src/http/one/Parser.h

   1 /*
   2  * Copyright (C) 1996-2015 The Squid Software Foundation and contributors
   3  *
   4  * Squid software is distributed under GPLv2+ license and includes
   5  * contributions from numerous individuals and organizations.
   6  * Please see the COPYING and CONTRIBUTORS files for details.
   7  */
   8
   9 #ifndef _SQUID_SRC_HTTP_ONE_PARSER_H
  10 #define _SQUID_SRC_HTTP_ONE_PARSER_H
  11
  12 #include "anyp/ProtocolVersion.h"
  13 #include "http/one/forward.h"
  14 #include "http/StatusCode.h"
  15 #include "SBuf.h"
  16
  17 namespace Parser {
  18 class Tokenizer;
  19 }
  20
  21 namespace Http {
  22 namespace One {
  23
  24 // Parser states
  25 enum ParseState {
  26     HTTP_PARSE_NONE,     ///< initialized, but nothing usefully parsed yet
  27     HTTP_PARSE_FIRST,    ///< HTTP/1 message first-line
  28     HTTP_PARSE_MIME,     ///< HTTP/1 mime-header block
  29     HTTP_PARSE_DONE      ///< parsed a message header, or reached a terminal syntax error
  30 };
  31
  32 /** HTTP/1.x protocol parser
  33  *
  34  * Works on a raw character I/O buffer and tokenizes the content into
  35  * the major CRLF delimited segments of an HTTP/1 procotol message:
  36  *
  37  * \item first-line (request-line / simple-request / status-line)
  38  * \item mime-header 0*( header-name ':' SP field-value CRLF)
  39  */
  40 class Parser : public RefCountable
  41 {
  42 public:
  43     typedef SBuf::size_type size_type;
  44
  45     Parser() : parseStatusCode(Http::scNone), parsingStage_(HTTP_PARSE_NONE) {}
  46     virtual ~Parser() {}
  47
  48     /// Set this parser back to a default state.
  49     /// Will DROP any reference to a buffer (does not free).
  50     virtual void clear() = 0;
  51
  52     /// attempt to parse a message from the buffer
  53     /// \retval true if a full message was found and parsed
  54     /// \retval false if incomplete, invalid or no message was found
  55     virtual bool parse(const SBuf &aBuf) = 0;
  56
  57     /** Whether the parser is waiting on more data to complete parsing a message.
  58      * Use to distinguish between incomplete data and error results
  59      * when parse() returns false.
  60      */
  61     bool needsMoreData() const {return parsingStage_!=HTTP_PARSE_DONE;}
  62
  63     /// size in bytes of the first line including CRLF terminator
  64     virtual size_type firstLineSize() const = 0;
  65
  66     /// size in bytes of the message headers including CRLF terminator(s)
  67     /// but excluding first-line bytes
  68     size_type headerBlockSize() const {return mimeHeaderBlock_.length();}
  69
  70     /// size in bytes of HTTP message block, includes first-line and mime headers
  71     /// excludes any body/entity/payload bytes
  72     /// excludes any garbage prefix before the first-line
  73     size_type messageHeaderSize() const {return firstLineSize() + headerBlockSize();}
  74
  75     /// buffer containing HTTP mime headers, excluding message first-line.
  76     SBuf mimeHeader() const {return mimeHeaderBlock_;}
  77
  78     /// the protocol label for this message
  79     const AnyP::ProtocolVersion & messageProtocol() const {return msgProtocol_;}
  80
  81     /**
  82      * Scan the mime header block (badly) for a header with the given name.
  83      *
  84      * BUG: omits lines when searching for headers with obs-fold or multiple entries.
  85      *
  86      * BUG: limits output to just 1KB when Squid accepts up to 64KB line length.
  87      *
  88      * \return A pointer to a field-value of the first matching field-name, or NULL.
  89      */
  90     char *getHeaderField(const char *name);
  91
  92     /// the remaining unprocessed section of buffer
  93     const SBuf &remaining() const {return buf_;}
  94
  95     /**
  96      * HTTP status code resulting from the parse process.
  97      * to be used on the invalid message handling.
  98      *
  99      * Http::scNone indicates incomplete parse,
 100      * Http::scOkay indicates no error,
 101      * other codes represent a parse error.
 102      */
 103     Http::StatusCode parseStatusCode;
 104
 105 protected:
 106     /// detect and skip the CRLF or (if tolerant) LF line terminator
 107     /// consume from the tokenizer and return true only if found
 108     bool skipLineTerminator(::Parser::Tokenizer &tok) const;
 109
 110     /**
 111      * Scan to find the mime headers block for current message.
 112      *
 113      * \retval true   If mime block (or a blocks non-existence) has been
 114      *                identified accurately within limit characters.
 115      *                mimeHeaderBlock_ has been updated and buf_ consumed.
 116      *
 117      * \retval false  An error occured, or no mime terminator found within limit.
 118      */
 119     bool grabMimeBlock(const char *which, const size_t limit);
 120
 121     /// RFC 7230 section 2.6 - 7 magic octets
 122     static const SBuf Http1magic;
 123
 124     /// bytes remaining to be parsed
 125     SBuf buf_;
 126
 127     /// what stage the parser is currently up to
 128     ParseState parsingStage_;
 129
 130     /// what protocol label has been found in the first line (if any)
 131     AnyP::ProtocolVersion msgProtocol_;
 132
 133     /// buffer holding the mime headers (if any)
 134     SBuf mimeHeaderBlock_;
 135 };
 136
 137 } // namespace One
 138 } // namespace Http
 139
 140 #endif /*  _SQUID_SRC_HTTP_ONE_PARSER_H */
 141