mdw@git.distorted.org.uk Git - mLib/blob - codec/url.c

   1 /* -*-c-*-
   2  *
   3  * Parsing and construction of url-encoded name/value pairs
   4  *
   5  * (c) 1999 Straylight/Edgeware
   6  */
   7
   8 /*----- Licensing notice --------------------------------------------------*
   9  *
  10  * This file is part of the mLib utilities library.
  11  *
  12  * mLib is free software; you can redistribute it and/or modify
  13  * it under the terms of the GNU Library General Public License as
  14  * published by the Free Software Foundation; either version 2 of the
  15  * License, or (at your option) any later version.
  16  *
  17  * mLib is distributed in the hope that it will be useful,
  18  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  19  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  20  * GNU Library General Public License for more details.
  21  *
  22  * You should have received a copy of the GNU Library General Public
  23  * License along with mLib; if not, write to the Free
  24  * Software Foundation, Inc., 59 Temple Place - Suite 330, Boston,
  25  * MA 02111-1307, USA.
  26  */
  27
  28 /*----- Header files ------------------------------------------------------*/
  29
  30 #include <ctype.h>
  31 #include <stdio.h>
  32 #include <stdlib.h>
  33 #include <string.h>
  34
  35 #include "dstr.h"
  36 #include "url.h"
  37
  38 /*----- Main code ---------------------------------------------------------*/
  39
  40 /* --- @url_initenc@ --- *
  41  *
  42  * Arguments:   @url_ectx *ctx@ = pointer to context block
  43  *
  44  * Returns:     ---
  45  *
  46  * Use:         Initializes a URL encoding context.
  47  */
  48
  49 void url_initenc(url_ectx *ctx) { ctx->f = 0; }
  50
  51 /* --- @encode@ --- *
  52  *
  53  * Arguments:   @url_ectx *ctx@ = encoding context
  54  *              @dstr *d@ = pointer to output string
  55  *              @const char *p@ = pointer to thing to encode
  56  *
  57  * Returns:     ---
  58  *
  59  * Use:         Encodes the input string into the output string.
  60  */
  61
  62 static void encode(url_ectx *ctx, dstr *d, const char *p)
  63 {
  64   while (*p) {
  65     switch (*p) {
  66       case ' ': DPUTC(d, '+');
  67         break;
  68       default:
  69         if (isspace((unsigned char)*p)) goto unsafe;
  70         else if (isalnum((unsigned char)*p)) goto safe;
  71         else if (ctx->f&URLF_LAX) goto safe;
  72         else goto unsafe;
  73       case '/': case '~':
  74         if (ctx->f&URLF_STRICT) goto unsafe; /* else fall through... */
  75       safe: case '-': case '.': case '_':
  76         DPUTC(d, *p);   break;
  77       unsafe: case '+': case '%': case '=': case '&': case ';':
  78         dstr_putf(d, "%%%02x", *p); break;
  79     }
  80     p++;
  81   }
  82 }
  83
  84 /* --- @url_enc@ --- *
  85  *
  86  * Arguments:   @url_ectx *ctx@ = pointer to encoding context
  87  *              @dstr *d@ = pointer to output string
  88  *              @const char *name@ = pointer to name
  89  *              @const char *value@ = pointer to value
  90  *
  91  * Returns:     ---
  92  *
  93  * Use:         Writes an assignment between @name@ and @value@ to the
  94  *              output string, encoding the values properly.
  95  */
  96
  97 void url_enc(url_ectx *ctx, dstr *d, const char *name, const char *value)
  98 {
  99   if (ctx->f & URLF_SEP)
 100     DPUTC(d, (ctx->f & URLF_SEMI) ? ';' : '&');
 101   encode(ctx, d, name);
 102   DPUTC(d, '=');
 103   encode(ctx, d, value);
 104   DPUTZ(d);
 105   ctx->f |= URLF_SEP;
 106 }
 107
 108 /* --- @url_initdec@ --- *
 109  *
 110  * Arguments:   @url_dctx *ctx@ = pointer to context block
 111  *              @const char *p@ = string to read data from
 112  *
 113  * Returns:     ---
 114  *
 115  * Use:         Initializes a URL decoding context.
 116  */
 117
 118 void url_initdec(url_dctx *ctx, const char *p) { ctx->p = p; ctx->f = 0; }
 119
 120 /* --- @decode@ --- *
 121  *
 122  * Arguments:   @url_dctx *ctx@ = pointer to the context
 123  *              @dstr *d@ = pointer to output string
 124  *              @const char *p@ = pointer to input data
 125  *              @int eq@ = whether to stop at `=' characters
 126  *
 127  * Returns:     Pointer to next available character.
 128  *
 129  * Use:         Does a URL decode.
 130  */
 131
 132 static const char *decode(url_dctx *ctx, dstr *d, const char *p, int eq)
 133 {
 134   if (!*p)
 135     return (0);
 136   for (;;) {
 137     switch (*p) {
 138       case '=':
 139         if (eq)
 140           return (p);
 141         goto boring;
 142       case ';':
 143         if (ctx->f & URLF_SEMI)
 144           return (p);
 145         goto boring;
 146       case 0:
 147       case '&':
 148         return (p);
 149       case '+':
 150         DPUTC(d, ' ');
 151         break;
 152       case '%': {
 153         unsigned int ch;
 154         int n;
 155         int x = sscanf(p + 1, "%2x%n", &ch, &n);
 156         if (x == 1) {
 157           DPUTC(d, ch);
 158           p += n;
 159           break;
 160         }
 161       }
 162       default:
 163       boring:
 164         DPUTC(d, *p);
 165         break;
 166     }
 167     p++;
 168   }
 169 }
 170
 171 /* --- @url_dec@ --- *
 172  *
 173  * Arguments:   @url_dctx *ctx@ = pointer to decode context
 174  *              @dstr *n@ = pointer to output string for name
 175  *              @dstr *v@ = pointer to output string for value
 176  *
 177  * Returns:     Nonzero if it read something, zero if there's nothing left
 178  *
 179  * Use:         Decodes the next name/value pair from a urlencoded string.
 180  */
 181
 182 int url_dec(url_dctx *ctx, dstr *n, dstr *v)
 183 {
 184   const char *p = ctx->p;
 185   size_t l = n->len;
 186
 187 again:
 188   if ((p = decode(ctx, n, p, 1)) == 0 || *p == 0)
 189     return (0);
 190   if (*p != '=') {
 191     p++;
 192     n->len = l;
 193     goto again;
 194   }
 195   p++;
 196   if ((p = decode(ctx, v, p, 0)) == 0)
 197     return (0);
 198   DPUTZ(n);
 199   DPUTZ(v);
 200   ctx->p = p;
 201   return (1);
 202 }
 203
 204 /*----- That's all, folks -------------------------------------------------*/