Speeds up font compilation by around 200%

Cython is used to compile some hot paths into native Python extensions.
These hot paths were identified through running ufocompile with the hotshot
profiler and then converting file by file to Cython, starting with the "hottest"
paths and continuing until returns were deminishing. This means that only a few
Python files were converted to Cython.

Closes #23
Closes #20 (really this time)

misc/pylib/robofab/tools/fontlabFeatureSplitter.py

@@ -0,0 +1,85 @@
+import re
+
+featureRE = re.compile(
+    "^"            # start of line
+    "\s*"          #
+    "feature"      # feature
+    "\s+"          #
+    "(\w{4})"      # four alphanumeric characters
+    "\s*"          #
+    "\{"           # {
+    , re.MULTILINE # run in multiline to preserve line seps
+)
+
+def splitFeaturesForFontLab(text):
+    """
+    >>> result = splitFeaturesForFontLab(testText)
+    >>> result == expectedTestResult
+    True
+    """
+    classes = ""
+    features = []
+    while text:
+        m = featureRE.search(text)
+        if m is None:
+            classes = text
+            text = ""
+        else:
+            start, end = m.span()
+            # if start is not zero, this is the first match
+            # and all previous lines are part of the "classes"
+            if start > 0:
+                assert not classes
+                classes = text[:start]
+            # extract the current feature
+            featureName = m.group(1)
+            featureText = text[start:end]
+            text = text[end:]
+            # grab all text before the next feature definition
+            # and add it to the current definition
+            if text:
+                m = featureRE.search(text)
+                if m is not None:
+                    start, end = m.span()
+                    featureText += text[:start]
+                    text = text[start:]
+                else:
+                    featureText += text
+                    text = ""
+            # store the feature
+            features.append((featureName, featureText))
+    return classes, features
+
+testText = """
+@class1 = [a b c d];
+
+feature liga {
+    sub f i by fi;
+} liga;
+
+@class2 = [x y z];
+
+feature salt {
+    sub a by a.alt;
+} salt; feature ss01 {sub x by x.alt} ss01;
+
+feature ss02 {sub y by y.alt} ss02;
+
+# feature calt {
+#     sub a b' by b.alt;
+# } calt;
+"""
+
+expectedTestResult = (
+    "\n@class1 = [a b c d];\n",
+    [
+        ("liga", "\nfeature liga {\n    sub f i by fi;\n} liga;\n\n@class2 = [x y z];\n"),
+        ("salt", "\nfeature salt {\n    sub a by a.alt;\n} salt; feature ss01 {sub x by x.alt} ss01;\n"),
+        ("ss02", "\nfeature ss02 {sub y by y.alt} ss02;\n\n# feature calt {\n#     sub a b' by b.alt;\n# } calt;\n")
+    ]
+)
+
+
+if __name__ == "__main__":
+    import doctest
+    doctest.testmod()