From c21a462d52bd32737c32c29b060da03b38f1c2e6 Mon Sep 17 00:00:00 2001 From: Andrew Dolgov Date: Fri, 1 Jun 2012 00:07:59 +0400 Subject: remove htmlpurifier --- .../library/HTMLPurifier/Strategy/Composite.php | 23 - .../library/HTMLPurifier/Strategy/Core.php | 18 - .../library/HTMLPurifier/Strategy/FixNesting.php | 328 ------------- .../HTMLPurifier/Strategy/MakeWellFormed.php | 532 --------------------- .../Strategy/RemoveForeignElements.php | 188 -------- .../HTMLPurifier/Strategy/ValidateAttributes.php | 39 -- 6 files changed, 1128 deletions(-) delete mode 100644 lib/htmlpurifier/library/HTMLPurifier/Strategy/Composite.php delete mode 100644 lib/htmlpurifier/library/HTMLPurifier/Strategy/Core.php delete mode 100644 lib/htmlpurifier/library/HTMLPurifier/Strategy/FixNesting.php delete mode 100644 lib/htmlpurifier/library/HTMLPurifier/Strategy/MakeWellFormed.php delete mode 100644 lib/htmlpurifier/library/HTMLPurifier/Strategy/RemoveForeignElements.php delete mode 100644 lib/htmlpurifier/library/HTMLPurifier/Strategy/ValidateAttributes.php (limited to 'lib/htmlpurifier/library/HTMLPurifier/Strategy') diff --git a/lib/htmlpurifier/library/HTMLPurifier/Strategy/Composite.php b/lib/htmlpurifier/library/HTMLPurifier/Strategy/Composite.php deleted file mode 100644 index 92aefd33e..000000000 --- a/lib/htmlpurifier/library/HTMLPurifier/Strategy/Composite.php +++ /dev/null @@ -1,23 +0,0 @@ -strategies as $strategy) { - $tokens = $strategy->execute($tokens, $config, $context); - } - return $tokens; - } - -} - -// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Strategy/Core.php b/lib/htmlpurifier/library/HTMLPurifier/Strategy/Core.php deleted file mode 100644 index d90e15860..000000000 --- a/lib/htmlpurifier/library/HTMLPurifier/Strategy/Core.php +++ /dev/null @@ -1,18 +0,0 @@ -strategies[] = new HTMLPurifier_Strategy_RemoveForeignElements(); - $this->strategies[] = new HTMLPurifier_Strategy_MakeWellFormed(); - $this->strategies[] = new HTMLPurifier_Strategy_FixNesting(); - $this->strategies[] = new HTMLPurifier_Strategy_ValidateAttributes(); - } - -} - -// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Strategy/FixNesting.php b/lib/htmlpurifier/library/HTMLPurifier/Strategy/FixNesting.php deleted file mode 100644 index f81802391..000000000 --- a/lib/htmlpurifier/library/HTMLPurifier/Strategy/FixNesting.php +++ /dev/null @@ -1,328 +0,0 @@ -getHTMLDefinition(); - - // insert implicit "parent" node, will be removed at end. - // DEFINITION CALL - $parent_name = $definition->info_parent; - array_unshift($tokens, new HTMLPurifier_Token_Start($parent_name)); - $tokens[] = new HTMLPurifier_Token_End($parent_name); - - // setup the context variable 'IsInline', for chameleon processing - // is 'false' when we are not inline, 'true' when it must always - // be inline, and an integer when it is inline for a certain - // branch of the document tree - $is_inline = $definition->info_parent_def->descendants_are_inline; - $context->register('IsInline', $is_inline); - - // setup error collector - $e =& $context->get('ErrorCollector', true); - - //####################################################################// - // Loop initialization - - // stack that contains the indexes of all parents, - // $stack[count($stack)-1] being the current parent - $stack = array(); - - // stack that contains all elements that are excluded - // it is organized by parent elements, similar to $stack, - // but it is only populated when an element with exclusions is - // processed, i.e. there won't be empty exclusions. - $exclude_stack = array(); - - // variable that contains the start token while we are processing - // nodes. This enables error reporting to do its job - $start_token = false; - $context->register('CurrentToken', $start_token); - - //####################################################################// - // Loop - - // iterate through all start nodes. Determining the start node - // is complicated so it has been omitted from the loop construct - for ($i = 0, $size = count($tokens) ; $i < $size; ) { - - //################################################################// - // Gather information on children - - // child token accumulator - $child_tokens = array(); - - // scroll to the end of this node, report number, and collect - // all children - for ($j = $i, $depth = 0; ; $j++) { - if ($tokens[$j] instanceof HTMLPurifier_Token_Start) { - $depth++; - // skip token assignment on first iteration, this is the - // token we currently are on - if ($depth == 1) continue; - } elseif ($tokens[$j] instanceof HTMLPurifier_Token_End) { - $depth--; - // skip token assignment on last iteration, this is the - // end token of the token we're currently on - if ($depth == 0) break; - } - $child_tokens[] = $tokens[$j]; - } - - // $i is index of start token - // $j is index of end token - - $start_token = $tokens[$i]; // to make token available via CurrentToken - - //################################################################// - // Gather information on parent - - // calculate parent information - if ($count = count($stack)) { - $parent_index = $stack[$count-1]; - $parent_name = $tokens[$parent_index]->name; - if ($parent_index == 0) { - $parent_def = $definition->info_parent_def; - } else { - $parent_def = $definition->info[$parent_name]; - } - } else { - // processing as if the parent were the "root" node - // unknown info, it won't be used anyway, in the future, - // we may want to enforce one element only (this is - // necessary for HTML Purifier to clean entire documents - $parent_index = $parent_name = $parent_def = null; - } - - // calculate context - if ($is_inline === false) { - // check if conditions make it inline - if (!empty($parent_def) && $parent_def->descendants_are_inline) { - $is_inline = $count - 1; - } - } else { - // check if we're out of inline - if ($count === $is_inline) { - $is_inline = false; - } - } - - //################################################################// - // Determine whether element is explicitly excluded SGML-style - - // determine whether or not element is excluded by checking all - // parent exclusions. The array should not be very large, two - // elements at most. - $excluded = false; - if (!empty($exclude_stack)) { - foreach ($exclude_stack as $lookup) { - if (isset($lookup[$tokens[$i]->name])) { - $excluded = true; - // no need to continue processing - break; - } - } - } - - //################################################################// - // Perform child validation - - if ($excluded) { - // there is an exclusion, remove the entire node - $result = false; - $excludes = array(); // not used, but good to initialize anyway - } else { - // DEFINITION CALL - if ($i === 0) { - // special processing for the first node - $def = $definition->info_parent_def; - } else { - $def = $definition->info[$tokens[$i]->name]; - - } - - if (!empty($def->child)) { - // have DTD child def validate children - $result = $def->child->validateChildren( - $child_tokens, $config, $context); - } else { - // weird, no child definition, get rid of everything - $result = false; - } - - // determine whether or not this element has any exclusions - $excludes = $def->excludes; - } - - // $result is now a bool or array - - //################################################################// - // Process result by interpreting $result - - if ($result === true || $child_tokens === $result) { - // leave the node as is - - // register start token as a parental node start - $stack[] = $i; - - // register exclusions if there are any - if (!empty($excludes)) $exclude_stack[] = $excludes; - - // move cursor to next possible start node - $i++; - - } elseif($result === false) { - // remove entire node - - if ($e) { - if ($excluded) { - $e->send(E_ERROR, 'Strategy_FixNesting: Node excluded'); - } else { - $e->send(E_ERROR, 'Strategy_FixNesting: Node removed'); - } - } - - // calculate length of inner tokens and current tokens - $length = $j - $i + 1; - - // perform removal - array_splice($tokens, $i, $length); - - // update size - $size -= $length; - - // there is no start token to register, - // current node is now the next possible start node - // unless it turns out that we need to do a double-check - - // this is a rought heuristic that covers 100% of HTML's - // cases and 99% of all other cases. A child definition - // that would be tricked by this would be something like: - // ( | a b c) where it's all or nothing. Fortunately, - // our current implementation claims that that case would - // not allow empty, even if it did - if (!$parent_def->child->allow_empty) { - // we need to do a double-check - $i = $parent_index; - array_pop($stack); - } - - // PROJECTED OPTIMIZATION: Process all children elements before - // reprocessing parent node. - - } else { - // replace node with $result - - // calculate length of inner tokens - $length = $j - $i - 1; - - if ($e) { - if (empty($result) && $length) { - $e->send(E_ERROR, 'Strategy_FixNesting: Node contents removed'); - } else { - $e->send(E_WARNING, 'Strategy_FixNesting: Node reorganized'); - } - } - - // perform replacement - array_splice($tokens, $i + 1, $length, $result); - - // update size - $size -= $length; - $size += count($result); - - // register start token as a parental node start - $stack[] = $i; - - // register exclusions if there are any - if (!empty($excludes)) $exclude_stack[] = $excludes; - - // move cursor to next possible start node - $i++; - - } - - //################################################################// - // Scroll to next start node - - // We assume, at this point, that $i is the index of the token - // that is the first possible new start point for a node. - - // Test if the token indeed is a start tag, if not, move forward - // and test again. - $size = count($tokens); - while ($i < $size and !$tokens[$i] instanceof HTMLPurifier_Token_Start) { - if ($tokens[$i] instanceof HTMLPurifier_Token_End) { - // pop a token index off the stack if we ended a node - array_pop($stack); - // pop an exclusion lookup off exclusion stack if - // we ended node and that node had exclusions - if ($i == 0 || $i == $size - 1) { - // use specialized var if it's the super-parent - $s_excludes = $definition->info_parent_def->excludes; - } else { - $s_excludes = $definition->info[$tokens[$i]->name]->excludes; - } - if ($s_excludes) { - array_pop($exclude_stack); - } - } - $i++; - } - - } - - //####################################################################// - // Post-processing - - // remove implicit parent tokens at the beginning and end - array_shift($tokens); - array_pop($tokens); - - // remove context variables - $context->destroy('IsInline'); - $context->destroy('CurrentToken'); - - //####################################################################// - // Return - - return $tokens; - - } - -} - -// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Strategy/MakeWellFormed.php b/lib/htmlpurifier/library/HTMLPurifier/Strategy/MakeWellFormed.php deleted file mode 100644 index c7aa1bb86..000000000 --- a/lib/htmlpurifier/library/HTMLPurifier/Strategy/MakeWellFormed.php +++ /dev/null @@ -1,532 +0,0 @@ -getHTMLDefinition(); - - // local variables - $generator = new HTMLPurifier_Generator($config, $context); - $escape_invalid_tags = $config->get('Core.EscapeInvalidTags'); - // used for autoclose early abortion - $global_parent_allowed_elements = array(); - if (isset($definition->info[$definition->info_parent])) { - // may be unset under testing circumstances - $global_parent_allowed_elements = $definition->info[$definition->info_parent]->child->getAllowedElements($config); - } - $e = $context->get('ErrorCollector', true); - $t = false; // token index - $i = false; // injector index - $token = false; // the current token - $reprocess = false; // whether or not to reprocess the same token - $stack = array(); - - // member variables - $this->stack =& $stack; - $this->t =& $t; - $this->tokens =& $tokens; - $this->config = $config; - $this->context = $context; - - // context variables - $context->register('CurrentNesting', $stack); - $context->register('InputIndex', $t); - $context->register('InputTokens', $tokens); - $context->register('CurrentToken', $token); - - // -- begin INJECTOR -- - - $this->injectors = array(); - - $injectors = $config->getBatch('AutoFormat'); - $def_injectors = $definition->info_injector; - $custom_injectors = $injectors['Custom']; - unset($injectors['Custom']); // special case - foreach ($injectors as $injector => $b) { - // XXX: Fix with a legitimate lookup table of enabled filters - if (strpos($injector, '.') !== false) continue; - $injector = "HTMLPurifier_Injector_$injector"; - if (!$b) continue; - $this->injectors[] = new $injector; - } - foreach ($def_injectors as $injector) { - // assumed to be objects - $this->injectors[] = $injector; - } - foreach ($custom_injectors as $injector) { - if (!$injector) continue; - if (is_string($injector)) { - $injector = "HTMLPurifier_Injector_$injector"; - $injector = new $injector; - } - $this->injectors[] = $injector; - } - - // give the injectors references to the definition and context - // variables for performance reasons - foreach ($this->injectors as $ix => $injector) { - $error = $injector->prepare($config, $context); - if (!$error) continue; - array_splice($this->injectors, $ix, 1); // rm the injector - trigger_error("Cannot enable {$injector->name} injector because $error is not allowed", E_USER_WARNING); - } - - // -- end INJECTOR -- - - // a note on reprocessing: - // In order to reduce code duplication, whenever some code needs - // to make HTML changes in order to make things "correct", the - // new HTML gets sent through the purifier, regardless of its - // status. This means that if we add a start token, because it - // was totally necessary, we don't have to update nesting; we just - // punt ($reprocess = true; continue;) and it does that for us. - - // isset is in loop because $tokens size changes during loop exec - for ( - $t = 0; - $t == 0 || isset($tokens[$t - 1]); - // only increment if we don't need to reprocess - $reprocess ? $reprocess = false : $t++ - ) { - - // check for a rewind - if (is_int($i) && $i >= 0) { - // possibility: disable rewinding if the current token has a - // rewind set on it already. This would offer protection from - // infinite loop, but might hinder some advanced rewinding. - $rewind_to = $this->injectors[$i]->getRewind(); - if (is_int($rewind_to) && $rewind_to < $t) { - if ($rewind_to < 0) $rewind_to = 0; - while ($t > $rewind_to) { - $t--; - $prev = $tokens[$t]; - // indicate that other injectors should not process this token, - // but we need to reprocess it - unset($prev->skip[$i]); - $prev->rewind = $i; - if ($prev instanceof HTMLPurifier_Token_Start) array_pop($this->stack); - elseif ($prev instanceof HTMLPurifier_Token_End) $this->stack[] = $prev->start; - } - } - $i = false; - } - - // handle case of document end - if (!isset($tokens[$t])) { - // kill processing if stack is empty - if (empty($this->stack)) break; - - // peek - $top_nesting = array_pop($this->stack); - $this->stack[] = $top_nesting; - - // send error [TagClosedSuppress] - if ($e && !isset($top_nesting->armor['MakeWellFormed_TagClosedError'])) { - $e->send(E_NOTICE, 'Strategy_MakeWellFormed: Tag closed by document end', $top_nesting); - } - - // append, don't splice, since this is the end - $tokens[] = new HTMLPurifier_Token_End($top_nesting->name); - - // punt! - $reprocess = true; - continue; - } - - $token = $tokens[$t]; - - //echo '
'; printTokens($tokens, $t); printTokens($this->stack); - //flush(); - - // quick-check: if it's not a tag, no need to process - if (empty($token->is_tag)) { - if ($token instanceof HTMLPurifier_Token_Text) { - foreach ($this->injectors as $i => $injector) { - if (isset($token->skip[$i])) continue; - if ($token->rewind !== null && $token->rewind !== $i) continue; - $injector->handleText($token); - $this->processToken($token, $i); - $reprocess = true; - break; - } - } - // another possibility is a comment - continue; - } - - if (isset($definition->info[$token->name])) { - $type = $definition->info[$token->name]->child->type; - } else { - $type = false; // Type is unknown, treat accordingly - } - - // quick tag checks: anything that's *not* an end tag - $ok = false; - if ($type === 'empty' && $token instanceof HTMLPurifier_Token_Start) { - // claims to be a start tag but is empty - $token = new HTMLPurifier_Token_Empty($token->name, $token->attr, $token->line, $token->col, $token->armor); - $ok = true; - } elseif ($type && $type !== 'empty' && $token instanceof HTMLPurifier_Token_Empty) { - // claims to be empty but really is a start tag - $this->swap(new HTMLPurifier_Token_End($token->name)); - $this->insertBefore(new HTMLPurifier_Token_Start($token->name, $token->attr, $token->line, $token->col, $token->armor)); - // punt (since we had to modify the input stream in a non-trivial way) - $reprocess = true; - continue; - } elseif ($token instanceof HTMLPurifier_Token_Empty) { - // real empty token - $ok = true; - } elseif ($token instanceof HTMLPurifier_Token_Start) { - // start tag - - // ...unless they also have to close their parent - if (!empty($this->stack)) { - - // Performance note: you might think that it's rather - // inefficient, recalculating the autoclose information - // for every tag that a token closes (since when we - // do an autoclose, we push a new token into the - // stream and then /process/ that, before - // re-processing this token.) But this is - // necessary, because an injector can make an - // arbitrary transformations to the autoclosing - // tokens we introduce, so things may have changed - // in the meantime. Also, doing the inefficient thing is - // "easy" to reason about (for certain perverse definitions - // of "easy") - - $parent = array_pop($this->stack); - $this->stack[] = $parent; - - if (isset($definition->info[$parent->name])) { - $elements = $definition->info[$parent->name]->child->getAllowedElements($config); - $autoclose = !isset($elements[$token->name]); - } else { - $autoclose = false; - } - - if ($autoclose && $definition->info[$token->name]->wrap) { - // Check if an element can be wrapped by another - // element to make it valid in a context (for - // example,