1.0.6

Security Issues

On 1.0.5 and before, a bug that has accessed deleted heap memory in chardet and chardet_r api has been fixed. (#18) Thanks for @gaoxiang-ut

Changes:

• fixed #9 configure.ac needs subdir-objects

• fixed #10 autogen failure because AM_PROG_AR with automake 1.11.1

• fixed #12 No include guard

• fixed #13 bom member has been added to the DetectObj structure

  • New unicode language model : BOCU-1, GB-18030, SCSU, UTF-1, UTF-7, UTF-EBCDIC

  diff --git a/src/chardet.h b/src/chardet.h
  index 84975a3..f603a37 100644
  --- a/src/chardet.h
  +++ b/src/chardet.h
  @@ -89,6 +89,7 @@ extern "C" {
      typedef struct DetectObject {
          char * encoding;
          float confidence;
  +       short bom;
      } DetectObj;
  
      CHARDET_API char * detect_version (void);

  #ifdef CHARDET_BOM_CHECK
      printf ("#1 %s : %s : %f : %d\n", string, obj->encoding, obj->confidence, obj->bom);
  #else
      printf ("#1 %s : %s : %f\n", string, obj->encoding, obj->confidence);
  #endif

• fixed #14 can't detect short euc-kr

• fixed #15 support automake style 'make check'

• fixed #18 SECURITY! Invalid memory approach (heap-use-after-free) (@gaoxiang-ut)

1.0.5

Changes:

• #8 fixed can not detect UTF-16/32.

  • This is binary safe problems
  • In order to solve this problems, support _detect_r_ and _detect_handledata_r_ API.
  • Support _CHARDET_BINARY_SAFE_ consantant whether support _detect_r_ or _detect_handledata_r_

  #ifdef CHARDET_BINARY_SAFE
          if ( detect_r (str[i], strlen (str[i]), &obj) == CHARDET_OUT_OF_MEMORY )
  #else
          if ( detect (str[i], &obj) == CHARDET_OUT_OF_MEMORY )
  #endif
          {
              fprintf (stderr, "On handle processing, occured out of memory\n");
              return CHARDET_OUT_OF_MEMORY;
          } 
  
  #ifdef CHARDET_BINARY_SAFE
          if ( detect_handledata_r (&d, str[i], strlen (str[i]), &obj) == CHARDET_OUT_OF_MEMORY )
  #else
          if ( detect_handledata (&d, str[i], &obj) == CHARDET_OUT_OF_MEMORY )
  #endif
          {
              fprintf (stderr, "On handle processing, occured out of memory\n");
              return CHARDET_OUT_OF_MEMORY;
          }

• Merge uchardet's improves

  • #6 fixed extended character range on EUT-KR and EUC-TW
    • can detect CP949 (for example, "똠방각하", "뷁")
    • can detect extended EUC-TW ("灣,是,台" and so on)
  • #2, #5 Improve single-byte charset detection confidence algorithm
  • #4 New single-byte language model
    • Arabic
    • Danash
    • Esperanto
    • German
    • Spanish
    • Turkish
    • Vietnamese

• #3 Update language model of Greek, Hungarian and Thai

• fixed man pages wrong macro bug (martin.gansser@gmail.com)

1.0.4

• fixed duplicated path on chardet.pc @8ddca58f
• Windows support
  • support windows native library with MinGW @215952ce
  • support Code::Blocks project @215952ce
  • fixed build error and shared library on cygwin @b201aa5

1.0.3

• add chardet.pc (Lee ByungYoung <darklin20@gamil.com>) @2fd56976
• applied automake @0fdd1e28
• add english man page @da00ca24
• fixed comparison on JpCntx.cpp @7f624fb2

1.0.2

• support visibility attribute on gcc4 @1f2d6ec5
• add version information api @d42e4c01

1.0.1

• fix wrong detect TIS-620 charset @f828c136