Adds misc/kernsample.py

CONTRIBUTING.md

@@ -228,6 +228,45 @@ You can now visit `http://localhost:2015/lab/`.
 After you rebuild some font files, reload the web page to refresh fonts.
 
 
+### Kerning
+
+Kerning is the concept of harmony in the pace of characters, defined as a set of distances
+between specific character pairs, like "A" & "c".
+Good kerning makes words more readable. Oftentimes this means that when adjusting kerning,
+you have to look at whole words and sentences when adjusting the kerning of a pair, since
+the spacing between two characters should work in harmony with the spacing of all other characters.
+
+All major font editors provide user interfaces for previewing and adjusting kerning.
+
+When adding or adjusting kerning:
+
+- Make sure to use kerning groups (`src/Inter-UI-*.ufo/groups.plist`)
+
+- If a glyphname is missing in kerning groups, define a new group for it.
+  Group naming scheme is: `@KERN_<DIRECTION>_<NAME>` where `<DIRECTION>`
+  is either `LEFT` or `RIGHT`, depending on if the group represents the left-hand side
+  of a pair or the right-hand side. `<NAME>` is the name of a glyph that most commonly
+  represents the group. For instance, for all glyphs that has a left-side shape similar
+  to "c", like "d", the group name is "c". This makes it easy to test kerning of groups
+  by just using the `<NAME>` part in previews.
+
+- Try to submit image samples of kerning adjustments with your pull requests whenever
+  feasible.
+
+The script `misc/kernsample.py` is helpful in generating samples for all existing
+right-hand side characters given a left-hand side glyphname.
+
+```txt
+$ misc/kernsample.py src/Inter-UI-Black.ufo P -suffix MOR
+PAMOR P/AE MOR PJMOR PXMOR PYMOR PZMOR P/ae mor P/ampersand mor P/backslash mor P/dzcaron mor P/eightsub mor P/ellipsis mor Pfmor P/four mor P/guilsinglleft mor P/idieresisacute mor P/periodcentered mor P/quotedblbase mor Psmor P/seven mor P/slash mor Ptmor P/two mor P/underscore mor Pymor
+```
+
+Type `misc/kernsample.py -h` for help on how to use the program.
+
+This only includes existing kerning and is thus only useful for adjustments.
+Additions must still be done manually.
+
+
 ## FAQ
 
 > Do I need RoboFont?

misc/kernsample.py

@@ -0,0 +1,141 @@
+#!/usr/bin/env python
+# encoding: utf8
+from __future__ import print_function
+import os, sys, plistlib
+from collections import OrderedDict
+from argparse import ArgumentParser
+from robofab.objects.objectsRF import OpenFont
+
+
+def mapGroups(groups):  # => { glyphname => set(groupname, ...), ... }
+  m = OrderedDict()
+  for groupname, glyphnames in groups.iteritems():
+    for glyphname in glyphnames:
+      m.setdefault(glyphname, set()).add(groupname)
+  return m
+
+
+def _addRightnames(groups, kerning, leftname, rightnames, includeAll=True):
+  if leftname in kerning:
+    for rightname in kerning[leftname]:
+      if rightname[0] == '@':
+        for rightname2 in groups[rightname]:
+          rightnames.add(rightname2)
+          if not includeAll:
+            # TODO: in this case, pick the one rightname that has the highest
+            # ranking in glyphorder
+            break
+      else:
+        rightnames.add(rightname)
+
+
+def fmtGlyphname(glyphname, glyph=None):
+  if glyph is not None and len(glyphname) == 1 and ord(glyphname[0]) == glyph.unicode:
+    # literal, e.g. "A"
+    return glyphname
+  else:
+    # named, e.g. "\Omega"
+    return '/' + glyphname + ' '
+
+
+def samplesForGlyphname(font, groups, groupmap, kerning, glyphname, args):
+  leftGlyph = font[glyphname]  # verify it's present
+  rightnames = set()
+
+  _addRightnames(groups, kerning, glyphname, rightnames, includeAll=args.includeAllInGroup)
+
+  if glyphname in groupmap:
+    for groupname in groupmap[glyphname]:
+      _addRightnames(groups, kerning, groupname, rightnames, includeAll=args.includeAllInGroup)
+
+  rightnames = sorted(rightnames)
+
+  out = []
+  if args.formatAsUnicode:
+    left = '\\u%04X' % leftGlyph.unicode
+    for rightname in rightnames:
+      if rightname in font:
+        rightGlyph = font[rightname]
+        if rightGlyph.unicode is not None:
+          out.append('%s\\u%04X' % (left, rightGlyph.unicode))
+  else:
+    left = fmtGlyphname(glyphname, leftGlyph)
+    suffix_uc = ''
+    suffix_lc = ''
+    if len(args.suffix) > 0:
+      s = unicode(args.suffix)
+      if s[0].isupper():
+        suffix_uc = args.suffix
+        suffix_lc = args.suffix.lower()
+      else:
+        suffix_uc = args.suffix.upper()
+        suffix_lc = args.suffix
+    for rightname in rightnames:
+      if rightname in font:
+        rightGlyph = None
+        if len(rightname) == 1:
+          rightGlyph = font[rightname]
+        suffix = suffix_lc
+        if unicode(rightname[0]).isupper():
+          suffix = suffix_uc
+        out.append('%s%s%s' % (left, fmtGlyphname(rightname, rightGlyph), suffix))
+
+  print(' '.join(out))
+
+
+
+
+def main():
+  argparser = ArgumentParser(
+    description='Generate kerning samples by providing the left-hand side glyph')
+
+  argparser.add_argument(
+    '-u', dest='formatAsUnicode', action='store_const', const=True, default=False,
+    help='Format output as unicode escape sequences instead of glyphnames. ' +
+         'E.g. "\\u2126" instead of "\\Omega"')
+
+  argparser.add_argument(
+    '-suffix', dest='suffix', metavar='<text>', type=str,
+    help='Text to append after each pair')
+
+  argparser.add_argument(
+    '-all-in-groups', dest='includeAllInGroup',
+    action='store_const', const=True, default=False,
+    help='Include all glyphs for groups rather than just the first glyph listed.')
+
+  argparser.add_argument(
+    'fontPath', metavar='<ufofile>', type=str, help='UFO font source')
+
+  argparser.add_argument(
+    'glyphnames', metavar='<glyphname>', type=str, nargs='+',
+    help='Name of glyphs to generate samples for. '+
+         'You can also provide a Unicode code point using the syntax "U+XXXX"')
+
+  args = argparser.parse_args()
+  dryRun = args.dryRun
+
+  font = OpenFont(args.fontPath)
+
+  groupsFilename = os.path.join(args.fontPath, 'groups.plist')
+  kerningFilename = os.path.join(args.fontPath, 'kerning.plist')
+
+  groups = plistlib.readPlist(groupsFilename)   # { groupName => [glyphName] }
+  kerning = plistlib.readPlist(kerningFilename) # { leftName => {rightName => kernVal} }
+  groupmap = mapGroups(groups)
+
+  # expand any unicode codepoints
+  glyphnames = []
+  for glyphname in args.glyphnames:
+    if len(glyphname) > 2 and glyphname[:2] == 'U+':
+      cp = int(glyphname[2:], 16)
+      ucmap = font.getCharacterMapping()  # { 2126: ['Omega', ...], ...}
+      for glyphname2 in ucmap[cp]:
+        glyphnames.append(glyphname2)
+    else:
+      glyphnames.append(glyphname)
+
+  for glyphname in glyphnames:
+    samplesForGlyphname(font, groups, groupmap, kerning, glyphname, args)
+
+
+main()