Merge pull request #83 from C1710/emoji13-1

Emoji 13.1

Blobmoji.gpl

@@ -0,0 +1,197 @@
+GIMP Palette
+Name: Blobmoji
+Columns: 10
+#
+255 235 238	red-50
+255 205 210	red-100
+239 154 154	red-200
+229 115 115	red-300
+227  95  92	red-400
+219  68  55	red-500
+211  48  44	red-600
+198  42  42	red-700
+178  36  36	red-800
+155  24  24	red-900
+252 228 236	pink-50
+249 195 213	pink-100
+244 164 191	pink-200
+241 147 179	pink-300
+235 122 164	pink-400
+215  89 139	pink-500
+209  71 122	pink-600
+202  47 109	pink-700
+179  35  98	pink-800
+149  24  91	pink-900
+243 229 245	purple-50
+225 190 231	purple-100
+202 152 211	purple-200
+181 111 193	purple-300
+165  79 180	purple-400
+149  49 166	purple-500
+136  46 160	purple-600
+119  41 152	purple-700
+104  37 144	purple-800
+ 75  29 131	purple-900
+237 231 246	indigo-50
+197 202 233	indigo-100
+160 168 217	indigo-200
+120 133 204	indigo-300
+ 92 106 192	indigo-400
+ 72  87 172	indigo-500
+ 66  79 162	indigo-600
+ 56  69 151	indigo-700
+ 48  59 139	indigo-800
+ 33  41 119	indigo-900
+232 240 246	blue-50
+173 206 230	blue-100
+145 188 219	blue-200
+ 82 156 210	blue-300
+ 46 130 189	blue-400
+ 25 108 162	blue-500
+ 28  97 146	blue-600
+ 24  89 135	blue-700
+ 19  81 124	blue-800
+ 12  66 104	blue-900
+225 245 254	light-blue-50
+186 227 245	light-blue-100
+148 212 241	light-blue-200
+129 205 238	light-blue-300
+ 98 192 234	light-blue-400
+ 66 173 231	light-blue-500
+ 30 157 219	light-blue-600
+ 29 140 200	light-blue-700
+ 25 124 183	light-blue-800
+ 12  94 159	light-blue-900
+225 250 254	sky-blue-50
+186 235 245	sky-blue-100
+148 224 241	sky-blue-200
+129 215 238	sky-blue-300
+ 98 205 234	sky-blue-400
+ 64 192 231	sky-blue-500
+ 30 175 219	sky-blue-600
+ 29 163 200	sky-blue-700
+ 25 149 183	sky-blue-800
+ 12 127 159	sky-blue-900
+220 246 245	teal-50
+177 238 232	teal-100
+ 93 221 209	teal-200
+ 29 206 189	teal-300
+ 27 187 172	teal-400
+ 19 171 150	teal-500
+ 11 138 125	teal-600
+  4 126 112	teal-700
+  4 107  94	teal-800
+  0  77  64	teal-900
+241 248 233	light-green-50
+216 235 194	light-green-100
+188 221 152	light-green-200
+164 208 113	light-green-300
+142 197  79	light-green-400
+124 179  66	light-green-500
+111 160  59	light-green-600
+ 91 140  49	light-green-700
+ 72 118  40	light-green-800
+ 40  83  24	light-green-900
+248 249 233	lime-50
+237 240 199	lime-100
+225 232 162	lime-200
+213 222 126	lime-300
+204 215  97	lime-400
+189 207  70	lime-500
+182 191  62	lime-600
+165 169  54	lime-700
+149 148  45	lime-800
+122 113  31	lime-900
+255 253 231	yellow-50
+255 250 209	yellow-100
+255 247 179	yellow-200
+255 244 143	yellow-300
+255 240 112	yellow-400
+255 238  88	yellow-500
+253 221  78	yellow-600
+251 199  70	yellow-700
+250 178  61	yellow-800
+246 142  50	yellow-900
+255 248 225	amber-50
+255 236 179	amber-100
+255 224 130	amber-200
+255 213  79	amber-300
+255 202  40	amber-400
+252 194  27	amber-500
+255 179   0	amber-600
+255 160   0	amber-700
+255 143   0	amber-800
+255 111   0	amber-900
+255 243 224	dark-amber-50
+255 224 178	dark-amber-100
+255 204 128	dark-amber-200
+255 183  77	dark-amber-300
+255 167  38	dark-amber-400
+247 147  41	dark-amber-500
+247 140   4	dark-amber-600
+241 124   4	dark-amber-700
+239 108   0	dark-amber-800
+230  81   0	dark-amber-900
+251 236 231	orange-50
+255 206 188	orange-100
+255 176 145	orange-200
+255 147 101	orange-300
+242 141  94	orange-400
+237 108  48	orange-500
+231 102  43	orange-600
+230  86  25	orange-700
+216  76  21	orange-800
+191  63  12	orange-900
+250 250 250	grey-50
+245 245 245	grey-100
+238 238 238	grey-200
+224 224 224	grey-300
+189 189 189	grey-400
+158 158 158	grey-500
+117 117 117	grey-600
+ 97  97  97	grey-700
+ 66  66  66	grey-800
+ 33  33  33	grey-900
+236 240 241	blue-grey-50
+203 219 224	blue-grey-100
+175 199 207	blue-grey-200
+147 177 187	blue-grey-300
+134 168 180	blue-grey-400
+101 147 162	blue-grey-500
+ 90 129 142	blue-grey-600
+ 75 107 117	blue-grey-700
+ 62  82  89	blue-grey-800
+ 43  59  64	blue-grey-900
+239 235 233	brown-50
+215 203 200	brown-100
+188 169 164	brown-200
+161 134 127	brown-300
+141 107  99	brown-400
+133  92  82	brown-500
+109  74  65	brown-600
+ 93  64  55	brown-700
+ 78  52  46	brown-800
+ 62  39  35	brown-900
+
+
+# Skin tones
+250 220 188 light-skin
+224 187 149 medium-light-skin
+191 143 104 medium-skin
+155 100  60 medium-dark-skin
+ 99  67  52 dark-skin
+
+# Skin tones (hair)
+ 49  45  45 light-hair
+191 160  85 medium-light-hair
+109  76  65 medium-hair
+ 71  53  45 medium-dark-hair
+ 35  32  32 dark-hair
+249 224 115 blond-hair
+
+# Skin tones (shadow/nose)
+219 166 137 light-shadow
+196 142 106 medium-light-shadow
+153 103  79 medium-shadow
+122  76  50 medium-dark-shadow
+ 86  62  55 dark-shadow

CHANGES.md

@@ -1,3 +1,5 @@
+_This file will not be updated anymore. Please look at `CHANGED.md` instead_
+
 Changes from Google Noto Color Emoji v2017-05-18 (approx.)
 =======
 

Dockerfile

@@ -1,22 +1,13 @@
-FROM python:slim
+FROM ghcr.io/c1710/emoji_builder
 
-RUN apt-get update \
-    && apt-get install --no-install-recommends -y \
-    make \
-    gcc \
-    zopfli \
-    libc-dev \
-    libpng-dev \
-    libcairo2-dev \
-    imagemagick \
-    && apt-get clean \
-    && rm -rf /var/lib/apt/lists/*
+COPY svg ./svg
+COPY third_party/region-flags/svg ./flags
+COPY emoji_aliases.txt NotoColorEmoji.tmpl.ttx.tmpl Blobmoji.gpl ./
+COPY AUTHORS CONTRIBUTORS CHANGES.md LICENSE ./
 
-RUN pip install --no-cache notofonttools
+VOLUME /build
+VOLUME /output
 
-ADD . /blobmoji
-WORKDIR /blobmoji
-
-RUN mkdir /output
-
-CMD make -j $(nproc) && cp NotoColorEmoji.ttf /output/
+CMD /github_workflow_setup.sh && \
+	/bin/emoji_builder -vv -b /build -o Blobmoji.ttf -O /output --flags ./flags blobmoji -w -a ./emoji_aliases.txt --ttx-tmpl ./NotoColorEmoji.tmpl.ttx.tmpl --palette ./Blobmoji.gpl && \
+	mv /output/Blobmoji_win.ttf /output/BlobmojiWindows.ttf

README.md

@@ -1,4 +1,3 @@
-
 #### DISCLAIMER:
 I am **neither** affiliated nor in _any_ relationship to the original creators or to Emojipedia or anything or anyone else.
 
@@ -24,24 +23,39 @@ But now to the original content of this Readme:
 
 Color and Black-and-White Noto emoji fonts, and tools for working with them.
 
-## Building NotoColorEmoji
+## ~~Building NotoColorEmoji~~
 
-Building NotoColorEmoji currently requires a Python 2.x wide build.  To build
+~~Building NotoColorEmoji currently requires a Python 2.x wide build.  To build
 the emoji font you will require a few files from nototools.  Clone a copy from
 https://github.com/googlei18n/nototools and either put it in your PYTHONPATH or
 use 'python setup.py develop' ('install' currently won't fully install all the
 data used by nototools).  You will also need fontTools, get it from
-https://github.com/behdad/fonttools.git.
+https://github.com/behdad/fonttools.git.~~
 
-Then run `make`.  NotoColorEmoji is the default target.  It's suggested to use `-j`,
+~~Then run `make`.  NotoColorEmoji is the default target.  It's suggested to use `-j`,
 especially if you are using zopflipng for compression.  Intermediate products
 (compressed image files, for example) will be put into a build subdirectory; the
-font will be at the top level.
+font will be at the top level.~~
 
-## Docker build
+## Building Blobmoji 13+
 
-Alternatively, you can also build the font within Docker through the provided Dockerfile.
-Just run `docker build . -t blobmoji && docker run --rm -it -v "$PWD/output:/output" blobmoji`. The resulting font will reside in the 'output' folder in your current working directory.
+_Building is now done using [emoji_builder](https://github.com/C1710/emoji_builder/) (name WIP). Once you have it running, you can build it using the following command (you'll need to replace `emoji_builder` by the executable you use, e.g. `emoji_builder.exe` and maybe including the path. I recommend copying it into the `blobmoji`-directory):_
+```
+emoji_builder --flags ./third_party/region-flags/svg blobmoji -w -a ./emoji_aliases.txt --ttx-tmpl ./NotoColorEmoji.tmpl.ttx.tmpl --palette ./Blobmoji.gpl --default_font "Comic Neue"
+```
+- `--flags`: Use the directory conaining the flags
+- `-w` add a wave-effect to the flags
+- `-a` use an alias file
+- `--ttx-tmpl` Use the template for the font metadata
+- `--palette` normalize the colors to a specific color palette in the GIMP format (which is a derivation of the color palette present in the [2014 _Material Design_](https://material.io/archive/guidelines/style/color.html#color-color-palette))
+- `--default_font` Because the graphics program I currently use (Affinity Designer) outputs font specifications in a format that `resvg`/`emoji_builder` has issues with, the font is explicitly specified here (note that if the font is correctly recognized, this one is not used. So as of now it is _not_ used to use a font for _all_ emojis)
+
+## ~~Docker build~~
+
+~~Alternatively, you can also build the font within Docker through the provided Dockerfile.
+Just run `docker build . -t blobmoji && docker run --rm -it -v "$PWD/output:/output" blobmoji`. The resulting font will reside in the 'output' folder in your current working directory.~~
+
+The Docker build method isn't available for `emoji_builder` yet
 
 ## Using NotoColorEmoji
 
@@ -95,3 +109,6 @@ _Please try to use the discussion feature for artistic topics, like the style of
 * 2017-09-13: Emoji redesign released.
 * 2015-12-09: Unicode 7 and 8 emoji image data (.png format) added.
 * 2015-09-29: All Noto fonts now licensed under the SIL Open Font License.
+
+
+_Microsoft, Windows are trademarks of the Microsoft group of companies._

add_aliases.py

@@ -1,224 +0,0 @@
-#!/usr/bin/env python3
-#
-# Copyright 2017 Google Inc. All rights reserved.
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-
-from __future__ import print_function
-import argparse
-import glob
-import os
-from os import path
-import shutil
-import sys
-
-from nototools import unicode_data
-
-"""Create aliases in target directory.
-
-In addition to links/copies named with aliased sequences, this can also
-create canonically named aliases/copies, if requested."""
-
-
-DATA_ROOT = path.dirname(path.abspath(__file__))
-
-def str_to_seq(seq_str):
-  res = [int(s, 16) for s in seq_str.split('_')]
-  if 0xfe0f in res:
-    print('0xfe0f in file name: %s' % seq_str)
-    res = [x for x in res if x != 0xfe0f]
-  return tuple(res)
-
-
-def seq_to_str(seq):
-  return '_'.join('%04x' % cp for cp in seq)
-
-
-def read_default_unknown_flag_aliases():
-  unknown_flag_path = path.join(DATA_ROOT, 'unknown_flag_aliases.txt')
-  return read_emoji_aliases(unknown_flag_path)
-
-
-def read_default_emoji_aliases():
-  alias_path = path.join(DATA_ROOT, 'emoji_aliases.txt')
-  return read_emoji_aliases(alias_path)
-
-
-def read_emoji_aliases(filename):
-  result = {}
-
-  with open(filename, 'r') as f:
-    for line in f:
-      ix = line.find('#')
-      if (ix > -1):
-        line = line[:ix]
-      line = line.strip()
-      if not line:
-        continue
-      als, trg = (s.strip() for s in line.split(';'))
-      try:
-        als_seq = tuple([int(x, 16) for x in als.split('_')])
-        trg_seq = tuple([int(x, 16) for x in trg.split('_')])
-      except:
-        print('cannot process alias %s -> %s' % (als, trg))
-        continue
-      result[als_seq] = trg_seq
-  return result
-
-
-def add_aliases(
-    srcdir, dstdir, aliasfile, prefix, ext, replace=False, copy=False,
-    canonical_names=False, dry_run=False):
-  """Use aliasfile to create aliases of files in srcdir matching prefix/ext in
-  dstdir.  If dstdir is null, use srcdir as dstdir.  If replace is false
-  and a file already exists in dstdir, report and do nothing.  If copy is false
-  create a symlink, else create a copy.
-
-  If canonical_names is true, check all source files and generate aliases/copies
-  using the canonical name if different from the existing name.
-
-  If dry_run is true, report what would be done.  Dstdir will be created if
-  necessary, even if dry_run is true."""
-
-  if not path.isdir(srcdir):
-    print('%s is not a directory' % srcdir, file=sys.stderr)
-    return
-
-  if not dstdir:
-    dstdir = srcdir
-  elif not path.isdir(dstdir):
-    os.makedirs(dstdir)
-
-  prefix_len = len(prefix)
-  suffix_len = len(ext) + 1
-  filenames = [path.basename(f)
-               for f in glob.glob(path.join(srcdir, '%s*.%s' % (prefix, ext)))]
-  seq_to_file = {
-      str_to_seq(name[prefix_len:-suffix_len]) : name
-      for name in filenames}
-
-  aliases = read_emoji_aliases(aliasfile)
-  aliases_to_create = {}
-  aliases_to_replace = []
-  alias_exists = False
-
-  def check_alias_seq(seq):
-    alias_str = seq_to_str(seq)
-    alias_name = '%s%s.%s' % (prefix, alias_str, ext)
-    alias_path = path.join(dstdir, alias_name)
-    if path.exists(alias_path):
-      if replace:
-        aliases_to_replace.append(alias_name)
-      else:
-        print('alias %s exists' % alias_str, file=sys.stderr)
-        alias_exists = True
-        return None
-    return alias_name
-
-  canonical_to_file = {}
-  for als, trg in sorted(aliases.items()):
-    if trg not in seq_to_file:
-      print('target %s for %s does not exist' % (
-          seq_to_str(trg), seq_to_str(als)), file=sys.stderr)
-      continue
-    alias_name = check_alias_seq(als)
-    if alias_name:
-      target_file = seq_to_file[trg]
-      aliases_to_create[alias_name] = target_file
-      if canonical_names:
-        canonical_seq = unicode_data.get_canonical_emoji_sequence(als)
-        if canonical_seq and canonical_seq != als:
-          canonical_alias_name = check_alias_seq(canonical_seq)
-          if canonical_alias_name:
-            canonical_to_file[canonical_alias_name] = target_file
-
-  if canonical_names:
-    print('adding %d canonical aliases' % len(canonical_to_file))
-    for seq, f in seq_to_file.iteritems():
-      canonical_seq = unicode_data.get_canonical_emoji_sequence(seq)
-      if canonical_seq and canonical_seq != seq:
-        alias_name = check_alias_seq(canonical_seq)
-        if alias_name:
-          canonical_to_file[alias_name] = f
-
-    print('adding %d total canonical sequences' % len(canonical_to_file))
-    aliases_to_create.update(canonical_to_file)
-
-  if replace:
-    if not dry_run:
-      for k in sorted(aliases_to_replace):
-        os.remove(path.join(dstdir, k))
-    print('replacing %d files' % len(aliases_to_replace))
-  elif alias_exists:
-    print('aborting, aliases exist.', file=sys.stderr)
-    return
-
-  for k, v in sorted(aliases_to_create.items()):
-    if dry_run:
-      msg = 'replace ' if k in aliases_to_replace else ''
-      print('%s%s -> %s' % (msg, k, v))
-    else:
-      try:
-        if copy:
-          shutil.copy2(path.join(srcdir, v), path.join(dstdir, k))
-        else:
-          # fix this to create relative symlinks
-          if srcdir == dstdir:
-            os.symlink(v, path.join(dstdir, k))
-          else:
-            raise Exception('can\'t create cross-directory symlinks yet')
-      except Exception as e:
-        print('failed to create %s -> %s' % (k, v), file=sys.stderr)
-        raise Exception('oops, ' + str(e))
-  print('created %d %s' % (
-      len(aliases_to_create), 'copies' if copy else 'symlinks'))
-
-
-def main():
-  parser = argparse.ArgumentParser()
-  parser.add_argument(
-      '-s', '--srcdir', help='directory containing files to alias',
-      required=True, metavar='dir')
-  parser.add_argument(
-      '-d', '--dstdir', help='directory to write aliases, default srcdir',
-      metavar='dir')
-  parser.add_argument(
-      '-a', '--aliasfile', help='alias file (default emoji_aliases.txt)',
-      metavar='file', default='emoji_aliases.txt')
-  parser.add_argument(
-      '-p', '--prefix', help='file name prefix (default emoji_u)',
-      metavar='pfx', default='emoji_u')
-  parser.add_argument(
-      '-e', '--ext', help='file name extension (default png)',
-      choices=['ai', 'png', 'svg'], default='png')
-  parser.add_argument(
-      '-r', '--replace', help='replace existing files/aliases',
-      action='store_true')
-  parser.add_argument(
-      '-c', '--copy', help='create a copy of the file, not a symlink',
-      action='store_true')
-  parser.add_argument(
-      '--canonical_names', help='include extra copies with canonical names '
-      '(including fe0f emoji presentation character)', action='store_true');
-  parser.add_argument(
-      '-n', '--dry_run', help='print out aliases to create only',
-      action='store_true')
-  args = parser.parse_args()
-
-  add_aliases(
-      args.srcdir, args.dstdir, args.aliasfile, args.prefix, args.ext,
-      args.replace, args.copy, args.canonical_names, args.dry_run)
-
-
-if __name__ == '__main__':
-  main()

add_emoji_gsub.py

@@ -1,195 +0,0 @@
-#!/usr/bin/env python3
-#
-# Copyright 2014 Google Inc. All rights reserved.
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-
-"""Modify the Noto Color Emoji font to use GSUB rules for flags and keycaps."""
-
-__author__ = "roozbeh@google.com (Roozbeh Pournader)"
-
-
-import sys
-
-from fontTools import agl
-from fontTools import ttLib
-from fontTools.ttLib.tables import otTables
-
-from nototools import font_data
-
-
-def create_script_list(script_tag='DFLT'):
-    """Create a ScriptList for the GSUB table."""
-    def_lang_sys = otTables.DefaultLangSys()
-    def_lang_sys.ReqFeatureIndex = 0xFFFF
-    def_lang_sys.FeatureCount = 1
-    def_lang_sys.FeatureIndex = [0]
-    def_lang_sys.LookupOrder = None
-
-    script_record = otTables.ScriptRecord()
-    script_record.ScriptTag = script_tag
-    script_record.Script = otTables.Script()
-    script_record.Script.DefaultLangSys = def_lang_sys
-    script_record.Script.LangSysCount = 0
-    script_record.Script.LangSysRecord = []
-
-    script_list = otTables.ScriptList()
-    script_list.ScriptCount = 1
-    script_list.ScriptRecord = [script_record]
-
-    return script_list
-
-
-def create_feature_list(feature_tag, lookup_count):
-    """Create a FeatureList for the GSUB table."""
-    feature_record = otTables.FeatureRecord()
-    feature_record.FeatureTag = feature_tag
-    feature_record.Feature = otTables.Feature()
-    feature_record.Feature.LookupCount = lookup_count
-    feature_record.Feature.LookupListIndex = range(lookup_count)
-    feature_record.Feature.FeatureParams = None
-
-    feature_list = otTables.FeatureList()
-    feature_list.FeatureCount = 1
-    feature_list.FeatureRecord = [feature_record]
-
-    return feature_list
-
-
-def create_lookup_list(lookups):
-    """Create a LookupList for the GSUB table."""
-    lookup_list = otTables.LookupList()
-    lookup_list.LookupCount = len(lookups)
-    lookup_list.Lookup = lookups
-
-    return lookup_list
-
-
-def get_glyph_name_or_create(char, font):
-    """Return the glyph name for a character, creating if it doesn't exist."""
-    cmap = font_data.get_cmap(font)
-    if char in cmap:
-        return cmap[char]
-
-    glyph_name = agl.UV2AGL[char]
-    assert glyph_name not in font.glyphOrder
-
-    font['hmtx'].metrics[glyph_name] = [0, 0]
-    cmap[char] = glyph_name
-
-    if 'glyf' in font:
-        from fontTools.ttLib.tables import _g_l_y_f
-        empty_glyph = _g_l_y_f.Glyph()
-        font['glyf'].glyphs[glyph_name] = empty_glyph
-
-    font.glyphOrder.append(glyph_name)
-    return glyph_name
-
-
-def create_lookup(table, font, flag=0):
-    """Create a Lookup based on mapping table."""
-    cmap = font_data.get_cmap(font)
-
-    ligatures = {}
-    for output, (ch1, ch2) in table.iteritems():
-        output = cmap[output]
-        ch1 = get_glyph_name_or_create(ch1, font)
-        ch2 = get_glyph_name_or_create(ch2, font)
-
-        ligature = otTables.Ligature()
-        ligature.CompCount = 2
-        ligature.Component = [ch2]
-        ligature.LigGlyph = output
-
-        try:
-            ligatures[ch1].append(ligature)
-        except KeyError:
-            ligatures[ch1] = [ligature]
-
-    ligature_subst = otTables.LigatureSubst()
-    ligature_subst.ligatures = ligatures
-
-    lookup = otTables.Lookup()
-    lookup.LookupType = 4
-    lookup.LookupFlag = flag
-    lookup.SubTableCount = 1
-    lookup.SubTable = [ligature_subst]
-
-    return lookup
-
-
-def create_simple_gsub(lookups, script='DFLT', feature='ccmp'):
-    """Create a simple GSUB table."""
-    gsub_class = ttLib.getTableClass('GSUB')
-    gsub = gsub_class('GSUB')
-
-    gsub.table = otTables.GSUB()
-    gsub.table.Version = 1.0
-    gsub.table.ScriptList = create_script_list(script)
-    gsub.table.FeatureList = create_feature_list(feature, len(lookups))
-    gsub.table.LookupList = create_lookup_list(lookups)
-    return gsub
-
-
-def reg_indicator(letter):
-    """Return a regional indicator charater from corresponing capital letter.
-    """
-    return 0x1F1E6 + ord(letter) - ord('A')
-
-
-EMOJI_FLAGS = {
-    0xFE4E5: (reg_indicator('J'), reg_indicator('P')),  # Japan
-    0xFE4E6: (reg_indicator('U'), reg_indicator('S')),  # United States
-    0xFE4E7: (reg_indicator('F'), reg_indicator('R')),  # France
-    0xFE4E8: (reg_indicator('D'), reg_indicator('E')),  # Germany
-    0xFE4E9: (reg_indicator('I'), reg_indicator('T')),  # Italy
-    0xFE4EA: (reg_indicator('G'), reg_indicator('B')),  # United Kingdom
-    0xFE4EB: (reg_indicator('E'), reg_indicator('S')),  # Spain
-    0xFE4EC: (reg_indicator('R'), reg_indicator('U')),  # Russia
-    0xFE4ED: (reg_indicator('C'), reg_indicator('N')),  # China
-    0xFE4EE: (reg_indicator('K'), reg_indicator('R')),  # Korea
-}
-
-KEYCAP = 0x20E3
-
-EMOJI_KEYCAPS = {
-    0xFE82C: (ord('#'), KEYCAP),
-    0xFE82E: (ord('1'), KEYCAP),
-    0xFE82F: (ord('2'), KEYCAP),
-    0xFE830: (ord('3'), KEYCAP),
-    0xFE831: (ord('4'), KEYCAP),
-    0xFE832: (ord('5'), KEYCAP),
-    0xFE833: (ord('6'), KEYCAP),
-    0xFE834: (ord('7'), KEYCAP),
-    0xFE835: (ord('8'), KEYCAP),
-    0xFE836: (ord('9'), KEYCAP),
-    0xFE837: (ord('0'), KEYCAP),
-}
-
-def main(argv):
-    """Modify all the fonts given in the command line."""
-    for font_name in argv[1:]:
-        font = ttLib.TTFont(font_name)
-
-        assert 'GSUB' not in font
-        font['GSUB'] = create_simple_gsub([
-            create_lookup(EMOJI_KEYCAPS, font),
-            create_lookup(EMOJI_FLAGS, font)])
-
-        font_data.delete_from_cmap(
-            font, EMOJI_FLAGS.keys() + EMOJI_KEYCAPS.keys())
-
-        font.save(font_name+'-fixed')
-
-if __name__ == '__main__':
-    main(sys.argv)

add_glyphs.py

@@ -1,405 +0,0 @@
-#!/usr/bin/env python3
-
-"""Extend a ttx file with additional data.
-
-Takes a ttx file and one or more directories containing image files named
-after sequences of codepoints, extends the cmap, hmtx, GSUB, and GlyphOrder
-tables in the source ttx file based on these sequences, and writes out a new
-ttx file.
-
-This can also apply aliases from an alias file."""
-
-import argparse
-import collections
-import os
-from os import path
-import re
-import sys
-
-from fontTools import ttx
-from fontTools.ttLib.tables import otTables
-
-import add_emoji_gsub
-import add_aliases
-
-sys.path.append(
-    path.join(os.path.dirname(__file__), 'third_party', 'color_emoji'))
-from png import PNG
-
-
-def get_seq_to_file(image_dir, prefix, suffix):
-  """Return a mapping from codepoint sequences to files in the given directory,
-  for files that match the prefix and suffix.  File names with this prefix and
-  suffix should consist of codepoints in hex separated by underscore.  'fe0f'
-  (the codepoint of the emoji presentation variation selector) is stripped from
-  the sequence.
-  """
-  start = len(prefix)
-  limit = -len(suffix)
-  seq_to_file = {}
-  for name in os.listdir(image_dir):
-    if not (name.startswith(prefix) and name.endswith(suffix)):
-      continue
-    try:
-      cps = [int(s, 16) for s in name[start:limit].split('_')]
-      seq = tuple(cp for cp in cps if cp != 0xfe0f)
-    except:
-      raise Exception('could not parse "%s"' % name)
-    for cp in cps:
-      if not (0 <= cp <= 0x10ffff):
-        raise Exception('bad codepoint(s) in "%s"' % name)
-    if seq in seq_to_file:
-      raise Exception('duplicate sequence for "%s" in %s' % (name, image_dir))
-    seq_to_file[seq] = path.join(image_dir, name)
-  return seq_to_file
-
-
-def collect_seq_to_file(image_dirs, prefix, suffix):
-  """Return a sequence to file mapping by calling get_seq_to_file on a list
-  of directories.  When sequences for files in later directories match those
-  from earlier directories, the later file replaces the earlier one.
-  """
-  seq_to_file = {}
-  for image_dir in image_dirs:
-    seq_to_file.update(get_seq_to_file(image_dir, prefix, suffix))
-  return seq_to_file
-
-
-def remap_values(seq_to_file, map_fn):
-  return {k: map_fn(v) for k, v in seq_to_file.items()}
-
-
-def get_png_file_to_advance_mapper(lineheight):
-  def map_fn(filename):
-    wid, ht = PNG(filename).get_size()
-    return int(round(float(lineheight) * wid / ht))
-  return map_fn
-
-
-def cp_name(cp):
-  """return uniXXXX or uXXXXX(X) as a name for the glyph mapped to this cp."""
-  return '%s%04X' % ('u' if cp > 0xffff else 'uni', cp)
-
-
-def seq_name(seq):
-  """Sequences of length one get the cp_name.  Others start with 'u' followed by
-  two or more 4-to-6-digit hex strings separated by underscore."""
-  if len(seq) == 1:
-    return cp_name(seq[0])
-  return 'u' + '_'.join('%04X' % cp for cp in seq)
-
-
-def collect_cps(seqs):
-  cps = set()
-  for seq in seqs:
-    cps.update(seq)
-  return cps
-
-
-def get_glyphorder_cps_and_truncate(glyphOrder):
-  """This scans glyphOrder for names that correspond to a single codepoint
-  using the 'u(ni)XXXXXX' syntax.  All names that don't match are moved
-  to the front the glyphOrder list in their original order, and the
-  list is truncated.  The ones that do match are returned as a set of
-  codepoints."""
-  glyph_name_re = re.compile(r'^u(?:ni)?([0-9a-fA-F]{4,6})$')
-  cps = set()
-  write_ix = 0
-  for ix, name in enumerate(glyphOrder):
-    m = glyph_name_re.match(name)
-    if m:
-      cps.add(int(m.group(1), 16))
-    else:
-      glyphOrder[write_ix] = name
-      write_ix += 1
-  del glyphOrder[write_ix:]
-  return cps
-
-
-def get_all_seqs(font, seq_to_advance):
-  """Copies the sequences from seq_to_advance and extends it with single-
-  codepoint sequences from the GlyphOrder table as well as those internal
-  to sequences in seq_to_advance.  Reduces the GlyphOrder table. """
-
-  all_seqs = set(seq_to_advance.keys())
-  # using collect_cps includes cps internal to a seq
-  cps = collect_cps(all_seqs)
-  glyphOrder = font.getGlyphOrder()
-  # extract cps in glyphOrder and reduce glyphOrder to only those that remain
-  glyphOrder_cps = get_glyphorder_cps_and_truncate(glyphOrder)
-  cps.update(glyphOrder_cps)
-  # add new single codepoint sequences from glyphOrder and sequences
-  all_seqs.update((cp,) for cp in cps)
-  return all_seqs
-
-
-def get_font_cmap(font):
-  """Return the first cmap in the font, we assume it exists and is a unicode
-  cmap."""
-  return font['cmap'].tables[0].cmap
-
-
-def add_glyph_data(font, seqs, seq_to_advance, vadvance):
-  """Add hmtx and GlyphOrder data for all sequences in seqs, and ensures there's
-  a cmap entry for each single-codepoint sequence.  Seqs not in seq_to_advance
-  will get a zero advance."""
-
-  # We allow the template cmap to omit mappings for single-codepoint glyphs
-  # defined in the template's GlyphOrder table.  Similarly, the hmtx table can
-  # omit advances.  We assume glyphs named 'uniXXXX' or 'uXXXXX(X)' in the
-  # GlyphOrder table correspond to codepoints based on the name; we don't
-  # attempt to handle other types of names and these must occur in the cmap and
-  # hmtx tables in the template.
-  #
-  # seq_to_advance maps sequences (including single codepoints) to advances.
-  # All codepoints in these sequences will be added to the cmap.  Some cps
-  # in these sequences have no corresponding single-codepoint sequence, they
-  # will also get added.
-  #
-  # The added codepoints have no advance information, so will get a zero
-  # advance.
-
-  cmap = get_font_cmap(font)
-  hmtx = font['hmtx'].metrics
-  vmtx = font['vmtx'].metrics
-
-  # We don't expect sequences to be in the glyphOrder, since we removed all the
-  # single-cp sequences from it and don't expect it to already contain names
-  # corresponding to multiple-cp sequencess.  But just in case, we use
-  # reverseGlyphMap to avoid duplicating names accidentally.
-
-  updatedGlyphOrder = False
-  reverseGlyphMap = font.getReverseGlyphMap()
-
-  # Order the glyphs by grouping all the single-codepoint sequences first,
-  # then order by sequence so that related sequences are together.  We group
-  # by single-codepoint sequence first in order to keep these glyphs together--
-  # they're used in the coverage tables for some of the substitutions, and
-  # those tables can be more compact this way.
-  for seq in sorted(seqs, key=lambda s: (0 if len(s) == 1 else 1, s)):
-    name = seq_name(seq)
-    if len(seq) == 1:
-      cmap[seq[0]] = name
-    advance = seq_to_advance.get(seq, 0)
-    hmtx[name] = [advance, 0]
-    vmtx[name] = [vadvance, 0]
-    if name not in reverseGlyphMap:
-      font.glyphOrder.append(name)
-      updatedGlyphOrder=True
-
-  if updatedGlyphOrder:
-    delattr(font, '_reverseGlyphOrderDict')
-
-
-def add_aliases_to_cmap(font, aliases):
-  """Some aliases might map a single codepoint to some other sequence.  These
-  should map directly to the glyph for that sequence in the cmap.  (Others will
-  map via GSUB).
-  """
-  if not aliases:
-    return
-
-  cp_aliases = [seq for seq in aliases if len(seq) == 1]
-  if not cp_aliases:
-    return
-
-  cmap = get_font_cmap(font)
-  for src_seq in cp_aliases:
-    cp = src_seq[0]
-    name = seq_name(aliases[src_seq])
-    cmap[cp] = name
-
-
-def get_rtl_seq(seq):
-  """Return the rtl variant of the sequence, if it has one, else the empty
-  sequence.
-  """
-  # Sequences with ZWJ or TAG_END in them will reflect.  Fitzpatrick modifiers
-  # however do not, so if we reflect we make a pass to swap them back into their
-  # logical order.
-
-  ZWJ = 0x200d
-  TAG_END = 0xe007f
-  def is_fitzpatrick(cp):
-    return 0x1f3fb <= cp <= 0x1f3ff
-
-  if not (ZWJ in seq or TAG_END in seq):
-    return ()
-
-  rev_seq = list(seq)
-  rev_seq.reverse()
-  for i in range(len(rev_seq)-1, 0, -1):
-    if is_fitzpatrick(rev_seq[i-1]):
-      rev_seq[i-1], rev_seq[i] = rev_seq[i], rev_seq[i-1]
-  return tuple(rev_seq)
-
-
-def get_gsub_ligature_lookup(font):
-  """If the font does not have a GSUB table, create one with a ligature
-  substitution lookup.  If it does, ensure the first lookup is a properly
-  initialized ligature substitution lookup.  Return the lookup."""
-
-  # The template might include more lookups after lookup 0, if it has a
-  # GSUB table.
-  if 'GSUB' not in font:
-    ligature_subst = otTables.LigatureSubst()
-    ligature_subst.ligatures = {}
-
-    lookup = otTables.Lookup()
-    lookup.LookupType = 4
-    lookup.LookupFlag = 0
-    lookup.SubTableCount = 1
-    lookup.SubTable = [ligature_subst]
-
-    font['GSUB'] = add_emoji_gsub.create_simple_gsub([lookup])
-  else:
-    lookup = font['GSUB'].table.LookupList.Lookup[0]
-    assert lookup.LookupFlag == 0
-
-    # importXML doesn't fully init GSUB structures, so help it out
-    st = lookup.SubTable[0]
-    if not hasattr(lookup, 'LookupType'):
-      assert st.LookupType == 4
-      setattr(lookup, 'LookupType', 4)
-
-    if not hasattr(st, 'ligatures'):
-      setattr(st, 'ligatures', {})
-
-  return lookup
-
-
-def add_ligature_sequences(font, seqs, aliases):
-  """Add ligature sequences."""
-
-  seq_to_target_name = {
-      seq: seq_name(seq) for seq in seqs if len(seq) > 1}
-  if aliases:
-    seq_to_target_name.update({
-        seq: seq_name(aliases[seq]) for seq in aliases if len(seq) > 1})
-  if not seq_to_target_name:
-    return
-
-  rtl_seq_to_target_name = {
-      get_rtl_seq(seq): name for seq, name in seq_to_target_name.items()}
-  seq_to_target_name.update(rtl_seq_to_target_name)
-  # sequences that don't have rtl variants get mapped to the empty sequence,
-  # delete it.
-  if () in seq_to_target_name:
-    del seq_to_target_name[()]
-
-  # organize by first codepoint in sequence
-  keyed_ligatures = collections.defaultdict(list)
-  for t in seq_to_target_name.items():
-    first_cp = t[0][0]
-    keyed_ligatures[first_cp].append(t)
-
-  def add_ligature(lookup, cmap, seq, name):
-    # The sequences consist of codepoints, but the entries in the ligature table
-    # are glyph names.  Aliasing can give single codepoints names based on
-    # sequences (e.g. 'guardsman' with 'male guardsman') so we map the
-    # codepoints through the cmap to get the glyph names.
-    glyph_names = [cmap[cp] for cp in seq]
-
-    lig = otTables.Ligature()
-    lig.CompCount = len(seq)
-    lig.Component = glyph_names[1:]
-    lig.LigGlyph = name
-
-    ligatures = lookup.SubTable[0].ligatures
-    first_name = glyph_names[0]
-    try:
-      ligatures[first_name].append(lig)
-    except KeyError:
-      ligatures[first_name] = [lig]
-
-  lookup = get_gsub_ligature_lookup(font)
-  cmap = get_font_cmap(font)
-  for first_cp in sorted(keyed_ligatures):
-    pairs = keyed_ligatures[first_cp]
-
-    # Sort longest first, this ensures longer sequences with common prefixes
-    # are handled before shorter ones.  The secondary sort is a standard
-    # sort on the codepoints in the sequence.
-    pairs.sort(key = lambda pair: (-len(pair[0]), pair[0]))
-    for seq, name in pairs:
-      add_ligature(lookup, cmap, seq, name)
-
-
-def update_font_data(font, seq_to_advance, vadvance, aliases):
-  """Update the font's cmap, hmtx, GSUB, and GlyphOrder tables."""
-  seqs = get_all_seqs(font, seq_to_advance)
-  add_glyph_data(font, seqs, seq_to_advance, vadvance)
-  add_aliases_to_cmap(font, aliases)
-  add_ligature_sequences(font, seqs, aliases)
-
-
-def apply_aliases(seq_dict, aliases):
-  """Aliases is a mapping from sequence to replacement sequence.  We can use
-  an alias if the target is a key in the dictionary.  Furthermore, if the
-  source is a key in the dictionary, we can delete it.  This updates the
-  dictionary and returns the usable aliases."""
-  usable_aliases = {}
-  for k, v in aliases.items():
-    if v in seq_dict:
-      usable_aliases[k] = v
-      if k in seq_dict:
-        del seq_dict[k]
-  return usable_aliases
-
-
-def update_ttx(in_file, out_file, image_dirs, prefix, ext, aliases_file):
-  if ext != '.png':
-    raise Exception('extension "%s" not supported' % ext)
-
-  seq_to_file = collect_seq_to_file(image_dirs, prefix, ext)
-  if not seq_to_file:
-    raise ValueError(
-        'no sequences with prefix "%s" and extension "%s" in %s' % (
-            prefix, ext, ', '.join(image_dirs)))
-
-  aliases = None
-  if aliases_file:
-    aliases = add_aliases.read_emoji_aliases(aliases_file)
-    aliases = apply_aliases(seq_to_file, aliases)
-
-  font = ttx.TTFont()
-  font.importXML(in_file)
-
-  lineheight = font['hhea'].ascent - font['hhea'].descent
-  map_fn = get_png_file_to_advance_mapper(lineheight)
-  seq_to_advance = remap_values(seq_to_file, map_fn)
-
-  vadvance = font['vhea'].advanceHeightMax if 'vhea' in font else lineheight
-
-  update_font_data(font, seq_to_advance, vadvance, aliases)
-
-  font.saveXML(out_file)
-
-
-def main():
-  parser = argparse.ArgumentParser()
-  parser.add_argument(
-      '-f', '--in_file', help='ttx input file', metavar='file', required=True)
-  parser.add_argument(
-      '-o', '--out_file', help='ttx output file', metavar='file', required=True)
-  parser.add_argument(
-      '-d', '--image_dirs', help='directories containing image files',
-      nargs='+', metavar='dir', required=True)
-  parser.add_argument(
-      '-p', '--prefix', help='file prefix (default "emoji_u")',
-      metavar='pfx', default='emoji_u')
-  parser.add_argument(
-      '-e', '--ext', help='file extension (default ".png", currently only '
-      '".png" is supported',  metavar='ext', default='.png')
-  parser.add_argument(
-      '-a', '--aliases', help='process alias table', const='emoji_aliases.txt',
-      nargs='?', metavar='file')
-  args = parser.parse_args()
-
-  update_ttx(
-      args.in_file, args.out_file, args.image_dirs, args.prefix, args.ext,
-      args.aliases)
-
-
-if __name__ == '__main__':
-  main()

add_svg_glyphs.py

@@ -1,295 +0,0 @@
-#!/usr/bin/env python3
-# Copyright 2015 Google, Inc. All Rights Reserved.
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-#
-# Google Author(s): Doug Felt
-
-"""Tool to update GSUB, hmtx, cmap, glyf tables with svg image glyphs."""
-from __future__ import print_function
-
-import argparse
-import glob
-import logging
-import os
-import re
-import sys
-
-from fontTools.ttLib.tables import otTables
-from fontTools.ttLib.tables import _g_l_y_f
-from fontTools.ttLib.tables import S_V_G_ as SVG
-from fontTools import ttx
-
-from nototools import tool_utils
-
-import add_emoji_gsub
-import svg_builder
-
-
-class FontBuilder(object):
-  """A utility for mutating a ttx font.  This maintains glyph_order, cmap, and
-  hmtx tables, and optionally GSUB, glyf, and SVN tables as well."""
-
-  def __init__(self, font):
-    self.font = font;
-    self.glyph_order = font.getGlyphOrder()
-    self.cmap = font['cmap'].tables[0].cmap
-    self.hmtx = font['hmtx'].metrics
-
-  def init_gsub(self):
-    """Call this if you are going to add ligatures to the font.  Creates a GSUB
-    table if there isn't one already."""
-
-    if hasattr(self, 'ligatures'):
-      return
-    font = self.font
-    if 'GSUB' not in font:
-      ligature_subst = otTables.LigatureSubst()
-      ligature_subst.ligatures = {}
-
-      lookup = otTables.Lookup()
-      lookup.LookupType = 4
-      lookup.LookupFlag = 0
-      lookup.SubTableCount = 1
-      lookup.SubTable = [ligature_subst]
-
-      font['GSUB'] = add_emoji_gsub.create_simple_gsub([lookup])
-    else:
-      lookup = font['GSUB'].table.LookupList.Lookup[0]
-      assert lookup.LookupType == 4
-      assert lookup.LookupFlag == 0
-    self.ligatures = lookup.SubTable[0].ligatures
-
-  def init_glyf(self):
-    """Call this if you need to create empty glyf entries in the font when you
-    add a new glyph."""
-
-    if hasattr(self, 'glyphs'):
-      return
-    font = self.font
-    if 'glyf' not in font:
-      glyf_table = _g_l_y_f.table__g_l_y_f()
-      glyf_table.glyphs = {}
-      glyf_table.glyphOrder = self.glyph_order
-      font['glyf'] = glyf_table
-    self.glyphs = font['glyf'].glyphs
-
-  def init_svg(self):
-    """Call this if you expect to add SVG images in the font. This calls
-    init_glyf since SVG support currently requires fallback glyf records for
-    each SVG image."""
-
-    if hasattr(self, 'svgs'):
-      return
-
-    # svg requires glyf
-    self.init_glyf()
-
-    font = self.font
-    if 'SVG ' not in font:
-      svg_table = SVG.table_S_V_G_()
-      svg_table.docList = []
-      svg_table.colorPalettes = None
-      font['SVG '] = svg_table
-    self.svgs = font['SVG '].docList
-
-  def glyph_name(self, string):
-    return "_".join(["u%04X" % ord(char) for char in string])
-
-  def glyph_name_to_index(self, name):
-    return self.glyph_order.index(name) if name in self.glyph_order else -1;
-
-  def glyph_index_to_name(self, glyph_index):
-    if glyph_index < len(self.glyph_order):
-      return self.glyph_order[glyph_index]
-    return ''
-
-  def have_glyph(self, name):
-    return self.name_to_glyph_index >= 0
-
-  def _add_ligature(self, glyphstr):
-    lig = otTables.Ligature()
-    lig.CompCount = len(glyphstr)
-    lig.Component = [self.glyph_name(ch) for ch in glyphstr[1:]]
-    lig.LigGlyph = self.glyph_name(glyphstr)
-
-    first = self.glyph_name(glyphstr[0])
-    try:
-      self.ligatures[first].append(lig)
-    except KeyError:
-      self.ligatures[first] = [lig]
-
-  def _add_empty_glyph(self, glyphstr, name):
-    """Create an empty glyph. If glyphstr is not a ligature, add a cmap entry
-    for it."""
-    if len(glyphstr) == 1:
-      self.cmap[ord(glyphstr)] = name
-    self.hmtx[name] = [0, 0]
-    self.glyph_order.append(name)
-    if hasattr(self, 'glyphs'):
-      self.glyphs[name] = _g_l_y_f.Glyph()
-
-  def add_components_and_ligature(self, glyphstr):
-    """Convert glyphstr to a name and check if it already exists. If not, check
-    if it is a ligature (longer than one codepoint), and if it is, generate
-    empty glyphs with cmap entries for any missing ligature components and add a
-    ligature record.  Then generate an empty glyph for the name.  Return a tuple
-    with the name, index, and a bool indicating whether the glyph already
-    existed."""
-
-    name = self.glyph_name(glyphstr)
-    index = self.glyph_name_to_index(name)
-    exists = index >= 0
-    if not exists:
-      if len(glyphstr) > 1:
-        for char in glyphstr:
-          if ord(char) not in self.cmap:
-            char_name = self.glyph_name(char)
-            self._add_empty_glyph(char, char_name)
-        self._add_ligature(glyphstr)
-      index = len(self.glyph_order)
-      self._add_empty_glyph(glyphstr, name)
-    return name, index, exists
-
-  def add_svg(self, doc, hmetrics, name, index):
-    """Add an svg table entry. If hmetrics is not None, update the hmtx table.
-    This expects the glyph has already been added."""
-    # sanity check to make sure name and index correspond.
-    assert name == self.glyph_index_to_name(index)
-    if hmetrics:
-      self.hmtx[name] = hmetrics
-    svg_record = (doc, index, index) # startGlyphId, endGlyphId are the same
-    self.svgs.append(svg_record)
-
-
-def collect_glyphstr_file_pairs(prefix, ext, include=None, exclude=None, verbosity=1):
-  """Scan files with the given prefix and extension, and return a list of
-  (glyphstr, filename) where glyphstr is the character or ligature, and filename
-  is the image file associated with it.  The glyphstr is formed by decoding the
-  filename (exclusive of the prefix) as a sequence of hex codepoints separated
-  by underscore. Include, if defined, is a regex string to include only matched
-  filenames. Exclude, if defined, is a regex string to exclude matched
-  filenames, and is applied after include."""
-
-  image_files = {}
-  glob_pat = "%s*.%s" % (prefix, ext)
-  leading = len(prefix)
-  trailing = len(ext) + 1 # include dot
-  logging.info("Looking for images matching '%s'.", glob_pat)
-  ex_count = 0
-  ex = re.compile(exclude) if exclude else None
-  inc = re.compile(include) if include else None
-  if inc:
-    logging.info("Including images matching '%s'.", include)
-  if ex:
-    logging.info("Excluding images matching '%s'.", exclude)
-
-  for image_file in glob.glob(glob_pat):
-    if inc and not inc.search(image_file):
-      continue
-
-    if ex and ex.search(image_file):
-      if verbosity > 1:
-        print("Exclude %s" % image_file)
-      ex_count += 1
-      continue
-
-    codes = image_file[leading:-trailing]
-    if "_" in codes:
-      pieces = codes.split ("_")
-      u = "".join ([unichr(int(code, 16)) for code in pieces])
-    else:
-      u = unichr(int(codes, 16))
-    image_files[u] = image_file
-
-  if ex_count:
-    logging.info("Excluded %d files.", ex_count)
-  if not image_files:
-    raise Exception ("No image files matching '%s'.", glob_pat)
-  logging.info("Matched %s files.", len(image_files))
-  return image_files.items()
-
-
-def sort_glyphstr_tuples(glyphstr_tuples):
-  """The list contains tuples whose first element is a string representing a
-  character or ligature.  It is sorted with shorter glyphstrs first, then
-  alphabetically. This ensures that ligature components are added to the font
-  before any ligatures that contain them."""
-  glyphstr_tuples.sort(key=lambda t: (len(t[0]), t[0]))
-
-
-def add_image_glyphs(in_file, out_file, pairs):
-  """Add images from pairs (glyphstr, filename) to .ttx file in_file and write
-  to .ttx file out_file."""
-
-  font = ttx.TTFont()
-  font.importXML(in_file)
-
-  sort_glyphstr_tuples(pairs)
-
-  font_builder = FontBuilder(font)
-  # we've already sorted by length, so the longest glyphstrs are at the end. To
-  # see if we have ligatures, we just need to check the last one.
-  if len(pairs[-1][0]) > 1:
-    font_builder.init_gsub()
-
-  img_builder = svg_builder.SvgBuilder(font_builder)
-  for glyphstr, filename in pairs:
-    logging.debug("Adding glyph for U+%s", ",".join(
-          ["%04X" % ord(char) for char in glyphstr]))
-    img_builder.add_from_filename(glyphstr, filename)
-
-  font.saveXML(out_file)
-  logging.info("Added %s images to %s", len(pairs), out_file)
-
-
-def main(argv):
-  usage = """This will search for files that have image_prefix followed by one
-  or more hex numbers (separated by underscore if more than one), and end in
-  ".svg". For example, if image_prefix is "icons/u", then files with names like
-  "icons/u1F4A9.svg" or "icons/u1F1EF_1F1F5.svg" will be loaded.  The script
-  then adds cmap, htmx, and potentially GSUB entries for the Unicode characters
-  found.  The advance width will be chosen based on image aspect ratio.  If
-  Unicode values outside the BMP are desired, the existing cmap table should be
-  of the appropriate (format 12) type.  Only the first cmap table and the first
-  GSUB lookup (if existing) are modified."""
-
-  parser = argparse.ArgumentParser(
-      description='Update cmap, glyf, GSUB, and hmtx tables from image glyphs.',
-      epilog=usage)
-  parser.add_argument(
-      'in_file', help='Input ttx file name.', metavar='fname')
-  parser.add_argument(
-      'out_file', help='Output ttx file name.', metavar='fname')
-  parser.add_argument(
-      'image_prefix', help='Location and prefix of image files.',
-      metavar='path')
-  parser.add_argument(
-      '-i', '--include', help='include files whoses name matches this regex',
-      metavar='regex')
-  parser.add_argument(
-      '-e', '--exclude', help='exclude files whose name matches this regex',
-      metavar='regex')
-  parser.add_argument(
-      '-l', '--loglevel', help='log level name', default='warning')
-  args = parser.parse_args(argv)
-
-  tool_utils.setup_logging(args.loglevel)
-
-  pairs = collect_glyphstr_file_pairs(
-      args.image_prefix, 'svg', include=args.include, exclude=args.exclude)
-  add_image_glyphs(args.in_file, args.out_file, pairs)
-
-
-if __name__ == '__main__':
-  main(sys.argv[1:])

check_emoji_sequences.py

@@ -1,463 +0,0 @@
-#!/usr/bin/env python3
-#
-# Copyright 2016 Google Inc. All rights reserved.
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-
-"""Compare emoji image file namings against unicode property data."""
-from __future__ import print_function
-
-import argparse
-import collections
-import glob
-import os
-from os import path
-import re
-import sys
-
-from nototools import unicode_data
-import add_aliases
-
-ZWJ = 0x200d
-EMOJI_VS = 0xfe0f
-
-END_TAG = 0xe007f
-
-def _make_tag_set():
-  tag_set = set()
-  tag_set |= set(range(0xe0030, 0xe003a))  # 0-9
-  tag_set |= set(range(0xe0061, 0xe007b))  # a-z
-  tag_set.add(END_TAG)
-  return tag_set
-
-TAG_SET = _make_tag_set()
-
-_namedata = None
-
-def seq_name(seq):
-  global _namedata
-
-  if not _namedata:
-    def strip_vs_map(seq_map):
-      return {
-          unicode_data.strip_emoji_vs(k): v
-          for k, v in seq_map.iteritems()}
-    _namedata = [
-        strip_vs_map(unicode_data.get_emoji_combining_sequences()),
-        strip_vs_map(unicode_data.get_emoji_flag_sequences()),
-        strip_vs_map(unicode_data.get_emoji_modifier_sequences()),
-        strip_vs_map(unicode_data.get_emoji_zwj_sequences()),
-        ]
-
-  if len(seq) == 1:
-    return unicode_data.name(seq[0], None)
-
-  for data in _namedata:
-    if seq in data:
-      return data[seq]
-  if EMOJI_VS in seq:
-    non_vs_seq = unicode_data.strip_emoji_vs(seq)
-    for data in _namedata:
-      if non_vs_seq in data:
-        return data[non_vs_seq]
-
-  return None
-
-
-def _check_no_vs(sorted_seq_to_filepath):
-  """Our image data does not use emoji presentation variation selectors."""
-  for seq, fp in sorted_seq_to_filepath.iteritems():
-    if EMOJI_VS in seq:
-      print('check no VS: FE0F in path: %s' % fp)
-
-
-def _check_valid_emoji_cps(sorted_seq_to_filepath, unicode_version):
-  """Ensure all cps in these sequences are valid emoji cps or specific cps
-  used in forming emoji sequences.  This is a 'pre-check' that reports
-  this specific problem."""
-
-  valid_cps = set(unicode_data.get_emoji())
-  if unicode_version is None or unicode_version >= unicode_data.PROPOSED_EMOJI_AGE:
-    valid_cps |= unicode_data.proposed_emoji_cps()
-  else:
-    valid_cps = set(
-        cp for cp in valid_cps if unicode_data.age(cp) <= unicode_version)
-  valid_cps.add(0x200d)  # ZWJ
-  valid_cps.add(0x20e3)  # combining enclosing keycap
-  valid_cps.add(0xfe0f)  # variation selector (emoji presentation)
-  valid_cps.add(0xfe82b)  # PUA value for unknown flag
-  valid_cps |= TAG_SET  # used in subregion tag sequences
-
-  not_emoji = {}
-  for seq, fp in sorted_seq_to_filepath.iteritems():
-    for cp in seq:
-      if cp not in valid_cps:
-        if cp not in not_emoji:
-          not_emoji[cp] = []
-        not_emoji[cp].append(fp)
-
-  if len(not_emoji):
-    print(
-        'check valid emoji cps: %d non-emoji cp found' % len(not_emoji),
-        file=sys.stderr)
-    for cp in sorted(not_emoji):
-      fps = not_emoji[cp]
-      print(
-          'check valid emoji cps: %04x (in %d sequences)' % (cp, len(fps)),
-          file=sys.stderr)
-
-
-def _check_zwj(sorted_seq_to_filepath):
-  """Ensure zwj is only between two appropriate emoji.  This is a 'pre-check'
-  that reports this specific problem."""
-
-  for seq, fp in sorted_seq_to_filepath.iteritems():
-    if ZWJ not in seq:
-      continue
-    if seq[0] == ZWJ:
-      print('check zwj: zwj at head of sequence in %s' % fp, file=sys.stderr)
-    if len(seq) == 1:
-      continue
-    if seq[-1] == ZWJ:
-      print('check zwj: zwj at end of sequence in %s' % fp, file=sys.stderr)
-    for i, cp in enumerate(seq):
-      if cp == ZWJ:
-        if i > 0:
-          pcp = seq[i-1]
-          if pcp != EMOJI_VS and not unicode_data.is_emoji(pcp):
-            print(
-                'check zwj: non-emoji %04x preceeds ZWJ in %s' % (pcp, fp),
-                file=sys.stderr)
-        if i < len(seq) - 1:
-          fcp = seq[i+1]
-          if not unicode_data.is_emoji(fcp):
-            print(
-                'check zwj: non-emoji %04x follows ZWJ in %s' % (fcp, fp),
-                file=sys.stderr)
-
-
-def _check_flags(sorted_seq_to_filepath):
-  """Ensure regional indicators are only in sequences of one or two, and
-  never mixed."""
-  for seq, fp in sorted_seq_to_filepath.iteritems():
-    have_reg = None
-    for cp in seq:
-      is_reg = unicode_data.is_regional_indicator(cp)
-      if have_reg == None:
-        have_reg = is_reg
-      elif have_reg != is_reg:
-        print(
-            'check flags: mix of regional and non-regional in %s' % fp,
-            file=sys.stderr)
-    if have_reg and len(seq) > 2:
-      # We provide dummy glyphs for regional indicators, so there are sequences
-      # with single regional indicator symbols, the len check handles this.
-      print(
-          'check flags: regional indicator sequence length != 2 in %s' % fp,
-          file=sys.stderr)
-
-def _check_tags(sorted_seq_to_filepath):
-  """Ensure tag sequences (for subregion flags) conform to the spec.  We don't
-  validate against CLDR, just that there's a sequence of 2 or more tags starting
-  and ending with the appropriate codepoints."""
-
-  BLACK_FLAG = 0x1f3f4
-  BLACK_FLAG_SET = set([BLACK_FLAG])
-  for seq, fp in sorted_seq_to_filepath.iteritems():
-    seq_set = set(cp for cp in seq)
-    overlap_set = seq_set & TAG_SET
-    if not overlap_set:
-      continue
-    if seq[0] != BLACK_FLAG:
-      print('check tags: bad start tag in %s' % fp)
-    elif seq[-1] != END_TAG:
-      print('check tags: bad end tag in %s' % fp)
-    elif len(seq) < 4:
-      print('check tags: sequence too short in %s' % fp)
-    elif seq_set - TAG_SET != BLACK_FLAG_SET:
-      print('check tags: non-tag items in %s' % fp)
-
-
-def _check_skintone(sorted_seq_to_filepath):
-  """Ensure skin tone modifiers are not applied to emoji that are not defined
-  to take them.  May appear standalone, though.  Also check that emoji that take
-  skin tone modifiers have a complete set."""
-  base_to_modifiers = collections.defaultdict(set)
-  for seq, fp in sorted_seq_to_filepath.iteritems():
-    for i, cp in enumerate(seq):
-      if unicode_data.is_skintone_modifier(cp):
-        if i == 0:
-          if len(seq) > 1:
-            print(
-                'check skintone: skin color selector first in sequence %s' % fp,
-                file=sys.stderr)
-          # standalone are ok
-          continue
-        pcp = seq[i-1]
-        if not unicode_data.is_emoji_modifier_base(pcp):
-          print(
-              'check skintone: emoji skintone modifier applied to non-base ' +
-              'at %d: %s' % (i, fp), file=sys.stderr)
-        else:
-          if pcp not in base_to_modifiers:
-            base_to_modifiers[pcp] = set()
-          base_to_modifiers[pcp].add(cp)
-
-  for cp, modifiers in sorted(base_to_modifiers.iteritems()):
-    if len(modifiers) != 5:
-      print(
-          'check skintone: base %04x has %d modifiers defined (%s) in %s' % (
-              cp, len(modifiers),
-              ', '.join('%04x' % cp for cp in sorted(modifiers)), fp),
-          file=sys.stderr)
-
-
-def _check_zwj_sequences(sorted_seq_to_filepath, unicode_version):
-  """Verify that zwj sequences are valid for the given unicode version."""
-  for seq, fp in sorted_seq_to_filepath.iteritems():
-    if ZWJ not in seq:
-      continue
-    age = unicode_data.get_emoji_sequence_age(seq)
-    if age is None or unicode_version is not None and age > unicode_version:
-      print('check zwj sequences: undefined sequence %s' % fp)
-
-
-def _check_no_alias_sources(sorted_seq_to_filepath):
-  """Check that we don't have sequences that we expect to be aliased to
-  some other sequence."""
-  aliases = add_aliases.read_default_emoji_aliases()
-  for seq, fp in sorted_seq_to_filepath.iteritems():
-    if seq in aliases:
-      print('check no alias sources: aliased sequence %s' % fp)
-
-
-def _check_coverage(seq_to_filepath, unicode_version):
-  """Ensure we have all and only the cps and sequences that we need for the
-  font as of this version."""
-
-  age = unicode_version
-
-  non_vs_to_canonical = {}
-  for k in seq_to_filepath:
-    if EMOJI_VS in k:
-      non_vs = unicode_data.strip_emoji_vs(k)
-      non_vs_to_canonical[non_vs] = k
-
-  aliases = add_aliases.read_default_emoji_aliases()
-  for k, v in sorted(aliases.items()):
-    if v not in seq_to_filepath and v not in non_vs_to_canonical:
-      alias_str = unicode_data.seq_to_string(k)
-      target_str = unicode_data.seq_to_string(v)
-      print('coverage: alias %s missing target %s' % (alias_str, target_str))
-      continue
-    if k in seq_to_filepath or k in non_vs_to_canonical:
-      alias_str = unicode_data.seq_to_string(k)
-      target_str = unicode_data.seq_to_string(v)
-      print('coverage: alias %s already exists as %s (%s)' % (
-          alias_str, target_str, seq_name(v)))
-      continue
-    filename = seq_to_filepath.get(v) or seq_to_filepath[non_vs_to_canonical[v]]
-    seq_to_filepath[k] = 'alias:' + filename
-
-  # check single emoji, this includes most of the special chars
-  emoji = sorted(unicode_data.get_emoji(age=age))
-  for cp in emoji:
-    if tuple([cp]) not in seq_to_filepath:
-      print(
-          'coverage: missing single %04x (%s)' % (
-              cp, unicode_data.name(cp, '<no name>')))
-
-  # special characters
-  # all but combining enclosing keycap are currently marked as emoji
-  for cp in [ord('*'), ord('#'), ord(u'\u20e3')] + range(0x30, 0x3a):
-    if cp not in emoji and tuple([cp]) not in seq_to_filepath:
-      print('coverage: missing special %04x (%s)' % (cp, unicode_data.name(cp)))
-
-  # combining sequences
-  comb_seq_to_name = sorted(
-      unicode_data.get_emoji_combining_sequences(age=age).iteritems())
-  for seq, name in comb_seq_to_name:
-    if seq not in seq_to_filepath:
-      # strip vs and try again
-      non_vs_seq = unicode_data.strip_emoji_vs(seq)
-      if non_vs_seq not in seq_to_filepath:
-        print('coverage: missing combining sequence %s (%s)' %
-              (unicode_data.seq_to_string(seq), name))
-
-  # flag sequences
-  flag_seq_to_name = sorted(
-      unicode_data.get_emoji_flag_sequences(age=age).iteritems())
-  for seq, name in flag_seq_to_name:
-    if seq not in seq_to_filepath:
-      print('coverage: missing flag sequence %s (%s)' %
-            (unicode_data.seq_to_string(seq), name))
-
-  # skin tone modifier sequences
-  mod_seq_to_name = sorted(
-      unicode_data.get_emoji_modifier_sequences(age=age).iteritems())
-  for seq, name in mod_seq_to_name:
-    if seq not in seq_to_filepath:
-      print('coverage: missing modifier sequence %s (%s)' % (
-          unicode_data.seq_to_string(seq), name))
-
-  # zwj sequences
-  # some of ours include the emoji presentation variation selector and some
-  # don't, and the same is true for the canonical sequences.  normalize all
-  # of them to omit it to test coverage, but report the canonical sequence.
-  zwj_seq_without_vs = set()
-  for seq in seq_to_filepath:
-    if ZWJ not in seq:
-      continue
-    if EMOJI_VS in seq:
-      seq = tuple(cp for cp in seq if cp != EMOJI_VS)
-    zwj_seq_without_vs.add(seq)
-
-  for seq, name in sorted(
-      unicode_data.get_emoji_zwj_sequences(age=age).iteritems()):
-    if EMOJI_VS in seq:
-      test_seq = tuple(s for s in seq if s != EMOJI_VS)
-    else:
-      test_seq = seq
-    if test_seq not in zwj_seq_without_vs:
-      print('coverage: missing (canonical) zwj sequence %s (%s)' % (
-          unicode_data.seq_to_string(seq), name))
-
-  # check for 'unknown flag'
-  # this is either emoji_ufe82b or 'unknown_flag', but we filter out things that
-  # don't start with our prefix so 'unknown_flag' would be excluded by default.
-  if tuple([0xfe82b]) not in seq_to_filepath:
-    print('coverage: missing unknown flag PUA fe82b')
-
-
-def check_sequence_to_filepath(seq_to_filepath, unicode_version, coverage):
-  sorted_seq_to_filepath = collections.OrderedDict(
-      sorted(seq_to_filepath.items()))
-  _check_no_vs(sorted_seq_to_filepath)
-  _check_valid_emoji_cps(sorted_seq_to_filepath, unicode_version)
-  _check_zwj(sorted_seq_to_filepath)
-  _check_flags(sorted_seq_to_filepath)
-  _check_tags(sorted_seq_to_filepath)
-  _check_skintone(sorted_seq_to_filepath)
-  _check_zwj_sequences(sorted_seq_to_filepath, unicode_version)
-  _check_no_alias_sources(sorted_seq_to_filepath)
-  if coverage:
-    _check_coverage(sorted_seq_to_filepath, unicode_version)
-
-
-def create_sequence_to_filepath(name_to_dirpath, prefix, suffix):
-  """Check names, and convert name to sequences for names that are ok,
-  returning a sequence to file path mapping.  Reports bad segments
-  of a name to stderr."""
-  segment_re = re.compile(r'^[0-9a-f]{4,6}$')
-  result = {}
-  for name, dirname in name_to_dirpath.iteritems():
-    if not name.startswith(prefix):
-      print('expected prefix "%s" for "%s"' % (prefix, name))
-      continue
-
-    segments = name[len(prefix): -len(suffix)].split('_')
-    segfail = False
-    seq = []
-    for s in segments:
-      if not segment_re.match(s):
-        print('bad codepoint name "%s" in %s/%s' % (s, dirname, name))
-        segfail = True
-        continue
-      n = int(s, 16)
-      if n > 0x10ffff:
-        print('codepoint "%s" out of range in %s/%s' % (s, dirname, name))
-        segfail = True
-        continue
-      seq.append(n)
-    if not segfail:
-      result[tuple(seq)] = path.join(dirname, name)
-  return result
-
-
-def collect_name_to_dirpath(directory, prefix, suffix, exclude=None):
-  """Return a mapping from filename to path rooted at directory, ignoring files
-  that don't match suffix, and subtrees with names in exclude.  Report when a
-  filename appears in more than one subdir; the first path found is kept."""
-  result = {}
-  for dirname, dirs, files in os.walk(directory, topdown=True):
-    if exclude:
-      dirs[:] = [d for d in dirs if d not in exclude]
-
-    if directory != '.':
-      dirname = path.join(directory, dirname)
-    for f in files:
-      if not f.endswith(suffix):
-        continue
-      if f in result:
-        print('duplicate file "%s" in %s and %s ' % (
-            f, dirname, result[f]), file=sys.stderr)
-        continue
-      result[f] = dirname
-  return result
-
-
-def collect_name_to_dirpath_with_override(dirs, prefix, suffix, exclude=None):
-  """Return a mapping from filename to a directory path rooted at a directory
-  in dirs, using collect_name_to_filepath.  The last directory is retained. This
-  does not report an error if a file appears under more than one root directory,
-  so lets later root directories override earlier ones.  Use 'exclude' to
-  name subdirectories (of any root) whose subtree you wish to skip."""
-  result = {}
-  for d in dirs:
-    result.update(collect_name_to_dirpath(d, prefix, suffix, exclude))
-  return result
-
-
-def run_check(dirs, prefix, suffix, exclude, unicode_version, coverage):
-  msg = ''
-  if unicode_version:
-    msg = ' (%3.1f)' % unicode_version
-  print('Checking files with prefix "%s" and suffix "%s"%s in:\n  %s' % (
-      prefix, suffix, msg, '\n  '.join(dirs)))
-  name_to_dirpath = collect_name_to_dirpath_with_override(
-      dirs, prefix=prefix, suffix=suffix, exclude=exclude)
-  print('checking %d names' % len(name_to_dirpath))
-  seq_to_filepath = create_sequence_to_filepath(name_to_dirpath, prefix, suffix)
-  print('checking %d sequences' % len(seq_to_filepath))
-  check_sequence_to_filepath(seq_to_filepath, unicode_version, coverage)
-  print('done.')
-
-
-def main():
-  parser = argparse.ArgumentParser()
-  parser.add_argument(
-      '-d', '--dirs', help='directory roots containing emoji images',
-      metavar='dir', nargs='+', required=True)
-  parser.add_argument(
-      '-e', '--exclude', help='names of source subdirs to exclude',
-      metavar='dir', nargs='+')
-  parser.add_argument(
-      '-c', '--coverage', help='test for complete coverage',
-      action='store_true')
-  parser.add_argument(
-      '-p', '--prefix', help='prefix to match, default "emoji_u"',
-      metavar='pfx', default='emoji_u')
-  parser.add_argument(
-      '-s', '--suffix', help='suffix to match, default ".png"', metavar='sfx',
-      default='.png')
-  parser.add_argument(
-      '-u', '--unicode_version', help='limit to this unicode version or before',
-      metavar='version', type=float)
-  args = parser.parse_args()
-  run_check(
-      args.dirs, args.prefix, args.suffix, args.exclude, args.unicode_version,
-      args.coverage)
-
-
-if __name__ == '__main__':
-  main()

collect_emoji_svg.py

@@ -1,150 +0,0 @@
-#!/usr/bin/env python3
-# Copyright 2015 Google, Inc. All Rights Reserved.
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-#
-# Google Author(s): Doug Felt
-
-"""Tool to collect emoji svg glyphs into one directory for processing
-by add_svg_glyphs.  There are two sources, noto/color_emoji/svg and
-noto/third_party/region-flags/svg.  The add_svg_glyphs file expects
-the file names to contain the character string that represents it
-represented as a sequence of hex-encoded codepoints separated by
-underscore.  The files in noto/color_emoji/svg do this, and have the
-prefix 'emoji_u', but the files in region-flags/svg just have the
-two-letter code.
-
-We create a directory and copy the files into it with the required
-naming convention. First we do this for region-flags/svg, converting
-the names, and then we do this for color_emoji/svg, so any duplicates
-will be overwritten by what we assume are the preferred svg.  We use
-copies instead of symlinks so we can continue to optimize or modify
-the files without messing with the originals."""
-
-import argparse
-import glob
-import logging
-import os
-import os.path
-import re
-import shutil
-import sys
-
-from nototools import tool_utils
-
-def _is_svg(f):
-  return f.endswith('.svg')
-
-
-def _is_svg_and_startswith_emoji(f):
-  return f.endswith('.svg') and f.startswith('emoji_u')
-
-
-def _flag_rename(f):
-  """Converts a file name from two-letter upper-case ASCII to our expected
-  'emoji_uXXXXX_XXXXX form, mapping each character to the corresponding
-  regional indicator symbol."""
-
-  cp_strs = []
-  name, ext = os.path.splitext(f)
-  if len(name) != 2:
-    raise ValueError('illegal flag name "%s"' % f)
-  for cp in name:
-    if not ('A' <= cp <= 'Z'):
-      raise ValueError('illegal flag name "%s"' % f)
-    ncp = 0x1f1e6 - 0x41 + ord(cp)
-    cp_strs.append("%04x" % ncp)
-  return 'emoji_u%s%s' % ('_'.join(cp_strs), ext)
-
-
-def copy_with_rename(src_dir, dst_dir, accept_pred=None, rename=None):
-  """Copy files from src_dir to dst_dir that match accept_pred (all if None) and
-  rename using rename (if not None), replacing existing files.  accept_pred
-  takes the filename and returns True if the file should be copied, rename takes
-  the filename and returns a new file name."""
-
-  count = 0
-  replace_count = 0
-  for src_filename in os.listdir(src_dir):
-    if accept_pred and not accept_pred(src_filename):
-      continue
-    dst_filename = rename(src_filename) if rename else src_filename
-    src = os.path.join(src_dir, src_filename)
-    dst = os.path.join(dst_dir, dst_filename)
-    if os.path.exists(dst):
-      logging.debug('Replacing existing file %s', dst)
-      os.unlink(dst)
-      replace_count += 1
-    shutil.copy2(src, dst)
-    logging.debug('cp -p %s %s', src, dst)
-    count += 1
-  if logging.getLogger().getEffectiveLevel() <= logging.INFO:
-    src_short = tool_utils.short_path(src_dir)
-    dst_short = tool_utils.short_path(dst_dir)
-    logging.info('Copied %d files (replacing %d) from %s to %s',
-        count, replace_count, src_short, dst_short)
-
-
-def build_svg_dir(dst_dir, clean=False, emoji_dir='', flags_dir=''):
-  """Copies/renames files from emoji_dir and then flags_dir, giving them the
-  standard format and prefix ('emoji_u' followed by codepoints expressed in hex
-  separated by underscore).  If clean, removes the target dir before proceding.
-  If either emoji_dir or flags_dir are empty, skips them."""
-
-  dst_dir = tool_utils.ensure_dir_exists(dst_dir, clean=clean)
-
-  if not emoji_dir and not flags_dir:
-    logging.warning('Nothing to do.')
-    return
-
-  if emoji_dir:
-    copy_with_rename(
-        emoji_dir, dst_dir, accept_pred=_is_svg_and_startswith_emoji)
-
-  if flags_dir:
-     copy_with_rename(
-        flags_dir, dst_dir, accept_pred=_is_svg, rename=_flag_rename)
-
-
-def main(argv):
-  DEFAULT_EMOJI_DIR = '[emoji]/svg'
-  DEFAULT_FLAGS_DIR = '[emoji]/third_party/region-flags/svg'
-
-  parser = argparse.ArgumentParser(
-      description='Collect svg files into target directory with prefix.')
-  parser.add_argument(
-      'dst_dir', help='Directory to hold copied files.', metavar='dir')
-  parser.add_argument(
-      '--clean', '-c', help='Replace target directory', action='store_true')
-  parser.add_argument(
-      '--flags_dir', '-f', metavar='dir', help='directory containing flag svg, '
-      'default %s' % DEFAULT_FLAGS_DIR, default=DEFAULT_FLAGS_DIR)
-  parser.add_argument(
-      '--emoji_dir', '-e', metavar='dir',
-      help='directory containing emoji svg, default %s' % DEFAULT_EMOJI_DIR,
-      default=DEFAULT_EMOJI_DIR)
-  parser.add_argument(
-      '-l', '--loglevel', help='log level name/value', default='warning')
-  args = parser.parse_args(argv)
-
-  tool_utils.setup_logging(args.loglevel)
-
-  args.flags_dir = tool_utils.resolve_path(args.flags_dir)
-  args.emoji_dir = tool_utils.resolve_path(args.emoji_dir)
-  build_svg_dir(
-      args.dst_dir, clean=args.clean, emoji_dir=args.emoji_dir,
-      flags_dir=args.flags_dir)
-
-
-if __name__ == '__main__':
-  main(sys.argv[1:])