeric array, merge individual elements
foreach (array_reverse($parsed) as $parsedItem) {
if (!is_array($parsedItem)) {
throw new ParseException('Merge items must be arrays.', $this->getRealCurrentLineNb() + 1, $parsedItem);
}
$merged = array_merge($parsedItem, $merged);
}
} else {
// Associative array, merge
$merged = array_merge($merged, $parsed);
}
$isProcessed = $merged;
}
} elseif (isset($values['value']) && preg_match('#^&(?P<ref>[^ ]+) *(?P<value>.*)#u', $values['value'], $matches)) {
$isRef = $matches['ref'];
$values['value'] = $matches['value'];
}
if ($isProcessed) {
// Merge keys
$data = $isProcessed;
// hash
} elseif (!isset($values['value']) || '' == trim($values['value'], ' ') || 0 === strpos(ltrim($values['value'], ' '), '#')) {
// if next line is less indented or equal, then it means that the current value is null
if (!$this->isNextLineIndented() && !$this->isNextLineUnIndentedCollection()) {
$data[$key] = null;
} else {
$c = $this->getRealCurrentLineNb() + 1;
$parser = new Parser($c);
$parser->refs =& $this->refs;
$data[$key] = $parser->parse($this->getNextEmbedBlock(), $exceptionOnInvalidType, $objectSupport);
}
} else {
if ($isInPlace) {
$data = $this->refs[$isInPlace];
} else {
$data[$key] = $this->parseValue($values['value'], $exceptionOnInvalidType, $objectSupport);
}
}
} else {
// 1-liner optionally followed by newline
$lineCount = count($this->lines);
if (1 === $lineCount || 2 === $lineCount && empty($this->lines[1])) {
try {
$value = Inline::parse($this->lines[0], $exceptionOnInvalidType, $objectSupport);
} catch (ParseException $e) {
$e->setParsedLine($this->getRealCurrentLineNb() + 1);
$e->setSnippet($this->currentLine);
throw $e;
}
if (is_array($value)) {
$first = reset($value);
if (is_string($first) && 0 === strpos($first, '*')) {
$data = array();
foreach ($value as $alias) {
$data[] = $this->refs[substr($alias, 1)];
}
$value = $data;
}
}
if (isset($mbEncoding)) {
mb_internal_encoding($mbEncoding);
}
return $value;
}
switch (preg_last_error()) {
case PREG_INTERNAL_ERROR:
$error = 'Internal PCRE error.';
break;
case PREG_BACKTRACK_LIMIT_ERROR:
$error = 'pcre.backtrack_limit reached.';
break;
case PREG_RECURSION_LIMIT_ERROR:
$error = 'pcre.recursion_limit reached.';
break;
case PREG_BAD_UTF8_ERROR:
$error = 'Malformed UTF-8 data.';
break;
case PREG_BAD_UTF8_OFFSET_ERROR:
$error = 'Offset doesn\'t correspond to the begin of a valid UTF-8 code point.';
break;
default:
$error = 'Unable to parse.';
}
throw new ParseException($error, $this->getRealCurrentLineNb() + 1, $this->currentLine);
}
if ($isRef) {
$this->refs[$isRef] = end($data);
}
}
if (isset($mbEncoding)) {
mb_internal_encoding($mbEncoding);
}
return empty($data) ? null : $data;
}
