|
| 1 | +/* |
| 2 | + * encode.c |
| 3 | + *Various data encoding/decoding things. |
| 4 | + * |
| 5 | + * Copyright (c) 2001 Marko Kreen |
| 6 | + * All rights reserved. |
| 7 | + * |
| 8 | + * Redistribution and use in source and binary forms, with or without |
| 9 | + * modification, are permitted provided that the following conditions |
| 10 | + * are met: |
| 11 | + * 1. Redistributions of source code must retain the above copyright |
| 12 | + * notice, this list of conditions and the following disclaimer. |
| 13 | + * 2. Redistributions in binary form must reproduce the above copyright |
| 14 | + * notice, this list of conditions and the following disclaimer in the |
| 15 | + * documentation and/or other materials provided with the distribution. |
| 16 | + * |
| 17 | + * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND |
| 18 | + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE |
| 19 | + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE |
| 20 | + * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE |
| 21 | + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL |
| 22 | + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS |
| 23 | + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) |
| 24 | + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT |
| 25 | + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY |
| 26 | + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF |
| 27 | + * SUCH DAMAGE. |
| 28 | + * |
| 29 | + * $Id: encode.c,v 1.1 2001/01/24 03:46:16 momjian Exp $ |
| 30 | + */ |
| 31 | + |
| 32 | +#include<postgres.h> |
| 33 | +#include<fmgr.h> |
| 34 | + |
| 35 | +#include"encode.h" |
| 36 | + |
| 37 | +/* |
| 38 | + * NAMEDATALEN is used for hash names |
| 39 | + */ |
| 40 | +#ifNAMEDATALEN<16 |
| 41 | +#error "NAMEDATALEN < 16: too small" |
| 42 | +#endif |
| 43 | + |
| 44 | +staticpg_coding* |
| 45 | +find_coding(pg_coding*hbuf,text*name,intsilent); |
| 46 | +staticpg_coding* |
| 47 | +pg_find_coding(pg_coding*res,char*name); |
| 48 | + |
| 49 | + |
| 50 | +/* SQL function: encode(bytea, text) returns text */ |
| 51 | +PG_FUNCTION_INFO_V1(encode); |
| 52 | + |
| 53 | +Datum |
| 54 | +encode(PG_FUNCTION_ARGS) |
| 55 | +{ |
| 56 | +text*arg; |
| 57 | +text*name; |
| 58 | +uintlen,rlen,rlen0; |
| 59 | +pg_coding*c,cbuf; |
| 60 | +text*res; |
| 61 | + |
| 62 | +if (PG_ARGISNULL(0)||PG_ARGISNULL(1)) |
| 63 | +PG_RETURN_NULL(); |
| 64 | + |
| 65 | +name=PG_GETARG_TEXT_P(1); |
| 66 | +c=find_coding(&cbuf,name,0);/* will give error if fails */ |
| 67 | + |
| 68 | +arg=PG_GETARG_TEXT_P(0); |
| 69 | +len=VARSIZE(arg)-VARHDRSZ; |
| 70 | + |
| 71 | +rlen0=c->encode_len(len); |
| 72 | + |
| 73 | +res= (text*)palloc(rlen0+VARHDRSZ); |
| 74 | + |
| 75 | +rlen=c->encode(VARDATA(arg),len,VARDATA(res)); |
| 76 | +VARATT_SIZEP(res)=rlen+VARHDRSZ; |
| 77 | + |
| 78 | +if (rlen>rlen0) |
| 79 | +elog(FATAL,"pg_encode: overflow, encode estimate too small"); |
| 80 | + |
| 81 | +PG_FREE_IF_COPY(arg,0); |
| 82 | +PG_FREE_IF_COPY(name,0); |
| 83 | + |
| 84 | +PG_RETURN_TEXT_P(res); |
| 85 | +} |
| 86 | + |
| 87 | +/* SQL function: decode(text, text) returns bytea */ |
| 88 | +PG_FUNCTION_INFO_V1(decode); |
| 89 | + |
| 90 | +Datum |
| 91 | +decode(PG_FUNCTION_ARGS) |
| 92 | +{ |
| 93 | +text*arg; |
| 94 | +text*name; |
| 95 | +uintlen,rlen,rlen0; |
| 96 | +pg_coding*c,cbuf; |
| 97 | +text*res; |
| 98 | + |
| 99 | +if (PG_ARGISNULL(0)||PG_ARGISNULL(1)) |
| 100 | +PG_RETURN_NULL(); |
| 101 | + |
| 102 | +name=PG_GETARG_TEXT_P(1); |
| 103 | +c=find_coding(&cbuf,name,0);/* will give error if fails */ |
| 104 | + |
| 105 | +arg=PG_GETARG_TEXT_P(0); |
| 106 | +len=VARSIZE(arg)-VARHDRSZ; |
| 107 | + |
| 108 | +rlen0=c->decode_len(len); |
| 109 | + |
| 110 | +res= (text*)palloc(rlen0+VARHDRSZ); |
| 111 | + |
| 112 | +rlen=c->decode(VARDATA(arg),len,VARDATA(res)); |
| 113 | +VARATT_SIZEP(res)=rlen+VARHDRSZ; |
| 114 | + |
| 115 | +if (rlen>rlen0) |
| 116 | +elog(FATAL,"pg_decode: overflow, decode estimate too small"); |
| 117 | + |
| 118 | +PG_FREE_IF_COPY(arg,0); |
| 119 | +PG_FREE_IF_COPY(name,0); |
| 120 | + |
| 121 | +PG_RETURN_TEXT_P(res); |
| 122 | +} |
| 123 | + |
| 124 | +staticpg_coding* |
| 125 | +find_coding(pg_coding*dst,text*name,intsilent) |
| 126 | +{ |
| 127 | +pg_coding*p; |
| 128 | +charbuf[NAMEDATALEN]; |
| 129 | +uintlen; |
| 130 | + |
| 131 | +len=VARSIZE(name)-VARHDRSZ; |
| 132 | +if (len >=NAMEDATALEN) { |
| 133 | +if (silent) |
| 134 | +returnNULL; |
| 135 | +elog(ERROR,"Encoding type does not exist (name too long)"); |
| 136 | +} |
| 137 | + |
| 138 | +memcpy(buf,VARDATA(name),len); |
| 139 | +buf[len]=0; |
| 140 | + |
| 141 | +p=pg_find_coding(dst,buf); |
| 142 | + |
| 143 | +if (p==NULL&& !silent) |
| 144 | +elog(ERROR,"Encoding type does not exist: '%s'",buf); |
| 145 | +returnp; |
| 146 | +} |
| 147 | + |
| 148 | +staticchar*hextbl="0123456789abcdef"; |
| 149 | + |
| 150 | +uint |
| 151 | +hex_encode(uint8*src,uintlen,uint8*dst) |
| 152 | +{ |
| 153 | +uint8*end=src+len; |
| 154 | +while (src<end) { |
| 155 | +*dst++=hextbl[(*src >>4)&0xF]; |
| 156 | +*dst++=hextbl[*src&0xF]; |
| 157 | +src++; |
| 158 | +} |
| 159 | +returnlen*2; |
| 160 | +} |
| 161 | + |
| 162 | +/* probably should use lookup table */ |
| 163 | +staticuint8 |
| 164 | +get_hex(charc) |
| 165 | +{ |
| 166 | +uint8res=0; |
| 167 | + |
| 168 | +if (c >='0'&&c <='9') |
| 169 | +res=c-'0'; |
| 170 | +elseif (c >='a'&&c <='f') |
| 171 | +res=c-'a'+10; |
| 172 | +elseif (c >='A'&&c <='F') |
| 173 | +res=c-'A'+10; |
| 174 | +else |
| 175 | +elog(ERROR,"Bad hex code: '%c'",c); |
| 176 | + |
| 177 | +returnres; |
| 178 | +} |
| 179 | + |
| 180 | +uint |
| 181 | +hex_decode(uint8*src,uintlen,uint8*dst) |
| 182 | +{ |
| 183 | +uint8*s,*srcend,v1,v2,*p=dst; |
| 184 | + |
| 185 | +srcend=src+len; |
| 186 | +s=src;p=dst; |
| 187 | +while (s<srcend) { |
| 188 | +if (*s==' '||*s=='\n'||*s=='\t'||*s=='\r') { |
| 189 | +s++; |
| 190 | +continue; |
| 191 | +} |
| 192 | +v1=get_hex(*s++) <<4; |
| 193 | +if (s >=srcend) |
| 194 | +elog(ERROR,"hex_decode: invalid data"); |
| 195 | +v2=get_hex(*s++); |
| 196 | +*p++=v1 |v2; |
| 197 | +} |
| 198 | + |
| 199 | +returnp-dst; |
| 200 | +} |
| 201 | + |
| 202 | + |
| 203 | +staticunsignedchar_base64[]= |
| 204 | +"ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/"; |
| 205 | + |
| 206 | +uint |
| 207 | +b64_encode(uint8*src,uintlen,uint8*dst) |
| 208 | +{ |
| 209 | +uint8*s,*p,*end=src+len,*lend=dst+76; |
| 210 | +intpos=2; |
| 211 | +unsigned longbuf=0; |
| 212 | + |
| 213 | +s=src;p=dst; |
| 214 | + |
| 215 | +while (s<end) { |
| 216 | +buf |=*s << (pos <<3); |
| 217 | +pos--; |
| 218 | +s++; |
| 219 | + |
| 220 | +/* write it out */ |
| 221 | +if (pos<0) { |
| 222 | +*p++=_base64[(buf >>18)&0x3f]; |
| 223 | +*p++=_base64[(buf >>12)&0x3f]; |
| 224 | +*p++=_base64[(buf >>6)&0x3f]; |
| 225 | +*p++=_base64[buf&0x3f]; |
| 226 | + |
| 227 | +pos=2; |
| 228 | +buf=0; |
| 229 | +} |
| 230 | +if (p >=lend) { |
| 231 | +*p++='\n'; |
| 232 | +lend=p+76; |
| 233 | +} |
| 234 | +} |
| 235 | +if (pos!=2) { |
| 236 | +*p++=_base64[(buf >>18)&0x3f]; |
| 237 | +*p++=_base64[(buf >>12)&0x3f]; |
| 238 | +*p++= (pos==0) ?_base64[(buf >>6)&0x3f] :'='; |
| 239 | +*p++='='; |
| 240 | +} |
| 241 | + |
| 242 | +returnp-dst; |
| 243 | +} |
| 244 | + |
| 245 | +/* probably should use lookup table */ |
| 246 | +uint |
| 247 | +b64_decode(uint8*src,uintlen,uint8*dst) |
| 248 | +{ |
| 249 | +char*srcend=src+len,*s=src; |
| 250 | +uint8*p=dst; |
| 251 | +charc; |
| 252 | +uintb=0; |
| 253 | +unsigned longbuf=0; |
| 254 | +intpos=0,end=0; |
| 255 | + |
| 256 | +while (s<srcend) { |
| 257 | +c=*s++; |
| 258 | +if (c >='A'&&c <='Z') |
| 259 | +b=c-'A'; |
| 260 | +elseif (c >='a'&&c <='z') |
| 261 | +b=c-'a'+26; |
| 262 | +elseif (c >='0'&&c <='9') |
| 263 | +b=c-'0'+52; |
| 264 | +elseif (c=='+') |
| 265 | +b=62; |
| 266 | +elseif (c=='/') |
| 267 | +b=63; |
| 268 | +elseif (c=='=') { |
| 269 | +/* end sequence */ |
| 270 | +if (!end) { |
| 271 | +if (pos==2)end=1; |
| 272 | +elseif (pos==3)end=2; |
| 273 | +else |
| 274 | +elog(ERROR,"base64: unexpected '='"); |
| 275 | +} |
| 276 | +b=0; |
| 277 | +}elseif (c==' '||c=='\t'||c=='\n'||c=='\r') |
| 278 | +continue; |
| 279 | +else |
| 280 | +elog(ERROR,"base64: Invalid symbol"); |
| 281 | + |
| 282 | +/* add it to buffer */ |
| 283 | +buf= (buf <<6)+b; |
| 284 | +pos++; |
| 285 | +if (pos==4) { |
| 286 | +*p++= (buf >>16)&255; |
| 287 | +if (end==0||end>1) |
| 288 | +*p++= (buf >>8)&255; |
| 289 | +if (end==0||end>2) |
| 290 | +*p++=buf&255; |
| 291 | +buf=0; |
| 292 | +pos=0; |
| 293 | +} |
| 294 | +} |
| 295 | + |
| 296 | +if (pos!=0) |
| 297 | +elog(ERROR,"base64: invalid end sequence"); |
| 298 | + |
| 299 | +returnp-dst; |
| 300 | +} |
| 301 | + |
| 302 | + |
| 303 | +uint |
| 304 | +hex_enc_len(uintsrclen) |
| 305 | +{ |
| 306 | +returnsrclen <<1; |
| 307 | +} |
| 308 | + |
| 309 | +uint |
| 310 | +hex_dec_len(uintsrclen) |
| 311 | +{ |
| 312 | +returnsrclen >>1; |
| 313 | +} |
| 314 | + |
| 315 | +uint |
| 316 | +b64_enc_len(uintsrclen) |
| 317 | +{ |
| 318 | +returnsrclen+ (srclen /3)+ (srclen / (76 /2)); |
| 319 | +} |
| 320 | + |
| 321 | +uint |
| 322 | +b64_dec_len(uintsrclen) |
| 323 | +{ |
| 324 | +return (srclen*3) >>2; |
| 325 | +} |
| 326 | + |
| 327 | +staticpg_coding |
| 328 | +encoding_list []= { |
| 329 | +{"hex",hex_enc_len,hex_dec_len,hex_encode,hex_decode}, |
| 330 | +{"base64",b64_enc_len,b64_dec_len,b64_encode,b64_decode}, |
| 331 | +{NULL,NULL,NULL,NULL,NULL} |
| 332 | +}; |
| 333 | + |
| 334 | + |
| 335 | +staticpg_coding* |
| 336 | +pg_find_coding(pg_coding*res,char*name) |
| 337 | +{ |
| 338 | +pg_coding*p; |
| 339 | +for (p=encoding_list;p->name;p++) { |
| 340 | +if (!strcasecmp(p->name,name)) |
| 341 | +returnp; |
| 342 | +} |
| 343 | +returnNULL; |
| 344 | +} |
| 345 | + |