diff options
author | Rasmus Andersson <rasmus@notion.se> | 2017-09-04 06:03:17 +0300 |
---|---|---|
committer | Rasmus Andersson <rasmus@notion.se> | 2017-09-04 18:12:34 +0300 |
commit | 8234b62ab762637ef24c3398b4204a8ce8db31a7 (patch) | |
tree | 1c8df547021cdb58951630a015e4101ede46dbf1 /misc/pylib/robofab/glifLib.pyx | |
parent | 31ae014e0c827dd76696fdab7e4ca3fed9f6402b (diff) | |
download | inter-8234b62ab762637ef24c3398b4204a8ce8db31a7.tar.xz |
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)
Diffstat (limited to 'misc/pylib/robofab/glifLib.pyx')
-rwxr-xr-x | misc/pylib/robofab/glifLib.pyx | 718 |
1 files changed, 718 insertions, 0 deletions
diff --git a/misc/pylib/robofab/glifLib.pyx b/misc/pylib/robofab/glifLib.pyx new file mode 100755 index 000000000..8a447fff4 --- /dev/null +++ b/misc/pylib/robofab/glifLib.pyx @@ -0,0 +1,718 @@ +# -*- coding: utf-8 -*- +"""glifLib.py -- Generic module for reading and writing the .glif format. + +More info about the .glif format (GLyphInterchangeFormat) can be found here: + + http://robofab.com/ufo/glif.html + +The main class in this module is GlyphSet. It manages a set of .glif files +in a folder. It offers two ways to read glyph data, and one way to write +glyph data. See the class doc string for details. +""" + +__all__ = ["GlyphSet", "GlifLibError", + "readGlyphFromString", "writeGlyphToString", + "glyphNameToFileName"] + +import os +from robofab.xmlTreeBuilder import buildTree, stripCharacterData +from robofab.pens.pointPen import AbstractPointPen +from cStringIO import StringIO + + +class GlifLibError(Exception): pass + + +if os.name == "mac": + WRITE_MODE = "wb" # use unix line endings, even with Classic MacPython + READ_MODE = "rb" +else: + WRITE_MODE = "w" + READ_MODE = "r" + + +class Glyph: + + """Minimal glyph object. It has no glyph attributes until either + the draw() or the drawPoint() method has been called. + """ + + def __init__(self, glyphName, glyphSet): + self.glyphName = glyphName + self.glyphSet = glyphSet + + def draw(self, pen): + """Draw this glyph onto a *FontTools* Pen.""" + from robofab.pens.adapterPens import PointToSegmentPen + pointPen = PointToSegmentPen(pen) + self.drawPoints(pointPen) + + def drawPoints(self, pointPen): + """Draw this glyph onto a PointPen.""" + self.glyphSet.readGlyph(self.glyphName, self, pointPen) + + +def glyphNameToFileName(glyphName, glyphSet): + """Default algorithm for making a file name out of a glyph name. + This one has limited support for case insensitive file systems: + it assumes glyph names are not case sensitive apart from the first + character: + 'a' -> 'a.glif' + 'A' -> 'A_.glif' + 'A.alt' -> 'A_.alt.glif' + 'A.Alt' -> 'A_.Alt.glif' + 'T_H' -> 'T__H_.glif' + 'T_h' -> 'T__h.glif' + 't_h' -> 't_h.glif' + 'F_F_I' -> 'F__F__I_.glif' + 'f_f_i' -> 'f_f_i.glif' + + """ + if glyphName.startswith("."): + # some OSes consider filenames such as .notdef "hidden" + glyphName = "_" + glyphName[1:] + parts = glyphName.split(".") + if parts[0].find("_")!=-1: + # it is a compound name, check the separate parts + bits = [] + for p in parts[0].split("_"): + if p != p.lower(): + bits.append(p+"_") + continue + bits.append(p) + parts[0] = "_".join(bits) + else: + # it is a single name + if parts[0] != parts[0].lower(): + parts[0] += "_" + for i in range(1, len(parts)): + # resolve additional, period separated parts, like alt / Alt + if parts[i] != parts[i].lower(): + parts[i] += "_" + return ".".join(parts) + ".glif" + + + +class GlyphSet: + + """GlyphSet manages a set of .glif files inside one directory. + + GlyphSet's constructor takes a path to an existing directory as it's + first argument. Reading glyph data can either be done through the + readGlyph() method, or by using GlyphSet's dictionary interface, where + the keys are glyph names and the values are (very) simple glyph objects. + + To write a glyph to the glyph set, you use the writeGlyph() method. + The simple glyph objects returned through the dict interface do not + support writing, they are just means as a convenient way to get at + the glyph data. + """ + + glyphClass = Glyph + + def __init__(self, dirName, glyphNameToFileNameFunc=None): + """'dirName' should be a path to an existing directory. + + The optional 'glyphNameToFileNameFunc' argument must be a callback + function that takes two arguments: a glyph name and the GlyphSet + instance. It should return a file name (including the .glif + extension). The glyphNameToFileName function is called whenever + a file name is created for a given glyph name. + """ + self.dirName = dirName + if glyphNameToFileNameFunc is None: + glyphNameToFileNameFunc = glyphNameToFileName + self.glyphNameToFileName = glyphNameToFileNameFunc + self.contents = self._findContents() + self._reverseContents = None + + def rebuildContents(self): + """Rebuild the contents dict by checking what glyphs are available + on disk. + """ + self.contents = self._findContents(forceRebuild=True) + self._reverseContents = None + + def getReverseContents(self): + """Return a reversed dict of self.contents, mapping file names to + glyph names. This is primarily an aid for custom glyph name to file + name schemes that want to make sure they don't generate duplicate + file names. The file names are converted to lowercase so we can + reliably check for duplicates that only differ in case, which is + important for case-insensitive file systems. + """ + if self._reverseContents is None: + d = {} + for k, v in self.contents.iteritems(): + d[v.lower()] = k + self._reverseContents = d + return self._reverseContents + + def writeContents(self): + """Write the contents.plist file out to disk. Call this method when + you're done writing glyphs. + """ + from plistlib import writePlistToString + contentsPath = os.path.join(self.dirName, "contents.plist") + # We need to force Unix line endings, even in OS9 MacPython in FL, + # so we do the writing to file ourselves. + plist = writePlistToString(self.contents) + f = open(contentsPath, WRITE_MODE) + f.write(plist) + f.close() + + # reading/writing API + + def readGlyph(self, glyphName, glyphObject=None, pointPen=None): + """Read a .glif file for 'glyphName' from the glyph set. The + 'glyphObject' argument can be any kind of object (even None); + the readGlyph() method will attempt to set the following + attributes on it: + "width" the advance with of the glyph + "unicodes" a list of unicode values for this glyph + "note" a string + "lib" a dictionary containing custom data + + All attributes are optional, in two ways: + 1) An attribute *won't* be set if the .glif file doesn't + contain data for it. 'glyphObject' will have to deal + with default values itself. + 2) If setting the attribute fails with an AttributeError + (for example if the 'glyphObject' attribute is read- + only), readGlyph() will not propagate that exception, + but ignore that attribute. + + To retrieve outline information, you need to pass an object + conforming to the PointPen protocol as the 'pointPen' argument. + This argument may be None if you don't need the outline data. + + readGlyph() will raise KeyError if the glyph is not present in + the glyph set. + """ + tree = self._getXMLTree(glyphName) + _readGlyphFromTree(tree, glyphObject, pointPen) + + def writeGlyph(self, glyphName, glyphObject=None, drawPointsFunc=None): + """Write a .glif file for 'glyphName' to the glyph set. The + 'glyphObject' argument can be any kind of object (even None); + the writeGlyph() method will attempt to get the following + attributes from it: + "width" the advance with of the glyph + "unicodes" a list of unicode values for this glyph + "note" a string + "lib" a dictionary containing custom data + + All attributes are optional: if 'glyphObject' doesn't + have the attribute, it will simply be skipped. + + To write outline data to the .glif file, writeGlyph() needs + a function (any callable object actually) that will take one + argument: an object that conforms to the PointPen protocol. + The function will be called by writeGlyph(); it has to call the + proper PointPen methods to transfer the outline to the .glif file. + """ + data = writeGlyphToString(glyphName, glyphObject, drawPointsFunc) + + fileName = self.contents.get(glyphName) + if fileName is None: + fileName = self.glyphNameToFileName(glyphName, self) + self.contents[glyphName] = fileName + if self._reverseContents is not None: + self._reverseContents[fileName.lower()] = glyphName + path = os.path.join(self.dirName, fileName) + if os.path.exists(path): + f = open(path, READ_MODE) + oldData = f.read() + f.close() + if data == oldData: + return + f = open(path, WRITE_MODE) + f.write(data) + f.close() + + def deleteGlyph(self, glyphName): + """Permanently delete the glyph from the glyph set on disk. Will + raise KeyError if the glyph is not present in the glyph set. + """ + fileName = self.contents[glyphName] + os.remove(os.path.join(self.dirName, fileName)) + if self._reverseContents is not None: + del self._reverseContents[self.contents[glyphName].lower()] + del self.contents[glyphName] + + # dict-like support + + def keys(self): + return self.contents.keys() + + def has_key(self, glyphName): + return glyphName in self.contents + + __contains__ = has_key + + def __len__(self): + return len(self.contents) + + def __getitem__(self, glyphName): + if glyphName not in self.contents: + raise KeyError, glyphName + return self.glyphClass(glyphName, self) + + # quickly fetching unicode values + + def getUnicodes(self): + """Return a dictionary that maps all glyph names to lists containing + the unicode value[s] for that glyph, if any. This parses the .glif + files partially, so is a lot faster than parsing all files completely. + """ + # XXX: This method is quite wasteful if we've already parsed many .glif + # files completely. We could collect unicodes values in readGlyph, + # and only do _fetchUnicodes() for those we haven't seen yet. + unicodes = {} + for glyphName, fileName in self.contents.iteritems(): + path = os.path.join(self.dirName, fileName) + unicodes[glyphName] = _fetchUnicodes(path) + return unicodes + + # internal methods + + def _findContents(self, forceRebuild=False): + contentsPath = os.path.join(self.dirName, "contents.plist") + if forceRebuild or not os.path.exists(contentsPath): + fileNames = os.listdir(self.dirName) + fileNames = [n for n in fileNames if n.endswith(".glif")] + contents = {} + for n in fileNames: + glyphPath = os.path.join(self.dirName, n) + contents[_fetchGlyphName(glyphPath)] = n + else: + from plistlib import readPlist + contents = readPlist(contentsPath) + return contents + + def _getXMLTree(self, glyphName): + fileName = self.contents[glyphName] + path = os.path.join(self.dirName, fileName) + if not os.path.exists(path): + raise KeyError, glyphName + return _glifTreeFromFile(path) + + +def readGlyphFromString(aString, glyphObject=None, pointPen=None): + """Read .glif data from a string into a glyph object. + + The 'glyphObject' argument can be any kind of object (even None); + the readGlyphFromString() method will attempt to set the following + attributes on it: + "width" the advance with of the glyph + "unicodes" a list of unicode values for this glyph + "note" a string + "lib" a dictionary containing custom data + + All attributes are optional, in two ways: + 1) An attribute *won't* be set if the .glif file doesn't + contain data for it. 'glyphObject' will have to deal + with default values itself. + 2) If setting the attribute fails with an AttributeError + (for example if the 'glyphObject' attribute is read- + only), readGlyphFromString() will not propagate that + exception, but ignore that attribute. + + To retrieve outline information, you need to pass an object + conforming to the PointPen protocol as the 'pointPen' argument. + This argument may be None if you don't need the outline data. + """ + tree = _glifTreeFromFile(StringIO(aString)) + _readGlyphFromTree(tree, glyphObject, pointPen) + + +def writeGlyphToString(glyphName, glyphObject=None, drawPointsFunc=None, writer=None): + """Return .glif data for a glyph as a UTF-8 encoded string. + The 'glyphObject' argument can be any kind of object (even None); + the writeGlyphToString() method will attempt to get the following + attributes from it: + "width" the advance with of the glyph + "unicodes" a list of unicode values for this glyph + "note" a string + "lib" a dictionary containing custom data + + All attributes are optional: if 'glyphObject' doesn't + have the attribute, it will simply be skipped. + + To write outline data to the .glif file, writeGlyphToString() needs + a function (any callable object actually) that will take one + argument: an object that conforms to the PointPen protocol. + The function will be called by writeGlyphToString(); it has to call the + proper PointPen methods to transfer the outline to the .glif file. + """ + if writer is None: + try: + from xmlWriter import XMLWriter + except ImportError: + # try the other location + from fontTools.misc.xmlWriter import XMLWriter + aFile = StringIO() + writer = XMLWriter(aFile, encoding="UTF-8") + else: + aFile = None + writer.begintag("glyph", [("name", glyphName), ("format", "1")]) + writer.newline() + + width = getattr(glyphObject, "width", None) + if width is not None: + if not isinstance(width, (int, float)): + raise GlifLibError, "width attribute must be int or float" + writer.simpletag("advance", width=repr(width)) + writer.newline() + + unicodes = getattr(glyphObject, "unicodes", None) + if unicodes: + if isinstance(unicodes, int): + unicodes = [unicodes] + for code in unicodes: + if not isinstance(code, int): + raise GlifLibError, "unicode values must be int" + hexCode = hex(code)[2:].upper() + if len(hexCode) < 4: + hexCode = "0" * (4 - len(hexCode)) + hexCode + writer.simpletag("unicode", hex=hexCode) + writer.newline() + + note = getattr(glyphObject, "note", None) + if note is not None: + if not isinstance(note, (str, unicode)): + raise GlifLibError, "note attribute must be str or unicode" + note = note.encode('utf-8') + writer.begintag("note") + writer.newline() + for line in note.splitlines(): + writer.write(line.strip()) + writer.newline() + writer.endtag("note") + writer.newline() + + if drawPointsFunc is not None: + writer.begintag("outline") + writer.newline() + pen = GLIFPointPen(writer) + drawPointsFunc(pen) + writer.endtag("outline") + writer.newline() + + lib = getattr(glyphObject, "lib", None) + if lib: + from robofab.plistlib import PlistWriter + if not isinstance(lib, dict): + lib = dict(lib) + writer.begintag("lib") + writer.newline() + plistWriter = PlistWriter(writer.file, indentLevel=writer.indentlevel, + indent=writer.indentwhite, writeHeader=False) + plistWriter.writeValue(lib) + writer.endtag("lib") + writer.newline() + + writer.endtag("glyph") + writer.newline() + if aFile is not None: + return aFile.getvalue() + else: + return None + + +# misc helper functions + +def _stripGlyphXMLTree(nodes): + for element, attrs, children in nodes: + # "lib" is formatted as a plist, so we need unstripped + # character data so we can support strings with leading or + # trailing whitespace. Do strip everything else. + recursive = (element != "lib") + stripCharacterData(children, recursive=recursive) + + +def _glifTreeFromFile(aFile): + try: + tree = buildTree(aFile, stripData=False) + stripCharacterData(tree[2], recursive=False) + assert tree[0] == "glyph" + _stripGlyphXMLTree(tree[2]) + return tree + except: + print "Problem with glif file", aFile + raise + return None + + +def _relaxedSetattr(object, attr, value): + try: + setattr(object, attr, value) + except AttributeError: + pass + + +def _number(s): + """Given a numeric string, return an integer or a float, whichever + the string indicates. _number("1") will return the integer 1, + _number("1.0") will return the float 1.0. + """ + try: + n = int(s) + except ValueError: + n = float(s) + return n + + + +def _readGlyphFromTree(tree, glyphObject=None, pointPen=None): + unicodes = [] + assert tree[0] == "glyph" + formatVersion = int(tree[1].get("format", "0")) + if formatVersion not in (0, 1): + raise GlifLibError, "unsupported glif format version: %s" % formatVersion + glyphName = tree[1].get("name") + if glyphName and glyphObject is not None: + _relaxedSetattr(glyphObject, "name", glyphName) + for element, attrs, children in tree[2]: + if element == "outline": + if pointPen is not None: + if formatVersion == 0: + buildOutline_Format0(pointPen, children) + else: + buildOutline_Format1(pointPen, children) + elif glyphObject is None: + continue + elif element == "advance": + width = _number(attrs["width"]) + _relaxedSetattr(glyphObject, "width", width) + elif element == "unicode": + unicodes.append(int(attrs["hex"], 16)) + elif element == "note": + rawNote = "\n".join(children) + lines = rawNote.split("\n") + lines = [line.strip() for line in lines] + note = "\n".join(lines) + _relaxedSetattr(glyphObject, "note", note) + elif element == "lib": + from plistFromTree import readPlistFromTree + assert len(children) == 1 + lib = readPlistFromTree(children[0]) + _relaxedSetattr(glyphObject, "lib", lib) + if unicodes: + _relaxedSetattr(glyphObject, "unicodes", unicodes) + + +class _DoneParsing(Exception): pass + +def _startElementHandler(tagName, attrs): + if tagName != "glyph": + # the top level element of any .glif file must be <glyph> + raise _DoneParsing(None) + glyphName = attrs["name"] + raise _DoneParsing(glyphName) + +def _fetchGlyphName(glyphPath): + # Given a path to an existing .glif file, get the glyph name + # from the XML data. + from xml.parsers.expat import ParserCreate + + p = ParserCreate() + p.StartElementHandler = _startElementHandler + p.returns_unicode = True + f = open(glyphPath) + try: + p.ParseFile(f) + except _DoneParsing, why: + glyphName = why.args[0] + if glyphName is None: + raise ValueError, (".glif file doen't have a <glyph> top-level " + "element: %r" % glyphPath) + else: + assert 0, "it's not expected that parsing the file ends normally" + return glyphName + + +def _fetchUnicodes(glyphPath): + # Given a path to an existing .glif file, get a list of all + # unicode values from the XML data. + # NOTE: this assumes .glif files written by glifLib, since + # we simply stop parsing as soon as we see anything else than + # <glyph>, <advance> or <unicode>. glifLib always writes those + # elements in that order, before anything else. + from xml.parsers.expat import ParserCreate + + unicodes = [] + def _startElementHandler(tagName, attrs, _unicodes=unicodes): + if tagName == "unicode": + _unicodes.append(int(attrs["hex"], 16)) + elif tagName not in ("glyph", "advance"): + raise _DoneParsing() + + p = ParserCreate() + p.StartElementHandler = _startElementHandler + p.returns_unicode = True + f = open(glyphPath) + try: + p.ParseFile(f) + except _DoneParsing: + pass + return unicodes + + +def buildOutline_Format0(pen, xmlNodes): + # This reads the "old" .glif format, retroactively named "format 0", + # later formats have a "format" attribute in the <glyph> element. + for element, attrs, children in xmlNodes: + if element == "contour": + pen.beginPath() + currentSegmentType = None + for subElement, attrs, dummy in children: + if subElement != "point": + continue + x = _number(attrs["x"]) + y = _number(attrs["y"]) + pointType = attrs.get("type", "onCurve") + if pointType == "bcp": + currentSegmentType = "curve" + elif pointType == "offCurve": + currentSegmentType = "qcurve" + elif currentSegmentType is None and pointType == "onCurve": + currentSegmentType = "line" + if pointType == "onCurve": + segmentType = currentSegmentType + currentSegmentType = None + else: + segmentType = None + smooth = attrs.get("smooth") == "yes" + pen.addPoint((x, y), segmentType=segmentType, smooth=smooth) + pen.endPath() + elif element == "component": + baseGlyphName = attrs["base"] + transformation = [] + for attr, default in _transformationInfo: + value = attrs.get(attr) + if value is None: + value = default + else: + value = _number(value) + transformation.append(value) + pen.addComponent(baseGlyphName, tuple(transformation)) + elif element == "anchor": + name, x, y = attrs["name"], _number(attrs["x"]), _number(attrs["y"]) + pen.beginPath() + pen.addPoint((x, y), segmentType="move", name=name) + pen.endPath() + + +def buildOutline_Format1(pen, xmlNodes): + for element, attrs, children in xmlNodes: + if element == "contour": + pen.beginPath() + for subElement, attrs, dummy in children: + if subElement != "point": + continue + x = _number(attrs["x"]) + y = _number(attrs["y"]) + segmentType = attrs.get("type", "offcurve") + if segmentType == "offcurve": + segmentType = None + smooth = attrs.get("smooth") == "yes" + name = attrs.get("name") + pen.addPoint((x, y), segmentType=segmentType, smooth=smooth, name=name) + pen.endPath() + elif element == "component": + baseGlyphName = attrs["base"] + transformation = [] + for attr, default in _transformationInfo: + value = attrs.get(attr) + if value is None: + value = default + else: + value = _number(value) + transformation.append(value) + pen.addComponent(baseGlyphName, tuple(transformation)) + + +_transformationInfo = [ + # field name, default value + ("xScale", 1), + ("xyScale", 0), + ("yxScale", 0), + ("yScale", 1), + ("xOffset", 0), + ("yOffset", 0), +] + +class GLIFPointPen(AbstractPointPen): + + """Helper class using the PointPen protocol to write the <outline> + part of .glif files. + """ + + def __init__(self, xmlWriter): + self.writer = xmlWriter + + def beginPath(self): + self.writer.begintag("contour") + self.writer.newline() + + def endPath(self): + self.writer.endtag("contour") + self.writer.newline() + + def addPoint(self, pt, segmentType=None, smooth=None, name=None, **kwargs): + attrs = [] + if pt is not None: + for coord in pt: + if not isinstance(coord, (int, float)): + raise GlifLibError, "coordinates must be int or float" + attrs.append(("x", repr(pt[0]))) + attrs.append(("y", repr(pt[1]))) + if segmentType is not None: + attrs.append(("type", segmentType)) + if smooth: + attrs.append(("smooth", "yes")) + if name is not None: + attrs.append(("name", name)) + self.writer.simpletag("point", attrs) + self.writer.newline() + + def addComponent(self, glyphName, transformation): + attrs = [("base", glyphName)] + for (attr, default), value in zip(_transformationInfo, transformation): + if not isinstance(value, (int, float)): + raise GlifLibError, "transformation values must be int or float" + if value != default: + attrs.append((attr, repr(value))) + self.writer.simpletag("component", attrs) + self.writer.newline() + + +if __name__ == "__main__": + from pprint import pprint + from robofab.pens.pointPen import PrintingPointPen + class TestGlyph: pass + gs = GlyphSet(".") + def drawPoints(pen): + pen.beginPath() + pen.addPoint((100, 200), name="foo") + pen.addPoint((200, 250), segmentType="curve", smooth=True) + pen.endPath() + pen.addComponent("a", (1, 0, 0, 1, 20, 30)) + glyph = TestGlyph() + glyph.width = 120 + glyph.unicodes = [1, 2, 3, 43215, 66666] + glyph.lib = {"a": "b", "c": [1, 2, 3, True]} + glyph.note = " hallo! " + if 0: + gs.writeGlyph("a", glyph, drawPoints) + g2 = TestGlyph() + gs.readGlyph("a", g2, PrintingPointPen()) + pprint(g2.__dict__) + else: + s = writeGlyphToString("a", glyph, drawPoints) + print s + g2 = TestGlyph() + readGlyphFromString(s, g2, PrintingPointPen()) + pprint(g2.__dict__) + |