summaryrefslogtreecommitdiffstats
path: root/src/retr.c
blob: 6a4318f8ed2b9791fed17ec472ff13d53d9932c1 (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
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
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
/* File retrieval.
   Copyright (C) 1996-2011, 2014-2015, 2018-2024 Free Software
   Foundation, Inc.

This file is part of GNU Wget.

GNU Wget is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 3 of the License, or (at
your option) any later version.

GNU Wget is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
along with Wget.  If not, see <http://www.gnu.org/licenses/>.

Additional permission under GNU GPL version 3 section 7

If you modify this program, or any covered work, by linking or
combining it with the OpenSSL project's OpenSSL library (or a
modified version of that library), containing parts covered by the
terms of the OpenSSL or SSLeay licenses, the Free Software Foundation
grants you additional permission to convey the resulting work.
Corresponding Source for a non-source form of such a combination
shall include the source code for the parts of OpenSSL used as well
as that of the covered work.  */

#include "wget.h"

#include <stdio.h>
#include <stdlib.h>
#include <unistd.h>
#include <errno.h>
#include <string.h>
#include <assert.h>
#ifdef VMS
# include <unixio.h>            /* For delete(). */
#endif

#ifdef HAVE_LIBZ
# include <zlib.h>
#endif

#ifdef HAVE_LIBPROXY
# include "proxy.h"
#endif

#include "exits.h"
#include "utils.h"
#include "retr.h"
#include "progress.h"
#include "url.h"
#include "recur.h"
#include "ftp.h"
#include "http.h"
#include "host.h"
#include "connect.h"
#include "convert.h"
#include "ptimer.h"
#include "html-url.h"
#include "iri.h"
#include "hsts.h"

/* Total size of downloaded files.  Used to enforce quota.  */
wgint total_downloaded_bytes;

/* Total download time in seconds. */
double total_download_time;

/* If non-NULL, the stream to which output should be written.  This
   stream is initialized when `-O' is used.  */
FILE *output_stream;

/* Whether output_document is a regular file we can manipulate,
   i.e. not `-' or a device file. */
bool output_stream_regular;

static struct {
  wgint chunk_bytes;
  double chunk_start;
  double sleep_adjust;
} limit_data;

static void
limit_bandwidth_reset (void)
{
  xzero (limit_data);
}

#ifdef HAVE_LIBZ
static voidpf
zalloc (voidpf opaque, unsigned int items, unsigned int size)
{
  (void) opaque;
  return (voidpf) xcalloc (items, size);
}

static void
zfree (voidpf opaque, voidpf address)
{
  (void) opaque;
  xfree (address);
}
#endif

/* Limit the bandwidth by pausing the download for an amount of time.
   BYTES is the number of bytes received from the network, and TIMER
   is the timer that started at the beginning of download.  */

static void
limit_bandwidth (wgint bytes, struct ptimer *timer)
{
  double delta_t = ptimer_read (timer) - limit_data.chunk_start;
  double expected;

  limit_data.chunk_bytes += bytes;

  /* Calculate the amount of time we expect downloading the chunk
     should take.  If in reality it took less time, sleep to
     compensate for the difference.  */
  expected = (double) limit_data.chunk_bytes / opt.limit_rate;

  if (expected > delta_t)
    {
      double slp = expected - delta_t + limit_data.sleep_adjust;
      double t0, t1;
      if (slp < 0.2)
        {
          DEBUGP (("deferring a %.2f ms sleep (%s/%.2f).\n",
                   slp * 1000, number_to_static_string (limit_data.chunk_bytes),
                   delta_t));
          return;
        }
      DEBUGP (("\nsleeping %.2f ms for %s bytes, adjust %.2f ms\n",
               slp * 1000, number_to_static_string (limit_data.chunk_bytes),
               limit_data.sleep_adjust));

      t0 = ptimer_read (timer);
      xsleep (slp);
      t1 = ptimer_measure (timer);

      /* Due to scheduling, we probably slept slightly longer (or
         shorter) than desired.  Calculate the difference between the
         desired and the actual sleep, and adjust the next sleep by
         that amount.  */
      limit_data.sleep_adjust = slp - (t1 - t0);
      /* If sleep_adjust is very large, it's likely due to suspension
         and not clock inaccuracy.  Don't enforce those.  */
      if (limit_data.sleep_adjust > 0.5)
        limit_data.sleep_adjust = 0.5;
      else if (limit_data.sleep_adjust < -0.5)
        limit_data.sleep_adjust = -0.5;
    }

  limit_data.chunk_bytes = 0;
  limit_data.chunk_start = ptimer_read (timer);
}

/* Write data in BUF to OUT.  However, if *SKIP is non-zero, skip that
   amount of data and decrease SKIP.  Increment *TOTAL by the amount
   of data written.  If OUT2 is not NULL, also write BUF to OUT2.
   In case of error writing to OUT, -2 is returned.  In case of error
   writing to OUT2, -3 is returned.  Return 1 if the whole BUF was
   skipped.  */

static int
write_data (FILE *out, FILE *out2, const char *buf, int bufsize,
            wgint *skip, wgint *written)
{
  if (out == NULL && out2 == NULL)
    return 1;

  if (skip)
    {
      if (*skip > bufsize)
        {
          *skip -= bufsize;
          return 1;
        }
      if (*skip)
        {
          buf += *skip;
          bufsize -= *skip;
          *skip = 0;
          if (bufsize == 0)
            return 1;
        }
    }

  if (out)
    fwrite (buf, 1, bufsize, out);
  if (out2)
    fwrite (buf, 1, bufsize, out2);

  if (written)
    *written += bufsize;

  /* Immediately flush the downloaded data.  This should not hinder
     performance: fast downloads will arrive in large 16K chunks
     (which stdio would write out immediately anyway), and slow
     downloads wouldn't be limited by disk speed.  */

  /* 2005-04-20 SMS.
     Perhaps it shouldn't hinder performance, but it sure does, at least
     on VMS (more than 2X).  Rather than speculate on what it should or
     shouldn't do, it might make more sense to test it.  Even better, it
     might be nice to explain what possible benefit it could offer, as
     it appears to be a clear invitation to poor performance with no
     actual justification.  (Also, why 16K?  Anyone test other values?)
  */
#ifndef __VMS
  if (out)
    fflush (out);
  if (out2)
    fflush (out2);
#endif /* ndef __VMS */

  if (out && ferror (out))
    return -2;
  else if (out2 && ferror (out2))
    return -3;

  return 0;
}

/* Read the contents of file descriptor FD until it the connection
   terminates or a read error occurs.  The data is read in portions of
   up to 16K and written to OUT as it arrives.  If opt.verbose is set,
   the progress is shown.

   TOREAD is the amount of data expected to arrive, normally only used
   by the progress gauge.

   STARTPOS is the position from which the download starts, used by
   the progress gauge.  If QTYREAD is non-NULL, the value it points to
   is incremented by the amount of data read from the network.  If
   QTYWRITTEN is non-NULL, the value it points to is incremented by
   the amount of data written to disk.  The time it took to download
   the data is stored to ELAPSED.

   If OUT2 is non-NULL, the contents is also written to OUT2.
   OUT2 will get an exact copy of the response: if this is a chunked
   response, everything -- including the chunk headers -- is written
   to OUT2.  (OUT will only get the unchunked response.)

   The function exits and returns the amount of data read.  In case of
   error while reading data, -1 is returned.  In case of error while
   writing data to OUT, -2 is returned.  In case of error while writing
   data to OUT2, -3 is returned.  */

int
fd_read_body (const char *downloaded_filename, int fd, FILE *out, wgint toread, wgint startpos,

              wgint *qtyread, wgint *qtywritten, double *elapsed, int flags,
              FILE *out2)
{
  int ret = 0;
  int dlbufsize = MAX (BUFSIZ, 64 * 1024);
  char *dlbuf = xmalloc (dlbufsize);

  struct ptimer *timer = NULL;
  double last_successful_read_tm = 0;

  /* The progress gauge, set according to the user preferences. */
  void *progress = NULL;

  /* Non-zero if the progress gauge is interactive, i.e. if it can
     continually update the display.  When true, smaller timeout
     values are used so that the gauge can update the display when
     data arrives slowly. */
  bool progress_interactive = false;

  bool exact = !!(flags & rb_read_exactly);

  /* Used only by HTTP/HTTPS chunked transfer encoding.  */
  bool chunked = flags & rb_chunked_transfer_encoding;
  wgint skip = 0;

  /* How much data we've read/written.  */
  wgint sum_read = 0;
  wgint sum_written = 0;
  wgint remaining_chunk_size = 0;

#ifdef HAVE_LIBZ
  /* try to minimize the number of calls to inflate() and write_data() per
     call to fd_read() */
  unsigned int gzbufsize = dlbufsize * 4;
  char *gzbuf = NULL;
  z_stream gzstream;

  if (flags & rb_compressed_gzip)
    {
      gzbuf = xmalloc (gzbufsize);
      gzstream.zalloc = zalloc;
      gzstream.zfree = zfree;
      gzstream.opaque = Z_NULL;
      gzstream.next_in = Z_NULL;
      gzstream.avail_in = 0;

      #define GZIP_DETECT 32 /* gzip format detection */
      #define GZIP_WINDOW 15 /* logarithmic window size (default: 15) */
      ret = inflateInit2 (&gzstream, GZIP_DETECT | GZIP_WINDOW);
      if (ret != Z_OK)
        {
          xfree (gzbuf);
          errno = (ret == Z_MEM_ERROR) ? ENOMEM : EINVAL;
          ret = -1;
          goto out;
        }
    }
#endif

  if (flags & rb_skip_startpos)
    skip = startpos;

  if (opt.show_progress)
    {
      const char *filename_progress;
      /* If we're skipping STARTPOS bytes, pass 0 as the INITIAL
         argument to progress_create because the indicator doesn't
         (yet) know about "skipping" data.  */
      wgint start = skip ? 0 : startpos;
      if (opt.dir_prefix)
        filename_progress = downloaded_filename + strlen (opt.dir_prefix) + 1;
      else
        filename_progress = downloaded_filename;
      progress = progress_create (filename_progress, start, start + toread);
      progress_interactive = progress_interactive_p (progress);
    }

  if (opt.limit_rate)
    limit_bandwidth_reset ();

  /* A timer is needed for tracking progress, for throttling, and for
     tracking elapsed time.  If either of these are requested, start
     the timer.  */
  if (progress || opt.limit_rate || elapsed)
    {
      timer = ptimer_new ();
      last_successful_read_tm = 0;
    }

  /* Use a smaller buffer for low requested bandwidths.  For example,
     with --limit-rate=2k, it doesn't make sense to slurp in 16K of
     data and then sleep for 8s.  With buffer size equal to the limit,
     we never have to sleep for more than one second.  */
  if (opt.limit_rate && opt.limit_rate < dlbufsize)
    dlbufsize = opt.limit_rate;

  /* Read from FD while there is data to read.  Normally toread==0
     means that it is unknown how much data is to arrive.  However, if
     EXACT is set, then toread==0 means what it says: that no data
     should be read.  */
  while (!exact || (sum_read < toread))
    {
      int rdsize;
      double tmout = opt.read_timeout;

      if (chunked)
        {
          if (remaining_chunk_size == 0)
            {
              char *line = fd_read_line (fd);
              char *endl;
              if (line == NULL)
                {
                  ret = -1;
                  break;
                }
              else if (out2 != NULL)
                fwrite (line, 1, strlen (line), out2);

              remaining_chunk_size = strtol (line, &endl, 16);
              xfree (line);

              if (remaining_chunk_size < 0)
                {
                  ret = -1;
                  break;
                }

              if (remaining_chunk_size == 0)
                {
                  ret = 0;
                  line = fd_read_line (fd);
                  if (line == NULL)
                    ret = -1;
                  else
                    {
                      if (out2 != NULL)
                        fwrite (line, 1, strlen (line), out2);
                      xfree (line);
                    }
                  break;
                }
            }

          rdsize = MIN (remaining_chunk_size, dlbufsize);
        }
      else
        rdsize = exact ? MIN (toread - sum_read, dlbufsize) : dlbufsize;

      if (progress_interactive)
        {
          /* For interactive progress gauges, always specify a ~1s
             timeout, so that the gauge can be updated regularly even
             when the data arrives very slowly or stalls.  */
          tmout = 0.95;
          /* avoid wrong 'interactive timeout' */
          errno = 0;
          if (opt.read_timeout)
            {
              double waittm;
              waittm = ptimer_read (timer) - last_successful_read_tm;
              if (waittm + tmout > opt.read_timeout)
                {
                  /* Don't let total idle time exceed read timeout. */
                  tmout = opt.read_timeout - waittm;
                  /* if 0 fd_read can be 'blocked read' */
                  if (tmout <= 0)
                    {
                      /* We've already exceeded the timeout. */
                      ret = -1, errno = ETIMEDOUT;
                      break;
                    }
                }
            }
        }
      ret = fd_read (fd, dlbuf, rdsize, tmout);

      if (progress_interactive && ret < 0 && errno == ETIMEDOUT)
        ret = 0;                /* interactive timeout, handled above */
      else if (ret <= 0)
        break;                  /* EOF or read error */

      if (progress || opt.limit_rate || elapsed)
        {
          ptimer_measure (timer);
          if (ret > 0)
            last_successful_read_tm = ptimer_read (timer);
        }

      if (ret > 0)
        {
          int write_res;

          sum_read += ret;

#ifdef HAVE_LIBZ
          if (gzbuf)
            {
              int err;
              int towrite;

              /* Write original data to WARC file */
              write_res = write_data (NULL, out2, dlbuf, ret, NULL, NULL);
              if (write_res < 0)
                {
                  ret = write_res;
                  goto out;
                }

              gzstream.avail_in = ret;
              gzstream.next_in = (unsigned char *) dlbuf;

              do
                {
                  gzstream.avail_out = gzbufsize;
                  gzstream.next_out = (unsigned char *) gzbuf;

                  err = inflate (&gzstream, Z_NO_FLUSH);

                  switch (err)
                    {
                    case Z_MEM_ERROR:
                      errno = ENOMEM;
                      ret = -1;
                      goto out;
                    case Z_NEED_DICT:
                    case Z_DATA_ERROR:
                      errno = EINVAL;
                      ret = -1;
                      goto out;
                    case Z_STREAM_END:
                      if (exact && sum_read != toread)
                        {
                          DEBUGP(("zlib stream ended unexpectedly after %"PRId64"/%"PRId64
                                  " bytes\n", sum_read, toread));
                        }
                    }

                  towrite = gzbufsize - gzstream.avail_out;
                  write_res = write_data (out, NULL, gzbuf, towrite, &skip,
                                          &sum_written);
                  if (write_res < 0)
                    {
                      ret = write_res;
                      goto out;
                    }
                }
              while (gzstream.avail_out == 0);
            }
          else
#endif
            {
              write_res = write_data (out, out2, dlbuf, ret, &skip,
                                      &sum_written);
              if (write_res < 0)
                {
                  ret = write_res;
                  goto out;
                }
            }

          if (chunked)
            {
              remaining_chunk_size -= ret;
              if (remaining_chunk_size == 0)
                {
                  char *line = fd_read_line (fd);
                  if (line == NULL)
                    {
                      ret = -1;
                      break;
                    }
                  else
                    {
                      if (out2 != NULL)
                        fwrite (line, 1, strlen (line), out2);
                      xfree (line);
                    }
                }
            }
        }

      if (opt.limit_rate)
        limit_bandwidth (ret, timer);

      if (progress)
        progress_update (progress, ret, ptimer_read (timer));
#ifdef WINDOWS
      if (toread > 0 && opt.show_progress)
        ws_percenttitle (100.0 *
                         (startpos + sum_read) / (startpos + toread));
#endif
    }
  if (ret < -1)
    ret = -1;

 out:
  if (progress)
    progress_finish (progress, ptimer_read (timer));

  if (timer)
    {
      if (elapsed)
        *elapsed = ptimer_read (timer);
      ptimer_destroy (timer);
    }

#ifdef HAVE_LIBZ
  if (gzbuf)
    {
      int err = inflateEnd (&gzstream);
      if (ret >= 0)
        {
          /* with compression enabled, ret must be 0 if successful */
          if (err == Z_OK)
            ret = 0;
          else
            {
              errno = EINVAL;
              ret = -1;
            }
        }
      xfree (gzbuf);

      if (gzstream.total_in != (uLong) sum_read)
        {
          DEBUGP(("zlib read size differs from raw read size (%lu/%"PRId64")\n",
                  gzstream.total_in, sum_read));
        }
    }
#endif

  if (qtyread)
    *qtyread += sum_read;
  if (qtywritten)
    *qtywritten += sum_written;

  xfree (dlbuf);

  return ret;
}

/* Read a hunk of data from FD, up until a terminator.  The hunk is
   limited by whatever the TERMINATOR callback chooses as its
   terminator.  For example, if terminator stops at newline, the hunk
   will consist of a line of data; if terminator stops at two
   newlines, it can be used to read the head of an HTTP response.
   Upon determining the boundary, the function returns the data (up to
   the terminator) in malloc-allocated storage.

   In case of read error, NULL is returned.  In case of EOF and no
   data read, NULL is returned and errno set to 0.  In case of having
   read some data, but encountering EOF before seeing the terminator,
   the data that has been read is returned, but it will (obviously)
   not contain the terminator.

   The TERMINATOR function is called with three arguments: the
   beginning of the data read so far, the beginning of the current
   block of peeked-at data, and the length of the current block.
   Depending on its needs, the function is free to choose whether to
   analyze all data or just the newly arrived data.  If TERMINATOR
   returns NULL, it means that the terminator has not been seen.
   Otherwise it should return a pointer to the charactre immediately
   following the terminator.

   The idea is to be able to read a line of input, or otherwise a hunk
   of text, such as the head of an HTTP request, without crossing the
   boundary, so that the next call to fd_read etc. reads the data
   after the hunk.  To achieve that, this function does the following:

   1. Peek at incoming data.

   2. Determine whether the peeked data, along with the previously
      read data, includes the terminator.

      2a. If yes, read the data until the end of the terminator, and
          exit.

      2b. If no, read the peeked data and goto 1.

   The function is careful to assume as little as possible about the
   implementation of peeking.  For example, every peek is followed by
   a read.  If the read returns a different amount of data, the
   process is retried until all data arrives safely.

   SIZEHINT is the buffer size sufficient to hold all the data in the
   typical case (it is used as the initial buffer size).  MAXSIZE is
   the maximum amount of memory this function is allowed to allocate,
   or 0 if no upper limit is to be enforced.

   This function should be used as a building block for other
   functions -- see fd_read_line as a simple example.  */

char *
fd_read_hunk (int fd, hunk_terminator_t terminator, long sizehint, long maxsize)
{
  long bufsize = sizehint;
  char *hunk = xmalloc (bufsize);
  int tail = 0;                 /* tail position in HUNK */

  assert (!maxsize || maxsize >= bufsize);

  while (1)
    {
      const char *end;
      int pklen, rdlen, remain;

      /* First, peek at the available data. */

      pklen = fd_peek (fd, hunk + tail, bufsize - 1 - tail, -1);
      if (pklen < 0)
        {
          xfree (hunk);
          return NULL;
        }
      end = terminator (hunk, hunk + tail, pklen);
      if (end)
        {
          /* The data contains the terminator: we'll drain the data up
             to the end of the terminator.  */
          remain = end - (hunk + tail);
          assert (remain >= 0);
          if (remain == 0)
            {
              /* No more data needs to be read. */
              hunk[tail] = '\0';
              return hunk;
            }
          if (bufsize - 1 < tail + remain)
            {
              bufsize = tail + remain + 1;
              hunk = xrealloc (hunk, bufsize);
            }
        }
      else
        /* No terminator: simply read the data we know is (or should
           be) available.  */
        remain = pklen;

      /* Now, read the data.  Note that we make no assumptions about
         how much data we'll get.  (Some TCP stacks are notorious for
         read returning less data than the previous MSG_PEEK.)  */

      rdlen = fd_read (fd, hunk + tail, remain, 0);
      if (rdlen < 0)
        {
          xfree (hunk);
          return NULL;
        }
      tail += rdlen;
      hunk[tail] = '\0';

      if (rdlen == 0)
        {
          if (tail == 0)
            {
              /* EOF without anything having been read */
              xfree (hunk);
              errno = 0;
              return NULL;
            }
          else
            /* EOF seen: return the data we've read. */
            return hunk;
        }
      if (end && rdlen == remain)
        /* The terminator was seen and the remaining data drained --
           we got what we came for.  */
        return hunk;

      /* Keep looping until all the data arrives. */

      if (tail == bufsize - 1)
        {
          /* Double the buffer size, but refuse to allocate more than
             MAXSIZE bytes.  */
          if (maxsize && bufsize >= maxsize)
            {
              xfree (hunk);
              errno = ENOMEM;
              return NULL;
            }
          bufsize <<= 1;
          if (maxsize && bufsize > maxsize)
            bufsize = maxsize;
          hunk = xrealloc (hunk, bufsize);
        }
    }
}

static const char *
line_terminator (const char *start _GL_UNUSED, const char *peeked, int peeklen)
{
  const char *p = memchr (peeked, '\n', peeklen);
  if (p)
    /* p+1 because the line must include '\n' */
    return p + 1;
  return NULL;
}

/* The maximum size of the single line we agree to accept.  This is
   not meant to impose an arbitrary limit, but to protect the user
   from Wget slurping up available memory upon encountering malicious
   or buggy server output.  Define it to 0 to remove the limit.  */
#define FD_READ_LINE_MAX 4096

/* Read one line from FD and return it.  The line is allocated using
   malloc, but is never larger than FD_READ_LINE_MAX.

   If an error occurs, or if no data can be read, NULL is returned.
   In the former case errno indicates the error condition, and in the
   latter case, errno is NULL.  */

char *
fd_read_line (int fd)
{
  return fd_read_hunk (fd, line_terminator, 128, FD_READ_LINE_MAX);
}

/* Return a printed representation of the download rate, along with
   the units appropriate for the download speed.  */

const char *
retr_rate (wgint bytes, double secs)
{
  static char res[20];
  static const char *rate_names[] = {"B/s", "KB/s", "MB/s", "GB/s", "TB/s" };
  static const char *rate_names_bits[] = {"b/s", "Kb/s", "Mb/s", "Gb/s", "Tb/s" };
  int units;

  double dlrate = calc_rate (bytes, secs, &units);
  /* Use more digits for smaller numbers (regardless of unit used),
     e.g. "1022", "247", "12.5", "2.38".  */
  snprintf (res, sizeof(res), "%.*f %s",
           dlrate >= 99.95 ? 0 : dlrate >= 9.995 ? 1 : 2,
           dlrate, !opt.report_bps ? rate_names[units]: rate_names_bits[units]);

  return res;
}

/* Calculate the download rate and trim it as appropriate for the
   speed.  Appropriate means that if rate is greater than 1K/s,
   kilobytes are used, and if rate is greater than 1MB/s, megabytes
   are used.

   UNITS is zero for B/s, one for KB/s, two for MB/s, and three for
   GB/s.  */

double
calc_rate (wgint bytes, double secs, int *units)
{
  double dlrate;
  double bibyte;

  if (!opt.report_bps)
    bibyte = 1024.0;
  else
    bibyte = 1000.0;

  if (secs == 0)
    /* If elapsed time is exactly zero, it means we're under the
       resolution of the timer.  This can easily happen on systems
       that use time() for the timer.  Since the interval lies between
       0 and the timer's resolution, assume half the resolution.  */
    secs = ptimer_resolution () / 2.0;

  dlrate = secs ? convert_to_bits (bytes) / secs : 0;
  if (dlrate < bibyte)
    *units = 0;
  else if (dlrate < (bibyte * bibyte))
    *units = 1, dlrate /= bibyte;
  else if (dlrate < (bibyte * bibyte * bibyte))
    *units = 2, dlrate /= (bibyte * bibyte);
  else if (dlrate < (bibyte * bibyte * bibyte * bibyte))
    *units = 3, dlrate /= (bibyte * bibyte * bibyte);
  else {
    *units = 4, dlrate /= (bibyte * bibyte * bibyte * bibyte);
    if (dlrate > 99.99)
		 dlrate = 99.99; // upper limit 99.99TB/s
  }

  return dlrate;
}


#define SUSPEND_METHOD do {                     \
  method_suspended = true;                      \
  saved_body_data = opt.body_data;              \
  saved_body_file_name = opt.body_file;         \
  saved_method = opt.method;                    \
  opt.body_data = NULL;                         \
  opt.body_file = NULL;                         \
  opt.method = NULL;                            \
} while (0)

#define RESTORE_METHOD do {                             \
  if (method_suspended)                                 \
    {                                                   \
      opt.body_data = saved_body_data;                  \
      opt.body_file = saved_body_file_name;             \
      opt.method = saved_method;                        \
      method_suspended = false;                         \
    }                                                   \
} while (0)

static char *getproxy (struct url *);

/* Retrieve the given URL.  Decides which loop to call -- HTTP, FTP,
   FTP, proxy, etc.  */

/* #### This function should be rewritten so it doesn't return from
   multiple points. */

uerr_t
retrieve_url (struct url * orig_parsed, const char *origurl, char **file,
              char **newloc, const char *refurl, int *dt, bool recursive,
              struct iri *iri, bool register_status)
{
  uerr_t result;
  char *url;
  bool location_changed;
  bool iri_fallbacked = 0;
  int dummy;
  char *mynewloc, *proxy;
  struct url *u = orig_parsed, *proxy_url;
  int up_error_code;            /* url parse error code */
  char *local_file = NULL;
  int redirection_count = 0;

  bool method_suspended = false;
  char *saved_body_data = NULL;
  char *saved_method = NULL;
  char *saved_body_file_name = NULL;

  /* If dt is NULL, use local storage.  */
  if (!dt)
    {
      dt = &dummy;
      dummy = 0;
    }
  url = xstrdup (origurl);
  if (newloc)
    *newloc = NULL;
  if (file)
    *file = NULL;

  if (!refurl)
    refurl = opt.referer;

 redirected:
  /* (also for IRI fallbacking) */

  result = NOCONERROR;
  mynewloc = NULL;
  xfree(local_file);
  proxy_url = NULL;

  proxy = getproxy (u);
  if (proxy)
    {
      struct iri *pi = iri_new ();
      set_uri_encoding (pi, opt.locale, true);
      pi->utf8_encode = false;

      /* Parse the proxy URL.  */
      proxy_url = url_parse (proxy, &up_error_code, pi, true);
      if (!proxy_url)
        {
          logprintf (LOG_NOTQUIET, _("Error parsing proxy URL %s: %s.\n"),
                     proxy, url_error (up_error_code));
          xfree (url);
          xfree (proxy);
          iri_free (pi);
          RESTORE_METHOD;
          result = PROXERR;
          if (orig_parsed != u)
            url_free (u);
          goto bail;
        }
      if (proxy_url->scheme != SCHEME_HTTP && proxy_url->scheme != u->scheme)
        {
          logprintf (LOG_NOTQUIET, _("Error in proxy URL %s: Must be HTTP.\n"), proxy);
          url_free (proxy_url);
          xfree (url);
          xfree (proxy);
          iri_free (pi);
          RESTORE_METHOD;
          result = PROXERR;
          if (orig_parsed != u)
            url_free (u);
          goto bail;
        }
      iri_free(pi);
      xfree (proxy);
    }

  if (u->scheme == SCHEME_HTTP
#ifdef HAVE_SSL
      || u->scheme == SCHEME_HTTPS
#endif
      || (proxy_url && proxy_url->scheme == SCHEME_HTTP))
    {
#ifdef HAVE_HSTS
#ifdef TESTING
      /* we don't link against main.o when we're testing */
      hsts_store_t hsts_store = NULL;
#else
      extern hsts_store_t hsts_store;
#endif

      if (opt.hsts && hsts_store)
	{
	  if (hsts_match (hsts_store, u))
	    logprintf (LOG_VERBOSE, "URL transformed to HTTPS due to an HSTS policy\n");
	}
#endif
      result = http_loop (u, orig_parsed, &mynewloc, &local_file, refurl, dt,
                          proxy_url, iri);
    }
  else if (u->scheme == SCHEME_FTP
#ifdef HAVE_SSL
      || u->scheme == SCHEME_FTPS
#endif
      )
    {
      /* If this is a redirection, temporarily turn off opt.ftp_glob
         and opt.recursive, both being undesirable when following
         redirects.  */
      bool oldrec = recursive, glob = opt.ftp_glob;
      if (redirection_count)
        oldrec = glob = false;

      result = ftp_loop (u, orig_parsed, &local_file, dt, proxy_url,
                         recursive, glob);
      recursive = oldrec;

      /* There is a possibility of having HTTP being redirected to
         FTP.  In these cases we must decide whether the text is HTML
         according to the suffix.  The HTML suffixes are `.html',
         `.htm' and a few others, case-insensitive.  */
      if (redirection_count && local_file && (u->scheme == SCHEME_FTP
#ifdef HAVE_SSL
          || u->scheme == SCHEME_FTPS
#endif
          ))
        {
          if (has_html_suffix_p (local_file))
            *dt |= TEXTHTML;
        }
    }

  if (proxy_url)
    {
      url_free (proxy_url);
      proxy_url = NULL;
    }

  location_changed = (result == NEWLOCATION || result == NEWLOCATION_KEEP_POST);
  if (location_changed)
    {
      char *construced_newloc;
      struct url *newloc_parsed;

      assert (mynewloc != NULL);

      xfree (local_file);

      /* The HTTP specs only allow absolute URLs to appear in
         redirects, but a ton of boneheaded webservers and CGIs out
         there break the rules and use relative URLs, and popular
         browsers are lenient about this, so wget should be too. */
      construced_newloc = uri_merge (url, mynewloc ? mynewloc : "");
      xfree (mynewloc);
      mynewloc = construced_newloc;

#ifdef ENABLE_IRI
      /* Reset UTF-8 encoding state, set the URI encoding and reset
         the content encoding. */
      iri->utf8_encode = opt.enable_iri;
      if (opt.encoding_remote)
       set_uri_encoding (iri, opt.encoding_remote, true);
      set_content_encoding (iri, NULL);
      xfree (iri->orig_url);
#endif

      /* Now, see if this new location makes sense. */
      newloc_parsed = url_parse (mynewloc, &up_error_code, iri, true);
      if (!newloc_parsed)
        {
          logprintf (LOG_NOTQUIET, "%s: %s.\n", escnonprint_uri (mynewloc),
                     url_error (up_error_code));
          if (orig_parsed != u)
            {
              url_free (u);
            }
          xfree (url);
          xfree (mynewloc);
          RESTORE_METHOD;
          goto bail;
        }

      /* Now mynewloc will become newloc_parsed->url, because if the
         Location contained relative paths like .././something, we
         don't want that propagating as url.  */
      xfree (mynewloc);
      mynewloc = xstrdup (newloc_parsed->url);

      /* Check for max. number of redirections.  */
      if (++redirection_count > opt.max_redirect)
        {
          logprintf (LOG_NOTQUIET, _("%d redirections exceeded.\n"),
                     opt.max_redirect);
          url_free (newloc_parsed);
          if (orig_parsed != u)
            {
              url_free (u);
            }
          xfree (url);
          xfree (mynewloc);
          RESTORE_METHOD;
          result = WRONGCODE;
          goto bail;
        }

      xfree (url);
      url = mynewloc;
      if (orig_parsed != u)
        {
          url_free (u);
        }
      u = newloc_parsed;

      /* If we're being redirected from POST, and we received a
         redirect code different than 307, we don't want to POST
         again.  Many requests answer POST with a redirection to an
         index page; that redirection is clearly a GET.  We "suspend"
         POST data for the duration of the redirections, and restore
         it when we're done.

         RFC2616 HTTP/1.1 introduces code 307 Temporary Redirect
         specifically to preserve the method of the request.
     */
      if (result != NEWLOCATION_KEEP_POST && !method_suspended)
        SUSPEND_METHOD;

      goto redirected;
    }
  else
    {
      xfree(mynewloc);
    }

  /* Try to not encode in UTF-8 if fetching failed */
  if (!(*dt & RETROKF) && iri->utf8_encode)
    {
      iri->utf8_encode = false;
      if (orig_parsed != u)
        {
          url_free (u);
        }
      u = url_parse (origurl, NULL, iri, true);
      if (u)
        {
          if (strcmp(u->url, orig_parsed->url))
            {
              DEBUGP (("[IRI fallbacking to non-utf8 for %s\n", quote (url)));
              xfree (url);
              url = xstrdup (u->url);
              iri_fallbacked = 1;
              goto redirected;
            }
          else
              DEBUGP (("[Needn't fallback to non-utf8 for %s\n", quote (url)));
        }
      else
          DEBUGP (("[Couldn't fallback to non-utf8 for %s\n", quote (url)));
    }

  if (local_file && u && (*dt & RETROKF || opt.content_on_error))
    {
      register_download (u->url, local_file);

      if (!opt.spider && redirection_count && 0 != strcmp (origurl, u->url))
        register_redirection (origurl, u->url);

      if (*dt & TEXTHTML)
        register_html (local_file);

      if (*dt & TEXTCSS)
        register_css (local_file);
    }

  if (file)
    *file = local_file ? local_file : NULL;
  else
    xfree (local_file);

  if (orig_parsed != u)
    url_free (u);

  if (redirection_count || iri_fallbacked)
    {
      if (newloc)
        *newloc = url;
      else
        xfree (url);
    }
  else
    {
      if (newloc)
        *newloc = NULL;
      xfree (url);
    }

  RESTORE_METHOD;

bail:
  if (register_status)
    inform_exit_status (result);

  return result;
}

/* Find the URLs in the file and call retrieve_url() for each of them.
   If HTML is true, treat the file as HTML, and construct the URLs
   accordingly.

   If opt.recursive is set, call retrieve_tree() for each file.  */

uerr_t
retrieve_from_file (const char *file, bool html, int *count)
{
  uerr_t status;
  struct urlpos *url_list, *cur_url;
  struct iri *iri = iri_new();

  char *input_file, *url_file = NULL;
  const char *url = file;

  status = RETROK;             /* Suppose everything is OK.  */
  *count = 0;                  /* Reset the URL count.  */

  /* sXXXav : Assume filename and links in the file are in the locale */
  set_uri_encoding (iri, opt.locale, true);
  set_content_encoding (iri, opt.locale);

  if (url_valid_scheme (url))
    {
      int dt,url_err;
      struct url *url_parsed = url_parse (url, &url_err, iri, true);
      if (!url_parsed)
        {
          logprintf (LOG_NOTQUIET, "%s: %s.\n", url, url_error (url_err));
          iri_free (iri);
          return URLERROR;
        }

      if (!opt.base_href)
        opt.base_href = xstrdup (url);

      status = retrieve_url (url_parsed, url, &url_file, NULL, NULL, &dt,
                             false, iri, true);
      url_free (url_parsed);

      if (!url_file || (status != RETROK))
        return status;

      if (dt & TEXTHTML)
        html = true;

#ifdef ENABLE_IRI
      /* If we have a found a content encoding, use it.
       * ( == is okay, because we're checking for identical object) */
      if (iri->content_encoding != opt.locale)
          set_uri_encoding (iri, iri->content_encoding, false);
#endif

      /* Reset UTF-8 encode status */
      iri->utf8_encode = opt.enable_iri;
      xfree (iri->orig_url);

      input_file = url_file;
    }
  else
    input_file = (char *) file;

  url_list = (html ? get_urls_html (input_file, NULL, NULL, iri)
              : get_urls_file (input_file));

  xfree (url_file);

  for (cur_url = url_list; cur_url; cur_url = cur_url->next, ++*count)
    {
      char *filename = NULL, *new_file = NULL, *proxy;
      int dt = 0;
      struct iri *tmpiri = iri_dup (iri);
      struct url *parsed_url = NULL;

      if (cur_url->ignore_when_downloading)
        continue;

      if (opt.quota && total_downloaded_bytes > opt.quota)
        {
          status = QUOTEXC;
          break;
        }

      parsed_url = url_parse (cur_url->url->url, NULL, tmpiri, true);

      proxy = getproxy (cur_url->url);
      if ((opt.recursive || opt.page_requisites)
          && ((cur_url->url->scheme != SCHEME_FTP
#ifdef HAVE_SSL
          && cur_url->url->scheme != SCHEME_FTPS
#endif
          ) || proxy))
        {
          int old_follow_ftp = opt.follow_ftp;

          /* Turn opt.follow_ftp on in case of recursive FTP retrieval */
          if (cur_url->url->scheme == SCHEME_FTP
#ifdef HAVE_SSL
              || cur_url->url->scheme == SCHEME_FTPS
#endif
              )
            opt.follow_ftp = 1;

          status = retrieve_tree (parsed_url ? parsed_url : cur_url->url,
                                  tmpiri);

          opt.follow_ftp = old_follow_ftp;
        }
      else
        status = retrieve_url (parsed_url ? parsed_url : cur_url->url,
                               cur_url->url->url, &filename,
                               &new_file, NULL, &dt, opt.recursive, tmpiri,
                               true);
      xfree (proxy);

      if (parsed_url)
          url_free (parsed_url);

      if (filename && opt.delete_after && file_exists_p (filename, NULL))
        {
          DEBUGP (("\
Removing file due to --delete-after in retrieve_from_file():\n"));
          logprintf (LOG_VERBOSE, _("Removing %s.\n"), filename);
          if (unlink (filename))
            logprintf (LOG_NOTQUIET, "Failed to unlink %s: (%d) %s\n", filename, errno, strerror (errno));
          dt &= ~RETROKF;
        }

      xfree (new_file);
      xfree (filename);
      iri_free (tmpiri);
    }

  /* Free the linked list of URL-s.  */
  free_urlpos (url_list);

  iri_free (iri);

  return status;
}

/* Print `giving up', or `retrying', depending on the impending
   action.  N1 and N2 are the attempt number and the attempt limit.  */
void
printwhat (int n1, int n2)
{
  logputs (LOG_VERBOSE, (n1 == n2) ? _("Giving up.\n\n") : _("Retrying.\n\n"));
}

/* If opt.wait or opt.waitretry are specified, and if certain
   conditions are met, sleep the appropriate number of seconds.  See
   the documentation of --wait and --waitretry for more information.

   COUNT is the count of current retrieval, beginning with 1. */

void
sleep_between_retrievals (int count)
{
  static bool first_retrieval = true;

  if (first_retrieval)
    {
      /* Don't sleep before the very first retrieval. */
      first_retrieval = false;
      return;
    }

  if (opt.waitretry && count > 1)
    {
      /* If opt.waitretry is specified and this is a retry, wait for
         COUNT-1 number of seconds, or for opt.waitretry seconds.  */
      if (count <= opt.waitretry)
        xsleep (count - 1);
      else
        xsleep (opt.waitretry);
    }
  else if (opt.wait)
    {
      if (!opt.random_wait || count > 1)
        /* If random-wait is not specified, or if we are sleeping
           between retries of the same download, sleep the fixed
           interval.  */
        xsleep (opt.wait);
      else
        {
          /* Sleep a random amount of time averaging in opt.wait
             seconds.  The sleeping amount ranges from 0.5*opt.wait to
             1.5*opt.wait.  */
          double waitsecs = (0.5 + random_float ()) * opt.wait;
          DEBUGP (("sleep_between_retrievals: avg=%f,sleep=%f\n",
                   opt.wait, waitsecs));
          xsleep (waitsecs);
        }
    }
}

/* Free the linked list of urlpos.  */
void
free_urlpos (struct urlpos *l)
{
  while (l)
    {
      struct urlpos *next = l->next;
      if (l->url)
        url_free (l->url);
      xfree (l->local_name);
      xfree (l);
      l = next;
    }
}

/* Rotate FNAME opt.backups times */
void
rotate_backups(const char *fname)
{
#ifdef __VMS
# define SEP "_"
# define AVS ";*"                       /* All-version suffix. */
# define AVSL (sizeof (AVS) - 1)
#else
# define SEP "."
# define AVSL 0
#endif
#define FILE_BUF_SIZE 1024

  /* avoid alloca() here */
  char from[FILE_BUF_SIZE], to[FILE_BUF_SIZE];
  struct stat sb;
  bool overflow;
  int i;

  if (stat (fname, &sb) == 0)
    if (S_ISREG (sb.st_mode) == 0)
      return;

  for (i = opt.backups; i > 1; i--)
    {
#ifdef VMS
      /* Delete (all versions of) any existing max-suffix file, to avoid
       * creating multiple versions of it.  (On VMS, rename() will
       * create a new version of an existing destination file, not
       * destroy/overwrite it.)
       */
      if (i == opt.backups)
        {
          if (((unsigned) snprintf (to, sizeof (to), "%s%s%d%s", fname, SEP, i, AVS)) >= sizeof (to))
            logprintf (LOG_NOTQUIET, "Failed to delete %s: File name truncation\n", to);
          else
            delete (to);
        }
#endif
      overflow = (unsigned) snprintf (to, FILE_BUF_SIZE, "%s%s%d", fname, SEP, i) >= FILE_BUF_SIZE;
      overflow |= (unsigned) snprintf (from, FILE_BUF_SIZE, "%s%s%d", fname, SEP, i - 1) >= FILE_BUF_SIZE;

      if (overflow)
          errno = ENAMETOOLONG;
      if (overflow || rename (from, to))
        {
          // The original file may not exist. In which case rename() will
          // return ENOENT. This is not a real error. We could make this better
          // by calling stat() first and making sure that the file exists.
          if (errno != ENOENT)
              logprintf (LOG_NOTQUIET, "Failed to rename %s to %s: (%d) %s\n",
                      from, to, errno, strerror (errno));
        }
    }

  overflow = (unsigned) snprintf (to, FILE_BUF_SIZE, "%s%s%d", fname, SEP, 1) >= FILE_BUF_SIZE;
  if (overflow)
    errno = ENAMETOOLONG;
  if (overflow || rename(fname, to))
    {
      if (errno != ENOENT)
          logprintf (LOG_NOTQUIET, "Failed to rename %s to %s: (%d) %s\n",
                  from, to, errno, strerror (errno));
    }

#undef FILE_BUF_SIZE
}

static bool no_proxy_match (const char *, const char **);

/* Return the URL of the proxy appropriate for url U.  */

static char *
getproxy (struct url *u)
{
  char *proxy = NULL;
  char *rewritten_url;

  if (!opt.use_proxy)
    return NULL;
  if (no_proxy_match (u->host, (const char **)opt.no_proxy))
    return NULL;

  switch (u->scheme)
    {
    case SCHEME_HTTP:
      proxy = opt.http_proxy ? opt.http_proxy : getenv ("http_proxy");
      break;
#ifdef HAVE_SSL
    case SCHEME_HTTPS:
      proxy = opt.https_proxy ? opt.https_proxy : getenv ("https_proxy");
      break;
    case SCHEME_FTPS:
      proxy = opt.ftp_proxy ? opt.ftp_proxy : getenv ("ftps_proxy");
      break;
#endif
    case SCHEME_FTP:
      proxy = opt.ftp_proxy ? opt.ftp_proxy : getenv ("ftp_proxy");
      break;
    case SCHEME_INVALID:
      break;
    }
  if (!proxy || !*proxy)
#ifdef HAVE_LIBPROXY
    {
      pxProxyFactory *pf = px_proxy_factory_new ();
      if (!pf)
        {
          debug_logprintf ("Allocating memory for libproxy failed");
         return NULL;
        }

      debug_logprintf ("asking libproxy about url '%s'\n", u->url);
      char **proxies = px_proxy_factory_get_proxies (pf, u->url);
      if (proxies)
        {
          if (proxies[0])
            {
              debug_logprintf ("libproxy suggest to use '%s'\n", proxies[0]);
              if (strcmp (proxies[0], "direct://") != 0)
                {
                  proxy = xstrdup (proxies[0]);
                  debug_logprintf ("libproxy setting to use '%s'\n", proxy);
                }
            }

           px_proxy_factory_free_proxies (proxies);
        }
        px_proxy_factory_free (pf);

      if (!proxy || !*proxy)
        return NULL;
    }
#else
    return NULL;
#endif

  /* Handle shorthands.  `rewritten_storage' is a kludge to allow
     getproxy() to return static storage. */
  rewritten_url = rewrite_shorthand_url (proxy);
  if (rewritten_url)
    return rewritten_url;

  return strdup(proxy);
}

/* Returns true if URL would be downloaded through a proxy. */

bool
url_uses_proxy (struct url * u)
{
  bool ret;
  char *proxy;

  if (!u)
    return false;
  proxy = getproxy (u);
  ret = proxy != NULL;
  xfree (proxy);
  return ret;
}

/* Should a host be accessed through proxy, concerning no_proxy?  */
static bool
no_proxy_match (const char *host, const char **no_proxy)
{
  if (!no_proxy)
    return false;
  else
    return sufmatch (no_proxy, host);
}

/* Set the file parameter to point to the local file string.  */
void
set_local_file (const char **file, const char *default_file)
{
  if (opt.output_document)
    {
      if (output_stream_regular)
        *file = opt.output_document;
    }
  else
    *file = default_file;
}

/* Return true for an input file's own URL, false otherwise.  */
bool
input_file_url (const char *input_file)
{
  static bool first = true;

  if (input_file
      && url_has_scheme (input_file)
      && first)
    {
      first = false;
      return true;
    }
  else
    return false;
}

#ifdef TESTING

#include <stdint.h>
#include "../tests/unit-tests.h"

const char *
test_retr_rate(void)
{
  static const struct test {
    wgint bytes;
    double secs;
    const char *expected;
  } tests[] = {
    { 0, 1, "0.00 B/s" },
    { INT64_MAX, 1, "100 TB/s" },
  };

  for (struct test *t = tests; t < tests+countof(tests); t++)
    {
      const char *result = retr_rate (t->bytes, t->secs);

      if (strcmp(result,t->expected))
        return aprintf("%s: Expected '%s', got '%s'", __func__, t->expected, result);
    }

  return NULL;
}

#endif /* TESTING */