Merged in protz/pygments-main/add-envname (pull request #235)

1 files changed, 49 insertions, 38 deletions

diff --git a/pygments/formatters/rtf.py b/pygments/formatters/rtf.py
index 3efda284..abecd484 100644
--- a/pygments/formatters/rtf.py
+++ b/pygments/formatters/rtf.py
@@ -5,11 +5,12 @@
 
     A formatter that generates RTF files.
 
-    :copyright: Copyright 2006-2013 by the Pygments team, see AUTHORS.
+    :copyright: Copyright 2006-2014 by the Pygments team, see AUTHORS.
     :license: BSD, see LICENSE for details.
 """
 
 from pygments.formatter import Formatter
+from pygments.util import get_int_opt, _surrogatepair
 
 
 __all__ = ['RtfFormatter']
@@ -21,7 +22,11 @@ class RtfFormatter(Formatter):
     documents with color information and other useful stuff. Perfect for Copy and
     Paste into Microsoft® Word® documents.
 
-    *New in Pygments 0.6.*
+    Please note that ``encoding`` and ``outencoding`` options are ignored.
+    The RTF format is ASCII natively, but handles unicode characters correctly
+    thanks to escape sequences.
+
+    .. versionadded:: 0.6
 
     Additional options accepted:
 
@@ -32,13 +37,17 @@ class RtfFormatter(Formatter):
     `fontface`
         The used font famliy, for example ``Bitstream Vera Sans``. Defaults to
         some generic font which is supposed to have fixed width.
+
+    `fontsize`
+        Size of the font used. Size is specified in half points. The
+        default is 24 half-points, giving a size 12 font.
+
+        .. versionadded:: 2.0
     """
     name = 'RTF'
     aliases = ['rtf']
     filenames = ['*.rtf']
 
-    unicodeoutput = False
-
     def __init__(self, **options):
         """
         Additional options accepted:
@@ -49,48 +58,48 @@ class RtfFormatter(Formatter):
             specification claims that ``\fmodern`` are "Fixed-pitch serif
             and sans serif fonts". Hope every RTF implementation thinks
             the same about modern...
+
         """
         Formatter.__init__(self, **options)
         self.fontface = options.get('fontface') or ''
+        self.fontsize = get_int_opt(options, 'fontsize', 0)
 
     def _escape(self, text):
-        return text.replace('\\', '\\\\') \
-                   .replace('{', '\\{') \
-                   .replace('}', '\\}')
+        return text.replace(u'\\', u'\\\\') \
+                   .replace(u'{', u'\\{') \
+                   .replace(u'}', u'\\}')
 
     def _escape_text(self, text):
         # empty strings, should give a small performance improvment
         if not text:
-            return ''
+            return u''
 
         # escape text
         text = self._escape(text)
-        if self.encoding in ('utf-8', 'utf-16', 'utf-32'):
-            encoding = 'iso-8859-15'
-        else:
-            encoding = self.encoding or 'iso-8859-15'
 
         buf = []
         for c in text:
-            if ord(c) > 128:
-                ansic = c.encode(encoding, 'ignore') or '?'
-                if ord(ansic) > 128:
-                    ansic = '\\\'%x' % ord(ansic)
-                else:
-                    ansic = c
-                buf.append(r'\ud{\u%d%s}' % (ord(c), ansic))
-            else:
+            cn = ord(c)
+            if cn < (2**7):
+                # ASCII character
                 buf.append(str(c))
+            elif (2**7) <= cn < (2**16):
+                # single unicode escape sequence
+                buf.append(u'{\\u%d}' % cn)
+            elif (2**16) <= cn:
+                # RTF limits unicode to 16 bits.
+                # Force surrogate pairs
+                buf.append(u'{\\u%d}{\\u%d}' % _surrogatepair(cn))
 
-        return ''.join(buf).replace('\n', '\\par\n')
+        return u''.join(buf).replace(u'\n', u'\\par\n')
 
     def format_unencoded(self, tokensource, outfile):
         # rtf 1.8 header
-        outfile.write(r'{\rtf1\ansi\deff0'
-                      r'{\fonttbl{\f0\fmodern\fprq1\fcharset0%s;}}'
-                      r'{\colortbl;' % (self.fontface and
-                                        ' ' + self._escape(self.fontface) or
-                                        ''))
+        outfile.write(u'{\\rtf1\\ansi\\uc0\\deff0'
+                      u'{\\fonttbl{\\f0\\fmodern\\fprq1\\fcharset0%s;}}'
+                      u'{\\colortbl;' % (self.fontface and
+                                         u' ' + self._escape(self.fontface) or
+                                         u''))
 
         # convert colors and save them in a mapping to access them later.
         color_mapping = {}
@@ -99,13 +108,15 @@ class RtfFormatter(Formatter):
             for color in style['color'], style['bgcolor'], style['border']:
                 if color and color not in color_mapping:
                     color_mapping[color] = offset
-                    outfile.write(r'\red%d\green%d\blue%d;' % (
+                    outfile.write(u'\\red%d\\green%d\\blue%d;' % (
                         int(color[0:2], 16),
                         int(color[2:4], 16),
                         int(color[4:6], 16)
                     ))
                     offset += 1
-        outfile.write(r'}\f0')
+        outfile.write(u'}\\f0 ')
+        if self.fontsize:
+            outfile.write(u'\\fs%d' % (self.fontsize))
 
         # highlight stream
         for ttype, value in tokensource:
@@ -114,23 +125,23 @@ class RtfFormatter(Formatter):
             style = self.style.style_for_token(ttype)
             buf = []
             if style['bgcolor']:
-                buf.append(r'\cb%d' % color_mapping[style['bgcolor']])
+                buf.append(u'\\cb%d' % color_mapping[style['bgcolor']])
             if style['color']:
-                buf.append(r'\cf%d' % color_mapping[style['color']])
+                buf.append(u'\\cf%d' % color_mapping[style['color']])
             if style['bold']:
-                buf.append(r'\b')
+                buf.append(u'\\b')
             if style['italic']:
-                buf.append(r'\i')
+                buf.append(u'\\i')
             if style['underline']:
-                buf.append(r'\ul')
+                buf.append(u'\\ul')
             if style['border']:
-                buf.append(r'\chbrdr\chcfpat%d' %
+                buf.append(u'\\chbrdr\\chcfpat%d' %
                            color_mapping[style['border']])
-            start = ''.join(buf)
+            start = u''.join(buf)
             if start:
-                outfile.write('{%s ' % start)
+                outfile.write(u'{%s ' % start)
             outfile.write(self._escape_text(value))
             if start:
-                outfile.write('}')
+                outfile.write(u'}')
 
-        outfile.write('}')
+        outfile.write(u'}')