|
""" This is a python interface to Adobe Font Metrics Files. Although a number of other python implementations exist, and may be more complete than this, it was decided not to go with them because they were either:
1) copyrighted or used a non-BSD compatible license
2) had too many dependencies and a free standing lib was needed
3) Did more than needed and it was easier to write afresh rather than figure out how to get just what was needed.
It is pretty easy to use, and requires only built-in python libs:
>>> from matplotlib import rcParams >>> import os.path >>> afm_fname = os.path.join(rcParams['datapath'], ... 'fonts', 'afm', 'ptmr8a.afm') >>> >>> from matplotlib.afm import AFM >>> with open(afm_fname, 'rb') as fh: ... afm = AFM(fh) >>> afm.string_width_height('What the heck?') (6220.0, 694) >>> afm.get_fontname() 'Times-Roman' >>> afm.get_kern_dist('A', 'f') 0 >>> afm.get_kern_dist('A', 'y') -92.0 >>> afm.get_bbox_char('!') [130, -9, 238, 676]
As in the Adobe Font Metrics File Format Specification, all dimensions are given in units of 1/1000 of the scale factor (point size) of the font being used. """
# some afm files have floats where we are expecting ints -- there is # probably a better way to handle this (support floats, round rather # than truncate). But I don't know what the best approach is now and # this change to _to_int should at least prevent mpl from crashing on # these JDH (2009-11-06)
else:
""" Check if the file at least looks like AFM. If not, raise :exc:`RuntimeError`. """
# Remember the file position in case the caller wants to # do something else with the file. finally:
# AFM spec, Section 4: The StartFontMetrics keyword [followed by a # version number] must be the first line in the file, and the # EndFontMetrics keyword must be the last non-empty line in the # file. We just check the first line. raise RuntimeError('Not an AFM file')
""" Reads the font metrics header (up to the char metrics) and returns a dictionary mapping *key* to *val*. *val* will be converted to the appropriate python type as necessary; e.g.:
* 'False'->False * '0'->0 * '-168 -218 1000 898'-> [-168, -218, 1000, 898]
Dictionary keys are
StartFontMetrics, FontName, FullName, FamilyName, Weight, ItalicAngle, IsFixedPitch, FontBBox, UnderlinePosition, UnderlineThickness, Version, Notice, EncodingScheme, CapHeight, XHeight, Ascender, Descender, StartCharMetrics
""" b'StartFontMetrics': _to_float, b'FontName': _to_str, b'FullName': _to_str, b'FamilyName': _to_str, b'Weight': _to_str, b'ItalicAngle': _to_float, b'IsFixedPitch': _to_bool, b'FontBBox': _to_list_of_ints, b'UnderlinePosition': _to_int, b'UnderlineThickness': _to_int, b'Version': _to_str, b'Notice': _to_str, b'EncodingScheme': _to_str, b'CapHeight': _to_float, # Is the second version a mistake, or b'Capheight': _to_float, # do some AFM files contain 'Capheight'? -JKS b'XHeight': _to_float, b'Ascender': _to_float, b'Descender': _to_float, b'StdHW': _to_float, b'StdVW': _to_float, b'StartCharMetrics': _to_int, b'CharacterSet': _to_str, b'Characters': _to_int, }
else: val = b''
except ValueError: print('Value error parsing header in AFM:', key, val, file=sys.stderr) continue except KeyError: print('Found an unknown keyword in AFM header (was %r)' % key, file=sys.stderr) continue raise RuntimeError('Bad parse')
Represents the character metrics of a single character.
Notes ----- The fields do currently only describe a subset of character metrics information defined in the AFM standard. """ The bbox of the character (B) as a tuple (*llx*, *lly*, *urx*, *ury*)."""
""" Parse the given filehandle for character metrics information and return the information as dicts.
It is assumed that the file cursor is on the line behind 'StartCharMetrics'.
Returns ------- ascii_d : dict A mapping "ASCII num of the character" to `.CharMetrics`. name_d : dict A mapping "character name" to `.CharMetrics`.
Notes ----- This function is incomplete per the standard, but thus far parses all the sample afm files tried. """
# We are defensively letting values be utf8. The spec requires # ascii, but there are non-compliant fonts in circulation # Split the metric line into a dictionary, keyed by metric identifiers # There may be other metrics present, but only these are needed raise RuntimeError('Bad char metrics line: %s' % line) # Workaround: If the character name is 'Euro', give it the # corresponding character code, according to WinAnsiEncoding (see PDF # Reference). raise RuntimeError('Bad parse')
""" Return a kern pairs dictionary; keys are (*char1*, *char2*) tuples and values are the kern pair value. For example, a kern pairs line like ``KPX A y -50``
will be represented as::
d[ ('A', 'y') ] = -50
"""
raise RuntimeError('Bad start of kern pairs data: %s' % line)
raise RuntimeError('Bad kern pairs line: %s' % line) raise RuntimeError('Bad kern pairs parse')
Represents the information on a composite element of a composite char."""
""" Parse the given filehandle for composites information return them as a dict.
It is assumed that the file cursor is on the line behind 'StartComposites'.
Returns ------- composites : dict A dict mapping composite character names to a parts list. The parts list is a list of `.CompositePart` entries describing the parts of the composite.
Example ------- A composite definition line::
CC Aacute 2 ; PCC A 0 0 ; PCC acute 160 170 ;
will be represented as::
composites['Aacute'] = [CompositePart(name='A', dx=0, dy=0), CompositePart(name='acute', dx=160, dy=170)]
""" continue
raise RuntimeError('Bad composites parse')
""" Parse the optional fields for kern pair data and composites.
Returns ------- kern_data : dict A dict containing kerning information. May be empty. See `._parse_kern_pairs`. composites : dict A dict containing composite information. May be empty. See `._parse_composites`. """ b'StartKernData': _parse_kern_pairs, b'StartComposites': _parse_composites, }
b'StartComposites': {}} continue
def parse_afm(fh): return _parse_afm(fh)
""" Parse the Adobe Font Metrics file in file handle *fh*.
Returns ------- header : dict A header dict. See :func:`_parse_header`. cmetrics_by_ascii : dict From :func:`_parse_char_metrics`. cmetrics_by_name : dict From :func:`_parse_char_metrics`. kernpairs : dict From :func:`_parse_kern_pairs`. composites : dict From :func:`_parse_composites`
"""
"""Parse the AFM file in file object *fh*.""" self._metrics, self._metrics_by_name, self._kern, self._composite) = _parse_afm(fh)
if not isord: c = ord(c) return self._metrics[c].bbox
""" Return the string width (including kerning) and string height as a (*w*, *h*) tuple. """ if not len(s): return 0, 0 total_width = 0 namelast = None miny = 1e9 maxy = 0 for c in s: if c == '\n': continue wx, name, bbox = self._metrics[ord(c)]
total_width += wx + self._kern.get((namelast, name), 0) l, b, w, h = bbox miny = min(miny, b) maxy = max(maxy, b + h)
namelast = name
return total_width, maxy - miny
"""Return the string bounding box and the maximal descent.""" if not len(s): return 0, 0, 0, 0, 0 total_width = 0 namelast = None miny = 1e9 maxy = 0 left = 0 if not isinstance(s, str): s = _to_str(s) for c in s: if c == '\n': continue name = uni2type1.get(ord(c), 'question') try: wx, _, bbox = self._metrics_by_name[name] except KeyError: name = 'question' wx, _, bbox = self._metrics_by_name[name] total_width += wx + self._kern.get((namelast, name), 0) l, b, w, h = bbox left = min(left, l) miny = min(miny, b) maxy = max(maxy, b + h)
namelast = name
return left, miny, total_width, maxy - miny, -miny
"""Return the string bounding box.""" return self.get_str_bbox_and_descent(s)[:4]
"""Get the name of the character, i.e., ';' is 'semicolon'.""" if not isord: c = ord(c) return self._metrics[c].name
""" Get the width of the character from the character metric WX field. """ if not isord: c = ord(c) return self._metrics[c].width
"""Get the width of the character from a type1 character name.""" return self._metrics_by_name[name].width
"""Get the bounding box (ink) height of character *c* (space is 0).""" if not isord: c = ord(c) return self._metrics[c].bbox[-1]
""" Return the kerning pair distance (possibly 0) for chars *c1* and *c2*. """ name1, name2 = self.get_name_char(c1), self.get_name_char(c2) return self.get_kern_dist_from_name(name1, name2)
""" Return the kerning pair distance (possibly 0) for chars *name1* and *name2*. """ return self._kern.get((name1, name2), 0)
"""Return the font name, e.g., 'Times-Roman'."""
"""Return the font full name, e.g., 'Times-Roman'.""" name = self._header.get(b'FullName') if name is None: # use FontName as a substitute name = self._header[b'FontName'] return name
"""Return the font family name, e.g., 'Times'."""
# FamilyName not specified so we'll make a guess name = self.get_fullname() extras = (r'(?i)([ -](regular|plain|italic|oblique|bold|semibold|' r'light|ultralight|extra|condensed))+$') return re.sub(extras, '', name)
def family_name(self): """The font family name, e.g., 'Times'.""" return self.get_familyname()
"""Return the font weight, e.g., 'Bold' or 'Roman'."""
"""Return the fontangle as float."""
"""Return the cap height as float.""" return self._header[b'CapHeight']
"""Return the xheight as float.""" return self._header[b'XHeight']
"""Return the underline thickness as float.""" return self._header[b'UnderlineThickness']
""" Return the standard horizontal stem width as float, or *None* if not specified in AFM file. """ return self._header.get(b'StdHW', None)
""" Return the standard vertical stem width as float, or *None* if not specified in AFM file. """ return self._header.get(b'StdVW', None) |