Re: [Gnash-commit] gnash ChangeLog libbase/utf8.cpp libbase/utf8.h...

[zou lunkai]

>                * libbase/utf8.{cpp,h}: decodeNextUnicodeCharacter should know
>                  where the end of the string is, so it doesn't deference
>                  .end() (thanks zou).

Confirmed, should be fixed, no runtime assertion under MSVC8 now, thanks.

--zou


On 4/28/08, Benjamin Wolsey <address@hidden> wrote:
> CVSROOT:        /sources/gnash
> Module name:    gnash
> Changes by:     Benjamin Wolsey <bwy>   08/04/28 07:51:05
>
> Modified files:
>        .              : ChangeLog
>        libbase        : utf8.cpp utf8.h
>        server/vm      : ASHandlers.cpp
>
> Log message:
>                * libbase/utf8.{cpp,h}: decodeNextUnicodeCharacter should know
>                  where the end of the string is, so it doesn't deference
>                  .end() (thanks zou).
>                * server/vm/ASHanders.cpp: pass end of string iterator to
>                  decodeNextUnicodeCharacter, cache value generally to save 
> repeated
>                  calls to .end().
>
> CVSWeb URLs:
> http://cvs.savannah.gnu.org/viewcvs/gnash/ChangeLog?cvsroot=gnash&r1=1.6425&r2=1.6426
> http://cvs.savannah.gnu.org/viewcvs/gnash/libbase/utf8.cpp?cvsroot=gnash&r1=1.11&r2=1.12
> http://cvs.savannah.gnu.org/viewcvs/gnash/libbase/utf8.h?cvsroot=gnash&r1=1.18&r2=1.19
> http://cvs.savannah.gnu.org/viewcvs/gnash/server/vm/ASHandlers.cpp?cvsroot=gnash&r1=1.232&r2=1.233
>
> Patches:
> Index: ChangeLog
> ===================================================================
> RCS file: /sources/gnash/gnash/ChangeLog,v
> retrieving revision 1.6425
> retrieving revision 1.6426
> diff -u -b -r1.6425 -r1.6426
> --- ChangeLog   28 Apr 2008 06:44:50 -0000      1.6425
> +++ ChangeLog   28 Apr 2008 07:51:03 -0000      1.6426
> @@ -1,3 +1,12 @@
> +2008-04-27 Benjamin Wolsey <address@hidden>
> +
> +       * libbase/utf8.{cpp,h}: decodeNextUnicodeCharacter should know
> +         where the end of the string is, so it doesn't deference
> +         .end() (thanks zou).
> +       * server/vm/ASHanders.cpp: pass end of string iterator to
> +         decodeNextUnicodeCharacter, cache value generally to save repeated
> +         calls to .end().
> +
>  2008-04-28 Zou Lunkai <address@hidden>
>
>        * server/swf/DisplayListTag.h: isRmove(), isPlace(), isReplace(), 
> isMove(),
>
> Index: libbase/utf8.cpp
> ===================================================================
> RCS file: /sources/gnash/gnash/libbase/utf8.cpp,v
> retrieving revision 1.11
> retrieving revision 1.12
> diff -u -b -r1.11 -r1.12
> --- libbase/utf8.cpp    17 Apr 2008 08:05:36 -0000      1.11
> +++ libbase/utf8.cpp    28 Apr 2008 07:51:04 -0000      1.12
> @@ -33,11 +33,11 @@
>
>        std::wstring wstr = L"";
>
> -       std::string::const_iterator it = str.begin();
> +       std::string::const_iterator it = str.begin(), e = str.end();
>
>        if (version > 5)
>        {
> -               while (boost::uint32_t code = decodeNextUnicodeCharacter(it))
> +               while (boost::uint32_t code = decodeNextUnicodeCharacter(it, 
> e))
>                {
>                    if (code == utf8::invalid)
>                    {
> @@ -88,7 +88,8 @@
>
>
>  boost::uint32_t
> -utf8::decodeNextUnicodeCharacter(std::string::const_iterator& it)
> +utf8::decodeNextUnicodeCharacter(std::string::const_iterator& it,
> +                                 const std::string::const_iterator& e)
>  {
>        boost::uint32_t uc;
>
> @@ -112,12 +113,12 @@
>
>  #define NEXT_BYTE(shift)                                               \
>                                        \
> -       if (*it == 0) return 0; /* end of buffer, do not advance */     \
> +       if (it == e || *it == 0) return 0; /* end of buffer, do not advance 
> */  \
>        if ((*it & 0xC0) != 0x80) return utf8::invalid; /* standard check */   
>  \
>        /* Post-increment iterator: */          \
>        uc |= (*it++ & 0x3F) << shift;
>
> -       if (*it == 0) return 0; // End of buffer.  Do not advance.
> +       if (it == e || *it == 0) return 0;      // End of buffer.  Do not 
> advance.
>
>        // Conventional 7-bit ASCII; return and increment iterator:
>        if ((*it & 0x80) == 0) return (boost::uint32_t) *it++;
>
> Index: libbase/utf8.h
> ===================================================================
> RCS file: /sources/gnash/gnash/libbase/utf8.h,v
> retrieving revision 1.18
> retrieving revision 1.19
> diff -u -b -r1.18 -r1.19
> --- libbase/utf8.h      17 Apr 2008 08:22:36 -0000      1.18
> +++ libbase/utf8.h      28 Apr 2008 07:51:04 -0000      1.19
> @@ -84,7 +84,8 @@
>        /// as output.  Advances string iterator past the character
>        /// returned, unless the returned character is '\0', in which
>        /// case the iterator does not advance.
> -       DSOEXPORT boost::uint32_t 
> decodeNextUnicodeCharacter(std::string::const_iterator& it);
> +       DSOEXPORT boost::uint32_t 
> decodeNextUnicodeCharacter(std::string::const_iterator& it,
> +                                                            const 
> std::string::const_iterator& e);
>
>        /// \brief Encodes the given wide character into a canonical
>        /// string, theoretically up to 6 chars in length.
>
> Index: server/vm/ASHandlers.cpp
> ===================================================================
> RCS file: /sources/gnash/gnash/server/vm/ASHandlers.cpp,v
> retrieving revision 1.232
> retrieving revision 1.233
> diff -u -b -r1.232 -r1.233
> --- server/vm/ASHandlers.cpp    25 Apr 2008 10:01:34 -0000      1.232
> +++ server/vm/ASHandlers.cpp    28 Apr 2008 07:51:04 -0000      1.233
> @@ -1570,17 +1570,19 @@
>     bool is_sought = true;
>
>     std::string::const_iterator it = str.begin();
> +    const std::string::const_iterator e = str.end();
> +
>     length = 0;
>
>     // First, assume it's UTF8 and try to be wrong.
> -    while (it != str.end() && is_sought)
> +    while (it != e && is_sought)
>     {
>         ++length;
>
>         offsets.push_back(it - str.begin()); // current position
>
>         // Advances the iterator to point to the next
> -        boost::uint32_t c = utf8::decodeNextUnicodeCharacter(it);
> +        boost::uint32_t c = utf8::decodeNextUnicodeCharacter(it, e);
>
>         if (c == utf8::invalid)
>         {
> @@ -1591,7 +1593,7 @@
>
>     offsets.push_back(it - str.begin()); // current position
>
> -    if (it == str.end() && is_sought)
> +    if (it == e && is_sought)
>     {
>         // No characters left, so it's almost certainly UTF8.
>         return ENCGUESS_UNICODE;
> @@ -1605,7 +1607,7 @@
>     bool was_odd = true;
>     bool was_even = true;
>     // Now, assume it's SHIFT_JIS and try to be wrong.
> -    while (it != str.end() && is_sought)
> +    while (it != e && is_sought)
>     {
>         int c = static_cast<int> (*it);
>
> @@ -1862,9 +1864,9 @@
>
>     const std::string s = env.top(0).to_string();
>
> -    std::string::const_iterator it = s.begin();
> +    std::string::const_iterator it = s.begin(), e = s.end();
>
> -    boost::uint32_t out = utf8::decodeNextUnicodeCharacter(it);
> +    boost::uint32_t out = utf8::decodeNextUnicodeCharacter(it, e);
>
>     /// Always valid, or can it be undefined?
>     env.top(0).set_int(out);
>
>
> _______________________________________________
> Gnash-commit mailing list
> address@hidden
> http://lists.gnu.org/mailman/listinfo/gnash-commit
>