Adds script for showing changes between branches (e.g. what glyphs has changed since last release)

1 files changed, 109 insertions, 0 deletions

diff --git a/misc/show-changes.py b/misc/show-changes.py
new file mode 100755
index 000000000..4e5a7ef37
--- /dev/null
+++ b/misc/show-changes.py
@@ -0,0 +1,109 @@
+#!/usr/bin/env python
+from __future__ import print_function
+import os, sys, subprocess
+from argparse import ArgumentParser
+from xml.dom.minidom import parse as xmlParseFile
+from collections import OrderedDict
+
+
+def main():
+  opts = ArgumentParser(description='Shows glyph-related changes.')
+
+  opts.add_argument(
+    'sinceCommit', metavar='<since-commit>', type=str,
+    help='Start commit.')
+
+  opts.add_argument(
+    'untilCommit', metavar='<until-commit>', type=str, nargs='?',
+    default='HEAD', help='End commit. Defaults to HEAD.')
+
+  opts.add_argument(
+    '-markdown', dest='markdown', action='store_const',
+    const=True, default=False,
+    help='Output text suitable for Markdown (rather than plain text.)')
+
+  a = opts.parse_args()
+
+  rootdir = os.path.abspath(os.path.join(
+    os.path.dirname(__file__),
+    os.pardir
+  ))
+
+  try:
+    out = subprocess.check_output(
+      [
+        'git',
+        '-C', rootdir,
+        'diff',
+        '--name-status',
+        a.sinceCommit + '..' + a.untilCommit,
+        '--', 'src'
+      ],
+      shell=False
+    ).strip()
+  except Exception as e:
+    print('Did you forget to `git fetch --tags` perhaps?', file=sys.stderr)
+    sys.exit(1)
+
+  ufoPrefix = 'src/Inter-UI-'
+  changes = OrderedDict()
+
+  for line in out.split('\n'):
+    changeType, name = line.split('\t')
+    if name.startswith(ufoPrefix) and name.endswith('.glif'):
+      weight = name[len(ufoPrefix):name.find('.ufo/')]
+      filename = os.path.join(rootdir, name)
+      doc = xmlParseFile(filename)
+      # print(changeType, weight, name)
+
+      g = doc.documentElement
+      gname = g.attributes['name'].value
+      unicodes = set([
+        'U+' + u.attributes['hex'].value
+          for u in g.getElementsByTagName('unicode')
+      ])
+      # print('gname', gname)
+      # print('unicodes:', unicodes)
+
+      c = changes.get(gname)
+      if c is None:
+        c = {
+          'unicodes': unicodes,
+          'weights': [(weight, changeType)]
+        }
+        changes[gname] = c
+      else:
+        c['unicodes'] = c['unicodes'].union(unicodes)
+        c['weights'].append((weight, changeType))
+
+      # break
+
+  longestName = 0
+  names = sorted(changes.keys())
+
+  if not a.markdown:
+    # find longest name
+    for name in names:
+      z = len(name)
+      if z > longestName:
+        longestName = z
+
+  for name in names:
+    c = changes[name]
+    weights = [ w[0] for w in c['weights'] ]
+    unicodes = c['unicodes']
+
+    if a.markdown:
+      unicodess = ''
+      if len(unicodes) != 0:
+        unicodess = ' %s' % ', '.join(['`%s`' % s for s in unicodes])
+      weightss = ' & '.join(weights)
+      print('- %s%s _%s_' % (name, unicodess, weightss))
+    else:
+      unicodess = ''
+      if len(unicodes) != 0:
+        unicodess = ' (%s)' % ', '.join(unicodes)
+      weightss = ' & '.join(weights)
+      print('%s%s %s' % (name.ljust(longestName), unicodess, weightss))
+
+main()
\ No newline at end of file