/**
* @dataProvider getErroneousTestDocuments
*/
public function testParseErroneousRstFile($file, $message)
{
$tokenizer = new ezcDocumentRstTokenizer();
$parser = new ezcDocumentRstParser();
try {
$document = $parser->parse($tokenizer->tokenizeFile($file));
$this->fail('Expected ezcDocumentRstParserException.');
} catch (ezcDocumentParserException $e) {
$this->assertSame($message, $e->getMessage(), 'Different parse error expected.');
}
}
/**
* Parse directive token list with RST parser
*
* This method is intended to parse the token list, provided for the RST
* contents using the standard RST parser. It will be visited afterwards by
* the provided RST-visitor implementation.
*
* The method returns the created document as a DOMDocument. You normally
* need to use DOMDocument::importNode to embed the conatined nodes in your
* target document.
*
* @param array $tokens
* @param ezcDocumentRstVisitor $visitor
* @return DOMDocument
*/
protected function parseTokens(array $tokens, ezcDocumentRstVisitor $visitor)
{
$parser = new ezcDocumentRstParser();
$ast = $parser->parse($tokens);
$doc = $visitor->visit($ast, $this->path);
return $doc;
}
/**
* Validate the input string
*
* Validate the input string against the specification of the current
* document format.
*
* Returns true, if the validation succeded, and an array with
* ezcDocumentValidationError objects otherwise.
*
* @param string $string
* @return mixed
*/
public function validateString($string)
{
$tokenizer = new ezcDocumentRstTokenizer();
$parser = new ezcDocumentRstParser();
// Only halt on parse errors, and collect all other errors.
$parser->options->errorReporting = E_PARSE;
$errors = array();
$ast = null;
try {
// Try to parse the document and keep the parse tree for evetual
// checking for decoration errors
$ast = $parser->parse($tokenizer->tokenizeString($string));
} catch (ezcDocumentParserException $e) {
$errors[] = $e;
}
// Get errors and notices from parsed document
$errors = array_merge($errors, $parser->errors);
// If we had no parse error until now, we also try to decorate the
// document, which may leed to another class of errors.
if ($ast !== null) {
$oldErrorReporting = $this->options->errorReporting;
$this->options->errorReporting = E_PARSE;
try {
$visitor = new ezcDocumentRstDocbookVisitor($this, $this->path);
$visitor->visit($ast, $this->path);
// Get errors and notices from parsed document
$errors = array_merge($errors, $visitor->getErrors());
} catch (ezcDocumentVisitException $e) {
$errors[] = $e;
}
// Reset error reporting
$this->options->errorReporting = $oldErrorReporting;
}
if (count($errors) === 0) {
// If no problem could be found, jsut return true
return true;
} else {
// Transform aggregated errors into validation errors
foreach ($errors as $nr => $error) {
$errors[$nr] = ezcDocumentValidationError::createFromException($error);
}
return $errors;
}
}
/**
* Reenter parser with a list of tokens
*
* Returns a parsed document created from the given tokens. With optional,
* but default, reindetation of the tokens relative to the first token.
*
* @param array $tokens
* @param bool $reindent
* @return ezcDocumentRstDocumentNode
*/
protected function reenterParser(array $tokens, $reindent = true)
{
if (count($tokens) < 1) {
return array();
}
// Fix indentation for all cell tokens, as they were a single document.
$fixedTokens = $reindent ? $this->realignTokens($tokens) : $tokens;
$parser = new ezcDocumentRstParser();
return $parser->parse($fixedTokens);
}
/**
* Reenter parser with a list of tokens
*
* Returns a parsed document created from the given tokens. With optional,
* but default, reindetation of the tokens relative to the first token.
*
* @param array $tokens
* @param bool $reindent
* @return ezcDocumentRstDocumentNode
*/
protected function reenterParser(array $tokens, $reindent = true)
{
if (count($tokens) < 1) {
return array();
}
/* DEBUG
static $c = 0;
file_put_contents( "tokens-reentered-$c.php", "<?php\n\n return " . var_export( $tokens, true ) . ";\n\n" );
// /DEBUG */
// Fix indentation for all cell tokens, as they were a single document.
if ($reindent) {
$firstToken = reset($tokens);
$offset = $firstToken->position + ($firstToken->type === ezcDocumentRstToken::WHITESPACE ? 0 : -1);
$fixedTokens = array();
foreach ($tokens as $nr => $token) {
if ($token->type === ezcDocumentRstToken::WHITESPACE && isset($tokens[$nr + 1]) && $tokens[$nr + 1]->type === ezcDocumentRstToken::WHITESPACE) {
// Skip multiple whitespace tokens in a row.
continue;
}
if ($token->type === ezcDocumentRstToken::WHITESPACE && $token->position <= $offset) {
if (strlen($token->content) <= 1) {
// Just skip token, completely out of tokens bounds
continue;
} else {
// Shorten starting whitespace token
$token = clone $token;
$token->position = 0;
$token->content = substr($token->content, 1);
$fixedTokens[] = $token;
}
} else {
$token = clone $token;
$token->position -= $offset;
$fixedTokens[] = $token;
}
}
// If required add a second newline, if the provided token array does
// not contain any newlines at the end.
if ($token->type !== ezcDocumentRstToken::NEWLINE) {
$fixedTokens[] = new ezcDocumentRstToken(ezcDocumentRstToken::NEWLINE, "\n", null, null);
}
} else {
$fixedTokens = $tokens;
}
$fixedTokens[] = new ezcDocumentRstToken(ezcDocumentRstToken::NEWLINE, "\n", null, null);
$fixedTokens[] = new ezcDocumentRstToken(ezcDocumentRstToken::EOF, null, null, null);
/* DEBUG
file_put_contents( "tokens-reentered-$c-fixed.php", "<?php\n\n return " . var_export( $fixedTokens, true ) . ";\n\n" );
++$c;
// /DEBUG */
$parser = new ezcDocumentRstParser();
return $parser->parse($fixedTokens);
}
