oldchangelog.html | searchcode

/htdocs/oldchangelog.html

Large files files are truncated, but you can click here to view the full file

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
    "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">

<html xmlns="http://www.w3.org/1999/xhtml">
<head>
  <meta name="generator" content=
  "HTML Tidy for Linux (vers 25 March 2009), see www.w3.org" />
  <!-- Created with AOLpress/2.0 -->
  <!-- AP: Created on: 16-Sep-2001 -->
  <!-- AP: Last modified: 29-Apr-2010 -->

  <title>Older changes to FontForge</title>
  <link rel="icon" href="fftype16.png" />
  <link rel="stylesheet" type="text/css" href="FontForge.css" />
  <style type="text/css">
/*<![CDATA[*/
  p.c3 {text-align: center}
  span.c2 {color: red}
  p.c1 {text-align: right}
  /*]]>*/
  </style>
</head>

<body>
  <div id="in">
    <table align="right">
      <tr>
        <td>
          <blockquote id="lit">
            <b><small>CHAPLAIN:</small></b><br />
            &nbsp;It was very interesting; I was dreaming I
            stood<br />
            &nbsp;On Jacob's ladder, waiting for the gates to
            open.<br />
            &nbsp;And the ladder was made entirely of diminished
            sevenths.<br />
            &nbsp;I was surprised, but not put out. Nothing<br />
            &nbsp;Is altogether what we suppose it to be.

            <p class="c1"><i>The Lady's not for Burning</i><br />
            Christopher Fry</p>
          </blockquote>
        </td>
      </tr>
    </table>

    <p><a href="changelog.html">Current changes</a><br />
    <a href="pfaeditchangelog.html">Changes to PfaEdit (predecessor
    to FontForge)</a><br />
    Changes to the <a href="sfdchangelog.html">sfd format</a>.</p>

    <ul>
      <li>24-August-2008

        <ul>
          <li>
            <table align="left">
              <tr>
                <td><img src="FF32.png" width="32" height=
                "32" /></td>

                <td><img src="FFsfd32.png" width="32" height=
                "32" /></td>
              </tr>
            </table>

            <p>Changes to create a stub Mac application. Add an
            apple event handler to fontforge so that it will notice
            when the finder wants it to open files. Add icons, etc.
            so it will look like a mac app.</p>
          </li>

          <li>Make it possible to drag and drop lookups

            <ul>
              <li>from one font info window to another<br />
              &nbsp; &nbsp; (copy lookups from one font to
              another)</li>

              <li>within a font info window<br />
              &nbsp; &nbsp; (reorder lookups)</li>
            </ul>
          </li>

          <li>Layers can now be marked as either background or
          foreground.

            <ul>
              <li>Foreground layers may be generated into
              fonts</li>

              <li>Background layers may contain images. Pasting
              into an empty glyph in a background layer will not
              set the width. Splines in background layers do not
              make a glyph worth outputting.</li>
            </ul>
          </li>

          <li>Add a color picker so users can select glyph label
          colors other than red, cyan, green, etc.</li>

          <li>Memory problems with references, introduced recently,
          have been fixed.</li>

          <li>I hope I'm now getting keysyms right for non-latin
          menu short cuts.</li>

          <li>Typo in background layer patch could cause a crash in
          fonts with many layers.</li>

          <li>Werner wants an $iscid variable to determine whether
          a font is a cid-keyed font or not.</li>

          <li>add font.iscid to python too.</li>

          <li>python font.cid* variables didn't work, fix
          them.</li>

          <li>Add $haspython variable to normal scripting so I can
          detect if python is available.</li>

          <li>When pasting a reference into an order2 layer, ff
          would sometimes miss the fact that the instructions were
          out of date.</li>

          <li>Replace with references had the same problem.</li>

          <li>In scripting, instructions did not get checked for
          up-to-dateness</li>

          <li>In the charview, unlink ref only worked on the
          foreground layer.</li>

          <li>Patch by Barry (fixed up by me) to fix up open
          charviews when a layer is added.</li>

          <li>Barry points out that Contour.round and Layer.round
          are supposed to have an optional argument, but instead
          have a required arg.</li>

          <li>Just as pasting references screws up instructions, so
          does replace with reference. Do the same thing that we
          just did to rplref.</li>

          <li>When scripting without a UI, the char changed routine
          did not check to see if it should clear the instructions
          on a change. It now does.</li>

          <li>Setting device table adjustments from the accent pane
          of the Math dlg didn't work.</li>

          <li>File-&gt;Revert File (from the UI) didn't work.</li>

          <li>If a contextual chaining pos/sub in a sfd file
          referred to a non-existant lookup we would insert a NULL
          lookup into a rule. This would cause almost anything that
          used that lookup to crash. Now, simply remove the
          sequence/lookup info.</li>

          <li>When importing lookups we might import a lookup twice
          if it were nested.</li>

          <li>When generating bitmaps automatically from a script
          it was possible that the freetype library got initialized
          too late and we'd pass a NULL pointer to FT_Memory_Face
          -- and things would crash.</li>

          <li>In a quadratic layer of a glyph, when selecting an
          on-curve point and trying to make it interpolated, if
          that point had no control points then behavior was
          odd.</li>

          <li>didn't create an input context for top level window
          char and font views. This meant could only use ASCII in
          menus if the focus were in the top level window (in the
          scroll bars, or the menu bar.</li>

          <li>Typo in the code to display side bearings in the
          charview.</li>

          <li>FontForge assumed that all anchor classes for a given
          subtable were in sequence, but Panov has a font where an
          anchor class for a different subtable appears in the
          middle of a list of classes for the desired subtable.
          Result was garbage output, and View-&gt;Show Att
          crashed.</li>

          <li>View-&gt;Next/Prev [Def] &amp; Find In Font View
          don't work in charviews when there are tabs active.</li>

          <li>R. Sander points out that if you select an ordered
          set of lookups in one font and import them into another
          they become disordered. Not good.</li>

          <li>When pasting into a charview we only set the width if
          there was nothing in foreground. When pasting into the
          fontview we always set the width because we were clearing
          the foreground before we pasted (unless we did paste into
          -- then we didn't set the width).

            <p>With layers that should all change. We should only
            set the width if there is nothing in any of the
            potential foreground layers. Don't count the background
            layers or the grid psuedo layer. Don't count open
            contours (might be a secondary guide line layer or
            something). If we're going to clear a layer before
            pasting (normal behavior in fontview) then don't count
            the layer cleared.</p>
          </li>

          <li>Add several different behavior patterms for the
          metrics view. Users complain that the current
          multi-functional window is confusing and would prefer a
          view which only does kerning, or only sets the advance
          widths, etc.</li>

          <li>Due to a typo, merging two splinesets at a curve
          point generally left the point marked with default
          adjustment enabled on its control points.</li>

          <li>Another layers problem. Only in the foreground layer
          could anchor points be moved.</li>

          <li>Python glyph.unlinkReference command didn't
          work.</li>

          <li>My intent was that ligature carets positioned at 0
          didn't count (0 isn't meaningful -- a ligature with a
          zero-width first component?) but that only got
          implemented for 'lcar' table not for 'GDEF'. So fix that.

            <p>Also allow the user to specify exactly how many
            ligature carets s/he wants rather than just setting a
            minimum bound.</p>

            <p>Also 'lcar' (Apple) output was broken.</p>

            <p>Also neither AAT nor OpenType supports negative lig
            carets, so if the user drags a lig caret line negative,
            have it stick at 0.</p>
          </li>

          <li>Moving ligature carets always marked for an undo in
          the foreground layer. Register the movement in the layer
          in which it happens.</li>

          <li>When placing an accent glyph about which we know
          nothing, assume it should be centered on top.</li>

          <li>Change sfd.c so that it will treat the sequence
          "backslash newline" as no characters on input (er, except
          in ASCII85 input). And so that it thinks it can parse
          version 4 or the sfd format.

            <p>We don't generate version 4 yet, and we won't for a
            while. Version 4 will include the ability to insert
            backslash newline into long lines. This is so we can
            conform to the internet requirements of fewer than 990
            bytes on a line. And the delay is so that people don't
            scream at me for making an incompatible change to the
            format -- if I give the parser time to perculate
            through the user community, with luck no one will even
            notice -- especially as long lines are rare.</p>
          </li>

          <li>Selections/Clipboard didn't work in textfields on
          64bit intel. There seems to be a bug in the 64bit Xlib
          which doesn't transfer 32 bit quantities properly. I was
          passing unicode strings as 32bit UCS4, but UTF8 will work
          just fine instead.</li>

          <li>Add the ability to clear all device tables associated
          with a kerning by classes table. (Needed because of a bug
          which introduced garbage).</li>

          <li>We didn't think we could read in a kerning subtable
          (class or pair) with device tables, so we converted them
          to pst data. We can read them in in optimized form and we
          should do so.</li>

          <li>Alexej points out that if a glyph has
          "instructions_out_of_date" set on it, then setting the
          instructions via a script should clear the bit -- but
          didn't.</li>

          <li>Werner points out that if a type2 charstring
          contained a sequence &lt;num&gt; endchar then ff would
          set the width of the charater even if the the width had
          already been set. Width can only be set on first stack
          clearing operation, any extra params on subsequent calls
          get ignored.

            <p>Actually it's worse than that, the spec says that if
            there are too many things on the stack then start
            parsing the stack from the unexpected end.</p>

            <p>Adobe have admitted that the obvious reading of
            their spec is incorrect (they can't quite bring
            themselves to say the spec is wrong). So revert some of
            the way we were parsing type2 charstrings.</p>
          </li>

          <li>Barry complains that when building ff does not apply
          the ".cap" suffix when using inverted breve.</li>

          <li>Valgrind finds a bug in drawing non-BMP characters in
          the metrics view.</li>

          <li>Alexej would like to be able to manipulate
          non-unicode glyphs with the text field of the metrics
          view. The textfield used to map all non-unicode code
          points to 0xfffd, but if the font doesn't make use of the
          supplementary private use areas (planes 15&amp;16) then
          create a mapping from the font into these areas so that
          we can represent non-unicode glyphs as funky private use
          glyphs.

            <p>(Only do this if font doesn't use these areas).</p>
          </li>

          <li>Alexej points out that in non-foreground quadratic
          fonts the "Points-&gt;Dont Interpolate" command doesn't
          do anything.

            <p>Fix a couple of other foreground specific things
            too.</p>
          </li>

          <li>Barry points out that setting a glyph's anchorPoints
          from python does not mark a glyph as changed nor force an
          update.</li>

          <li>Autoinstructor: large patch from Alexey:

            <ul>
              <li>handlles conflicting H/V stems much better;</li>

              <li>tries to place HStems between blues better;</li>

              <li>adds optional apparatus for instructing
              serifs;</li>

              <li>relies on generic point-to-stem snapping routines
              from stemdb.c;</li>

              <li>needs FPGM changed (older FF's glyph programs
              should still work, but FPGMs need updating!).</li>
            </ul>
          </li>

          <li>If the user has opened a backup sfd file (one that
          ends in .sfd~) and does a save on it, then pop up a save
          as dlg instead of blindly saving it. Saving it back to
          the backup is probably wrong, yet saving it onto the
          non-backup version without prompting the user is
          questionable.</li>

          <li>glyph.appendAccents didn't mark the glyph as changed
          nor did it force an update.</li>

          <li>Barry wants a way to control the use_my_metrics bit
          of a reference from python.<br />
          Give him a hack<br />
          glyph.useRefsMetrics(name[,set])</li>

          <li>Add a glyph.appendAccent() command.</li>

          <li>Ralf from end-if.de points out that when trying to
          change the kerning between two glyphs which have a kern
          class setting of 0, then ff keeps asking the same
          question several times before a non-zero value is
          entered. (once when the field is clicked in, once when a
          minus sign is typed, and once more when a number is
          entered).</li>

          <li>The python stuff won't compile under cygwin because
          it used pointers to stuff in shared libraries as
          initializers of static variables. It seems windows can't
          handle that. This patch should provide a work
          around.</li>

          <li>Add a font.importLookups (from another font) function
          to python scripting.</li>

          <li>When joining two splinesets (during glyph editing) we
          didn't pay attention to any user settings for the control
          points, and we'd just always refigure default control
          points. Now use user specified directions when
          possible.</li>

          <li>Add a problem (Find Problems) which will look through
          all glyphs for those with both contours and references,
          and which will create a new glyph to contain the contours
          and make a reference to it in the original.

            <p>Do something similar for bad transformation matrices
            (those with members 2 or more (or less than -2)).</p>
          </li>

          <li>Importing a kerning class from one font to another
          (and perhaps merging fonts) could cause memory problems
          if the number of second classes were bigger than the
          number of first classes.</li>

          <li>Retain sizes of bitmap and metrics views. People seem
          to like this even if I don't.</li>

          <li>Still problems with the guide lines layer

            <ol>
              <li>Displaying a point read bad memory (probably
              harmless, might cause visual artifacts).</li>

              <li>Edit-&gt;Undo in guide layer trampled
              memory.</li>
            </ol>
          </li>

          <li>Find Problems would occasionally pop up a blank
          problem dlg.</li>

          <li>Only show points of inflection in the active
          layer.</li>

          <li>If a glyph window were displaying something other
          than the foreground layer then changes in that window
          would not be reflected in dependant glyph windows (a
          change to tilde didn't show in atilde) nor in the
          fontview.</li>

          <li>In a cubic glyph with references, turning on point
          numbering did not renumber the references. They seem to
          be numbered for TrueType. Not sure where that comes
          from.</li>

          <li>Alexey points out that after deleting the active
          layer, ff does not update the internal bitmap font it
          used to display glyphs in the font view.</li>

          <li>Make "-lang ff" work for "-lang=ff", it's more
          consistent with the rest of the arguments.</li>

          <li>Add a -unique argument which means that FF checks to
          see if another fontforge is running on the current
          screen. If no other fontforge exists things proceed
          normally, but if there is another, then we send it our
          arguments and ask it to open them, and then we exit.</li>

          <li>Make the OFLib dialog have an option to download a
          preview when you click on a line (rather than requiring
          you press the [Preview] button for each).</li>

          <li>The BitmapsAvail and BitmapsRegen scripting functions
          were broken by the layers patch (they generated bitmaps
          of whatever was in the background layer).</li>

          <li>Put in some checks to warn the user when a 'kern'
          table subtable uses more than 10920 pairs.</li>

          <li>Give the user one of those "This operation cannot be
          undone" warnings when deleting a layer in Font Info.
          (Similar warning about instructions if user changes last
          quadratic layer(s) to cubic without deleting any).</li>

          <li>When deleting the last quadratic layer, delete the tt
          instructions also.</li>

          <li>When deleting a layer from a font, ff very carefully
          went through all open glyph views to make sure none was
          looking at the layer being deleted (and fixing them up if
          they were) -- but it totally ignored the font views.
          Their layers needed to be updated too. And the metrics
          views.</li>

          <li>The Import menu item in Font Info-&gt;Lookups popup
          menu was always enabled. It should not have been.</li>

          <li>Add menu items in Font Info-&gt;Lookups (popup menu)
          to do a global add/remove language from a specified
          script in all selected lookups.</li>

          <li>Get Info (maybe a few other things) didn't work on a
          selected reference which wasn't in the foreground
          layer.</li>

          <li>Copy/Paste of references only worked well in
          foreground layer.</li>

          <li>Only in the foreground layer was it possible to
          select a reference by clicking on it.</li>

          <li>Snapping to blues was broken due to a typo in the
          code.</li>

          <li>Still were a number of places where behavior depended
          on the spline order in the foreground layer:

            <ul>
              <li>Could only debug a glyph if the foreground layer
              were quadratic</li>

              <li>Show gridfit used some point numbers from the
              foreground layer and some from the layer being
              shown.</li>

              <li>Gridfit didn't change when changing layers.</li>
            </ul>
          </li>

          <li>Peter Lunicks points out that a tooltip in the prefs
          dlg actually was for a different preference item.</li>
        </ul>
      </li>

      <li>20-July-2008

        <ul>
          <li>Updated Russian UI</li>

          <li>Updated Simplified Chinese UI</li>

          <li>Small Caps dialog greatly rewritten and merged into
          the sub/superscript dialog and a new change glyph
          dialog.</li>

          <li>Alexey Kryukov makes Small Caps and friends deal with
          diagonal stems.</li>

          <li>Update the upload mechanism to OFLib to send a
          preview image, license text (if specified in our
          database) and fontlog (again, if in our database).
          Currently OFLib throws away both the license info and the
          fontlog, but I hope they will fix that in time.</li>

          <li>Add a dialog to browse fonts on the OFLib.</li>

          <li>Fix some bugs in feature file generation of
          contextual chaining lookups which refer to marks. Note
          these are not supported in the spec. This is a FontForge
          only extension.</li>

          <li>If a script contain no default language features (but
          did contain some languages) then ff produced a garbage
          script table for GPOS/GSUB.</li>

          <li>A number of Dialogs still did not autosize themselves
          for different languages Create Hints, Kern (Anchor) Pair,
          Review Hints.</li>

          <li>Still places where ff uses the foreground layer
          rather than the current layer to determine if a glyph is
          quadratic.</li>

          <li>On systems using different font metrics, the debug
          pane looks bad with the buttons partly covered by the
          instruction pane. If the font is large, and the cvt dlg
          empty, then I'm told ff complains.</li>

          <li>Barry points out that Add Encoding Slots scrolls to
          the wrong place.</li>

          <li>Adding glyphs in compacted encodings caused
          problems.</li>

          <li>When the user resizes the outline glyph window,
          retain that size and use it for the default size of
          subsequent invocations of the window.</li>

          <li>It used to be that whenever ff opened an sfd file
          then all subsequent new fonts (or fonts opened from real
          font files) would be displayed in windows the same size
          as the one requested in the sfd file. Only set the window
          size preference when the user explicitly changes it, not
          upon openning something with a different setting.</li>

          <li>In a CID-keyed font, when opening an empty CID slot,
          ff could trample memory.</li>

          <li>BlueFuzz support for autohinter and metafont.</li>

          <li>Add some checks to make sure that we can survive bad
          font data without crashing.</li>

          <li>When trying to change kerning between the last two
          glyphs in the metrics view using the text field (rather
          than dragging the line) then the textfield would change,
          but the kerning would not.</li>

          <li>In the metrics view, give the user several options
          about the grid (which shows the advance widths of the
          glyphs)

            <ul>
              <li>hidden</li>

              <li>fully displayed</li>

              <li>small lines at the top and bottom of the
              pane</li>

              <li>fully displayed except when the user is adjusting
              kerning or the advance width, etc. When adjusting
              grid vanishes until user releases the mouse
              button.</li>
            </ul>

            <p>(setting is sticky).</p>
          </li>

          <li>Allow the user to customize the colors in the metrics
          view.</li>

          <li>In the metrics view (or in the print display dlg)
          when glyphs are overlapping (e asier to see when glyphs
          are large) then the places where one glyph ends and
          another begins often have a faint halo of lighter pixels.
          Try to improve things and make these joins be more
          uniform in color.</li>

          <li>Add minimal support for alpha channels (treat them as
          a bitmask for transparent).</li>

          <li>I find it really hard to believe that this is an
          error, but I have a second hand report that MicroSoft
          claims that it is. To me making it erroneous is pointless
          and annoying, but I don't write the spec.<br />
          Add to Find Problems and Validate a check to see that if
          a base glyph (or mark) contains an anchor point in one
          anchor class in a lookup subtable, then it must have
          anchor points for ALL anchor classes in that
          subtable.</li>

          <li>Fontimage would crash if asked to process a font with
          glyphs not in BMP. Make fontimage better about fonts
          which contain no standard scripts (Apostolos's Phaistos
          font).</li>

          <li>A few tweaks to make stemdb.c usable for resizing
          glyphs and stem control.</li>

          <li>When importing lookups into a font, append the
          fontname to the lookup name only if there is already a
          lookup in the new font with the original name.</li>

          <li>Empty glyphs containing a "." in the name could cause
          ff to crash when doing or checking to see if it could do
          a build accented.</li>

          <li>Gasp editing was broken (still) from the change to
          the cleartype stuff.</li>

          <li>Andrey P. would like to see a reference's bounding
          box in the Element-&gt;Get Info dlg for the
          reference.</li>

          <li>Revert glyph had some stray memory references.</li>

          <li>Add two proposed language tags for IPA use (one for
          IPA and one for Americanist).</li>

          <li>In an empty python layer layer.boundingBox should
          return a tuple of 4 zeros, instead it returned a tuple of
          2 zeros and 2 random garbage values.</li>

          <li>Validate dlg's scrollbar was broken, and you couldn't
          drag the thumb up to the top (you could get it to the top
          by other ways).</li>

          <li>Add a revalidate all menu item to the window to force
          a reevaluation of the entire font.</li>

          <li>Add ability to import graphics from a page of a pdf
          file.</li>

          <li>Python interface does not categorize points.</li>

          <li>The (python) psMat.skew command skewed vertically,
          rather than horizontally and by the wrong amoung.</li>

          <li>When doing a font sample from a script which did not
          set the page width/height explicitly, then ff would use a
          size of 0 for each. All text drawn was out of the imaging
          area, lines were too short (as well as invisible), too
          many pages (all blank).</li>

          <li>Barry points out that "print" is a python keyword. So
          font.print() doesn't work. Call it printSample()
          instead.</li>

          <li>Freetype now (or in the next release, 2.3.7) supports
          dropout modes, which means fontforge must do a bit of
          work when generating rasters for the debugger.</li>

          <li>Move some logic out of fontimage.pe and into
          fontforge to make a default string for a font image more
          widely available.</li>

          <li>When saving an image fromt the print dlg, the top
          line was off the top of the image.</li>

          <li>Interpolated points need not be on integral
          boundaries, but ff was checking for them. Different
          problems in validator and find problems, but both
          wrong.</li>

          <li>In the point near hint check (find problems), ff was
          leaving the hint selected.</li>

          <li>Fix some rounding errors in spline/line
          intersection.</li>

          <li>in Font Info, after user pressed [OK] FF set the
          state of the subscript superscript default checkbox to
          the panose default checkbox by mistake. This means if the
          user defined panose and not sub/super garbage would
          appear in the OS/2 sub/super fields.</li>

          <li>fontimage didn't get the script right, so features
          weren't applied properly.</li>

          <li>More work on conflicting stems.</li>

          <li>Pressing [Done] in the print dialog would right to a
          random address on the stack. Depending on what was going
          on when this was done it could cause a crash.</li>

          <li>Fix an uninitialized variable introduced by the last
          change to order2 conversion.</li>

          <li>Add the ability to open an archive and peek inside it
          for fonts. Useful if we want to open stuff from the open
          font library which is often packaged in zip files.</li>

          <li>Add a new validation check to see if a glyph has
          non-integral coordinates. Only an error in TrueType (and
          can't happen when fontlinting a truetype file, but could
          happen on an sfd file). Add some logic to ask user if
          they intend to generate truetype in cases where we aren't
          sure.</li>

          <li>Add a preference which allows people to change the
          pixmap dir -- and do what we can to reload images from
          the new dir. Some old windows may not be updated, but new
          ones should reflect the new stuff.</li>

          <li>When ff changed the name of a glyph, it fixed up all
          the tabs in charviews to the new name -- but in neglected
          to refresh them. So they still showed on the screen with
          the old name (until something forced a refresh).</li>

          <li>The Anchor Class dialog (which lists all anchor
          classes of a subtable) only let you create an Anchor
          Control dialog for the first anchor class.</li>

          <li>Guard (harder) against generating pathological
          results when converting cubic to quadratic splines.</li>

          <li>If we had a spline which almost doubled back on
          itself, sufficiently so that the spline intersector found
          lots of really close points, which it called
          intersections, but not sufficiently so that our collinear
          detector caught it, then we'd generate a ton of
          intersections, overflow certain arrays, and trash the
          stack. =&gt; ff would crash. I didn't think this could
          happen, until someone provided a test case, so put in
          guards against this.</li>

          <li>provide python access to

            <blockquote>
              glyph.altuni
            </blockquote>

            <p>either None, or a tuple of tuples, one sub-tuple for
            each alternate. each sub-tuple is (alternate-unicode,
            variation selector (or -1), reserved for future=0
            now).</p>
          </li>

          <li>Trying to build dieresis with "PreferSpacingAccents"
          set meant that we tried to build it out of itself, and
          crashed.</li>

          <li>Also fix so that 0x22EF (ellipsis variant) is built
          out of centered dot in a spacing mode, while Ldot uses it
          in a non-spacing mode.</li>

          <li>Merge fonts broken again. This time references didn't
          work.</li>

          <li>Provide a mechanism for loading image files to make
          menu item icons (instead of expecting the images to live
          in memory inside of fontforge). Then provide a bunch of
          these icons so the menus will look more interesting.

            <p>Provide some icons</p>

            <p>Also update the install procedure to put them in the
            right place.</p>

            <p>Preference item containing a directory from which to
            read icon files</p>
          </li>

          <li>FF crashed when opening CID keyed fonts with
          compact_on_open set.</li>

          <li>If fontforge got an end of file in the middle of
          reading a (corrupt) cff font, then ff would sit there
          reading endlessly and ignoring the error.</li>

          <li>Off by one error in drawing the large "pixel"
          rectangles that show grid fit generated.</li>

          <li>Werner points out that my instruction glosses are out
          of date:

            <ol>
              <li>Was following Apple's docs for SCANTYPE, not MS.
              MS has a number of new values.</li>

              <li>Was using an old convention for accessing PPEM
              which broke after freetype 2.3.5</li>

              <li>Added the FreeType return for GETINFO.</li>
            </ol>
          </li>

          <li>Featurefile patch from Barry.</li>

          <li>I used to think that the max glyph instruction length
          field of 'maxp' refered to fpgm and prep. I thought this
          because there was a footnote (which I now see was for a
          different field) saying the field also applied to fpgm
          and prep.</li>
        </ul>
      </li>

      <li>07-June-2008

        <ul>
          <li>New Russian translation of the UI</li>

          <li>Sometimes FF didn't set the message locale directory
          properly.</li>

          <li>Since FF will crash if you load a feature file when
          the Font Info window is open for that font, close the
          Font Info window when loading a feature file.</li>

          <li>New <a href="Styles.html#Italic">dialog</a> to help
          users convert an upright font into an italic one.</li>

          <li>New <a href="Styles.html#Smallcaps">dialog</a> to
          help users create small caps glyphs.</li>

          <li>New <a href="Styles.html#SubSuper">dialog</a> to help
          users create subscript/superscript glyphs.</li>

          <li>And matching python commands: font.italicize(),
          font.addSmallCaps(), layer.stemControl()</li>

          <li>Make it clear that the transform dlg from the
          fontview will move the width of a glyph by default, and
          provide a checkbox so the user can turn this off.</li>

          <li>Fix various bugs in argument handling of
          fontforge.askChoices</li>

          <li>Change font.removeGlyph to accept a single string
          argument to remove the glyph with that name.</li>

          <li>FF didn't like decimal points in the metrics view l/r
          bearing fields -- even though it would put them there
          itself.</li>

          <li>Metrics windows weren't being remetricked after a
          feature file was loaded (nor an afm file for that
          matter). Nor were they remetricked after lookups changed
          in fontinfo.</li>

          <li>Add to python

            <ul>
              <li>font.revert()</li>

              <li>font.revertFromBackup()</li>

              <li>font.buildOrReplaceAALTFeatures()</li>
            </ul>
          </li>

          <li>Changing a glyph's width from a python script didn't
          update the metrics window.</li>

          <li>Adding two languages in the same script to a lookup
          in a feature file didn't work.</li>

          <li>It wasn't possible to rename an anchor class (at
          least not by the obvious method through the dlg).
          Attempting to delete an anchor class popped up an error
          message which didn't provide all its own arguments and so
          could crash on some systems.</li>

          <li>Better error recovery in feature file parsing.</li>

          <li>Werner points out that since I allow the user to see
          points in references transformed to their ultimate
          position, I should also transform the raster.</li>

          <li>Patch by Barry to fix kerning classes (ie. figuring
          out WHAT glyphs go into a class).</li>

          <li>When parsing a feature file which used a default
          languagesystem setup, ff only used the most recent
          script.</li>

          <li>If a lookup contained no data (no subtables, or empty
          subtables) then when generating a feature file ff would
          output stuff to declare a lookup, but when there was no
          content that frame could not be parsed as it would be
          impossible to figure out the lookup type. Try to avoid
          that.</li>

          <li>FF did not put contextual lookups on all the lists
          they needed to be on when loading them from a feature
          file.</li>

          <li>Sometimes the Open Font dlg opened with the files
          shifted left so the first characters of the name were
          outside the window.</li>

          <li>Patch from Barry Schwartz. "include" syntax in
          feature files didn't work.</li>

          <li>Sort button in the lookups pane of font info didn't
          refresh the window after it finished (so things looked
          unsorted even after pressing it).</li>

          <li>When doing a Build Accented on dotted glyphname
          (agrave.sc, etc.) the accent was positioned using anchors
          in the undotted base glyph (a rather than a.sc).</li>

          <li>Fix various problems with tt debugging.</li>

          <li>The XUID and UniqueID fields of a PostScript font are
          deprecated, yet FF has continued to output them by
          default. Make it so FF no longer does so.</li>

          <li>When outputing an sfd file, ff would sometimes output
          a truetype spline as a line if it had a collinear
          control-point. I thought I'd taken care of that, but
          here's another case which needed fixing.</li>

          <li>Typo when loading the underline position field from a
          'post' table. I was subtracting half the underline width.
          I should have been adding it instead. (Why did adobe use
          a different definition of the same concept in two places?
          GRRR).</li>

          <li>It wasn't possible to make the foreground layer
          invisible.</li>

          <li>When the file chooser bookmark list got too long ff
          would crash.</li>

          <li>I wasn't doing the obvious checks to the sfnt header
          in font lint. So now:

            <ol>
              <li>Check that there are no duplicate table
              entries</li>

              <li>Check that there are no overlapping tables</li>

              <li>Check that no tables extend beyond the end of
              file</li>

              <li>Check that the file's checksum is correct</li>

              <li>Check that each table's checksum is correct</li>

              <li>Check that the number of tables is reasonable and
              that the binary search header is consistent with the
              number of tables.</li>

              <li>Check that required tables are present</li>

              <li>Check that certain fixed length tables have the
              correct length.</li>
            </ol>
          </li>

          <li>Werner would like to be able to debug (truetype
          instructions) when using non-square pixels (x_ppem !=
          y_ppem).</li>

          <li>Barry wants to be able to close a tab. Add a menu
          item.</li>

          <li>When a glyph's name changed, I did not update the
          names in any tabs in the charviews -- so those tabs would
          stop working after the change.</li>

          <li>Scrollbar in points debug window wasn't set
          right.</li>

          <li>A. Panov provides a patch. Prefs was losing
          bookmarks.</li>

          <li>the new release of volt reveals some new opentype
          tags.</li>

          <li>If we have a glyph named "agrave.sc" or "eacute.sc"
          then FF will now let you build this as an accented
          character. It will remove the suffix from then name, and
          look up the components of the bare glyph. Then for the
          first component it will search for a glyph named "a.sc"
          and insert it. For the accents it will use "grave.sc" if
          present, and "grave" if it is not.</li>

          <li>FF wasn't handling a FontMatrix in the top dict of a
          cid font.</li>

          <li>Trying to filter for fig files in the Import dlg
          caused a crash.</li>

          <li>Check for bad OS/2 versions (that is, versions
          windows will reject) when loading a font.<br />
          Connect validation's "Points too far" with Find Problem's
          version.<br />
          Make fontlint aware of the new validation states.<br />
          When generating a font we complained about bad os/2
          version only if they asked for a validation. Move it
          somewhere more prominant.</li>

          <li>Add two new problems to find

            <ol>
              <li>Points too far apart</li>

              <li>Bitmap advance width different from scaled
              outline advance widths</li>
            </ol>

            <p>(also boxify the Find Problems dlg).</p>
          </li>

          <li>Add a new validation state: Test that points aren't
          too far appart (must be within 32767 of each other or we
          can't output them).</li>

          <li>Try a little harder to get some of the obscure code
          pages right in the OS/2 table.</li>

          <li>Was getting the first glyph index of OS/2 wrong for
          OTF fonts.</li>

          <li>If an sfnt contains embedded bitmaps, and if those
          bitmaps contain a glyph whose advance width is different
          from that expected by scaling the horizontal advance of
          the vector data, then windows will only get the correct
          metrics if the "Instructions modify width" bit is set in
          the 'head'.flags field.</li>

          <li>If the "changedsincelasthinted" bit got set on a
          glyph by the unlink/remove overlap flag on a glyph during
          font saving, then the glyph did not get restored to its
          original state.</li>

          <li>Doing an Undo in the Guidelines layer didn't work,
          usually crashed.</li>

          <li>Validate didn't work on CID keyed fonts.</li>
        </ul>
      </li>

      <li>29-Apr-2008

        <ul>
          <li>Validate didn't work in CID keyed fonts.</li>

          <li>Pierre Hanser wants an X Resource
          (fontforge.CharView.Rulers.FontSize) to specify the font
          size of text displayed in the info line and the rulers. I
          have applied his patch, but have made no effort to insure
          that text will fit in the space allowed for it.</li>

          <li>Hmm. Freetype seems to dislike rasterizing some
          complex outlines at large pixelsizes in black and white,
          but seems fine with doing them in greyscale.</li>

          <li>When opening a ttc file (and posting a dlg which
          lists all the fonts in the file) try to find a fontname
          in the current locale. We used to look first for English
          and then anything. Now try: First current locale, then
          english, then anything.</li>

          <li>Make the Open Font dlg (and similar file chooser
          dlgs) be customizable, and store that info in prefs.
          Specifically control whether directories are displayed in
          alphabetical order among other files, before normal
          files, or in a separate pane.

            <p>Also add a bookmark (and back/forward)
            mechanism.</p>
          </li>

          <li>Add support for opening a font from http &amp; ftp
          URLs. Can browse ftp URLs, and save fonts back to
          ftp.<br />
          <code>$ fontforge
          http://openfontlibrary.org/people/Nasenbaer/Nasenbaer_-_Widelands.ttf</code></li>

          <li>The validation patch broke the save mac family
          dialog, and then the upload to open font library broke it
          further.</li>

          <li>When doing expand stroke, it is possible to have two
          points so close together that there is no data on the
          curve in between the two. If this happens
          ApproximateFromPointsSlopes can't do a good job. So be
          careful of this case.</li>

          <li>Layers change introduced a potential crash in the
          Embolden command.</li>

          <li>fontimage didn't draw the first line.<br />
          Add a cgi front end to <a href=
          "fontimage.html">fontimage</a> so now it is possible to
          get a font sample of any font on my website.</li>

          <li>In the Display dlg, if you changed the font, or the
          point size without selecting something, then that change
          would vanish instantly. Fix that.

            <p>Also, was a crash bug when mixing a font with a
            baseline table and a font without one (in the disply
            dlg).</p>
          </li>

          <li>Add a command to insert text into a glyph.
          (<code>Element-&gt;Insert Text Outlines)</code> It adds
          the text outlines directly (instead of a draw command
          using another font) because most fonts can't include text
          drawing in their glyphs. Also include a Bind to Path
          variant.</li>

          <li>GLyph lists in kerning classes (in show att) were
          displayed as ligatures.</li>

          <li>Upgrade to unicode 5.1</li>

          <li>If an sfd file were saved with spiros active, and
          then loaded into a version of fontforge without libspiro,
          then bad things would happen. Whenever we check that
          spiros are active also check that they can be active, and
          add some warnings for the user to let him/her know that
          spiros aren't available.</li>

          <li>Add a tile pattern command (if --enable-tile-path)
          this allows the user to replicate any given pattern in
          another glyph.

            <p>This could be used to produce a grid of guidelines,
            or, in conjunction with
            <code>Element-&gt;Overlap-&gt;Intersec</code>t it could
            provide a poor man's "Pattern Fill" (Since Pattern
            fills are only available in type3 fonts this sort of
            makes them work for any font type).</p>
          </li>
        </ul>
      </li>

      <li>30-Mar-2008

        <ul>
          <li>Add support for the OpenType 'BASE' table, and to a
          lesser extent for Apple's 'bsln' table.

            <p>Add python scripting (font.horizontalBaseline,
            font.verticalBaseline) too.</p>

            <p>Add support under Show ATT.</p>

            <p>Remove the old vertical origin field.</p>
          </li>

          <li>Add the ability to create a dummy 'DSIG' table (one
          containing no signatures) so that Windows will show an
          opentype icon for a font with extension 'ttf'</li>

          <li>Add support for gradient and pattern fills to type3
          fonts (only if <code>./configure --enable-type3</code> is
          specified, of course.

            <p>Also add support for cliping paths</p>

            <p>Make fonts containing images work in pdf output.</p>
          </li>

          <li>Bump the sfd version number. This should have
          happened in the last release, but I didn't realize it was
          needed then, thus there will be sfd files with an
          incorrect version number (2) when it should be 3. However
          that should not cause the current parser problems.</li>

          <li>New Greek translation from Apostolos</li>

          <li>Updated Polish translation from Michal</li>

          <li>Apostolos points out that compiling on systems
          without freetype (Solaris boxes) was broken after last
          months layers patch.</li>

          <li>Werner suggested various improvements to Show ATT:

            <ul>
              <li>Lookups should be sorted</li>

              <li>Glyph images should be drawn in more cases
              (before, just for substitutions)</li>
            </ul>
          </li>

          <li>Patch from Alexey, for better handling conflicts
          between diagonal and H/V stems.</li>

          <li>Do a more consistent job of handlong curved stem
          segments. Also collect point data for off-curve points
          too (may be useful in future for TTF hinting).</li>

          <li>Last month I put in a patch to make sure the lsb as
          stored in the 'hmtx' table was really the glyph's lsb if
          bit 1 of the head.flags field were set. Unfortunately I
          read in the lsb as an unsigned number which shifted
          negative lsbs off to around 65535. FF sometimes refused
          to rasterize them. Of course when the font was generated
          they got truncated back to negatives in the output.</li>

          <li>Add a preference item so that fonts will be
          automagically compacted when they are opened.</li>

          <li>Add a new preference item (SeekCharacter) to allow
          users to control which character will be displayed when
          opening a new font. Normally FontForge displays 'A', but
          that's a bit latin-alphabet-centric.</li>

          <li>If palettes were not docked, then a complicated
          series of events could lead to a crash after closing a
          TilePath (Math kerning, Gradient Fill, etc.) dialog.</li>

          <li>Was not outputting the %%LanguageLevel DSC
          comment.</li>

          <li>Generating a CID-keyed otf font was broken by the
          layers patch. (Only the background layer …
Large files files are truncated, but you can click here to view the full file