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 ' ':
  67         DPUTC(d, '+');
  68         break;
  69       default:
  70         if ((ctx->f & URLF_LAX) || isalnum((unsigned char)*p))
  71           goto safe;
  72         else
  73           goto unsafe;
  74       case '/':
  75       case '~':
  76         if (ctx->f & URLF_STRICT)
  77           goto unsafe;
  78       case '-':
  79       case '.':
  80       case '_':
  81       safe:
  82         DPUTC(d, *p);
  83         break;
  84       unsafe:
  85       case '+':
  86       case '%':
  87       case '=':
  88       case '&':
  89       case ';':
  90         dstr_putf(d, "%%%02x", *p);
  91         break;
  92     }
  93     p++;
  94   }
  95 }
  96
  97 /* --- @url_enc@ --- *
  98  *
  99  * Arguments:   @url_ectx *ctx@ = pointer to encoding context
 100  *              @dstr *d@ = pointer to output string
 101  *              @const char *name@ = pointer to name
 102  *              @const char *value@ = pointer to value
 103  *
 104  * Returns:     ---
 105  *
 106  * Use:         Writes an assignment between @name@ and @value@ to the
 107  *              output string, encoding the values properly.
 108  */
 109
 110 void url_enc(url_ectx *ctx, dstr *d, const char *name, const char *value)
 111 {
 112   if (ctx->f & URLF_SEP)
 113     DPUTC(d, (ctx->f & URLF_SEMI) ? ';' : '&');
 114   encode(ctx, d, name);
 115   DPUTC(d, '=');
 116   encode(ctx, d, value);
 117   DPUTZ(d);
 118   ctx->f |= URLF_SEP;
 119 }
 120
 121 /* --- @url_initdec@ --- *
 122  *
 123  * Arguments:   @url_dctx *ctx@ = pointer to context block
 124  *              @const char *p@ = string to read data from
 125  *
 126  * Returns:     ---
 127  *
 128  * Use:         Initializes a URL decoding context.
 129  */
 130
 131 void url_initdec(url_dctx *ctx, const char *p) { ctx->p = p; ctx->f = 0; }
 132
 133 /* --- @decode@ --- *
 134  *
 135  * Arguments:   @url_dctx *ctx@ = pointer to the context
 136  *              @dstr *d@ = pointer to output string
 137  *              @const char *p@ = pointer to input data
 138  *              @int eq@ = whether to stop at `=' characters
 139  *
 140  * Returns:     Pointer to next available character.
 141  *
 142  * Use:         Does a URL decode.
 143  */
 144
 145 static const char *decode(url_dctx *ctx, dstr *d, const char *p, int eq)
 146 {
 147   if (!*p)
 148     return (0);
 149   for (;;) {
 150     switch (*p) {
 151       case '=':
 152         if (eq)
 153           return (p);
 154         goto boring;
 155       case ';':
 156         if (ctx->f & URLF_SEMI)
 157           return (p);
 158         goto boring;
 159       case 0:
 160       case '&':
 161         return (p);
 162       case '+':
 163         DPUTC(d, ' ');
 164         break;
 165       case '%': {
 166         unsigned int ch;
 167         int n;
 168         int x = sscanf(p + 1, "%2x%n", &ch, &n);
 169         if (x == 1) {
 170           DPUTC(d, ch);
 171           p += n;
 172           break;
 173         }
 174       }
 175       default:
 176       boring:
 177         DPUTC(d, *p);
 178         break;
 179     }
 180     p++;
 181   }
 182 }
 183
 184 /* --- @url_dec@ --- *
 185  *
 186  * Arguments:   @url_dctx *ctx@ = pointer to decode context
 187  *              @dstr *n@ = pointer to output string for name
 188  *              @dstr *v@ = pointer to output string for value
 189  *
 190  * Returns:     Nonzero if it read something, zero if there's nothing left
 191  *
 192  * Use:         Decodes the next name/value pair from a urlencoded string.
 193  */
 194
 195 int url_dec(url_dctx *ctx, dstr *n, dstr *v)
 196 {
 197   const char *p = ctx->p;
 198   size_t l = n->len;
 199
 200 again:
 201   if ((p = decode(ctx, n, p, 1)) == 0 || *p == 0)
 202     return (0);
 203   if (*p != '=') {
 204     p++;
 205     n->len = l;
 206     goto again;
 207   }
 208   p++;
 209   if ((p = decode(ctx, v, p, 0)) == 0)
 210     return (0);
 211   DPUTZ(n);
 212   DPUTZ(v);
 213   ctx->p = p;
 214   return (1);
 215 }
 216
 217 /*----- That's all, folks -------------------------------------------------*/