ovn/lib/lex.h

   1 /*
   2  * Copyright (c) 2015, 2016 Nicira, Inc.
   3  *
   4  * Licensed under the Apache License, Version 2.0 (the "License");
   5  * you may not use this file except in compliance with the License.
   6  * You may obtain a copy of the License at:
   7  *
   8  *     http://www.apache.org/licenses/LICENSE-2.0
   9  *
  10  * Unless required by applicable law or agreed to in writing, software
  11  * distributed under the License is distributed on an "AS IS" BASIS,
  12  * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  13  * See the License for the specific language governing permissions and
  14  * limitations under the License.
  15  */
  16
  17 #ifndef OVN_LEX_H
  18 #define OVN_LEX_H 1
  19
  20 /* OVN lexical analyzer
  21  * ====================
  22  *
  23  * This is a simple lexical analyzer (or tokenizer) for OVN match expressions
  24  * and ACLs. */
  25
  26 #include "openvswitch/meta-flow.h"
  27
  28 struct ds;
  29
  30 /* Token type. */
  31 enum lex_type {
  32     LEX_T_END,                  /* end of input */
  33
  34     /* Tokens with auxiliary data. */
  35     LEX_T_ID,                   /* foo */
  36     LEX_T_STRING,               /* "foo" */
  37     LEX_T_INTEGER,              /* 12345 or 1.2.3.4 or ::1 or 01:02:03:04:05 */
  38     LEX_T_MASKED_INTEGER,       /* 12345/10 or 1.2.0.0/16 or ::2/127 or... */
  39     LEX_T_MACRO,                /* $NAME */
  40     LEX_T_ERROR,                /* invalid input */
  41
  42     /* Bare tokens. */
  43     LEX_T_LPAREN,               /* ( */
  44     LEX_T_RPAREN,               /* ) */
  45     LEX_T_LCURLY,               /* { */
  46     LEX_T_RCURLY,               /* } */
  47     LEX_T_LSQUARE,              /* [ */
  48     LEX_T_RSQUARE,              /* ] */
  49     LEX_T_EQ,                   /* == */
  50     LEX_T_NE,                   /* != */
  51     LEX_T_LT,                   /* < */
  52     LEX_T_LE,                   /* <= */
  53     LEX_T_GT,                   /* > */
  54     LEX_T_GE,                   /* >= */
  55     LEX_T_LOG_NOT,              /* ! */
  56     LEX_T_LOG_AND,              /* && */
  57     LEX_T_LOG_OR,               /* || */
  58     LEX_T_ELLIPSIS,             /* .. */
  59     LEX_T_COMMA,                /* , */
  60     LEX_T_SEMICOLON,            /* ; */
  61     LEX_T_EQUALS,               /* = */
  62     LEX_T_EXCHANGE,             /* <-> */
  63     LEX_T_DECREMENT,            /* -- */
  64     LEX_T_COLON,                /* : */
  65 };
  66
  67 /* Subtype for LEX_T_INTEGER and LEX_T_MASKED_INTEGER tokens.
  68  *
  69  * These do not change the semantics of a token; instead, they determine the
  70  * format used when a token is serialized back to a text form.  That's
  71  * important because 3232268289 is meaningless to a human whereas 192.168.128.1
  72  * has some actual significance. */
  73 enum lex_format {
  74     LEX_F_DECIMAL,
  75     LEX_F_HEXADECIMAL,
  76     LEX_F_IPV4,
  77     LEX_F_IPV6,
  78     LEX_F_ETHERNET,
  79 };
  80 const char *lex_format_to_string(enum lex_format);
  81
  82 /* A token. */
  83 struct lex_token {
  84     /* One of LEX_*. */
  85     enum lex_type type;
  86
  87     /* Meaningful for LEX_T_ID, LEX_T_STRING, LEX_T_ERROR, LEX_T_MACRO only.
  88      * For these token types, 's' may point to 'buffer'; otherwise, it points
  89      * to malloc()ed memory owned by the token.
  90      *
  91      * Must be NULL for other token types.
  92      *
  93      * For LEX_T_MACRO, 's' does not include the leading $. */
  94     char *s;
  95
  96     /* LEX_T_INTEGER, LEX_T_MASKED_INTEGER only. */
  97     enum lex_format format;
  98
  99     union {
 100         /* LEX_T_INTEGER, LEX_T_MASKED_INTEGER only. */
 101         struct {
 102             union mf_subvalue value; /* LEX_T_INTEGER, LEX_T_MASKED_INTEGER. */
 103             union mf_subvalue mask;  /* LEX_T_MASKED_INTEGER only. */
 104         };
 105
 106         /* LEX_T_ID, LEX_T_STRING, LEX_T_ERROR, LEX_T_MACRO only. */
 107         char buffer[256];
 108     };
 109 };
 110
 111 void lex_token_init(struct lex_token *);
 112 void lex_token_destroy(struct lex_token *);
 113 void lex_token_swap(struct lex_token *, struct lex_token *);
 114 void lex_token_strcpy(struct lex_token *, const char *s, size_t length);
 115 void lex_token_strset(struct lex_token *, char *s);
 116 void lex_token_vsprintf(struct lex_token *, const char *format, va_list args);
 117
 118 void lex_token_format(const struct lex_token *, struct ds *);
 119 const char *lex_token_parse(struct lex_token *, const char *input,
 120                             const char **startp);
 121
 122 /* A lexical analyzer. */
 123 struct lexer {
 124     const char *input;          /* Remaining input (not owned by lexer). */
 125     const char *start;          /* Start of current token in 'input'. */
 126     struct lex_token token;     /* Current token (owned by lexer). */
 127 };
 128
 129 void lexer_init(struct lexer *, const char *input);
 130 void lexer_destroy(struct lexer *);
 131
 132 enum lex_type lexer_get(struct lexer *);
 133 enum lex_type lexer_lookahead(const struct lexer *);
 134 bool lexer_match(struct lexer *, enum lex_type);
 135 bool lexer_match_id(struct lexer *, const char *id);
 136 bool lexer_is_int(const struct lexer *);
 137 bool lexer_get_int(struct lexer *, int *value);
 138
 139 #endif /* ovn/lex.h */