First release of open-appsec source code

components/security_apps/waap/waap_clib/ParserHTML.h

@@ -0,0 +1,84 @@
+// Copyright (C) 2022 Check Point Software Technologies Ltd. All rights reserved.
+
+// Licensed under the Apache License, Version 2.0 (the "License");
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+#pragma once
+
+#include "ParserBase.h"
+#include "KeyStack.h"
+#include <libxml/xmlstring.h>
+#include <libxml/xmlerror.h>
+#include <libxml/parser.h>
+#include <libxml/HTMLparser.h>
+
+#define FIRST_BUFFER_SIZE 5 // must buffer at least 4 first bytes to allow unicode autodetection (BOM).
+
+class ParserHTML : public ParserBase {
+public:
+    ParserHTML(IParserStreamReceiver &receiver);
+    virtual ~ParserHTML();
+    size_t push(const char *data, size_t data_len);
+    void finish();
+    virtual const std::string &name() const;
+    bool error() const;
+    virtual size_t depth() { return (m_key.depth() > 0) ? m_key.depth()-1 : m_key.depth(); }
+private:
+    enum state {
+        s_start,
+        s_accumulate_first_bytes,
+        s_start_parsing,
+        s_parsing,
+        s_error
+    };
+
+    // Information tracked per each element in current stack of tracked HTML elements
+    struct ElemTrackInfo {
+        std::string value;
+        bool hasChildren;
+        ElemTrackInfo():hasChildren(false) {
+            // when element is just opened - we still didn't see any children,
+            // hence start with the "hasChildren" flag as false.
+            // This flag will be enabled once we meet opening of the a subelement.
+            // Also, we start from empty value string and gradually append to it each
+            // time we receive next piece of text from HTML parser.
+            // The collected value is then emitted when element finishes.
+        }
+    };
+
+    static void onStartElement(
+        void *ctx,
+        const xmlChar *localname,
+        const xmlChar **attributes);
+    static void onEndElement(
+        void* ctx,
+        const xmlChar* localname);
+    static void onCharacters(
+        void *ctx,
+        const xmlChar *ch,
+        int len);
+
+    // Filter out errors that should be ignored. Returns true if error should be treated,
+    // false if an error should be ignored
+    bool filterErrors(xmlErrorPtr xmlError);
+
+    IParserStreamReceiver &m_receiver;
+    enum state m_state;
+    // buffer first few bytes of stream (required before calling SAX parser for the first time)
+    char m_buf[FIRST_BUFFER_SIZE];
+    int m_bufLen;
+    KeyStack m_key;
+    std::vector<ElemTrackInfo> m_elemTrackStack;
+    htmlSAXHandler m_saxHandler;
+    htmlParserCtxtPtr m_pushParserCtxPtr;
+
+    static const std::string m_parserName;
+};