/usr/lib/python2.7/dist-packages/reportlab/pdfbase/ttfonts.py is in python-reportlab 3.0-1build1.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 | #Copyright ReportLab Europe Ltd. 2000-2012
#see license.txt for license details
__version__ = '$Id$'
__doc__="""TrueType font support
This defines classes to represent TrueType fonts. They know how to calculate
their own width and how to write themselves into PDF files. They support
subsetting and embedding and can represent all 16-bit Unicode characters.
Note on dynamic fonts
---------------------
Usually a Font in ReportLab corresponds to a fixed set of PDF objects (Font,
FontDescriptor, Encoding). But with dynamic font subsetting a single TTFont
will result in a number of Font/FontDescriptor/Encoding object sets, and the
contents of those will depend on the actual characters used for printing.
To support dynamic font subsetting a concept of "dynamic font" was introduced.
Dynamic Fonts have a _dynamicFont attribute set to 1.
Dynamic fonts have the following additional functions::
def splitString(self, text, doc):
'''Splits text into a number of chunks, each of which belongs to a
single subset. Returns a list of tuples (subset, string). Use
subset numbers with getSubsetInternalName. Doc is used to identify
a document so that different documents may have different dynamically
constructed subsets.'''
def getSubsetInternalName(self, subset, doc):
'''Returns the name of a PDF Font object corresponding to a given
subset of this dynamic font. Use this function instead of
PDFDocument.getInternalFontName.'''
You must never call PDFDocument.getInternalFontName for dynamic fonts.
If you have a traditional static font, mapping to PDF text output operators
is simple::
'%s 14 Tf (%s) Tj' % (getInternalFontName(psfontname), text)
If you have a dynamic font, use this instead::
for subset, chunk in font.splitString(text, doc):
'%s 14 Tf (%s) Tj' % (font.getSubsetInternalName(subset, doc), chunk)
(Tf is a font setting operator and Tj is a text ouput operator. You should
also escape invalid characters in Tj argument, see TextObject._formatText.
Oh, and that 14 up there is font size.)
Canvas and TextObject have special support for dynamic fonts.
"""
from struct import pack, unpack, error as structError
from reportlab.lib.utils import getBytesIO, isPy3, bytestr, isUnicode, char2int
from reportlab.pdfbase import pdfmetrics, pdfdoc
from reportlab import rl_config
from reportlab.lib.rl_accel import hex32, add32, calcChecksum, instanceStringWidthTTF
class TTFError(pdfdoc.PDFError):
"TrueType font exception"
pass
if isPy3:
def SUBSETN(n,table=bytes.maketrans(b'0123456789',b'ABCDEFGIJK')):
return bytes('%6.6d'%n,'ASCII').translate(table)
else:
import string
def SUBSETN(n,table=string.maketrans(b'0123456789',b'ABCDEFGIJK'),translate=string.translate):
return translate('%6.6d'%n,table)
#
# Helpers
#
def makeToUnicodeCMap(fontname, subset):
"""Creates a ToUnicode CMap for a given subset. See Adobe
_PDF_Reference (ISBN 0-201-75839-3) for more information."""
cmap = [
"/CIDInit /ProcSet findresource begin",
"12 dict begin",
"begincmap",
"/CIDSystemInfo",
"<< /Registry (%s)" % fontname,
"/Ordering (%s)" % fontname,
"/Supplement 0",
">> def",
"/CMapName /%s def" % fontname,
"/CMapType 2 def",
"1 begincodespacerange",
"<00> <%02X>" % (len(subset) - 1),
"endcodespacerange",
"%d beginbfchar" % len(subset)
] + ["<%02X> <%04X>" % (i,v) for i,v in enumerate(subset)] + [
"endbfchar",
"endcmap",
"CMapName currentdict /CMap defineresource pop",
"end",
"end"
]
return '\n'.join(cmap)
def splice(stream, offset, value):
"""Splices the given value into stream at the given offset and
returns the resulting stream (the original is unchanged)"""
return stream[:offset] + value + stream[offset + len(value):]
def _set_ushort(stream, offset, value):
"""Writes the given unsigned short value into stream at the given
offset and returns the resulting stream (the original is unchanged)"""
return splice(stream, offset, pack(">H", value))
#
# TrueType font handling
#
GF_ARG_1_AND_2_ARE_WORDS = 1 << 0
GF_ARGS_ARE_XY_VALUES = 1 << 1
GF_ROUND_XY_TO_GRID = 1 << 2
GF_WE_HAVE_A_SCALE = 1 << 3
GF_RESERVED = 1 << 4
GF_MORE_COMPONENTS = 1 << 5
GF_WE_HAVE_AN_X_AND_Y_SCALE = 1 << 6
GF_WE_HAVE_A_TWO_BY_TWO = 1 << 7
GF_WE_HAVE_INSTRUCTIONS = 1 << 8
GF_USE_MY_METRICS = 1 << 9
GF_OVERLAP_COMPOUND = 1 << 10
GF_SCALED_COMPONENT_OFFSET = 1 << 11
GF_UNSCALED_COMPONENT_OFFSET = 1 << 12
def TTFOpenFile(fn):
'''Opens a TTF file possibly after searching TTFSearchPath
returns (filename,file)
'''
from reportlab.lib.utils import rl_isfile, open_for_read
try:
f = open_for_read(fn,'rb')
return fn, f
except IOError:
import os
if not os.path.isabs(fn):
for D in rl_config.TTFSearchPath:
tfn = os.path.join(D,fn)
if rl_isfile(tfn):
f = open_for_read(tfn,'rb')
return tfn, f
raise TTFError('Can\'t open file "%s"' % fn)
class TTFontParser:
"Basic TTF file parser"
ttfVersions = (0x00010000,0x74727565,0x74746366)
ttcVersions = (0x00010000,0x00020000)
fileKind='TTF'
def __init__(self, file, validate=0,subfontIndex=0):
"""Loads and parses a TrueType font file. file can be a filename or a
file object. If validate is set to a false values, skips checksum
validation. This can save time, especially if the font is large.
"""
self.validate = validate
self.readFile(file)
isCollection = self.readHeader()
if isCollection:
self.readTTCHeader()
self.getSubfont(subfontIndex)
else:
if self.validate: self.checksumFile()
self.readTableDirectory()
self.subfontNameX = b''
def readTTCHeader(self):
self.ttcVersion = self.read_ulong()
self.fileKind = 'TTC'
self.ttfVersions = self.ttfVersions[:-1]
if self.ttcVersion not in self.ttcVersions:
raise TTFError('"%s" is not a %s file: can\'t read version 0x%8.8x' %(self.filename,self.fileKind,self.ttcVersion))
self.numSubfonts = self.read_ulong()
self.subfontOffsets = []
a = self.subfontOffsets.append
for i in range(self.numSubfonts):
a(self.read_ulong())
def getSubfont(self,subfontIndex):
if self.fileKind!='TTC':
raise TTFError('"%s" is not a TTC file: use this method' % (self.filename,self.fileKind))
try:
pos = self.subfontOffsets[subfontIndex]
except IndexError:
raise TTFError('TTC file "%s": bad subfontIndex %s not in [0,%d]' % (self.filename,subfontIndex,self.numSubfonts-1))
self.seek(pos)
self.readHeader()
self.readTableDirectory()
self.subfontNameX = bytestr('-'+str(subfontIndex))
def readTableDirectory(self):
try:
self.numTables = self.read_ushort()
self.searchRange = self.read_ushort()
self.entrySelector = self.read_ushort()
self.rangeShift = self.read_ushort()
# Read table directory
self.table = {}
self.tables = []
for n in range(self.numTables):
record = {}
record['tag'] = self.read_tag()
record['checksum'] = self.read_ulong()
record['offset'] = self.read_ulong()
record['length'] = self.read_ulong()
self.tables.append(record)
self.table[record['tag']] = record
except:
raise TTFError('Corrupt %s file "%s" cannot read Table Directory' % (self.fileKind, self.filename))
if self.validate: self.checksumTables()
def readHeader(self):
'''read the sfnt header at the current position'''
try:
self.version = version = self.read_ulong()
except:
raise TTFError('"%s" is not a %s file: can\'t read version' %(self.filename,self.fileKind))
if version==0x4F54544F:
raise TTFError('%s file "%s": postscript outlines are not supported'%(self.fileKind,self.filename))
if version not in self.ttfVersions:
raise TTFError('Not a TrueType font: version=0x%8.8X' % version)
return version==self.ttfVersions[-1]
def readFile(self,f):
if hasattr(f,'read'):
self.filename = '(ttf)'
else:
self.filename, f = TTFOpenFile(f)
self._ttf_data = f.read()
self._pos = 0
def checksumTables(self):
# Check the checksums for all tables
for t in self.tables:
table = self.get_chunk(t['offset'], t['length'])
checksum = calcChecksum(table)
if t['tag'] == 'head':
adjustment = unpack('>l', table[8:8+4])[0]
checksum = add32(checksum, -adjustment)
xchecksum = t['checksum']
if xchecksum != checksum:
raise TTFError('TTF file "%s": invalid checksum %s table: %s (expected %s)' % (self.filename,hex32(checksum),t['tag'],hex32(xchecksum)))
def checksumFile(self):
# Check the checksums for the whole file
checksum = calcChecksum(self._ttf_data)
if 0xB1B0AFBA!=checksum:
raise TTFError('TTF file "%s": invalid checksum %s (expected 0xB1B0AFBA) len: %d &3: %d' % (self.filename,hex32(checksum),len(self._ttf_data),(len(self._ttf_data)&3)))
def get_table_pos(self, tag):
"Returns the offset and size of a given TTF table."
offset = self.table[tag]['offset']
length = self.table[tag]['length']
return (offset, length)
def seek(self, pos):
"Moves read pointer to a given offset in file."
self._pos = pos
def skip(self, delta):
"Skip the given number of bytes."
self._pos = self._pos + delta
def seek_table(self, tag, offset_in_table = 0):
"""Moves read pointer to the given offset within a given table and
returns absolute offset of that position in the file."""
self._pos = self.get_table_pos(tag)[0] + offset_in_table
return self._pos
if isPy3:
def read_tag(self):
"Read a 4-character tag"
self._pos += 4
return str(self._ttf_data[self._pos - 4:self._pos],'utf8')
def get_chunk(self, pos, length):
"Return a chunk of raw data at given position"
return bytes(self._ttf_data[pos:pos+length])
else:
def read_tag(self):
"Read a 4-character tag"
self._pos += 4
return self._ttf_data[self._pos - 4:self._pos]
def get_chunk(self, pos, length):
"Return a chunk of raw data at given position"
return self._ttf_data[pos:pos+length]
def read_ushort(self):
"Reads an unsigned short"
self._pos += 2
return unpack('>H',self._ttf_data[self._pos-2:self._pos])[0]
def read_ulong(self):
"Reads an unsigned long"
self._pos += 4
return unpack('>L',self._ttf_data[self._pos - 4:self._pos])[0]
def read_short(self):
"Reads a signed short"
self._pos += 2
try:
return unpack('>h',self._ttf_data[self._pos-2:self._pos])[0]
except structError as error:
raise TTFError(error)
def get_ushort(self, pos):
"Return an unsigned short at given position"
return unpack('>H',self._ttf_data[pos:pos+2])[0]
def get_ulong(self, pos):
"Return an unsigned long at given position"
return unpack('>L',self._ttf_data[pos:pos+4])[0]
def get_table(self, tag):
"Return the given TTF table"
pos, length = self.get_table_pos(tag)
return self._ttf_data[pos:pos+length]
class TTFontMaker:
"Basic TTF file generator"
def __init__(self):
"Initializes the generator."
self.tables = {}
def add(self, tag, data):
"Adds a table to the TTF file."
if tag == 'head':
data = splice(data, 8, b'\0\0\0\0')
self.tables[tag] = data
def makeStream(self):
"Finishes the generation and returns the TTF file as a string"
stm = getBytesIO()
write = stm.write
tables = self.tables
numTables = len(tables)
searchRange = 1
entrySelector = 0
while searchRange * 2 <= numTables:
searchRange = searchRange * 2
entrySelector = entrySelector + 1
searchRange = searchRange * 16
rangeShift = numTables * 16 - searchRange
# Header
write(pack(">lHHHH", 0x00010000, numTables, searchRange,
entrySelector, rangeShift))
# Table directory
offset = 12 + numTables * 16
wStr = (lambda x:write(bytes(tag,'latin1'))) if isPy3 else write
tables_items = list(sorted(tables.items()))
for tag, data in tables_items:
if tag == 'head':
head_start = offset
checksum = calcChecksum(data)
wStr(tag)
write(pack(">LLL", checksum, offset, len(data)))
paddedLength = (len(data)+3)&~3
offset = offset + paddedLength
# Table data
for tag, data in tables_items:
data += b"\0\0\0"
write(data[:len(data)&~3])
checksum = calcChecksum(stm.getvalue())
checksum = add32(0xB1B0AFBA, -checksum)
stm.seek(head_start + 8)
write(pack('>L', checksum))
return stm.getvalue()
class TTFontFile(TTFontParser):
"TTF file parser and generator"
def __init__(self, file, charInfo=1, validate=0,subfontIndex=0):
"""Loads and parses a TrueType font file.
file can be a filename or a file object. If validate is set to a false
values, skips checksum validation. This can save time, especially if
the font is large. See TTFontFile.extractInfo for more information.
"""
TTFontParser.__init__(self, file, validate=validate,subfontIndex=subfontIndex)
self.extractInfo(charInfo)
def extractInfo(self, charInfo=1):
"""
Extract typographic information from the loaded font file.
The following attributes will be set::
name PostScript font name
flags Font flags
ascent Typographic ascender in 1/1000ths of a point
descent Typographic descender in 1/1000ths of a point
capHeight Cap height in 1/1000ths of a point (0 if not available)
bbox Glyph bounding box [l,t,r,b] in 1/1000ths of a point
_bbox Glyph bounding box [l,t,r,b] in unitsPerEm
unitsPerEm Glyph units per em
italicAngle Italic angle in degrees ccw
stemV stem weight in 1/1000ths of a point (approximate)
If charInfo is true, the following will also be set::
defaultWidth default glyph width in 1/1000ths of a point
charWidths dictionary of character widths for every supported UCS character
code
This will only work if the font has a Unicode cmap (platform 3,
encoding 1, format 4 or platform 0 any encoding format 4). Setting
charInfo to false avoids this requirement
"""
# name - Naming table
name_offset = self.seek_table("name")
format = self.read_ushort()
if format != 0:
raise TTFError("Unknown name table format (%d)" % format)
numRecords = self.read_ushort()
string_data_offset = name_offset + self.read_ushort()
names = {1:None,2:None,3:None,4:None,6:None}
K = list(names.keys())
nameCount = len(names)
for i in range(numRecords):
platformId = self.read_ushort()
encodingId = self.read_ushort()
languageId = self.read_ushort()
nameId = self.read_ushort()
length = self.read_ushort()
offset = self.read_ushort()
if nameId not in K: continue
N = None
if platformId == 3 and encodingId == 1 and languageId == 0x409: # Microsoft, Unicode, US English, PS Name
opos = self._pos
try:
self.seek(string_data_offset + offset)
if length % 2 != 0:
raise TTFError("PostScript name is UTF-16BE string of odd length")
length /= 2
N = []
A = N.append
while length > 0:
char = self.read_ushort()
A(chr(char))
length -= 1
N = ''.join(N)
finally:
self._pos = opos
elif platformId == 1 and encodingId == 0 and languageId == 0: # Macintosh, Roman, English, PS Name
# According to OpenType spec, if PS name exists, it must exist
# both in MS Unicode and Macintosh Roman formats. Apparently,
# you can find live TTF fonts which only have Macintosh format.
N = self.get_chunk(string_data_offset + offset, length)
if N and names[nameId]==None:
names[nameId] = N
nameCount -= 1
if nameCount==0: break
if names[6] is not None:
psName = names[6].replace(b" ", b"-") #Dinu Gherman's fix for font names with spaces
elif names[4] is not None:
psName = names[4].replace(b" ", b"-")
# Fine, one last try before we bail.
elif names[1] is not None:
psName = names[1].replace(b" ", b"-")
else:
psName = None
# Don't just assume, check for None since some shoddy fonts cause crashes here...
if not psName:
raise TTFError("Could not find PostScript font name")
for c in psName:
if char2int(c)>126 or c in b' [](){}<>/%':
raise TTFError("psName=%r contains invalid character %s" % (psName,ascii(c)))
self.name = psName
self.familyName = names[1] or psName
self.styleName = names[2] or 'Regular'
self.fullName = names[4] or psName
self.uniqueFontID = names[3] or psName
# head - Font header table
self.seek_table("head")
ver_maj, ver_min = self.read_ushort(), self.read_ushort()
if ver_maj != 1:
raise TTFError('Unknown head table version %d.%04x' % (ver_maj, ver_min))
self.fontRevision = self.read_ushort(), self.read_ushort()
self.skip(4)
magic = self.read_ulong()
if magic != 0x5F0F3CF5:
raise TTFError('Invalid head table magic %04x' % magic)
self.skip(2)
self.unitsPerEm = unitsPerEm = self.read_ushort()
scale = lambda x, unitsPerEm=unitsPerEm: x * 1000. / unitsPerEm
self.skip(16)
xMin = self.read_short()
yMin = self.read_short()
xMax = self.read_short()
yMax = self.read_short()
self.bbox = list(map(scale, [xMin, yMin, xMax, yMax]))
self.skip(3*2)
indexToLocFormat = self.read_ushort()
glyphDataFormat = self.read_ushort()
# OS/2 - OS/2 and Windows metrics table
# (needs data from head table)
if "OS/2" in self.table:
self.seek_table("OS/2")
version = self.read_ushort()
self.skip(2)
usWeightClass = self.read_ushort()
self.skip(2)
fsType = self.read_ushort()
if fsType == 0x0002 or (fsType & 0x0300) != 0:
raise TTFError('Font does not allow subsetting/embedding (%04X)' % fsType)
self.skip(58) #11*2 + 10 + 4*4 + 4 + 3*2
sTypoAscender = self.read_short()
sTypoDescender = self.read_short()
self.ascent = scale(sTypoAscender) # XXX: for some reason it needs to be multiplied by 1.24--1.28
self.descent = scale(sTypoDescender)
if version > 1:
self.skip(16) #3*2 + 2*4 + 2
sCapHeight = self.read_short()
self.capHeight = scale(sCapHeight)
else:
self.capHeight = self.ascent
else:
# Microsoft TTFs require an OS/2 table; Apple ones do not. Try to
# cope. The data is not very important anyway.
usWeightClass = 500
self.ascent = scale(yMax)
self.descent = scale(yMin)
self.capHeight = self.ascent
# There's no way to get stemV from a TTF file short of analyzing actual outline data
# This fuzzy formula is taken from pdflib sources, but we could just use 0 here
self.stemV = 50 + int((usWeightClass / 65.0) ** 2)
# post - PostScript table
# (needs data from OS/2 table)
self.seek_table("post")
ver_maj, ver_min = self.read_ushort(), self.read_ushort()
if ver_maj not in (1, 2, 3, 4):
# Adobe/MS documents 1, 2, 2.5, 3; Apple also has 4.
# From Apple docs it seems that we do not need to care
# about the exact version, so if you get this error, you can
# try to remove this check altogether.
raise TTFError('Unknown post table version %d.%04x' % (ver_maj, ver_min))
self.italicAngle = self.read_short() + self.read_ushort() / 65536.0
self.underlinePosition = self.read_short()
self.underlineThickness = self.read_short()
isFixedPitch = self.read_ulong()
self.flags = FF_SYMBOLIC # All fonts that contain characters
# outside the original Adobe character
# set are considered "symbolic".
if self.italicAngle!= 0:
self.flags = self.flags | FF_ITALIC
if usWeightClass >= 600: # FW_REGULAR == 500, FW_SEMIBOLD == 600
self.flags = self.flags | FF_FORCEBOLD
if isFixedPitch:
self.flags = self.flags | FF_FIXED
# XXX: FF_SERIF? FF_SCRIPT? FF_ALLCAP? FF_SMALLCAP?
# hhea - Horizontal header table
self.seek_table("hhea")
ver_maj, ver_min = self.read_ushort(), self.read_ushort()
if ver_maj != 1:
raise TTFError('Unknown hhea table version %d.%04x' % (ver_maj, ver_min))
self.skip(28)
metricDataFormat = self.read_ushort()
if metricDataFormat != 0:
raise TTFError('Unknown horizontal metric data format (%d)' % metricDataFormat)
numberOfHMetrics = self.read_ushort()
if numberOfHMetrics == 0:
raise TTFError('Number of horizontal metrics is 0')
# maxp - Maximum profile table
self.seek_table("maxp")
ver_maj, ver_min = self.read_ushort(), self.read_ushort()
if ver_maj != 1:
raise TTFError('Unknown maxp table version %d.%04x' % (ver_maj, ver_min))
numGlyphs = self.read_ushort()
if not charInfo:
self.charToGlyph = None
self.defaultWidth = None
self.charWidths = None
return
if glyphDataFormat != 0:
raise TTFError('Unknown glyph data format (%d)' % glyphDataFormat)
# cmap - Character to glyph index mapping table
cmap_offset = self.seek_table("cmap")
self.skip(2)
cmapTableCount = self.read_ushort()
unicode_cmap_offset = None
for n in range(cmapTableCount):
platformID = self.read_ushort()
encodingID = self.read_ushort()
offset = self.read_ulong()
if platformID == 3 and encodingID == 1: # Microsoft, Unicode
format = self.get_ushort(cmap_offset + offset)
if format == 4:
unicode_cmap_offset = cmap_offset + offset
break
elif platformID == 0: # Unicode -- assume all encodings are compatible
format = self.get_ushort(cmap_offset + offset)
if format == 4:
unicode_cmap_offset = cmap_offset + offset
break
if unicode_cmap_offset is None:
raise TTFError('Font does not have cmap for Unicode (platform 3, encoding 1, format 4 or platform 0 any encoding format 4)')
self.seek(unicode_cmap_offset + 2)
length = self.read_ushort()
limit = unicode_cmap_offset + length
self.skip(2)
segCount = int(self.read_ushort() / 2.0)
self.skip(6)
endCount = list(map(lambda x, self=self: self.read_ushort(), range(segCount)))
self.skip(2)
startCount = list(map(lambda x, self=self: self.read_ushort(), range(segCount)))
idDelta = list(map(lambda x, self=self: self.read_short(), range(segCount)))
idRangeOffset_start = self._pos
idRangeOffset = list(map(lambda x, self=self: self.read_ushort(), range(segCount)))
# Now it gets tricky.
glyphToChar = {}
charToGlyph = {}
for n in range(segCount):
for unichar in range(startCount[n], endCount[n] + 1):
if idRangeOffset[n] == 0:
glyph = (unichar + idDelta[n]) & 0xFFFF
else:
offset = (unichar - startCount[n]) * 2 + idRangeOffset[n]
offset = idRangeOffset_start + 2 * n + offset
if offset >= limit:
# workaround for broken fonts (like Thryomanes)
glyph = 0
else:
glyph = self.get_ushort(offset)
if glyph != 0:
glyph = (glyph + idDelta[n]) & 0xFFFF
charToGlyph[unichar] = glyph
if glyph in glyphToChar:
glyphToChar[glyph].append(unichar)
else:
glyphToChar[glyph] = [unichar]
self.charToGlyph = charToGlyph
# hmtx - Horizontal metrics table
# (needs data from hhea, maxp, and cmap tables)
self.seek_table("hmtx")
aw = None
self.charWidths = {}
self.hmetrics = []
for glyph in range(numberOfHMetrics):
# advance width and left side bearing. lsb is actually signed
# short, but we don't need it anyway (except for subsetting)
aw, lsb = self.read_ushort(), self.read_ushort()
self.hmetrics.append((aw, lsb))
aw = scale(aw)
if glyph == 0:
self.defaultWidth = aw
if glyph in glyphToChar:
for char in glyphToChar[glyph]:
self.charWidths[char] = aw
for glyph in range(numberOfHMetrics, numGlyphs):
# the rest of the table only lists advance left side bearings.
# so we reuse aw set by the last iteration of the previous loop
lsb = self.read_ushort()
self.hmetrics.append((aw, lsb))
if glyph in glyphToChar:
for char in glyphToChar[glyph]:
self.charWidths[char] = aw
# loca - Index to location
self.seek_table('loca')
self.glyphPos = []
if indexToLocFormat == 0:
for n in range(numGlyphs + 1):
self.glyphPos.append(self.read_ushort() << 1)
elif indexToLocFormat == 1:
for n in range(numGlyphs + 1):
self.glyphPos.append(self.read_ulong())
else:
raise TTFError('Unknown location table format (%d)' % indexToLocFormat)
# Subsetting
def makeSubset(self, subset):
"""Create a subset of a TrueType font"""
output = TTFontMaker()
# Build a mapping of glyphs in the subset to glyph numbers in
# the original font. Also build a mapping of UCS codes to
# glyph values in the new font.
# Start with 0 -> 0: "missing character"
glyphMap = [0] # new glyph index -> old glyph index
glyphSet = {0:0} # old glyph index -> new glyph index
codeToGlyph = {} # unicode -> new glyph index
for code in subset:
if code in self.charToGlyph:
originalGlyphIdx = self.charToGlyph[code]
else:
originalGlyphIdx = 0
if originalGlyphIdx not in glyphSet:
glyphSet[originalGlyphIdx] = len(glyphMap)
glyphMap.append(originalGlyphIdx)
codeToGlyph[code] = glyphSet[originalGlyphIdx]
# Also include glyphs that are parts of composite glyphs
start = self.get_table_pos('glyf')[0]
n = 0
while n < len(glyphMap):
originalGlyphIdx = glyphMap[n]
glyphPos = self.glyphPos[originalGlyphIdx]
glyphLen = self.glyphPos[originalGlyphIdx + 1] - glyphPos
n += 1
if not glyphLen: continue
self.seek(start + glyphPos)
numberOfContours = self.read_short()
if numberOfContours < 0:
# composite glyph
self.skip(8)
flags = GF_MORE_COMPONENTS
while flags & GF_MORE_COMPONENTS:
flags = self.read_ushort()
glyphIdx = self.read_ushort()
if glyphIdx not in glyphSet:
glyphSet[glyphIdx] = len(glyphMap)
glyphMap.append(glyphIdx)
if flags & GF_ARG_1_AND_2_ARE_WORDS:
self.skip(4)
else:
self.skip(2)
if flags & GF_WE_HAVE_A_SCALE:
self.skip(2)
elif flags & GF_WE_HAVE_AN_X_AND_Y_SCALE:
self.skip(4)
elif flags & GF_WE_HAVE_A_TWO_BY_TWO:
self.skip(8)
numGlyphs = n = len(glyphMap)
while n > 1 and self.hmetrics[n][0] == self.hmetrics[n - 1][0]:
n -= 1
numberOfHMetrics = n
# The following tables are simply copied from the original
for tag in ('name', 'OS/2', 'cvt ', 'fpgm', 'prep'):
try:
output.add(tag, self.get_table(tag))
except KeyError:
# Apparently some of the tables are optional (cvt, fpgm, prep).
# The lack of the required ones (name, OS/2) would have already
# been caught before.
pass
# post - PostScript
post = b"\x00\x03\x00\x00" + self.get_table('post')[4:16] + b"\x00" * 16
output.add('post', post)
# hhea - Horizontal Header
hhea = self.get_table('hhea')
hhea = _set_ushort(hhea, 34, numberOfHMetrics)
output.add('hhea', hhea)
# maxp - Maximum Profile
maxp = self.get_table('maxp')
maxp = _set_ushort(maxp, 4, numGlyphs)
output.add('maxp', maxp)
# cmap - Character to glyph mapping
# XXX maybe use format 0 if possible, not 6?
entryCount = len(subset)
length = 10 + entryCount * 2
cmap = [0, 1, # version, number of tables
1, 0, 0,12, # platform, encoding, offset (hi,lo)
6, length, 0, # format, length, language
0,
entryCount] + \
list(map(codeToGlyph.get, subset))
cmap = pack(*([">%dH" % len(cmap)] + cmap))
output.add('cmap', cmap)
# hmtx - Horizontal Metrics
hmtx = []
for n in range(numGlyphs):
originalGlyphIdx = glyphMap[n]
aw, lsb = self.hmetrics[originalGlyphIdx]
if n < numberOfHMetrics:
hmtx.append(int(aw))
hmtx.append(int(lsb))
hmtx = pack(*([">%dH" % len(hmtx)] + hmtx))
output.add('hmtx', hmtx)
# glyf - Glyph data
glyphData = self.get_table('glyf')
offsets = []
glyf = []
pos = 0
for n in range(numGlyphs):
offsets.append(pos)
originalGlyphIdx = glyphMap[n]
glyphPos = self.glyphPos[originalGlyphIdx]
glyphLen = self.glyphPos[originalGlyphIdx + 1] - glyphPos
data = glyphData[glyphPos:glyphPos+glyphLen]
# Fix references in composite glyphs
if glyphLen > 2 and unpack(">h", data[:2])[0] < 0:
# composite glyph
pos_in_glyph = 10
flags = GF_MORE_COMPONENTS
while flags & GF_MORE_COMPONENTS:
flags = unpack(">H", data[pos_in_glyph:pos_in_glyph+2])[0]
glyphIdx = unpack(">H", data[pos_in_glyph+2:pos_in_glyph+4])[0]
data = _set_ushort(data, pos_in_glyph + 2, glyphSet[glyphIdx])
pos_in_glyph = pos_in_glyph + 4
if flags & GF_ARG_1_AND_2_ARE_WORDS:
pos_in_glyph = pos_in_glyph + 4
else:
pos_in_glyph = pos_in_glyph + 2
if flags & GF_WE_HAVE_A_SCALE:
pos_in_glyph = pos_in_glyph + 2
elif flags & GF_WE_HAVE_AN_X_AND_Y_SCALE:
pos_in_glyph = pos_in_glyph + 4
elif flags & GF_WE_HAVE_A_TWO_BY_TWO:
pos_in_glyph = pos_in_glyph + 8
glyf.append(data)
pos = pos + glyphLen
if pos % 4 != 0:
padding = 4 - pos % 4
glyf.append(b'\0' * padding)
pos = pos + padding
offsets.append(pos)
output.add('glyf', b''.join(glyf))
# loca - Index to location
loca = []
if (pos + 1) >> 1 > 0xFFFF:
indexToLocFormat = 1 # long format
for offset in offsets:
loca.append(offset)
loca = pack(*([">%dL" % len(loca)] + loca))
else:
indexToLocFormat = 0 # short format
for offset in offsets:
loca.append(offset >> 1)
loca = pack(*([">%dH" % len(loca)] + loca))
output.add('loca', loca)
# head - Font header
head = self.get_table('head')
head = _set_ushort(head, 50, indexToLocFormat)
output.add('head', head)
return output.makeStream()
#
# TrueType font embedding
#
# PDF font flags (see PDF Reference Guide table 5.19)
FF_FIXED = 1 << 1-1
FF_SERIF = 1 << 2-1
FF_SYMBOLIC = 1 << 3-1
FF_SCRIPT = 1 << 4-1
FF_NONSYMBOLIC = 1 << 6-1
FF_ITALIC = 1 << 7-1
FF_ALLCAP = 1 << 17-1
FF_SMALLCAP = 1 << 18-1
FF_FORCEBOLD = 1 << 19-1
class TTFontFace(TTFontFile, pdfmetrics.TypeFace):
"""TrueType typeface.
Conceptually similar to a single byte typeface, but the glyphs are
identified by UCS character codes instead of glyph names."""
def __init__(self, filename, validate=0, subfontIndex=0):
"Loads a TrueType font from filename."
pdfmetrics.TypeFace.__init__(self, None)
TTFontFile.__init__(self, filename, validate=validate, subfontIndex=subfontIndex)
def getCharWidth(self, code):
"Returns the width of character U+<code>"
return self.charWidths.get(code, self.defaultWidth)
def addSubsetObjects(self, doc, fontname, subset):
"""Generate a TrueType font subset and add it to the PDF document.
Returns a PDFReference to the new FontDescriptor object."""
fontFile = pdfdoc.PDFStream()
fontFile.content = self.makeSubset(subset)
fontFile.dictionary['Length1'] = len(fontFile.content)
if doc.compression:
fontFile.filters = [pdfdoc.PDFZCompress]
fontFileRef = doc.Reference(fontFile, 'fontFile:%s(%s)' % (self.filename, fontname))
flags = self.flags & ~ FF_NONSYMBOLIC
flags = flags | FF_SYMBOLIC
fontDescriptor = pdfdoc.PDFDictionary({
'Type': '/FontDescriptor',
'Ascent': self.ascent,
'CapHeight': self.capHeight,
'Descent': self.descent,
'Flags': flags,
'FontBBox': pdfdoc.PDFArray(self.bbox),
'FontName': pdfdoc.PDFName(fontname),
'ItalicAngle': self.italicAngle,
'StemV': self.stemV,
'FontFile2': fontFileRef,
})
return doc.Reference(fontDescriptor, 'fontDescriptor:' + fontname)
class TTEncoding:
"""Encoding for TrueType fonts (always UTF-8).
TTEncoding does not directly participate in PDF object creation, since
we need a number of different 8-bit encodings for every generated font
subset. TTFont itself cares about that."""
def __init__(self):
self.name = "UTF-8"
class TTFont:
"""Represents a TrueType font.
Its encoding is always UTF-8.
Note: you cannot use the same TTFont object for different documents
at the same time.
Example of usage:
font = ttfonts.TTFont('PostScriptFontName', '/path/to/font.ttf')
pdfmetrics.registerFont(font)
canvas.setFont('PostScriptFontName', size)
canvas.drawString(x, y, "Some text encoded in UTF-8")
"""
class State:
namePrefix = 'F'
def __init__(self,asciiReadable=None):
self.assignments = {}
self.nextCode = 0
self.internalName = None
self.frozen = 0
if asciiReadable is None:
asciiReadable = rl_config.ttfAsciiReadable
if asciiReadable:
# Let's add the first 128 unicodes to the 0th subset, so ' '
# always has code 32 (for word spacing to work) and the ASCII
# output is readable
subset0 = list(range(128))
self.subsets = [subset0]
for n in subset0:
self.assignments[n] = n
self.nextCode = 128
else:
self.subsets = [[32]*33]
self.assignments[32] = 32
_multiByte = 1 # We want our own stringwidth
_dynamicFont = 1 # We want dynamic subsetting
def __init__(self, name, filename, validate=0, subfontIndex=0,asciiReadable=None):
"""Loads a TrueType font from filename.
If validate is set to a false values, skips checksum validation. This
can save time, especially if the font is large.
"""
self.fontName = name
self.face = TTFontFace(filename, validate=validate, subfontIndex=subfontIndex)
self.encoding = TTEncoding()
from weakref import WeakKeyDictionary
self.state = WeakKeyDictionary()
if asciiReadable is None:
asciiReadable = rl_config.ttfAsciiReadable
self._asciiReadable = asciiReadable
def stringWidth(self,text,size,encoding='utf8'):
return instanceStringWidthTTF(self,text,size,encoding)
def _assignState(self,doc,asciiReadable=None,namePrefix=None):
'''convenience function for those wishing to roll their own state properties'''
if asciiReadable is None:
asciiReadable = self._asciiReadable
try:
state = self.state[doc]
except KeyError:
state = self.state[doc] = TTFont.State(asciiReadable)
if namePrefix is not None:
state.namePrefix = namePrefix
return state
def splitString(self, text, doc, encoding='utf-8'):
"""Splits text into a number of chunks, each of which belongs to a
single subset. Returns a list of tuples (subset, string). Use subset
numbers with getSubsetInternalName. Doc is needed for distinguishing
subsets when building different documents at the same time."""
asciiReadable = self._asciiReadable
try: state = self.state[doc]
except KeyError: state = self.state[doc] = TTFont.State(asciiReadable)
curSet = -1
cur = []
results = []
if not isUnicode(text):
text = text.decode('utf-8') # encoding defaults to utf-8
assignments = state.assignments
subsets = state.subsets
for code in map(ord,text):
if code in assignments:
n = assignments[code]
else:
if state.frozen:
raise pdfdoc.PDFError("Font %s is already frozen, cannot add new character U+%04X" % (self.fontName, code))
n = state.nextCode
if n&0xFF==32:
# make code 32 always be a space character
if n!=32: subsets[n >> 8].append(32)
state.nextCode += 1
n = state.nextCode
state.nextCode += 1
assignments[code] = n
if n>32:
if not(n&0xFF): subsets.append([])
subsets[n >> 8].append(code)
else:
subsets[0][n] = code
if (n >> 8) != curSet:
if cur:
results.append((curSet,bytes(cur) if isPy3 else ''.join(chr(c) for c in cur)))
curSet = (n >> 8)
cur = []
cur.append(n & 0xFF)
if cur:
results.append((curSet,bytes(cur) if isPy3 else ''.join(chr(c) for c in cur)))
return results
def getSubsetInternalName(self, subset, doc):
"""Returns the name of a PDF Font object corresponding to a given
subset of this dynamic font. Use this function instead of
PDFDocument.getInternalFontName."""
try: state = self.state[doc]
except KeyError: state = self.state[doc] = TTFont.State(self._asciiReadable)
if subset < 0 or subset >= len(state.subsets):
raise IndexError('Subset %d does not exist in font %s' % (subset, self.fontName))
if state.internalName is None:
state.internalName = state.namePrefix +repr(len(doc.fontMapping) + 1)
doc.fontMapping[self.fontName] = '/' + state.internalName
doc.delayedFonts.append(self)
return '/%s+%d' % (state.internalName, subset)
def addObjects(self, doc):
"""Makes one or more PDF objects to be added to the document. The
caller supplies the internal name to be used (typically F1, F2, ... in
sequence).
This method creates a number of Font and FontDescriptor objects. Every
FontDescriptor is a (no more than) 256 character subset of the original
TrueType font."""
try: state = self.state[doc]
except KeyError: state = self.state[doc] = TTFont.State(self._asciiReadable)
state.frozen = 1
for n,subset in enumerate(state.subsets):
internalName = self.getSubsetInternalName(n, doc)[1:]
baseFontName = (b''.join((SUBSETN(n),b'+',self.face.name,self.face.subfontNameX))).decode('pdfdoc')
pdfFont = pdfdoc.PDFTrueTypeFont()
pdfFont.__Comment__ = 'Font %s subset %d' % (self.fontName, n)
pdfFont.Name = internalName
pdfFont.BaseFont = baseFontName
pdfFont.FirstChar = 0
pdfFont.LastChar = len(subset) - 1
widths = list(map(self.face.getCharWidth, subset))
pdfFont.Widths = pdfdoc.PDFArray(widths)
cmapStream = pdfdoc.PDFStream()
cmapStream.content = makeToUnicodeCMap(baseFontName, subset)
if doc.compression:
cmapStream.filters = [pdfdoc.PDFZCompress]
pdfFont.ToUnicode = doc.Reference(cmapStream, 'toUnicodeCMap:' + baseFontName)
pdfFont.FontDescriptor = self.face.addSubsetObjects(doc, baseFontName, subset)
# link it in
ref = doc.Reference(pdfFont, internalName)
fontDict = doc.idToObject['BasicFonts'].dict
fontDict[internalName] = pdfFont
del self.state[doc]
|