Parsing for snips templates

src/calibre/gui2/tweak_book/snips.py

@@ -6,6 +6,13 @@ from __future__ import (unicode_literals, division, absolute_import,
 __license__ = 'GPL v3'
 __copyright__ = '2014, Kovid Goyal <kovid at kovidgoyal.net>'
 
+import re
+from collections import OrderedDict
+from itertools import groupby
+from operator import attrgetter
+
+from calibre.utils.icu import string_length
+
 builtin_snips = {
     '<<' : {
         'description': _('Insert a HTML tag'),
@@ -27,5 +34,76 @@ builtin_snips = {
         'template': '<img src="$1" alt="${2*}" />$3',
     },
 
+    '<c' : {
+        'description': _('Insert a HTML tag with a class'),
+        'template': '<$1 class="$2">${3*}</$1>',
+    },
+
 }
 
+escape = unescape = None
+def escape_funcs():
+    global escape, unescape
+    if escape is None:
+        escapem = {('\\' + x):unichr(i+1) for i, x in enumerate('\\${}')}
+        escape_pat = re.compile('|'.join(map(re.escape, escapem)))
+        escape = lambda x: escape_pat.sub(lambda m: escapem[m.group()], x.replace(r'\\', '\x01'))
+        unescapem = {v:k[1] for k, v in escapem.iteritems()}
+        unescape_pat = re.compile('|'.join(unescapem))
+        unescape = lambda x:unescape_pat.sub(lambda m:unescapem[m.group()], x)
+    return escape, unescape
+
+class TabStop(unicode):
+
+    def __new__(self, raw, start_offset, tab_stops, is_toplevel=True):
+        if raw.endswith('}'):
+            unescape = escape_funcs()[1]
+            num, default = raw[2:-1].partition(':')[0::2]
+            # Look for tab stops defined in the default text
+            uraw, child_stops = parse_template(unescape(default), start_offset=start_offset, is_toplevel=False, grouped=False)
+            tab_stops.extend(child_stops)
+            self = unicode.__new__(self, uraw)
+            if num.endswith('*'):
+                self.takes_selection = True
+                num = num[:-1]
+            else:
+                self.takes_selection = False
+            self.num = int(num)
+        else:
+            self = unicode.__new__(self, '')
+            self.num = int(raw[1:])
+            self.takes_selection = False
+        self.start = start_offset
+        self.is_toplevel = is_toplevel
+        self.is_mirror = False
+        tab_stops.append(self)
+        return self
+
+    def __repr__(self):
+        return 'TabStop(text=%s, num=%d, start=%d, is_mirror=%s takes_selection=%s is_toplevel=%s)' % (
+            unicode.__repr__(self), self.num, self.start, self.is_mirror, self.takes_selection, self.is_toplevel)
+    __str__ = __unicode__ = __repr__
+
+def parse_template(template, start_offset=0, is_toplevel=True, grouped=True):
+    escape, unescape = escape_funcs()
+    template = escape(template)
+    pos, parts, tab_stops = start_offset, [], []
+    for part in re.split(r'(\$(?:\d+|\{[^}]+\}))', template):
+        is_tab_stop = part.startswith('$')
+        if is_tab_stop:
+            ts = TabStop(part, pos, tab_stops, is_toplevel=is_toplevel)
+            parts.append(ts)
+        else:
+            parts.append(unescape(part))
+        pos += string_length(parts[-1])
+    if grouped:
+        key = attrgetter('num')
+        tab_stops.sort(key=key)
+        ans = OrderedDict()
+        for num, stops in groupby(tab_stops, key):
+            stops = tuple(stops)
+            for ts in stops[1:]:
+                ts.is_mirror = True
+            ans[num] = stops
+        tab_stops = ans
+    return ''.join(parts), tab_stops