Problem report #110: base/src/emacs/src/coding.c (decode_coding_utf_8); UNINIT

[Dan Nicolaescu]

CID: 110
Checker: UNINIT (help)
File: base/src/emacs/src/coding.c
Function: decode_coding_utf_8
Description: Using uninitialized value "consumed_chars_base"

Event var_decl: Declared variable "consumed_chars_base" without initializer
Also see events: [uninit_use]

1329      int consumed_chars = 0, consumed_chars_base;
1330      int multibytep = coding->src_multibyte;
1331      enum utf_bom_type bom = CODING_UTF_8_BOM (coding);
1332      Lisp_Object attr, charset_list;

At conditional (1): "((0), (((0), ((Vcoding_system_hash_table & 
-8)->key_and_value & -8))->contents[((2 * (coding)->id) + 1)] & 
-8))->contents[2] == Qdos" taking true path

1333      int eol_crlf = EQ (CODING_ID_EOL_TYPE (coding->id), Qdos);
1334      int byte_after_cr = -1;
1335    

At conditional (2): "0" taking false path

1336      CODING_GET_INFO (coding, attr, charset_list);
1337    

At conditional (3): "bom != 1" taking true path

1338      if (bom != utf_without_bom)
1339        {
1340          int c1, c2, c3;
1341    
1342          src_base = src;

At conditional (4): "src == src_end" taking true path
At conditional (5): "src_base < src" taking true path

1343          ONE_MORE_BYTE (c1);
1344          if (! UTF_8_3_OCTET_LEADING_P (c1))
1345            src = src_base;
1346          else
1347            {
1348              ONE_MORE_BYTE (c2);      
1349              if (! UTF_8_EXTRA_OCTET_P (c2))
1350                src = src_base;
1351              else
1352                {
1353                  ONE_MORE_BYTE (c3);      
1354                  if (! UTF_8_EXTRA_OCTET_P (c3))
1355                    src = src_base;
1356                  else
1357                    {
1358                      if ((c1 != UTF_8_BOM_1)
1359                          || (c2 != UTF_8_BOM_2) || (c3 != UTF_8_BOM_3))
1360                        src = src_base;
1361                      else
1362                        CODING_UTF_8_BOM (coding) = utf_without_bom;
1363                    }
1364                }
1365            }
1366        }
1367      CODING_UTF_8_BOM (coding) = utf_without_bom;
1368    
1369    
1370    
1371      while (1)
1372        {
1373          int c, c1, c2, c3, c4, c5;
1374    
1375          src_base = src;
1376          consumed_chars_base = consumed_chars;
1377    
1378          if (charbuf >= charbuf_end)
1379            break;
1380    
1381          if (byte_after_cr >= 0)
1382            c1 = byte_after_cr, byte_after_cr = -1;
1383          else
1384            ONE_MORE_BYTE (c1);
1385          if (c1 < 0)
1386            {
1387              c = - c1;
1388            }
1389          else if (UTF_8_1_OCTET_P(c1))
1390            {
1391              if (eol_crlf && c1 == '\r')
1392                ONE_MORE_BYTE (byte_after_cr);
1393              c = c1;
1394            }
1395          else
1396            {
1397              ONE_MORE_BYTE (c2);
1398              if (c2 < 0 || ! UTF_8_EXTRA_OCTET_P (c2))
1399                goto invalid_code;
1400              if (UTF_8_2_OCTET_LEADING_P (c1))
1401                {
1402                  c = ((c1 & 0x1F) << 6) | (c2 & 0x3F);
1403                  /* Reject overlong sequences here and below.  Encoders
1404                     producing them are incorrect, they can be misleading,
1405                     and they mess up read/write invariance.  */
1406                  if (c < 128)
1407                    goto invalid_code;
1408                }
1409              else
1410                {
1411                  ONE_MORE_BYTE (c3);
1412                  if (c3 < 0 || ! UTF_8_EXTRA_OCTET_P (c3))
1413                    goto invalid_code;
1414                  if (UTF_8_3_OCTET_LEADING_P (c1))
1415                    {
1416                      c = (((c1 & 0xF) << 12)
1417                           | ((c2 & 0x3F) << 6) | (c3 & 0x3F));
1418                      if (c < 0x800
1419                          || (c >= 0xd800 && c < 0xe000)) /* surrogates 
(invalid) */
1420                        goto invalid_code;
1421                    }
1422                  else
1423                    {
1424                      ONE_MORE_BYTE (c4);
1425                      if (c4 < 0 || ! UTF_8_EXTRA_OCTET_P (c4))
1426                        goto invalid_code;
1427                      if (UTF_8_4_OCTET_LEADING_P (c1))
1428                        {
1429                        c = (((c1 & 0x7) << 18) | ((c2 & 0x3F) << 12)
1430                             | ((c3 & 0x3F) << 6) | (c4 & 0x3F));
1431                        if (c < 0x10000)
1432                          goto invalid_code;
1433                        }
1434                      else
1435                        {
1436                          ONE_MORE_BYTE (c5);
1437                          if (c5 < 0 || ! UTF_8_EXTRA_OCTET_P (c5))
1438                            goto invalid_code;
1439                          if (UTF_8_5_OCTET_LEADING_P (c1))
1440                            {
1441                              c = (((c1 & 0x3) << 24) | ((c2 & 0x3F) << 18)
1442                                   | ((c3 & 0x3F) << 12) | ((c4 & 0x3F) << 
6)
1443                                   | (c5 & 0x3F));
1444                              if ((c > MAX_CHAR) || (c < 0x200000))
1445                                goto invalid_code;
1446                            }
1447                          else
1448                            goto invalid_code;
1449                        }
1450                    }
1451                }
1452            }
1453    
1454          *charbuf++ = c;
1455          continue;
1456    
1457        invalid_code:
1458          src = src_base;
1459          consumed_chars = consumed_chars_base;
1460          ONE_MORE_BYTE (c);
1461          *charbuf++ = ASCII_BYTE_P (c) ? c : BYTE8_TO_CHAR (c);
1462          coding->errors++;
1463        }
1464    
1465     no_more_source:

Event uninit_use: Using uninitialized value "consumed_chars_base"
Also see events: [var_decl]

1466      coding->consumed_char += consumed_chars_base;
1467      coding->consumed = src_base - coding->source;
1468      coding->charbuf_used = charbuf - coding->charbuf;