reportlab: comparison src/reportlab/pdfbase/pdfmetrics.py

equal deleted inserted replaced

-:29c11b905751
+:99aa837b6703
 a registry of Font, TypeFace and Encoding objects.  Ideally these
 would be pre-loaded, but due to a nasty circularity problem we
 trap attempts to access them and do it on first access.
 """
 import string, os, sys
-from types import StringType, ListType, TupleType
 from reportlab.pdfbase import _fontdata
 from reportlab.lib.logger import warnOnce
-from reportlab.lib.utils import rl_isfile, rl_glob, rl_isdir, open_and_read, open_and_readlines, findInPaths
+from reportlab.lib.utils import rl_isfile, rl_glob, rl_isdir, open_and_read, open_and_readlines, findInPaths, isSeqType, isStrType, isUnicodeType, isPython3
 from reportlab.rl_config import defaultEncoding, T1SearchPath
 from . import rl_codecs
-_notdefChar = chr(110)
+_notdefChar = b'n'
 rl_codecs.RL_Codecs.register()
 standardFonts = _fontdata.standardFonts
 standardEncodings = _fontdata.standardEncodings
 _encodings = {}
 _fonts = {}
 def _py_unicode2T1(utext,fonts):
 '''return a list of (font,string) pairs representing the unicode text'''
-#print 'unicode2t1(%s, %s): %s' % (utext, fonts, type(utext))
-#if type(utext)
 R = []
 font, fonts = fonts[0], fonts[1:]
 enc = font.encName
 if 'UCS-2' in enc:
 enc = 'UTF16'
 while utext:
 try:
-R.append((font,utext.encode(enc)))
+if isUnicodeType(utext):
+s = utext.encode(enc)
+else:
+s = utext
+R.append((font,s))
 break
 except UnicodeEncodeError as e:
 i0, il = e.args[2:4]
 if i0:
 R.append((font,utext[:i0].encode(enc)))
 order."""
 lines = open_and_readlines(afmFileName, 'r')
 if len(lines)<=1:
 #likely to be a MAC file
-if lines: lines = string.split(lines[0],'\r')
+if lines: lines = lines[0].split('\r')
 if len(lines)<=1:
 raise ValueError('AFM file %s hasn\'t enough data' % afmFileName)
 topLevel = {}
 glyphLevel = []
-lines = [l for l in map(string.strip, lines) if not l.lower().startswith('comment')]
+lines = [l.strip() for l in lines]
+lines = [l for l in lines if not l.lower().startswith('comment')]
 #pass 1 - get the widths
 inMetrics = 0  # os 'TOP', or 'CHARMETRICS'
 for line in lines:
 if line[0:16] == 'StartCharMetrics':
 inMetrics = 1
 elif line[0:14] == 'EndCharMetrics':
 inMetrics = 0
 elif inMetrics:
-chunks = string.split(line, ';')
+chunks = line.split(';')
-chunks = list(map(string.strip, chunks))
+chunks = [chunk.strip() for chunk in chunks]
 cidChunk, widthChunk, nameChunk = chunks[0:3]
 # character ID
-l, r = string.split(cidChunk)
+l, r = cidChunk.split()
 assert l == 'C', 'bad line in font file %s' % line
-cid = string.atoi(r)
+cid = int(r)
 # width
-l, r = string.split(widthChunk)
+l, r = widthChunk.split()
 assert l == 'WX', 'bad line in font file %s' % line
-width = string.atoi(r)
+width = int(r)
 # name
-l, r = string.split(nameChunk)
+l, r = nameChunk.split()
 assert l == 'N', 'bad line in font file %s' % line
 name = r
 glyphLevel.append((cid, width, name))
 if line[0:16] == 'StartCharMetrics':
 inHeader = 0
 elif inHeader:
 if line[0:7] == 'Comment': pass
 try:
-left, right = string.split(line,' ',1)
+left, right = line.split(' ',1)
 except:
 raise ValueError("Header information error in afm %s: line='%s'" % (afmFileName, line))
 try:
-right = string.atoi(right)
+right = int(right)
 except:
 pass
 topLevel[left] = right
 def getFontFiles(self):
 "Info function, return list of the font files this depends on."
 return []
 def findT1File(self, ext='.pfb'):
-possible_exts = (string.lower(ext), string.upper(ext))
+possible_exts = (ext.lower(), ext.upper())
 if hasattr(self,'pfbFileName'):
 r_basename = os.path.splitext(self.pfbFileName)[0]
 for e in possible_exts:
 if rl_isfile(r_basename + e):
 return r_basename + e
 try:
 r = _fontdata.findT1File(self.name)
 except:
 afm = bruteForceSearchForAFM(self.name)
 if afm:
-if string.lower(ext) == '.pfb':
+if ext.lower() == '.pfb':
 for e in possible_exts:
 pfb = os.path.splitext(afm)[0] + e
 if rl_isfile(pfb):
 r = pfb
 else:
 r = None
-elif string.lower(ext) == '.afm':
+elif ext.lower() == '.afm':
 r = afm
 else:
 r = None
 if r is None:
 warnOnce("Can't find %s for face '%s'" % (ext, self.name))
 self.vector = _fontdata.encodings[name]
 elif base == None:
 # assume based on the usual one
 self.baseEncodingName = defaultEncoding
 self.vector = _fontdata.encodings[defaultEncoding]
-elif type(base) is StringType:
+elif isStrType(base):
 baseEnc = getEncoding(base)
 self.baseEncodingName = baseEnc.name
 self.vector = baseEnc.vector[:]
-elif type(base) in (ListType, TupleType):
+elif isSeqType(base):
 self.baseEncodingName = defaultEncoding
 self.vector = base[:]
 elif isinstance(base, Encoding):
 # accept a vector
 self.baseEncodingName = base.name
 def _py_stringWidth(self, text, size, encoding='utf8'):
 """This is the "purist" approach to width.  The practical approach
 is to use the stringWidth function, which may be swapped in for one
 written in C."""
-if not isinstance(text,str): text = text.decode(encoding)
+if not isUnicodeType(text): text = text.decode(encoding)
 return sum([sum(map(f.widths.__getitem__,list(map(ord,t)))) for f, t in unicode2T1(text,[self]+self.substitutionFonts)])*0.001*size
 stringWidth = _py_stringWidth
 def _formatWidths(self):
 "returns a pretty block in PDF Array format to aid inspection"
-text = '['
+text = b'['
 for i in range(256):
-text = text + ' ' + str(self.widths[i])
+text = text + b' ' + bytes(str(self.widths[i]),'utf8')
 if i == 255:
-text = text + ' ]'
+text = text + b' ]'
 if i % 16 == 15:
-text = text + '\n'
+text = text + b'\n'
 return text
 def addObjects(self, doc):
 """Makes and returns one or more PDF objects to be added
 to the document.  The caller supplies the internal name
 PFB_MARKER=chr(0x80)
 PFB_ASCII=chr(1)
 PFB_BINARY=chr(2)
 PFB_EOF=chr(3)
-def _pfbSegLen(p,d):
-'''compute a pfb style length from the first 4 bytes of string d'''
+if isPython3:
-return ((((ord(d[p+3])<<8)|ord(d[p+2])<<8)|ord(d[p+1]))<<8)|ord(d[p])
+def _pfbCheck(p,d,m,fn):
+if chr(d[p])!=PFB_MARKER or chr(d[p+1])!=m:
-def _pfbCheck(p,d,m,fn):
+raise ValueError('Bad pfb file\'%s\' expected chr(%d)chr(%d) at char %d, got chr(%d)chr(%d)' % (fn,ord(PFB_MARKER),ord(m),p,d[p],d[p+1]))
-if d[p]!=PFB_MARKER or d[p+1]!=m:
+if m==PFB_EOF: return
-raise ValueError('Bad pfb file\'%s\' expected chr(%d)chr(%d) at char %d, got chr(%d)chr(%d)' % (fn,ord(PFB_MARKER),ord(m),p,ord(d[p]),ord(d[p+1])))
+p = p + 2
-if m==PFB_EOF: return
+l = (((((d[p+3])<<8)|(d[p+2])<<8)|(d[p+1]))<<8)|(d[p])
-p = p + 2
+p = p + 4
-l = _pfbSegLen(p,d)
+if p+l>len(d):
-p = p + 4
+raise ValueError('Bad pfb file\'%s\' needed %d+%d bytes have only %d!' % (fn,p,l,len(d)))
-if p+l>len(d):
+return p, p+l
-raise ValueError('Bad pfb file\'%s\' needed %d+%d bytes have only %d!' % (fn,p,l,len(d)))
+else:
-return p, p+l
+def _pfbSegLen(p,d):
+'''compute a pfb style length from the first 4 bytes of string d'''
+return ((((ord(d[p+3])<<8)|ord(d[p+2])<<8)|ord(d[p+1]))<<8)|ord(d[p])
+def _pfbCheck(p,d,m,fn):
+if d[p]!=PFB_MARKER or d[p+1]!=m:
+raise ValueError('Bad pfb file\'%s\' expected chr(%d)chr(%d) at char %d, got chr(%d)chr(%d)' % (fn,ord(PFB_MARKER),ord(m),p,ord(d[p]),ord(d[p+1])))
+if m==PFB_EOF: return
+p = p + 2
+l = _pfbSegLen(p,d)
+p = p + 4
+if p+l>len(d):
+raise ValueError('Bad pfb file\'%s\' needed %d+%d bytes have only %d!' % (fn,p,l,len(d)))
+return p, p+l
 class EmbeddedType1Face(TypeFace):
 """A Type 1 font other than one of the basic 14.
 Its glyph data will be embedded in the PDF file."""
 self.italicAngle = topLevel.get('ItalicAngle', 0)
 self.stemV = topLevel.get('stemV', 0)
 self.xHeight = topLevel.get('XHeight', 1000)
 strBbox = topLevel.get('FontBBox', [0,0,1000,1000])
-tokens = string.split(strBbox)
+tokens = strBbox.split()
 self.bbox = []
 for tok in tokens:
-self.bbox.append(string.atoi(tok))
+self.bbox.append(int(tok))
 glyphWidths = {}
 for (cid, width, name) in glyphData:
 glyphWidths[name] = width
 self.glyphWidths = glyphWidths
 else:
 raise
 def findFontAndRegister(fontName):
 '''search for and register a font given its name'''
+assert type(fontName) is str
 #it might have a font-specific encoding e.g. Symbol
 # or Dingbats.  If not, take the default.
 face = getTypeFace(fontName)
 if face.requiredEncoding:
 font = Font(fontName, fontName, face.requiredEncoding)
 rawdata = open('../../rlextra/rml2pdf/doc/rml_user_guide.prep').read()
 print('rawdata length %d' % len(rawdata))
 print('test one huge string...')
 test3widths([rawdata])
 print()
-words = string.split(rawdata)
+words = rawdata.split()
 print('test %d shorter strings (average length %0.2f chars)...' % (len(words), 1.0*len(rawdata)/len(words)))
 test3widths(words)
 def test():

branch	py33
changeset 3723	99aa837b6703
parent 3721	0c93dd8ff567
child 3725	ca840494f9dd