src/http/one/Parser.h

   1 #ifndef _SQUID_SRC_HTTP_ONE_PARSER_H
   2 #define _SQUID_SRC_HTTP_ONE_PARSER_H
   3
   4 #include "anyp/ProtocolVersion.h"
   5 #include "http/one/forward.h"
   6 #include "SBuf.h"
   7
   8 namespace Http {
   9 namespace One {
  10
  11 // Parser states
  12 enum ParseState {
  13     HTTP_PARSE_NONE,     ///< initialized, but nothing usefully parsed yet
  14     HTTP_PARSE_FIRST,    ///< HTTP/1 message first-line
  15     HTTP_PARSE_MIME,     ///< HTTP/1 mime-header block
  16     HTTP_PARSE_DONE      ///< parsed a message header, or reached a terminal syntax error
  17 };
  18
  19 /** HTTP/1.x protocol parser
  20  *
  21  * Works on a raw character I/O buffer and tokenizes the content into
  22  * the major CRLF delimited segments of an HTTP/1 procotol message:
  23  *
  24  * \item first-line (request-line / simple-request / status-line)
  25  * \item mime-header 0*( header-name ':' SP field-value CRLF)
  26  */
  27 class Parser : public RefCountable
  28 {
  29     explicit Parser(const Parser&); // do not implement
  30     Parser& operator =(const Parser&); // do not implement
  31
  32 public:
  33     typedef SBuf::size_type size_type;
  34
  35     Parser() { clear(); }
  36     virtual ~Parser() {}
  37
  38     /// Set this parser back to a default state.
  39     /// Will DROP any reference to a buffer (does not free).
  40     virtual void clear();
  41
  42     /// attempt to parse a message from the buffer
  43     /// \retval true if a full message was found and parsed
  44     /// \retval false if incomplete, invalid or no message was found
  45     virtual bool parse(const SBuf &aBuf) = 0;
  46
  47     /** Whether the parser is waiting on more data to complete parsing a message.
  48      * Use to distinguish between incomplete data and error results
  49      * when parse() returns false.
  50      */
  51     bool needsMoreData() const {return parsingStage_!=HTTP_PARSE_DONE;}
  52
  53     /// size in bytes of the first line including CRLF terminator
  54     virtual size_type firstLineSize() const = 0;
  55
  56     /// size in bytes of the message headers including CRLF terminator(s)
  57     /// but excluding first-line bytes
  58     size_type headerBlockSize() const {return mimeHeaderBlock_.length();}
  59
  60     /// size in bytes of HTTP message block, includes first-line and mime headers
  61     /// excludes any body/entity/payload bytes
  62     /// excludes any garbage prefix before the first-line
  63     size_type messageHeaderSize() const {return firstLineSize() + headerBlockSize();}
  64
  65     /// buffer containing HTTP mime headers, excluding message first-line.
  66     SBuf mimeHeader() const {return mimeHeaderBlock_;}
  67
  68     /// the protocol label for this message
  69     const AnyP::ProtocolVersion & messageProtocol() const {return msgProtocol_;}
  70
  71     /**
  72      * Scan the mime header block (badly) for a header with teh given name.
  73      *
  74      * BUG: omits lines when searching for headers with obs-fold or multiple entries.
  75      *
  76      * BUG: limits output to just 1KB when Squid accepts up to 64KB line length.
  77      *
  78      * \return A pointer to a field-value of the first matching field-name, or NULL.
  79      */
  80     char *getHeaderField(const char *name);
  81
  82     /// the remaining unprocessed section of buffer
  83     const SBuf &remaining() const {return buf_;}
  84
  85 protected:
  86     /// RFC 7230 section 2.6 - 7 magic octets
  87     static const SBuf Http1magic;
  88
  89     /// bytes remaining to be parsed
  90     SBuf buf_;
  91
  92     /// what stage the parser is currently up to
  93     ParseState parsingStage_;
  94
  95     /// what protocol label has been found in the first line (if any)
  96     AnyP::ProtocolVersion msgProtocol_;
  97
  98     /// buffer holding the mime headers (if any)
  99     SBuf mimeHeaderBlock_;
 100 };
 101
 102 } // namespace One
 103 } // namespace Http
 104
 105 #endif /*  _SQUID_SRC_HTTP_ONE_PARSER_H */