changes

来自「SumatraPDF是一款小型开源的pdf阅读工具。虽然玲珑小巧（只有800多K」· 代码 · 共 1,627 行 · 第 1/5 页

CHANGES BETWEEN 2.1.5 and 2.1.4

  I. IMPORTANT BUG FIXES

    - Parsing the /CIDFontName field  now removes the leading slash to
      be in sync with other font drivers.

    - gzip support was buggy.  Some fonts could not be read.

    - Fonts which  have nested subglyphs  more than one level  deep no
      longer cause a segfault.

    - Creation of synthetic  cmaps for fonts in CFF  format was broken
      partially.

    - Numeric  font  dictionary entries  for  synthetic  fonts are  no
      longer overwritten.

    - The font matrix  wasn't applied to the advance  width for Type1,
      CID, and  CFF fonts.  This caused problems  when loading certain
      synthetic Type 1 fonts like `Helvetica Narrow'.

    - The test  for the charset registry  in BDF and PCF  fonts is now
      case-insensitive.

    - FT_Vector_Rotate  sometimes  returned   strange  values  due  to
      rounding errors.

    - The  PCF  driver  now  returns  the  correct  number  of  glyphs
      (including an artificial `notdef' glyph at index 0).

    - FreeType now  supports buggy CMaps  which are contained  in many
      CJK fonts from Dynalab.

    - Opening  an invalid  font  on a  Mac  caused a  segfault due  to
      double-freeing memory.

    - BDF  fonts  with  more   than  32768  glyphs  weren't  supported
      properly.


  II. IMPORTANT CHANGES

    - Accessing bitmap font formats has been synchronized.  To do that
      the FT_Bitmap_Size  structure has  been extended to  contain new
      fields `size', `x_ppem', and `y_ppem'.

    - The FNT driver now returns multiple faces, not multiple strikes.

    - The `psnames'  module has been  updated to the Adobe  Glyph List
      version 2.0.

    - The `psnames' module now understands `uXXXX[X[X]]' glyph names.

    - The algorithm for guessing the font style has been improved.

    - For fonts in SFNT format, root->height is no longer increased if
      the line gap  is zero.  There exist fonts  (containing e.g. form
      drawing  characters) which  intentionally have  a zero  line gap
      value.

    - ft_glyph_bbox_xxx  flags   are  now  deprecated   in  favour  of
      FT_GLYPH_BBOX_XXX.

    - ft_module_xxx   flags   are   now   deprecated  in   favour   of
      FT_MODULE_XXX.

    - FT_ENCODING_MS_{SJIS,GB2312,BIG5,WANSUNG,JOHAB}      are     now
      deprecated               in               favour              of
      FT_ENCODING_{SJIS,GB2312,GIB5,WANSONG,JOHAB}  -- those encodings
      are not specific to Microsoft.


  III. MISCELLANEOUS

    - The  autohinter  has been  further  improved;  for example,  `m'
      glyphs now retain its vertical symmetry.

    - Partial support of Mac fonts on non-Mac platforms.

    - `make   refdoc'   (after   first   `make')   builds   the   HTML
      documentation.  You need Python for this.

    - The make build system should  now work more reliably on DOS-like
      platforms.

    - Support for  EMX gcc  and Watson C/C++  compilers on  MS-DOS has
      been added.

    - Better VMS build support.

    - Support for the pkg-config  package by providing a `freetype.pc'
      file.

    - New configure option --with-old-mac-fonts for Darwin.

    - Some source files have been  renamed (mainly to fit into the 8.3
      naming scheme).


======================================================================

CHANGES BETWEEN 2.1.4 and 2.1.3

  I. IMPORTANT BUG FIXES

    - Updated  to newest  libtool  version, fixing  build problems  on
      various platforms.

    - A fix  in the Gzip stream  reader: It couldn't  read certain .gz
      files properly due to a  small typo.  In certain cases, FreeType
      could  also loop  endlessly  when trying  to  load tiny  gzipped
      files.

    - The configure script now tries  to use the system-wide zlib when
      it  finds one  (instead of  the  copy found  in src/gzip).   And
      `freetype-config' has  been updated to return  relevant flags in
      this case when invoked with `--libs' (e.g. `-lzlib').

    - Certain fonts couldn't be loaded  by 2.1.3 because they lacked a
      Unicode   charmap  (e.g.   SYMBOL.TTF).    FreeType  erroneously
      rejected them.

    - The CFF loader was modified to accept fonts which only contain a
      subset of  their reference charset.  This  prevented the correct
      use of PDF-embedded fonts.

    - The logic to detect Unicode charmaps has been modified.  This is
      required to  support fonts which include both  16-bit and 32-bit
      charmaps (like very  recent asian ones) using the  new 10 and 12
      SFNT formats.

    - The TrueType  loader now limits  the depth of  composite glyphs.
      This is necessary to prevent broken fonts to break the engine by
      blowing the stack with recursive glyph definitions.

    - The CMap cache is now  capable of managing UCS-4 character codes
      that   are   mapped   through   extended  charmaps   in   recent
      TrueType/OpenType fonts.

    - The   cache  sub-system   now  properly   manages  out-of-memory
      conditions  instead of  blindly  reporting them  to the  caller.
      This means that it will try to empty the cache before restarting
      its allocations to see if that can help.

    - The  PFR driver  didn't return  the list  of  available embedded
      bitmaps properly.

    - There was  a nasty  memory leak when  using embedded  bitmaps in
      certain font formats.


  II. IMPORTANT CHANGES

    - David Chester  contributed some enhancements  to the auto-hinter
      that  significantly increase  the  quality of  its output.   The
      Postscript hinter was also improved in several ways.

    - The FT_RENDER_MODE_LIGHT render mode was implemented.

    - A new  API function called `FT_Get_BDF_Property'  has been added
      to FT_BDF_H to  retrieve BDF properties from BDF  _and_ PCF font
      files.   THIS  IS  STILL  EXPERIMENTAL,  since  it  hasn't  been
      properly tested yet.

    - A Windows FNT specific API has been added, mostly to access font
      headers.  This is used by Wine.

    - TrueType tables  without an `hmtx' table are  now tolerated when
      an  incremental interface  is  used.  This  happens for  certain
      Type42 fonts passed from Ghostscript to FreeType.

    - The PFR font driver is  now capable of returning the font family
      and style  names when  they are available  (instead of  the sole
      `FontID').   This  is  performed  by parsing  an  *undocumented*
      portion of the font file!


  III. MISCELLANEOUS

    - The path stroker in FT_STROKER_H has entered beta stage.  It now
      works very  well, but  its interface might  change a bit  in the
      future.  More on this in later releases.

    - The documentation for  FT_Size_Metrics didn't appear properly in
      the API reference.

    - The file docs/VERSION.DLL has been updated to explain versioning
      with FreeType  (i.e., comparing release/libtool/so  numbers, and
      how to use them in autoconf scripts).

    - The  installation  documentation  has been  seriously  revamped.
      Everything is now in the `docs' directory.


======================================================================

CHANGES BETWEEN 2.1.3 and 2.1.2

  I. IMPORTANT BUG FIXES

    - FT_Vector_Transform  had  been  incorrectly modified  in  2.1.2,
      resulting  in  incorrect   transformations  being  applied  (for
      example, rotations were processed in opposite angles).

    - The format  8 and 12 TrueType charmap  enumeration routines have
      been fixed (FT_Get_Next_Char returned invalid values).

    - The  PFR font driver  returned incorrect  advance widths  if the
      outline  and metrics resolution  defined in  the font  file were
      different.

    - FT_Glyph_To_Bitmap now returns  successfully when called with an
      FT_BitmapGlyph argument (it previously returned an error).

    - A bug  in the Type 1  loader that prevented  valid font bounding
      boxes to be loaded from multiple master fonts.

    - The SFNT  validation code has been rewritten.   FreeType can now
      load `broken'  fonts that were  usable on Windows, but  not with
      previous versions of the library.

    - The computation of bearings in the BDF driver has been fixed.

    - The Postscript hinter crashed when trying to hint certain glyphs
      (more precisely,  when trying to  apply hints to an  empty glyph
      outline).

    - The  TrueType glyph  loader  now supports  composites in  `Apple
      format'  (they differ slightly  from Microsoft/OpenType  ones in
      the way transformation offsets are computed).

    - FreeType was  very slow at opening certain  asian CID/CFF fonts,
      due to  fixed increment  in dynamic array  re-allocations.  This
      has  been changed  to  exponential behaviour  to get  acceptable
      performance.



  II. IMPORTANT CHANGES

    - The PCF driver now supports gzip-compressed font files natively.
      This means that  you will be able to use  all these bitmap fonts
      that  come with  XFree86 with  FreeType (and  libXft/libXft2, by
      extension).

    - The  automatic and  postscript hinters  have both  been updated.
      This  results in  a relatively  important increase  of rendering
      quality since  many nasty defaults have been suppressed.  Please
      visit the web page:

        http://www.freetype.org/hinting/smooth-hinting.html

      for additional details on this topic.

    - The `load_flags' parameter of `FT_Load_Glyph' is now an FT_Int32
      (instead  of just  being  an FT_Int).   This  breaks source  and
      binary  compatibility for  16bit systems  only,  while retaining
      both of them for 32 and 64 bit ones.

      Some new flags have been added consequently:

        FT_LOAD_NO_AUTOHINT   :: Disable the use of the auto-hinter
                                 (but not native format hinters).

        FT_LOAD_TARGET_NORMAL :: Hint and render for normal
                                 anti-aliased displays.

        FT_LOAD_TARGET_MONO   :: Hint and render for 1-bit displays.

        FT_LOAD_TARGET_LCD    :: Hint and render for horizontal RGB or
                                 BGR sub-pixel displays (like LCD
                                 screens).  THIS IS STILL
                                 EXPERIMENTAL!

        FT_LOAD_TARGET_LCD_V  :: Same as FT_LOAD_TARGET_LCD, for
                                 vertical sub-pixel displays (like
                                 rotated LCD screens).  THIS IS STILL
                                 EXPERIMENTAL!

      FT_LOAD_MONOCHROME   is  still   supported,  but   only  affects
      rendering, not the hinting.

      Note that the `ftview'  demo program available in the `ft2demos'
      package  has been  updated to  support LCD-optimized  display on
      non-paletted displays (under Win32 and X11).

    - The  PFR  driver  now  supports embedded  bitmaps  (all  formats
      supported), and returns correct kerning metrics for all glyphs.

    - The TrueType charmap loader  now supports certain `broken' fonts
      that load under Windows without problems.

    - The cache API has been slightly modified (it's still a beta!):

       - The type  FTC_ImageDesc has been removed; it  is now replaced
         by  FTC_ImageTypeRec.   Note that  one  of  its  fields is  a
         `load_flag' parameter for FT_Load_Glyph.

       - The  field  `num_grays' of  FT_SBitRec  has  been changed  to
         `max_grays'  in  order to  fit  within  a  single byte.   Its
         maximum value is thus 255 (instead of 256 as previously).


  III. MISCELLANEOUS

    - Added support  for the  DESTDIR variable during  `make install'.
      This simplifies packaging of FreeType.

    - Included modified  copies of the  ZLib sources in  `src/gzip' in
      order to support  gzip-compressed PCF fonts.  We do  not use the
      system-provided  zlib  for  now,   though  this  is  a  probable
      enhancement for future releases.

    - The DocMaker tool used to generate the on-line API reference has
      been   completely    rewritten.    It   is    now   located   in
      `src/tools/docmaker/docmaker.py'.  Features:

        - better cross-referenced output
        - more polished output
        - uses Python regular expressions  (though it didn't speed the
          program)
        - much  more  modular structure,  which  allows for  different
          `backends'  in  order to  generate  HTML,  XML, or  whatever
          format.