aboutsummaryrefslogtreecommitdiff
path: root/freetype/src/type1/t1parse.c
blob: 093a1bd3b01a2d22b23eddf4b01723ef12678cb3 (plain)
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
/***************************************************************************/
/*                                                                         */
/*  t1parse.c                                                              */
/*                                                                         */
/*    Type 1 parser (body).                                                */
/*                                                                         */
/*  Copyright 1996-2005, 2008, 2009, 2012-2014 by                          */
/*  David Turner, Robert Wilhelm, and Werner Lemberg.                      */
/*                                                                         */
/*  This file is part of the FreeType project, and may only be used,       */
/*  modified, and distributed under the terms of the FreeType project      */
/*  license, LICENSE.TXT.  By continuing to use, modify, or distribute     */
/*  this file you indicate that you have read the license and              */
/*  understand and accept it fully.                                        */
/*                                                                         */
/***************************************************************************/


  /*************************************************************************/
  /*                                                                       */
  /* The Type 1 parser is in charge of the following:                      */
  /*                                                                       */
  /*  - provide an implementation of a growing sequence of objects called  */
  /*    a `T1_Table' (used to build various tables needed by the loader).  */
  /*                                                                       */
  /*  - opening .pfb and .pfa files to extract their top-level and private */
  /*    dictionaries.                                                      */
  /*                                                                       */
  /*  - read numbers, arrays & strings from any dictionary.                */
  /*                                                                       */
  /* See `t1load.c' to see how data is loaded from the font file.          */
  /*                                                                       */
  /*************************************************************************/


#include <ft2build.h>
#include <internal/ftdebug.h>
#include <internal/ftstream.h>
#include <internal/psaux.h>

#include "t1parse.h"

#include "t1errors.h"


  /*************************************************************************/
  /*                                                                       */
  /* The macro FT_COMPONENT is used in trace mode.  It is an implicit      */
  /* parameter of the FT_TRACE() and FT_ERROR() macros, used to print/log  */
  /* messages during execution.                                            */
  /*                                                                       */
#undef  FT_COMPONENT
#define FT_COMPONENT  trace_t1parse


  /*************************************************************************/
  /*************************************************************************/
  /*************************************************************************/
  /*****                                                               *****/
  /*****                   INPUT STREAM PARSER                         *****/
  /*****                                                               *****/
  /*************************************************************************/
  /*************************************************************************/
  /*************************************************************************/


  /* see Adobe Technical Note 5040.Download_Fonts.pdf */

  static FT_Error
  read_pfb_tag( FT_Stream   stream,
                FT_UShort  *atag,
                FT_ULong   *asize )
  {
    FT_Error   error;
    FT_UShort  tag;
    FT_ULong   size;


    *atag  = 0;
    *asize = 0;

    if ( !FT_READ_USHORT( tag ) )
    {
      if ( tag == 0x8001U || tag == 0x8002U )
      {
        if ( !FT_READ_ULONG_LE( size ) )
          *asize = size;
      }

      *atag = tag;
    }

    return error;
  }


  static FT_Error
  check_type1_format( FT_Stream    stream,
                      const char*  header_string,
                      size_t       header_length )
  {
    FT_Error   error;
    FT_UShort  tag;
    FT_ULong   dummy;


    if ( FT_STREAM_SEEK( 0 ) )
      goto Exit;

    error = read_pfb_tag( stream, &tag, &dummy );
    if ( error )
      goto Exit;

    /* We assume that the first segment in a PFB is always encoded as   */
    /* text.  This might be wrong (and the specification doesn't insist */
    /* on that), but we have never seen a counterexample.               */
    if ( tag != 0x8001U && FT_STREAM_SEEK( 0 ) )
      goto Exit;

    if ( !FT_FRAME_ENTER( header_length ) )
    {
      error = FT_Err_Ok;

      if ( ft_memcmp( stream->cursor, header_string, header_length ) != 0 )
        error = FT_THROW( Unknown_File_Format );

      FT_FRAME_EXIT();
    }

  Exit:
    return error;
  }


  FT_LOCAL_DEF( FT_Error )
  T1_New_Parser( T1_Parser      parser,
                 FT_Stream      stream,
                 FT_Memory      memory,
                 PSAux_Service  psaux )
  {
    FT_Error   error;
    FT_UShort  tag;
    FT_ULong   size;


    psaux->ps_parser_funcs->init( &parser->root, 0, 0, memory );

    parser->stream       = stream;
    parser->base_len     = 0;
    parser->base_dict    = 0;
    parser->private_len  = 0;
    parser->private_dict = 0;
    parser->in_pfb       = 0;
    parser->in_memory    = 0;
    parser->single_block = 0;

    /* check the header format */
    error = check_type1_format( stream, "%!PS-AdobeFont", 14 );
    if ( error )
    {
      if ( FT_ERR_NEQ( error, Unknown_File_Format ) )
        goto Exit;

      error = check_type1_format( stream, "%!FontType", 10 );
      if ( error )
      {
        FT_TRACE2(( "  not a Type 1 font\n" ));
        goto Exit;
      }
    }

    /******************************************************************/
    /*                                                                */
    /* Here a short summary of what is going on:                      */
    /*                                                                */
    /*   When creating a new Type 1 parser, we try to locate and load */
    /*   the base dictionary if this is possible (i.e., for PFB       */
    /*   files).  Otherwise, we load the whole font into memory.      */
    /*                                                                */
    /*   When `loading' the base dictionary, we only setup pointers   */
    /*   in the case of a memory-based stream.  Otherwise, we         */
    /*   allocate and load the base dictionary in it.                 */
    /*                                                                */
    /*   parser->in_pfb is set if we are in a binary (`.pfb') font.   */
    /*   parser->in_memory is set if we have a memory stream.         */
    /*                                                                */

    /* try to compute the size of the base dictionary;     */
    /* look for a Postscript binary file tag, i.e., 0x8001 */
    if ( FT_STREAM_SEEK( 0L ) )
      goto Exit;

    error = read_pfb_tag( stream, &tag, &size );
    if ( error )
      goto Exit;

    if ( tag != 0x8001U )
    {
      /* assume that this is a PFA file for now; an error will */
      /* be produced later when more things are checked        */
      if ( FT_STREAM_SEEK( 0L ) )
        goto Exit;
      size = stream->size;
    }
    else
      parser->in_pfb = 1;

    /* now, try to load `size' bytes of the `base' dictionary we */
    /* found previously                                          */

    /* if it is a memory-based resource, set up pointers */
    if ( !stream->read )
    {
      parser->base_dict = (FT_Byte*)stream->base + stream->pos;
      parser->base_len  = size;
      parser->in_memory = 1;

      /* check that the `size' field is valid */
      if ( FT_STREAM_SKIP( size ) )
        goto Exit;
    }
    else
    {
      /* read segment in memory -- this is clumsy, but so does the format */
      if ( FT_ALLOC( parser->base_dict, size )       ||
           FT_STREAM_READ( parser->base_dict, size ) )
        goto Exit;
      parser->base_len = size;
    }

    parser->root.base   = parser->base_dict;
    parser->root.cursor = parser->base_dict;
    parser->root.limit  = parser->root.cursor + parser->base_len;

  Exit:
    if ( error && !parser->in_memory )
      FT_FREE( parser->base_dict );

    return error;
  }


  FT_LOCAL_DEF( void )
  T1_Finalize_Parser( T1_Parser  parser )
  {
    FT_Memory  memory = parser->root.memory;


    /* always free the private dictionary */
    FT_FREE( parser->private_dict );

    /* free the base dictionary only when we have a disk stream */
    if ( !parser->in_memory )
      FT_FREE( parser->base_dict );

    parser->root.funcs.done( &parser->root );
  }


  FT_LOCAL_DEF( FT_Error )
  T1_Get_Private_Dict( T1_Parser      parser,
                       PSAux_Service  psaux )
  {
    FT_Stream  stream = parser->stream;
    FT_Memory  memory = parser->root.memory;
    FT_Error   error  = FT_Err_Ok;
    FT_ULong   size;


    if ( parser->in_pfb )
    {
      /* in the case of the PFB format, the private dictionary can be  */
      /* made of several segments.  We thus first read the number of   */
      /* segments to compute the total size of the private dictionary  */
      /* then re-read them into memory.                                */
      FT_Long    start_pos = FT_STREAM_POS();
      FT_UShort  tag;


      parser->private_len = 0;
      for (;;)
      {
        error = read_pfb_tag( stream, &tag, &size );
        if ( error )
          goto Fail;

        if ( tag != 0x8002U )
          break;

        parser->private_len += size;

        if ( FT_STREAM_SKIP( size ) )
          goto Fail;
      }

      /* Check that we have a private dictionary there */
      /* and allocate private dictionary buffer        */
      if ( parser->private_len == 0 )
      {
        FT_ERROR(( "T1_Get_Private_Dict:"
                   " invalid private dictionary section\n" ));
        error = FT_THROW( Invalid_File_Format );
        goto Fail;
      }

      if ( FT_STREAM_SEEK( start_pos )                           ||
           FT_ALLOC( parser->private_dict, parser->private_len ) )
        goto Fail;

      parser->private_len = 0;
      for (;;)
      {
        error = read_pfb_tag( stream, &tag, &size );
        if ( error || tag != 0x8002U )
        {
          error = FT_Err_Ok;
          break;
        }

        if ( FT_STREAM_READ( parser->private_dict + parser->private_len,
                             size ) )
          goto Fail;

        parser->private_len += size;
      }
    }
    else
    {
      /* We have already `loaded' the whole PFA font file into memory; */
      /* if this is a memory resource, allocate a new block to hold    */
      /* the private dict.  Otherwise, simply overwrite into the base  */
      /* dictionary block in the heap.                                 */

      /* first of all, look at the `eexec' keyword */
      FT_Byte*    cur   = parser->base_dict;
      FT_Byte*    limit = cur + parser->base_len;
      FT_Byte     c;
      FT_Pointer  pos_lf;
      FT_Bool     test_cr;


    Again:
      for (;;)
      {
        c = cur[0];
        if ( c == 'e' && cur + 9 < limit )  /* 9 = 5 letters for `eexec' + */
                                            /* whitespace + 4 chars        */
        {
          if ( cur[1] == 'e' &&
               cur[2] == 'x' &&
               cur[3] == 'e' &&
               cur[4] == 'c' )
            break;
        }
        cur++;
        if ( cur >= limit )
        {
          FT_ERROR(( "T1_Get_Private_Dict:"
                     " could not find `eexec' keyword\n" ));
          error = FT_THROW( Invalid_File_Format );
          goto Exit;
        }
      }

      /* check whether `eexec' was real -- it could be in a comment */
      /* or string (as e.g. in u003043t.gsf from ghostscript)       */

      parser->root.cursor = parser->base_dict;
      /* set limit to `eexec' + whitespace + 4 characters */
      parser->root.limit  = cur + 10;

      cur   = parser->root.cursor;
      limit = parser->root.limit;

      while ( cur < limit )
      {
        if ( *cur == 'e' && ft_strncmp( (char*)cur, "eexec", 5 ) == 0 )
          goto Found;

        T1_Skip_PS_Token( parser );
        if ( parser->root.error )
          break;
        T1_Skip_Spaces  ( parser );
        cur = parser->root.cursor;
      }

      /* we haven't found the correct `eexec'; go back and continue */
      /* searching                                                  */

      cur   = limit;
      limit = parser->base_dict + parser->base_len;
      goto Again;

      /* now determine where to write the _encrypted_ binary private  */
      /* dictionary.  We overwrite the base dictionary for disk-based */
      /* resources and allocate a new block otherwise                 */

    Found:
      parser->root.limit = parser->base_dict + parser->base_len;

      T1_Skip_PS_Token( parser );
      cur   = parser->root.cursor;
      limit = parser->root.limit;

      /* According to the Type 1 spec, the first cipher byte must not be */
      /* an ASCII whitespace character code (blank, tab, carriage return */
      /* or line feed).  We have seen Type 1 fonts with two line feed    */
      /* characters...  So skip now all whitespace character codes.      */
      /*                                                                 */
      /* On the other hand, Adobe's Type 1 parser handles fonts just     */
      /* fine that are violating this limitation, so we add a heuristic  */
      /* test to stop at \r only if it is not used for EOL.              */

      pos_lf  = ft_memchr( cur, '\n', limit - cur );
      test_cr = FT_BOOL( !pos_lf                                      ||
                         pos_lf > ft_memchr( cur, '\r', limit - cur ) );

      while ( cur < limit                    &&
              ( *cur == ' '                ||
                *cur == '\t'               ||
                (test_cr && *cur == '\r' ) ||
                *cur == '\n'               ) )
        ++cur;
      if ( cur >= limit )
      {
        FT_ERROR(( "T1_Get_Private_Dict:"
                   " `eexec' not properly terminated\n" ));
        error = FT_THROW( Invalid_File_Format );
        goto Exit;
      }

      size = (FT_ULong)( parser->base_len - ( cur - parser->base_dict ) );

      if ( parser->in_memory )
      {
        /* note that we allocate one more byte to put a terminating `0' */
        if ( FT_ALLOC( parser->private_dict, size + 1 ) )
          goto Fail;
        parser->private_len = size;
      }
      else
      {
        parser->single_block = 1;
        parser->private_dict = parser->base_dict;
        parser->private_len  = size;
        parser->base_dict    = 0;
        parser->base_len     = 0;
      }

      /* now determine whether the private dictionary is encoded in binary */
      /* or hexadecimal ASCII format -- decode it accordingly              */

      /* we need to access the next 4 bytes (after the final whitespace */
      /* following the `eexec' keyword); if they all are hexadecimal    */
      /* digits, then we have a case of ASCII storage                   */

      if ( cur + 3 < limit                                &&
           ft_isxdigit( cur[0] ) && ft_isxdigit( cur[1] ) &&
           ft_isxdigit( cur[2] ) && ft_isxdigit( cur[3] ) )
      {
        /* ASCII hexadecimal encoding */
        FT_Long  len;


        parser->root.cursor = cur;
        (void)psaux->ps_parser_funcs->to_bytes( &parser->root,
                                                parser->private_dict,
                                                parser->private_len,
                                                &len,
                                                0 );
        parser->private_len = len;

        /* put a safeguard */
        parser->private_dict[len] = '\0';
      }
      else
        /* binary encoding -- copy the private dict */
        FT_MEM_MOVE( parser->private_dict, cur, size );
    }

    /* we now decrypt the encoded binary private dictionary */
    psaux->t1_decrypt( parser->private_dict, parser->private_len, 55665U );

    if ( parser->private_len < 4 )
    {
      FT_ERROR(( "T1_Get_Private_Dict:"
                 " invalid private dictionary section\n" ));
      error = FT_THROW( Invalid_File_Format );
      goto Fail;
    }

    /* replace the four random bytes at the beginning with whitespace */
    parser->private_dict[0] = ' ';
    parser->private_dict[1] = ' ';
    parser->private_dict[2] = ' ';
    parser->private_dict[3] = ' ';

    parser->root.base   = parser->private_dict;
    parser->root.cursor = parser->private_dict;
    parser->root.limit  = parser->root.cursor + parser->private_len;

  Fail:
  Exit:
    return error;
  }


/* END */