src/http/one/Tokenizer.h

   1 /*
   2  * Copyright (C) 1996-2017 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_TOKENIZER_H
  10 #define SQUID_SRC_HTTP_ONE_TOKENIZER_H
  11
  12 #include "parser/Tokenizer.h"
  13
  14 namespace Http {
  15 namespace One {
  16
  17 /**
  18  * Lexical processor extended to tokenize HTTP/1.x syntax.
  19  *
  20  * \see ::Parser::Tokenizer for more detail
  21  */
  22 class Tokenizer : public ::Parser::Tokenizer
  23 {
  24 public:
  25     Tokenizer(SBuf &s) : ::Parser::Tokenizer(s), savedStats_(0) {}
  26
  27     /**
  28      * Attempt to parse a quoted-string lexical construct.
  29      *
  30      * Governed by:
  31      *  - RFC 1945 section 2.1
  32      *  "
  33      *    A string of text is parsed as a single word if it is quoted using
  34      *    double-quote marks.
  35      *
  36      *        quoted-string  = ( <"> *(qdtext) <"> )
  37      *
  38      *        qdtext         = <any CHAR except <"> and CTLs,
  39      *                         but including LWS>
  40      *
  41      *    Single-character quoting using the backslash ("\") character is not
  42      *    permitted in HTTP/1.0.
  43      *  "
  44      *
  45      *  - RFC 7230 section 3.2.6
  46      *  "
  47      *    A string of text is parsed as a single value if it is quoted using
  48      *    double-quote marks.
  49      *
  50      *    quoted-string  = DQUOTE *( qdtext / quoted-pair ) DQUOTE
  51      *    qdtext         = HTAB / SP /%x21 / %x23-5B / %x5D-7E / obs-text
  52      *    obs-text       = %x80-FF
  53      *  "
  54      *
  55      * \param escaped HTTP/1.0 does not permit \-escaped characters
  56      */
  57     bool quotedString(SBuf &value, const bool http1p0 = false);
  58
  59     /**
  60      * Attempt to parse a (token / quoted-string ) lexical construct.
  61      */
  62     bool quotedStringOrToken(SBuf &value, const bool http1p0 = false);
  63
  64 private:
  65     /// parse the internal component of a quote-string, and terminal DQUOTE
  66     bool qdText(SBuf &value, const bool http1p0);
  67
  68     void checkpoint() { savedCheckpoint_ = buf(); savedStats_ = parsedSize(); }
  69     void restoreLastCheckpoint() { undoParse(savedCheckpoint_, savedStats_); }
  70
  71     SBuf savedCheckpoint_;
  72     SBuf::size_type savedStats_;
  73 };
  74
  75 } // namespace One
  76 } // namespace Http
  77
  78 #endif /* SQUID_SRC_HTTP_ONE_TOKENIZER_H */
  79