From b7e8f69ece40b13d8442556ec40c4ed4c649fc7b Mon Sep 17 00:00:00 2001
From: Kovid Goyal <kovid@kovidgoyal.net>
Date: Thu, 8 Nov 2012 15:54:59 +0530
Subject: [PATCH] Start work on subsetting CFF tables

---
 src/calibre/utils/fonts/sfnt/cff/table.py  | 23 ++++++++++++++++++++++
 src/calibre/utils/fonts/sfnt/cff/writer.py | 16 +++++++++++++++
 src/calibre/utils/fonts/sfnt/subset.py     |  1 +
 3 files changed, 40 insertions(+)
 create mode 100644 src/calibre/utils/fonts/sfnt/cff/writer.py

diff --git a/src/calibre/utils/fonts/sfnt/cff/table.py b/src/calibre/utils/fonts/sfnt/cff/table.py
index 11f24aa98d..accb5f91f3 100644
--- a/src/calibre/utils/fonts/sfnt/cff/table.py
+++ b/src/calibre/utils/fonts/sfnt/cff/table.py
@@ -268,6 +268,12 @@ class Charset(list):
             return self[glyph_id]
         return self.STANDARD_CHARSETS[self.standard_charset][glyph_id].encode('ascii')
 
+    def safe_lookup(self, glyph_id):
+        try:
+            return self.lookup(glyph_id)
+        except (KeyError, IndexError, ValueError):
+            return None
+
 class Subrs(Index):
     pass
 
@@ -279,6 +285,23 @@ class CFFTable(UnknownTable):
     def decompile(self):
         self.cff = CFF(self.raw)
 
+    def subset(self, character_map):
+        from calibre.utils.fonts.sfnt.cff.writer import Subset
+        # Map codes from the cmap table to glyph names, this will be used to
+        # reconstruct character_map for the subset font
+        charset_map = {code:self.cff.charset.safe_lookup(glyph_id) for code,
+                glyph_id in character_map.iteritems()}
+        charset = set(charset_map.itervalues())
+        charset.discard(None)
+        s = Subset(self.cff, charset)
+
+        # Rebuild character_map with the glyph ids from the subset font
+        character_map.clear()
+        for code, charname in charset_map:
+            glyph_id = s.charname_map.get(charname, None)
+            if glyph_id:
+                character_map[code] = glyph_id
+
 # cff_standard_strings {{{
 # The 391 Standard Strings as used in the CFF format.
 # from Adobe Technical None #5176, version 1.0, 18 March 1998
diff --git a/src/calibre/utils/fonts/sfnt/cff/writer.py b/src/calibre/utils/fonts/sfnt/cff/writer.py
new file mode 100644
index 0000000000..cab5707fa0
--- /dev/null
+++ b/src/calibre/utils/fonts/sfnt/cff/writer.py
@@ -0,0 +1,16 @@
+#!/usr/bin/env python
+# vim:fileencoding=UTF-8:ts=4:sw=4:sta:et:sts=4:fdm=marker:ai
+from __future__ import (unicode_literals, division, absolute_import,
+                        print_function)
+
+__license__   = 'GPL v3'
+__copyright__ = '2012, Kovid Goyal <kovid at kovidgoyal.net>'
+__docformat__ = 'restructuredtext en'
+
+class Subset(object):
+
+    def __init__(self, cff, character_map):
+        self.cff = cff
+        self.character_map = character_map
+        self.charname_map = {}
+
diff --git a/src/calibre/utils/fonts/sfnt/subset.py b/src/calibre/utils/fonts/sfnt/subset.py
index c7e3ec8c34..4c4fbfa7c1 100644
--- a/src/calibre/utils/fonts/sfnt/subset.py
+++ b/src/calibre/utils/fonts/sfnt/subset.py
@@ -69,6 +69,7 @@ def subset_truetype(sfnt, character_map):
 def subset_postscript(sfnt, character_map):
     cff = sfnt[b'CFF ']
     cff.decompile()
+    cff.subset(character_map)
     raise Exception('TODO: Implement CFF subsetting')
 
 def subset(raw, individual_chars, ranges=()):