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
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
|
/* sort - sort lines of text (with all kinds of options).
Copyright (C) 88, 1991-2000 Free Software Foundation, Inc.
This program 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 2, or (at your option)
any later version.
This program 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 this program; if not, write to the Free Software
Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
Written December 1988 by Mike Haertel.
The author may be reached (Email) at the address mike@gnu.ai.mit.edu,
or (US mail) as Mike Haertel c/o Free Software Foundation.
Ørn E. Hansen added NLS support in 1997. */
#include <config.h>
#include <sys/types.h>
#include <signal.h>
#include <stdio.h>
#include <assert.h>
#include "system.h"
#include "closeout.h"
#include "long-options.h"
#include "error.h"
#include "hard-locale.h"
#include "memcoll.h"
#include "xalloc.h"
/* The official name of this program (e.g., no `g' prefix). */
#define PROGRAM_NAME "sort"
#define AUTHORS "Mike Haertel"
#if defined ENABLE_NLS && HAVE_LANGINFO_H
# include <langinfo.h>
#endif
#if HAVE_PATHCONF && defined _PC_NAME_MAX
# define NAME_MAX_IN_DIR(Dir) pathconf (Dir, _PC_NAME_MAX)
#else
# define NAME_MAX_IN_DIR(Dir) 255
#endif
#ifndef STDC_HEADERS
double strtod ();
#endif
/* Undefine, to avoid warning about redefinition on some systems. */
#undef min
#define min(a, b) ((a) < (b) ? (a) : (b))
#undef max
#define max(a, b) ((a) > (b) ? (a) : (b))
#define UCHAR_LIM (UCHAR_MAX + 1)
#define UCHAR(c) ((unsigned char) (c))
#ifndef DEFAULT_TMPDIR
# define DEFAULT_TMPDIR "/tmp"
#endif
/* Use this as exit status in case of error, not EXIT_FAILURE. This
is necessary because EXIT_FAILURE is usually 1 and POSIX requires
that sort exit with status 1 IFF invoked with -c and the input is
not properly sorted. Any other irregular exit must exit with a
status code greater than 1. */
#define SORT_FAILURE 2
#define SORT_OUT_OF_ORDER 1
#define C_DECIMAL_POINT '.'
#define NEGATION_SIGN '-'
#define NUMERIC_ZERO '0'
#ifdef ENABLE_NLS
static char decimal_point;
static int th_sep; /* if CHAR_MAX + 1, then there is no thousands separator */
/* Nonzero if the corresponding locales are hard. */
static int hard_LC_COLLATE;
# if HAVE_NL_LANGINFO
static int hard_LC_TIME;
# endif
# define IS_THOUSANDS_SEP(x) ((x) == th_sep)
#else
# define decimal_point C_DECIMAL_POINT
# define IS_THOUSANDS_SEP(x) 0
#endif
/* The kind of blanks for '-b' to skip in various options. */
enum blanktype { bl_start, bl_end, bl_both };
/* The character marking end of line. Default to \n. */
int eolchar = '\n';
/* Lines are held in core as counted strings. */
struct line
{
char *text; /* Text of the line. */
int length; /* Length including final newline. */
char *keybeg; /* Start of first key. */
char *keylim; /* Limit of first key. */
};
/* Arrays of lines. */
struct lines
{
struct line *lines; /* Dynamically allocated array of lines. */
int used; /* Number of slots used. */
int alloc; /* Number of slots allocated. */
int limit; /* Max number of slots to allocate. */
};
/* Input buffers. */
struct buffer
{
char *buf; /* Dynamically allocated buffer. */
int used; /* Number of bytes used. */
int alloc; /* Number of bytes allocated. */
int left; /* Number of bytes left from previous reads. */
int newline_free; /* Number of left bytes that are known
to be newline-free. */
};
struct keyfield
{
int sword; /* Zero-origin 'word' to start at. */
int schar; /* Additional characters to skip. */
int skipsblanks; /* Skip leading white space at start. */
int eword; /* Zero-origin first word after field. */
int echar; /* Additional characters in field. */
int skipeblanks; /* Skip trailing white space at finish. */
int *ignore; /* Boolean array of characters to ignore. */
char *translate; /* Translation applied to characters. */
int numeric; /* Flag for numeric comparison. Handle
strings of digits with optional decimal
point, but no exponential notation. */
int general_numeric; /* Flag for general, numeric comparison.
Handle numbers in exponential notation. */
int month; /* Flag for comparison by month name. */
int reverse; /* Reverse the sense of comparison. */
struct keyfield *next; /* Next keyfield to try. */
};
struct month
{
char *name;
int val;
};
/* The name this program was run with. */
char *program_name;
/* FIXME: None of these tables work with multibyte character sets.
Also, there are many other bugs when handling multibyte characters,
or even unibyte encodings where line boundaries are not in the
initial shift state. One way to fix this is to rewrite `sort' to
use wide characters internally, but doing this with good
performance is a bit tricky. */
/* Table of white space. */
static int blanks[UCHAR_LIM];
/* Table of non-printing characters. */
static int nonprinting[UCHAR_LIM];
/* Table of non-dictionary characters (not letters, digits, or blanks). */
static int nondictionary[UCHAR_LIM];
/* Translation table folding lower case to upper. */
static char fold_toupper[UCHAR_LIM];
#define MONTHS_PER_YEAR 12
#if defined ENABLE_NLS && HAVE_NL_LANGINFO
# define MONTHTAB_CONST /* empty */
#else
# define MONTHTAB_CONST const
#endif
/* Table mapping month names to integers.
Alphabetic order allows binary search. */
static MONTHTAB_CONST struct month monthtab[] =
{
{"APR", 4},
{"AUG", 8},
{"DEC", 12},
{"FEB", 2},
{"JAN", 1},
{"JUL", 7},
{"JUN", 6},
{"MAR", 3},
{"MAY", 5},
{"NOV", 11},
{"OCT", 10},
{"SEP", 9}
};
/* During the merge phase, the number of files to merge at once. */
#define NMERGE 16
/* Initial buffer size for in-core sorting. The buffer will grow only
if a line longer than this is seen. */
#define SORTALLOC (8 * 1024 * 1024)
static int const sortalloc = SORTALLOC;
/* Initial buffer size for in core merge buffers. Bear in mind that
up to NMERGE * mergealloc bytes may be allocated for merge buffers. */
static int const mergealloc = SORTALLOC / NMERGE / 2;
/* Guess of average line length. */
static int const linelength = 30;
/* Maximum number of elements for the array(s) of struct line's, in bytes. */
#define LINEALLOC (SORTALLOC / 2)
/* Array of directory names in which any temporary files are to be created. */
static char const **temp_dirs;
/* Number of temporary directory names used. */
static size_t temp_dir_count;
/* Number of allocated slots in temp_dirs. */
static size_t temp_dir_alloc;
/* Our process ID. */
static pid_t process_id;
/* Flag to reverse the order of all comparisons. */
static int reverse;
/* Flag for stable sort. This turns off the last ditch bytewise
comparison of lines, and instead leaves lines in the same order
they were read if all keys compare equal. */
static int stable;
/* Tab character separating fields. If NUL, then fields are separated
by the empty string between a non-whitespace character and a whitespace
character. */
static char tab;
/* Flag to remove consecutive duplicate lines from the output.
Only the last of a sequence of equal lines will be output. */
static int unique;
/* Nonzero if any of the input files are the standard input. */
static int have_read_stdin;
/* List of key field comparisons to be tried. */
static struct keyfield *keylist;
void
usage (int status)
{
if (status != 0)
fprintf (stderr, _("Try `%s --help' for more information.\n"),
program_name);
else
{
printf (_("\
Usage: %s [OPTION]... [FILE]...\n\
"),
program_name);
printf (_("\
Write sorted concatenation of all FILE(s) to standard output.\n\
\n\
+POS1 [-POS2] start a key at POS1, end it *before* POS2 (obsolescent)\n\
field numbers and character offsets are numbered\n\
starting with zero (contrast with the -k option)\n\
-b ignore leading blanks in sort fields or keys\n\
-c check if given files already sorted, do not sort\n\
-d consider only blanks and alphanumeric characters in keys\n\
-f fold lower case to upper case characters in keys\n\
-g compare according to general numerical value, imply -b\n\
-i consider only printable characters in keys\n\
-k POS1[,POS2] start a key at POS1, end it *at* POS2\n\
field numbers and character offsets are numbered\n\
starting with one (contrast with zero-based +POS form)\n\
-m merge already sorted files, do not sort\n\
-M compare (unknown) < `JAN' < ... < `DEC', imply -b\n\
-n compare according to string numerical value, imply -b\n\
-o FILE write result on FILE instead of standard output\n\
-r reverse the result of comparisons\n\
-s stabilize sort by disabling last resort comparison\n\
-t SEP use SEParator instead of non- to whitespace transition\n\
-T DIRECTORY use DIRECTORY for temporary files, not $TMPDIR or %s\n\
multiple -T options specify multiple directories\n\
-u with -c, check for strict ordering;\n\
with -m, only output the first of an equal sequence\n\
-z end lines with 0 byte, not newline, for find -print0\n\
--help display this help and exit\n\
--version output version information and exit\n\
\n\
"),
DEFAULT_TMPDIR);
printf (_("\
POS is F[.C][OPTS], where F is the field number and C the character position\n\
in the field, both counted from one with -k, from zero with the obsolescent\n\
form. OPTS is made up of one or more of Mbdfinr; this effectively disables\n\
global -Mbdfinr settings for that key. If no key is given, use the entire\n\
line as the key. With no FILE, or when FILE is -, read standard input.\n\
\n\
*** WARNING ***\n\
The locale specified by the environment affects sort order.\n\
Set LC_ALL=C to get the traditional sort order that uses native byte values.\n\
")
);
puts (_("\nReport bugs to <bug-textutils@gnu.org>."));
}
/* Don't use EXIT_FAILURE here in case it is defined to be 1.
POSIX requires that sort return 1 IFF invoked with -c and
the input is not properly sorted. */
assert (status == 0 || status == SORT_FAILURE);
exit (status);
}
/* The list of temporary files. */
static struct tempnode
{
char *name;
struct tempnode *next;
} temphead;
/* Clean up any remaining temporary files. */
static void
cleanup (void)
{
struct tempnode *node;
for (node = temphead.next; node; node = node->next)
unlink (node->name);
}
static FILE *
xtmpfopen (const char *file)
{
FILE *fp;
int fd;
/* Open temporary file exclusively, to foil a common
denial-of-service attack. */
fd = open (file, O_WRONLY | O_CREAT | O_TRUNC | O_EXCL, 0600);
if (fd < 0 || (fp = fdopen (fd, "w")) == NULL)
{
error (0, errno, "%s", file);
cleanup ();
exit (SORT_FAILURE);
}
return fp;
}
static FILE *
xfopen (const char *file, const char *how)
{
FILE *fp;
if (STREQ (file, "-"))
{
fp = stdin;
}
else
{
if ((fp = fopen (file, how)) == NULL)
{
error (0, errno, "%s", file);
cleanup ();
exit (SORT_FAILURE);
}
}
if (fp == stdin)
have_read_stdin = 1;
return fp;
}
static void
xfclose (FILE *fp)
{
if (fp == stdin)
{
/* Allow reading stdin from tty more than once. */
if (feof (fp))
clearerr (fp);
}
else if (fp == stdout)
{
if (fflush (fp) != 0)
{
error (0, errno, _("flushing file"));
cleanup ();
exit (SORT_FAILURE);
}
}
else
{
if (fclose (fp) != 0)
{
error (0, errno, _("error closing file"));
cleanup ();
exit (SORT_FAILURE);
}
}
}
static void
write_bytes (const char *buf, size_t n_bytes, FILE *fp, const char *output_file)
{
if (fwrite (buf, 1, n_bytes, fp) != n_bytes)
{
error (0, errno, _("%s: write error"), output_file);
cleanup ();
exit (SORT_FAILURE);
}
}
/* Append DIR to the array of temporary directory names. */
static void
add_temp_dir (char const *dir)
{
if (temp_dir_count == temp_dir_alloc)
{
temp_dir_alloc = temp_dir_alloc ? temp_dir_alloc * 2 : 16;
temp_dirs = xrealloc (temp_dirs, sizeof (temp_dirs) * temp_dir_alloc);
}
temp_dirs[temp_dir_count++] = dir;
}
/* Return a name for a temporary file. */
static char *
tempname (void)
{
static unsigned long sequence_number;
unsigned long seq = sequence_number++;
unsigned long pid = process_id;
char const *temp_dir = temp_dirs[seq % temp_dir_count];
size_t len = strlen (temp_dir);
char const *slash = "/" + (len == 0 || temp_dir[len - 1] == '/');
char *name = xmalloc (len + 1 + sizeof "sort" - 1
+ sizeof pid * CHAR_BIT / 3 + 1
+ sizeof seq * CHAR_BIT / 3 + 1);
int long_file_names = NAME_MAX_IN_DIR (temp_dir) > 12;
struct tempnode *node;
if (long_file_names)
sprintf (name, "%s%ssort%lu.%.5lu", temp_dir, slash, pid, seq);
else
{
/* Make sure the file name is safe for an 8.3 filesystem. */
sprintf (name, "%s%ss%.5d%.2d.%.3d", temp_dir, slash,
(int) (pid % 100000), (int) (seq / 1000 % 100),
(int) (seq % 1000));
}
node = (struct tempnode *) xmalloc (sizeof (struct tempnode));
node->name = name;
node->next = temphead.next;
temphead.next = node;
return name;
}
/* Search through the list of temporary files for NAME;
remove it if it is found on the list. */
static void
zaptemp (const char *name)
{
struct tempnode *node, *temp;
for (node = &temphead; node->next; node = node->next)
if (STREQ (name, node->next->name))
break;
if (node->next)
{
temp = node->next;
unlink (temp->name);
free (temp->name);
node->next = temp->next;
free ((char *) temp);
}
}
#ifdef ENABLE_NLS
static int
struct_month_cmp (const void *m1, const void *m2)
{
return strcmp (((const struct month *) m1)->name,
((const struct month *) m2)->name);
}
#endif /* NLS */
/* Initialize the character class tables. */
static void
inittables (void)
{
int i;
for (i = 0; i < UCHAR_LIM; ++i)
{
if (ISBLANK (i))
blanks[i] = 1;
if (!ISPRINT (i))
nonprinting[i] = 1;
if (!ISALNUM (i) && !ISBLANK (i))
nondictionary[i] = 1;
if (ISLOWER (i))
fold_toupper[i] = toupper (i);
else
fold_toupper[i] = i;
}
#if defined ENABLE_NLS && HAVE_NL_LANGINFO
/* If we're not in the "C" locale, read different names for months. */
if (hard_LC_TIME)
{
for (i = 0; i < MONTHS_PER_YEAR; i++)
{
char *s;
size_t s_len;
size_t j;
char *name;
s = (char *) nl_langinfo (ABMON_1 + i);
s_len = strlen (s);
monthtab[i].name = name = (char *) xmalloc (s_len + 1);
monthtab[i].val = i + 1;
for (j = 0; j < s_len; j++)
name[j] = fold_toupper[UCHAR (s[j])];
name[j] = '\0';
}
qsort ((void *) monthtab, MONTHS_PER_YEAR,
sizeof (struct month), struct_month_cmp);
}
#endif /* NLS */
}
/* Initialize BUF, allocating ALLOC bytes initially. */
static void
initbuf (struct buffer *buf, int alloc)
{
buf->alloc = alloc;
buf->buf = xmalloc (buf->alloc);
buf->used = buf->left = buf->newline_free = 0;
}
/* Fill BUF reading from FP, moving buf->left bytes from the end
of buf->buf to the beginning first. If EOF is reached and the
file wasn't terminated by a newline, supply one. Return a count
of bytes buffered. */
static int
fillbuf (struct buffer *buf, FILE *fp)
{
if (buf->used != buf->left)
{
memmove (buf->buf, buf->buf + buf->used - buf->left, buf->left);
buf->used = buf->left;
}
while (!feof (fp))
{
int cc;
if (buf->used == buf->alloc)
{
buf->alloc *= 2;
buf->buf = xrealloc (buf->buf, buf->alloc);
}
cc = fread (buf->buf + buf->used, 1, buf->alloc - buf->used, fp);
if (ferror (fp))
{
error (0, errno, _("read error"));
cleanup ();
exit (SORT_FAILURE);
}
buf->used += cc;
if (memchr (buf->buf + buf->used - cc, eolchar, cc))
break;
}
if (feof (fp) && buf->used && buf->buf[buf->used - 1] != eolchar)
{
if (buf->used == buf->alloc)
{
/* If the buffer isn't parsed yet, don't increase the buffer
size to append a newline, as this won't be needed unless
the buffer turns out to be newline-free. */
if (buf->newline_free != buf->used)
return buf->used;
buf->alloc *= 2;
buf->buf = xrealloc (buf->buf, buf->alloc);
}
buf->buf[buf->used++] = eolchar;
}
return buf->used;
}
/* Initialize LINES, allocating space for ALLOC lines initially.
LIMIT is the maximum possible number of lines to allocate space
for, ever. */
static void
initlines (struct lines *lines, int alloc, int limit)
{
lines->alloc = alloc;
lines->lines = (struct line *) xmalloc (lines->alloc * sizeof (struct line));
lines->used = 0;
lines->limit = limit;
}
/* Return a pointer to the first character of the field specified
by KEY in LINE. */
static char *
begfield (const struct line *line, const struct keyfield *key)
{
register char *ptr = line->text, *lim = ptr + line->length - 1;
register int sword = key->sword, schar = key->schar;
if (tab)
while (ptr < lim && sword--)
{
while (ptr < lim && *ptr != tab)
++ptr;
if (ptr < lim)
++ptr;
}
else
while (ptr < lim && sword--)
{
while (ptr < lim && blanks[UCHAR (*ptr)])
++ptr;
while (ptr < lim && !blanks[UCHAR (*ptr)])
++ptr;
}
if (key->skipsblanks)
while (ptr < lim && blanks[UCHAR (*ptr)])
++ptr;
if (ptr + schar <= lim)
ptr += schar;
else
ptr = lim;
return ptr;
}
/* Return the limit of (a pointer to the first character after) the field
in LINE specified by KEY. */
static char *
limfield (const struct line *line, const struct keyfield *key)
{
register char *ptr = line->text, *lim = ptr + line->length - 1;
register int eword = key->eword, echar = key->echar;
/* Note: from the POSIX spec:
The leading field separator itself is included in
a field when -t is not used. FIXME: move this comment up... */
/* Move PTR past EWORD fields or to one past the last byte on LINE,
whichever comes first. If there are more than EWORD fields, leave
PTR pointing at the beginning of the field having zero-based index,
EWORD. If a delimiter character was specified (via -t), then that
`beginning' is the first character following the delimiting TAB.
Otherwise, leave PTR pointing at the first `blank' character after
the preceding field. */
if (tab)
while (ptr < lim && eword--)
{
while (ptr < lim && *ptr != tab)
++ptr;
if (ptr < lim && (eword || echar > 0))
++ptr;
}
else
while (ptr < lim && eword--)
{
while (ptr < lim && blanks[UCHAR (*ptr)])
++ptr;
while (ptr < lim && !blanks[UCHAR (*ptr)])
++ptr;
}
#ifdef POSIX_UNSPECIFIED
/* The following block of code makes GNU sort incompatible with
standard Unix sort, so it's ifdef'd out for now.
The POSIX spec isn't clear on how to interpret this.
FIXME: request clarification.
From: kwzh@gnu.ai.mit.edu (Karl Heuer)
Date: Thu, 30 May 96 12:20:41 -0400
[...]I believe I've found another bug in `sort'.
$ cat /tmp/sort.in
a b c 2 d
pq rs 1 t
$ textutils-1.15/src/sort +0.6 -0.7 </tmp/sort.in
a b c 2 d
pq rs 1 t
$ /bin/sort +0.6 -0.7 </tmp/sort.in
pq rs 1 t
a b c 2 d
Unix sort produced the answer I expected: sort on the single character
in column 6. GNU sort produced different results, because it disagrees
on the interpretation of the key-end spec "-M.N". Unix sort reads this
as "skip M fields, then N characters"; but GNU sort wants it to mean
"skip M fields, then either N characters or the rest of the current
field, whichever comes first". This extra clause applies only to
key-ends, not key-starts.
*/
/* Make LIM point to the end of (one byte past) the current field. */
if (tab)
{
char *newlim;
newlim = memchr (ptr, tab, lim - ptr);
if (newlim)
lim = newlim;
}
else
{
char *newlim;
newlim = ptr;
while (newlim < lim && blanks[UCHAR (*newlim)])
++newlim;
while (newlim < lim && !blanks[UCHAR (*newlim)])
++newlim;
lim = newlim;
}
#endif
/* If we're skipping leading blanks, don't start counting characters
until after skipping past any leading blanks. */
if (key->skipsblanks)
while (ptr < lim && blanks[UCHAR (*ptr)])
++ptr;
/* Advance PTR by ECHAR (if possible), but no further than LIM. */
if (ptr + echar <= lim)
ptr += echar;
else
ptr = lim;
return ptr;
}
/* FIXME */
void
trim_trailing_blanks (const char *a_start, char **a_end)
{
while (*a_end > a_start && blanks[UCHAR (*(*a_end - 1))])
--(*a_end);
}
/* Find the lines in BUF, storing pointers and lengths in LINES. */
static void
findlines (struct buffer *buf, struct lines *lines)
{
register char *beg = buf->buf, *lim = buf->buf + buf->used, *ptr;
struct keyfield *key = keylist;
lines->used = 0;
while (beg < lim && (ptr = memchr (beg, eolchar, lim - beg))
&& lines->used < lines->limit)
{
struct line *line;
if (lines->used == lines->alloc)
{
lines->alloc *= 2;
lines->lines = (struct line *)
xrealloc ((char *) lines->lines,
lines->alloc * sizeof (struct line));
}
line = &lines->lines[lines->used];
line->text = beg;
line->length = ptr + 1 - beg;
if (key)
{
/* Precompute the position of the first key for efficiency. */
line->keylim = 0 <= key->eword ? limfield (line, key) : ptr;
if (key->sword >= 0)
line->keybeg = begfield (line, key);
else
{
if (key->skipsblanks)
while (blanks[UCHAR (*beg)])
++beg;
line->keybeg = beg;
}
if (key->skipeblanks)
trim_trailing_blanks (line->keybeg, &line->keylim);
}
++lines->used;
beg = ptr + 1;
}
buf->newline_free = buf->left = lim - beg;
}
/* Compare strings A and B containing decimal fractions < 1. Each string
should begin with a decimal point followed immediately by the digits
of the fraction. Strings not of this form are considered to be zero. */
/* The goal here, is to take two numbers a and b... compare these
in parallel. Instead of converting each, and then comparing the
outcome. Most likely stopping the comparison before the conversion
is complete. The algorithm used, in the old sort:
Algorithm: fraccompare
Action : compare two decimal fractions
accepts : char *a, char *b
returns : -1 if a<b, 0 if a=b, 1 if a>b.
implement:
if *a == decimal_point AND *b == decimal_point
find first character different in a and b.
if both are digits, return the difference *a - *b.
if *a is a digit
skip past zeros
if digit return 1, else 0
if *b is a digit
skip past zeros
if digit return -1, else 0
if *a is a decimal_point
skip past decimal_point and zeros
if digit return 1, else 0
if *b is a decimal_point
skip past decimal_point and zeros
if digit return -1, else 0
return 0 */
static int
fraccompare (register const char *a, register const char *b)
{
if (*a == decimal_point && *b == decimal_point)
{
while (*++a == *++b)
if (! ISDIGIT (*a))
return 0;
if (ISDIGIT (*a) && ISDIGIT (*b))
return *a - *b;
if (ISDIGIT (*a))
goto a_trailing_nonzero;
if (ISDIGIT (*b))
goto b_trailing_nonzero;
return 0;
}
else if (*a++ == decimal_point)
{
a_trailing_nonzero:
while (*a == NUMERIC_ZERO)
a++;
return ISDIGIT (*a);
}
else if (*b++ == decimal_point)
{
b_trailing_nonzero:
while (*b == NUMERIC_ZERO)
b++;
return - ISDIGIT (*b);
}
return 0;
}
/* Compare strings A and B as numbers without explicitly converting them to
machine numbers. Comparatively slow for short strings, but asymptotically
hideously fast. */
static int
numcompare (register const char *a, register const char *b)
{
register int tmpa, tmpb, loga, logb, tmp;
tmpa = *a;
tmpb = *b;
while (blanks[UCHAR (tmpa)])
tmpa = *++a;
while (blanks[UCHAR (tmpb)])
tmpb = *++b;
if (tmpa == NEGATION_SIGN)
{
do
tmpa = *++a;
while (tmpa == NUMERIC_ZERO || IS_THOUSANDS_SEP (tmpa));
if (tmpb != NEGATION_SIGN)
{
if (tmpa == decimal_point)
do
tmpa = *++a;
while (tmpa == NUMERIC_ZERO);
if (ISDIGIT (tmpa))
return -1;
while (tmpb == NUMERIC_ZERO || IS_THOUSANDS_SEP (tmpb))
tmpb = *++b;
if (tmpb == decimal_point)
do
tmpb = *++b;
while (tmpb == NUMERIC_ZERO);
if (ISDIGIT (tmpb))
return -1;
return 0;
}
do
tmpb = *++b;
while (tmpb == NUMERIC_ZERO || IS_THOUSANDS_SEP (tmpb));
while (tmpa == tmpb && ISDIGIT (tmpa))
{
do
tmpa = *++a;
while (IS_THOUSANDS_SEP (tmpa));
do
tmpb = *++b;
while (IS_THOUSANDS_SEP (tmpb));
}
if ((tmpa == decimal_point && !ISDIGIT (tmpb))
|| (tmpb == decimal_point && !ISDIGIT (tmpa)))
return -fraccompare (a, b);
tmp = tmpb - tmpa;
for (loga = 0; ISDIGIT (tmpa); ++loga)
do
tmpa = *++a;
while (IS_THOUSANDS_SEP (tmpa));
for (logb = 0; ISDIGIT (tmpb); ++logb)
do
tmpb = *++b;
while (IS_THOUSANDS_SEP (tmpb));
if (logb - loga != 0)
return logb - loga;
if (!loga)
return 0;
return tmp;
}
else if (tmpb == NEGATION_SIGN)
{
do
tmpb = *++b;
while (tmpb == NUMERIC_ZERO || IS_THOUSANDS_SEP (tmpb));
if (tmpb == decimal_point)
do
tmpb = *++b;
while (tmpb == NUMERIC_ZERO);
if (ISDIGIT (tmpb))
return 1;
while (tmpa == NUMERIC_ZERO || IS_THOUSANDS_SEP (tmpa))
tmpa = *++a;
if (tmpa == decimal_point)
do
tmpa = *++a;
while (tmpa == NUMERIC_ZERO);
if (ISDIGIT (tmpa))
return 1;
return 0;
}
else
{
while (tmpa == NUMERIC_ZERO || IS_THOUSANDS_SEP (tmpa))
tmpa = *++a;
while (tmpb == NUMERIC_ZERO || IS_THOUSANDS_SEP (tmpb))
tmpb = *++b;
while (tmpa == tmpb && ISDIGIT (tmpa))
{
do
tmpa = *++a;
while (IS_THOUSANDS_SEP (tmpa));
do
tmpb = *++b;
while (IS_THOUSANDS_SEP (tmpb));
}
if ((tmpa == decimal_point && !ISDIGIT (tmpb))
|| (tmpb == decimal_point && !ISDIGIT (tmpa)))
return fraccompare (a, b);
tmp = tmpa - tmpb;
for (loga = 0; ISDIGIT (tmpa); ++loga)
do
tmpa = *++a;
while (IS_THOUSANDS_SEP (tmpa));
for (logb = 0; ISDIGIT (tmpb); ++logb)
do
tmpb = *++b;
while (IS_THOUSANDS_SEP (tmpb));
if (loga - logb != 0)
return loga - logb;
if (!loga)
return 0;
return tmp;
}
}
static int
general_numcompare (const char *sa, const char *sb)
{
/* FIXME: add option to warn about failed conversions. */
/* FIXME: maybe add option to try expensive FP conversion
only if A and B can't be compared more cheaply/accurately. */
char *ea;
char *eb;
double a = strtod (sa, &ea);
double b = strtod (sb, &eb);
/* Put conversion errors at the start of the collating sequence. */
if (sa == ea)
return sb == eb ? 0 : -1;
if (sb == eb)
return 1;
/* Sort numbers in the usual way, where -0 == +0. Put NaNs after
conversion errors but before numbers; sort them by internal
bit-pattern, for lack of a more portable alternative. */
return (a < b ? -1
: a > b ? 1
: a == b ? 0
: b == b ? -1
: a == a ? 1
: memcmp ((char *) &a, (char *) &b, sizeof a));
}
/* Return an integer in 1..12 of the month name S with length LEN.
Return 0 if the name in S is not recognized. */
static int
getmonth (const char *s, int len)
{
char *month;
register int i, lo = 0, hi = MONTHS_PER_YEAR, result;
while (len > 0 && blanks[UCHAR (*s)])
{
++s;
--len;
}
if (len == 0)
return 0;
month = (char *) alloca (len + 1);
for (i = 0; i < len; ++i)
month[i] = fold_toupper[UCHAR (s[i])];
while (blanks[UCHAR (month[i - 1])])
--i;
month[i] = '\0';
do
{
int ix = (lo + hi) / 2;
if (strncmp (month, monthtab[ix].name, strlen (monthtab[ix].name)) < 0)
hi = ix;
else
lo = ix;
}
while (hi - lo > 1);
result = (!strncmp (month, monthtab[lo].name, strlen (monthtab[lo].name))
? monthtab[lo].val : 0);
return result;
}
/* Compare two lines A and B trying every key in sequence until there
are no more keys or a difference is found. */
static int
keycompare (const struct line *a, const struct line *b)
{
struct keyfield *key = keylist;
/* For the first iteration only, the key positions have been
precomputed for us. */
register char *texta = a->keybeg;
register char *textb = b->keybeg;
register char *lima = a->keylim;
register char *limb = b->keylim;
int diff, lena, lenb;
for (;;)
{
register unsigned char *translate = (unsigned char *) key->translate;
register int *ignore = key->ignore;
/* Find the lengths. */
lena = lima - texta, lenb = limb - textb;
if (lena < 0)
lena = 0;
if (lenb < 0)
lenb = 0;
if (key->skipeblanks)
{
char *a_end = texta + lena;
char *b_end = textb + lenb;
trim_trailing_blanks (texta, &a_end);
trim_trailing_blanks (textb, &b_end);
lena = a_end - texta;
lenb = b_end - textb;
}
/* Actually compare the fields. */
if (key->numeric | key->general_numeric)
{
char savea = *lima, saveb = *limb;
*lima = *limb = '\0';
diff = ((key->numeric ? numcompare : general_numcompare)
(texta, textb));
*lima = savea, *limb = saveb;
}
else if (key->month)
diff = getmonth (texta, lena) - getmonth (textb, lenb);
#ifdef ENABLE_NLS
/* Sorting like this may become slow, so in a simple locale the user
can select a faster sort that is similar to ascii sort */
else if (hard_LC_COLLATE)
{
if (ignore || translate)
{
char *copy_a = (char *) alloca (lena + 1 + lenb + 1);
char *copy_b = copy_a + lena + 1;
int new_len_a, new_len_b, i;
/* Ignore and/or translate chars before comparing. */
for (new_len_a = new_len_b = i = 0; i < max (lena, lenb); i++)
{
if (i < lena)
{
copy_a[new_len_a] = (translate
? translate[UCHAR (texta[i])]
: texta[i]);
if (!ignore || !ignore[UCHAR (texta[i])])
++new_len_a;
}
if (i < lenb)
{
copy_b[new_len_b] = (translate
? translate[UCHAR (textb[i])]
: textb [i]);
if (!ignore || !ignore[UCHAR (textb[i])])
++new_len_b;
}
}
diff = memcoll (copy_a, new_len_a, copy_b, new_len_b);
}
else if (lena == 0)
diff = -lenb;
else if (lenb == 0)
goto greater;
else
diff = memcoll (texta, lena, textb, lenb);
}
#endif
else if (ignore)
{
#define CMP_WITH_IGNORE(A, B) \
do \
{ \
for (;;) \
{ \
while (texta < lima && ignore[UCHAR (*texta)]) \
++texta; \
while (textb < limb && ignore[UCHAR (*textb)]) \
++textb; \
if (! (texta < lima && textb < limb)) \
break; \
diff = UCHAR (A) - UCHAR (B); \
if (diff) \
goto not_equal; \
++texta; \
++textb; \
} \
\
diff = (lima - texta) - (limb - textb); \
} \
while (0)
if (translate)
CMP_WITH_IGNORE (translate[UCHAR (*texta)],
translate[UCHAR (*textb)]);
else
CMP_WITH_IGNORE (UCHAR (*texta), UCHAR (*textb));
}
else if (lena == 0)
diff = -lenb;
else if (lenb == 0)
goto greater;
else
{
if (translate)
{
while (texta < lima && textb < limb)
{
diff = (UCHAR (translate[UCHAR (*texta++)])
- UCHAR (translate[UCHAR (*textb++)]));
if (diff)
goto not_equal;
}
}
else
{
diff = memcmp (texta, textb, min (lena, lenb));
if (diff)
goto not_equal;
}
diff = lena - lenb;
}
if (diff)
goto not_equal;
key = key->next;
if (! key)
break;
/* Find the beginning and limit of the next field. */
if (key->eword >= 0)
lima = limfield (a, key), limb = limfield (b, key);
else
lima = a->text + a->length - 1, limb = b->text + b->length - 1;
if (key->sword >= 0)
texta = begfield (a, key), textb = begfield (b, key);
else
{
texta = a->text, textb = b->text;
if (key->skipsblanks)
{
while (texta < lima && blanks[UCHAR (*texta)])
++texta;
while (textb < limb && blanks[UCHAR (*textb)])
++textb;
}
}
}
return 0;
greater:
diff = 1;
not_equal:
return key->reverse ? -diff : diff;
}
/* Compare two lines A and B, returning negative, zero, or positive
depending on whether A compares less than, equal to, or greater than B. */
static int
compare (register const struct line *a, register const struct line *b)
{
int diff, alen, blen;
/* First try to compare on the specified keys (if any).
The only two cases with no key at all are unadorned sort,
and unadorned sort -r. */
if (keylist)
{
diff = keycompare (a, b);
alloca (0);
if (diff != 0 || unique || stable)
return diff;
}
/* If the keys all compare equal (or no keys were specified)
fall through to the default comparison. */
alen = a->length - 1, blen = b->length - 1;
if (alen == 0)
diff = - blen;
else if (blen == 0)
diff = alen;
#ifdef ENABLE_NLS
else if (hard_LC_COLLATE)
diff = memcoll (a->text, alen, b->text, blen);
#endif
else if (! (diff = memcmp (a->text, b->text, min (alen, blen))))
diff = alen - blen;
return reverse ? -diff : diff;
}
/* Check that the lines read from the given FP come in order. Print a
diagnostic (FILE_NAME, line number, contents of line) to stderr and return
the line number of the first out-of-order line (counting from 1) if they
are not in order. Otherwise, print no diagnostic and return zero. */
static int
checkfp (FILE *fp, const char *file_name)
{
struct buffer buf; /* Input buffer. */
struct lines lines; /* Lines scanned from the buffer. */
struct line temp; /* Copy of previous line. */
int cc; /* Character count. */
int alloc;
int line_number = 1;
struct line *disorder_line IF_LINT (= NULL);
int disorder_line_number = 0;
struct keyfield *key = keylist;
initbuf (&buf, mergealloc);
initlines (&lines, mergealloc / linelength + 1,
LINEALLOC / ((NMERGE + NMERGE) * sizeof (struct line)));
alloc = linelength;
temp.text = xmalloc (alloc);
cc = fillbuf (&buf, fp);
if (cc == 0)
goto finish;
findlines (&buf, &lines);
while (1)
{
struct line *prev_line; /* Pointer to previous line. */
int cmp; /* Result of calling compare. */
int i;
/* Compare each line in the buffer with its successor. */
for (i = 0; i < lines.used - 1; ++i)
{
cmp = compare (&lines.lines[i], &lines.lines[i + 1]);
if ((unique && cmp >= 0) || (cmp > 0))
{
disorder_line = &lines.lines[i + 1];
disorder_line_number = line_number + i + 1;
goto finish;
}
}
line_number += lines.used;
/* Save the last line of the buffer and refill the buffer. */
prev_line = lines.lines + (lines.used - 1);
if (alloc < prev_line->length)
{
while ((alloc *= 2) < prev_line->length)
continue;
temp.text = xrealloc (temp.text, alloc);
}
memcpy (temp.text, prev_line->text, prev_line->length);
temp.length = prev_line->length;
if (key)
{
temp.keybeg = temp.text + (prev_line->keybeg - prev_line->text);
temp.keylim = temp.text + (prev_line->keylim - prev_line->text);
}
cc = fillbuf (&buf, fp);
if (cc == 0)
break;
findlines (&buf, &lines);
/* Make sure the line saved from the old buffer contents is
less than or equal to the first line of the new buffer. */
cmp = compare (&temp, &lines.lines[0]);
if ((unique && cmp >= 0) || (cmp > 0))
{
disorder_line = &lines.lines[0];
disorder_line_number = line_number;
break;
}
}
finish:
xfclose (fp);
if (disorder_line_number)
{
fprintf (stderr, _("%s: %s:%d: disorder: "), program_name, file_name,
disorder_line_number);
write_bytes (disorder_line->text, disorder_line->length, stderr,
_("standard error"));
}
free (buf.buf);
free ((char *) lines.lines);
free (temp.text);
return disorder_line_number;
}
/* Merge lines from FPS onto OFP. NFPS cannot be greater than NMERGE.
Close FPS before returning. */
static void
mergefps (FILE **fps, register int nfps, FILE *ofp, const char *output_file)
{
struct buffer buffer[NMERGE]; /* Input buffers for each file. */
struct lines lines[NMERGE]; /* Line tables for each buffer. */
struct line saved; /* Saved line storage for unique check. */
struct line const *savedline IF_LINT (= NULL);
/* &saved if there is a saved line. */
int savealloc IF_LINT (= 0); /* Size allocated for the saved line. */
int cur[NMERGE]; /* Current line in each line table. */
int ord[NMERGE]; /* Table representing a permutation of fps,
such that lines[ord[0]].lines[cur[ord[0]]]
is the smallest line and will be next
output. */
register int i, j, t;
struct keyfield *key = keylist;
/* Allocate space for a saved line if necessary. */
if (unique)
{
savedline = NULL;
savealloc = linelength;
saved.text = xmalloc (savealloc);
}
/* Read initial lines from each input file. */
for (i = 0; i < nfps; ++i)
{
initbuf (&buffer[i], mergealloc);
/* If a file is empty, eliminate it from future consideration. */
while (i < nfps && !fillbuf (&buffer[i], fps[i]))
{
xfclose (fps[i]);
--nfps;
for (j = i; j < nfps; ++j)
fps[j] = fps[j + 1];
}
if (i == nfps)
free (buffer[i].buf);
else
{
initlines (&lines[i], mergealloc / linelength + 1,
LINEALLOC / ((NMERGE + NMERGE) * sizeof (struct line)));
findlines (&buffer[i], &lines[i]);
cur[i] = 0;
}
}
/* Set up the ord table according to comparisons among input lines.
Since this only reorders two items if one is strictly greater than
the other, it is stable. */
for (i = 0; i < nfps; ++i)
ord[i] = i;
for (i = 1; i < nfps; ++i)
if (compare (&lines[ord[i - 1]].lines[cur[ord[i - 1]]],
&lines[ord[i]].lines[cur[ord[i]]]) > 0)
t = ord[i - 1], ord[i - 1] = ord[i], ord[i] = t, i = 0;
/* Repeatedly output the smallest line until no input remains. */
while (nfps)
{
struct line const *smallest = &lines[ord[0]].lines[cur[ord[0]]];
/* If uniquified output is turned on, output only the first of
an identical series of lines. */
if (unique)
{
if (savedline && compare (savedline, smallest))
{
savedline = 0;
write_bytes (saved.text, saved.length, ofp, output_file);
}
if (!savedline)
{
savedline = &saved;
if (savealloc < smallest->length)
{
while ((savealloc *= 2) < smallest->length)
continue;
saved.text = xrealloc (saved.text, savealloc);
}
saved.length = smallest->length;
memcpy (saved.text, smallest->text, saved.length);
if (key)
{
saved.keybeg =
saved.text + (smallest->keybeg - smallest->text);
saved.keylim =
saved.text + (smallest->keylim - smallest->text);
}
}
}
else
write_bytes (smallest->text, smallest->length, ofp, output_file);
/* Check if we need to read more lines into core. */
if (++cur[ord[0]] == lines[ord[0]].used)
{
if (fillbuf (&buffer[ord[0]], fps[ord[0]]))
{
findlines (&buffer[ord[0]], &lines[ord[0]]);
cur[ord[0]] = 0;
}
else
{
/* We reached EOF on fps[ord[0]]. */
for (i = 1; i < nfps; ++i)
if (ord[i] > ord[0])
--ord[i];
--nfps;
xfclose (fps[ord[0]]);
free (buffer[ord[0]].buf);
free ((char *) lines[ord[0]].lines);
for (i = ord[0]; i < nfps; ++i)
{
fps[i] = fps[i + 1];
buffer[i] = buffer[i + 1];
lines[i] = lines[i + 1];
cur[i] = cur[i + 1];
}
for (i = 0; i < nfps; ++i)
ord[i] = ord[i + 1];
continue;
}
}
/* The new line just read in may be larger than other lines
already in core; push it back in the queue until we encounter
a line larger than it. */
for (i = 1; i < nfps; ++i)
{
t = compare (&lines[ord[0]].lines[cur[ord[0]]],
&lines[ord[i]].lines[cur[ord[i]]]);
if (!t)
t = ord[0] - ord[i];
if (t < 0)
break;
}
t = ord[0];
for (j = 1; j < i; ++j)
ord[j - 1] = ord[j];
ord[i - 1] = t;
}
if (unique && savedline)
{
write_bytes (saved.text, saved.length, ofp, output_file);
free (saved.text);
}
}
/* Sort the array LINES with NLINES members, using TEMP for temporary space. */
static void
sortlines (struct line *lines, int nlines, struct line *temp)
{
register struct line *lo, *hi, *t;
register int nlo, nhi;
if (nlines == 2)
{
if (compare (&lines[0], &lines[1]) > 0)
{
*temp = lines[0];
lines[0] = lines[1];
lines[1] = *temp;
}
return;
}
nlo = nlines / 2;
lo = lines;
nhi = nlines - nlo;
hi = lines + nlo;
if (nlo > 1)
sortlines (lo, nlo, temp);
if (nhi > 1)
sortlines (hi, nhi, temp);
t = temp;
while (nlo && nhi)
if (compare (lo, hi) <= 0)
*t++ = *lo++, --nlo;
else
*t++ = *hi++, --nhi;
while (nlo--)
*t++ = *lo++;
for (lo = lines, nlo = nlines - nhi, t = temp; nlo; --nlo)
*lo++ = *t++;
}
/* Check that each of the NFILES FILES is ordered.
Return a count of disordered files. */
static int
check (char **files, int nfiles)
{
int i, disorders = 0;
FILE *fp;
for (i = 0; i < nfiles; ++i)
{
fp = xfopen (files[i], "r");
if (checkfp (fp, files[i]))
{
++disorders;
}
}
return disorders;
}
/* Merge NFILES FILES onto OFP. */
static void
merge (char **files, int nfiles, FILE *ofp, const char *output_file)
{
int i, j, t;
char *temp;
FILE *fps[NMERGE], *tfp;
while (nfiles > NMERGE)
{
t = 0;
for (i = 0; i < nfiles / NMERGE; ++i)
{
for (j = 0; j < NMERGE; ++j)
fps[j] = xfopen (files[i * NMERGE + j], "r");
tfp = xtmpfopen (temp = tempname ());
mergefps (fps, NMERGE, tfp, temp);
xfclose (tfp);
for (j = 0; j < NMERGE; ++j)
zaptemp (files[i * NMERGE + j]);
files[t++] = temp;
}
for (j = 0; j < nfiles % NMERGE; ++j)
fps[j] = xfopen (files[i * NMERGE + j], "r");
tfp = xtmpfopen (temp = tempname ());
mergefps (fps, nfiles % NMERGE, tfp, temp);
xfclose (tfp);
for (j = 0; j < nfiles % NMERGE; ++j)
zaptemp (files[i * NMERGE + j]);
files[t++] = temp;
nfiles = t;
}
for (i = 0; i < nfiles; ++i)
fps[i] = xfopen (files[i], "r");
mergefps (fps, i, ofp, output_file);
for (i = 0; i < nfiles; ++i)
zaptemp (files[i]);
}
/* Sort NFILES FILES onto OFP. */
static void
sort (char **files, int nfiles, FILE *ofp, const char *output_file)
{
struct buffer buf;
struct lines lines;
struct line *tmp;
int i, ntmp;
FILE *fp, *tfp;
struct tempnode *node;
int n_temp_files = 0;
char **tempfiles;
initbuf (&buf, sortalloc);
initlines (&lines, sortalloc / linelength + 1,
LINEALLOC / sizeof (struct line));
ntmp = lines.alloc;
tmp = (struct line *) xmalloc (ntmp * sizeof (struct line));
while (nfiles--)
{
const char *temp_output;
fp = xfopen (*files++, "r");
while (fillbuf (&buf, fp))
{
if (nfiles && buf.used != buf.alloc && feof (fp))
{
/* End of file, but there is more input and buffer room.
Concatenate the next input file; this is faster in
the usual case. */
buf.left = buf.used;
break;
}
findlines (&buf, &lines);
if (ntmp < lines.used)
{
while ((ntmp *= 2) < lines.used)
continue;
tmp = (struct line *)
xrealloc ((char *) tmp, ntmp * sizeof (struct line));
}
sortlines (lines.lines, lines.used, tmp);
if (feof (fp) && !nfiles && !n_temp_files && !buf.left)
{
tfp = ofp;
temp_output = output_file;
}
else
{
++n_temp_files;
tfp = xtmpfopen (temp_output = tempname ());
}
for (i = 0; i < lines.used; ++i)
if (!unique || i == 0
|| compare (&lines.lines[i], &lines.lines[i - 1]))
write_bytes (lines.lines[i].text, lines.lines[i].length, tfp,
temp_output);
if (tfp != ofp)
xfclose (tfp);
}
xfclose (fp);
}
free (buf.buf);
free ((char *) lines.lines);
free ((char *) tmp);
if (n_temp_files)
{
tempfiles = (char **) xmalloc (n_temp_files * sizeof (char *));
i = n_temp_files;
for (node = temphead.next; i > 0; node = node->next)
tempfiles[--i] = node->name;
merge (tempfiles, n_temp_files, ofp, output_file);
free ((char *) tempfiles);
}
}
/* Insert key KEY at the end of the key list. */
static void
insertkey (struct keyfield *key)
{
struct keyfield **p;
for (p = &keylist; *p; p = &(*p)->next)
continue;
*p = key;
key->next = NULL;
}
static void
badfieldspec (const char *s)
{
error (SORT_FAILURE, 0, _("invalid field specification `%s'"), s);
}
/* Handle interrupts and hangups. */
static void
sighandler (int sig)
{
#ifdef SA_NOCLDSTOP
struct sigaction sigact;
sigact.sa_handler = SIG_DFL;
sigemptyset (&sigact.sa_mask);
sigact.sa_flags = 0;
sigaction (sig, &sigact, NULL);
#else
signal (sig, SIG_DFL);
#endif
cleanup ();
kill (process_id, sig);
}
/* Set the ordering options for KEY specified in S.
Return the address of the first character in S that
is not a valid ordering option.
BLANKTYPE is the kind of blanks that 'b' should skip. */
static char *
set_ordering (register const char *s, struct keyfield *key,
enum blanktype blanktype)
{
while (*s)
{
switch (*s)
{
case 'b':
if (blanktype == bl_start || blanktype == bl_both)
key->skipsblanks = 1;
if (blanktype == bl_end || blanktype == bl_both)
key->skipeblanks = 1;
break;
case 'd':
key->ignore = nondictionary;
break;
case 'f':
key->translate = fold_toupper;
break;
case 'g':
key->general_numeric = 1;
break;
case 'i':
key->ignore = nonprinting;
break;
case 'M':
key->month = 1;
break;
case 'n':
key->numeric = 1;
break;
case 'r':
key->reverse = 1;
break;
default:
return (char *) s;
}
++s;
}
return (char *) s;
}
static void
key_init (struct keyfield *key)
{
memset (key, 0, sizeof (*key));
key->eword = -1;
}
int
main (int argc, char **argv)
{
struct keyfield *key = NULL, gkey;
char *s;
int i, t, t2;
int checkonly = 0, mergeonly = 0, nfiles = 0;
char *minus = "-", *outfile = minus, **files, *tmp;
FILE *ofp;
#ifdef SA_NOCLDSTOP
struct sigaction oldact, newact;
#endif
program_name = argv[0];
process_id = getpid ();
setlocale (LC_ALL, "");
bindtextdomain (PACKAGE, LOCALEDIR);
textdomain (PACKAGE);
close_stdout_set_status (SORT_FAILURE);
atexit (close_stdout);
#ifdef ENABLE_NLS
hard_LC_COLLATE = hard_locale (LC_COLLATE);
# if HAVE_NL_LANGINFO
hard_LC_TIME = hard_locale (LC_TIME);
# endif
/* Let's get locale's representation of the decimal point */
{
struct lconv *lconvp = localeconv ();
/* If the locale doesn't define a decimal point, or if the decimal
point is multibyte, use the C decimal point. We don't support
multibyte decimal points yet. */
decimal_point = *lconvp->decimal_point;
if (! decimal_point || lconvp->decimal_point[1])
decimal_point = C_DECIMAL_POINT;
/* We don't support multibyte thousands separators yet. */
th_sep = *lconvp->thousands_sep;
if (! th_sep || lconvp->thousands_sep[1])
th_sep = CHAR_MAX + 1;
}
#endif /* NLS */
parse_long_options (argc, argv, PROGRAM_NAME, GNU_PACKAGE, VERSION,
AUTHORS, usage);
have_read_stdin = 0;
inittables ();
/* Change the way xmalloc and xrealloc fail. */
xalloc_exit_failure = SORT_FAILURE;
xalloc_fail_func = cleanup;
#ifdef SA_NOCLDSTOP
newact.sa_handler = sighandler;
sigemptyset (&newact.sa_mask);
newact.sa_flags = 0;
sigaction (SIGINT, NULL, &oldact);
if (oldact.sa_handler != SIG_IGN)
sigaction (SIGINT, &newact, NULL);
sigaction (SIGHUP, NULL, &oldact);
if (oldact.sa_handler != SIG_IGN)
sigaction (SIGHUP, &newact, NULL);
sigaction (SIGPIPE, NULL, &oldact);
if (oldact.sa_handler != SIG_IGN)
sigaction (SIGPIPE, &newact, NULL);
sigaction (SIGTERM, NULL, &oldact);
if (oldact.sa_handler != SIG_IGN)
sigaction (SIGTERM, &newact, NULL);
#else
if (signal (SIGINT, SIG_IGN) != SIG_IGN)
signal (SIGINT, sighandler);
if (signal (SIGHUP, SIG_IGN) != SIG_IGN)
signal (SIGHUP, sighandler);
if (signal (SIGPIPE, SIG_IGN) != SIG_IGN)
signal (SIGPIPE, sighandler);
if (signal (SIGTERM, SIG_IGN) != SIG_IGN)
signal (SIGTERM, sighandler);
#endif
gkey.sword = gkey.eword = -1;
gkey.ignore = NULL;
gkey.translate = NULL;
gkey.numeric = gkey.general_numeric = gkey.month = gkey.reverse = 0;
gkey.skipsblanks = gkey.skipeblanks = 0;
files = (char **) xmalloc (sizeof (char *) * argc);
for (i = 1; i < argc; ++i)
{
if (argv[i][0] == '+')
{
if (key)
insertkey (key);
key = (struct keyfield *) xmalloc (sizeof (struct keyfield));
key_init (key);
s = argv[i] + 1;
if (! (ISDIGIT (*s) || (*s == '.' && ISDIGIT (s[1]))))
badfieldspec (argv[i]);
for (t = 0; ISDIGIT (*s); ++s)
t = 10 * t + *s - '0';
t2 = 0;
if (*s == '.')
for (++s; ISDIGIT (*s); ++s)
t2 = 10 * t2 + *s - '0';
if (t2 || t)
{
key->sword = t;
key->schar = t2;
}
else
key->sword = -1;
s = set_ordering (s, key, bl_start);
if (*s)
badfieldspec (argv[i]);
}
else if (argv[i][0] == '-' && argv[i][1])
{
s = argv[i] + 1;
if (ISDIGIT (*s) || (*s == '.' && ISDIGIT (s[1])))
{
if (!key)
{
/* Provoke with `sort -9'. */
error (0, 0, _("when using the old-style +POS and -POS \
key specifiers,\nthe +POS specifier must come first"));
usage (SORT_FAILURE);
}
for (t = 0; ISDIGIT (*s); ++s)
t = t * 10 + *s - '0';
t2 = 0;
if (*s == '.')
for (++s; ISDIGIT (*s); ++s)
t2 = t2 * 10 + *s - '0';
key->eword = t;
key->echar = t2;
s = set_ordering (s, key, bl_end);
if (*s)
badfieldspec (argv[i]);
insertkey (key);
key = NULL;
}
else
while (*s)
{
s = set_ordering (s, &gkey, bl_both);
switch (*s)
{
case '\0':
break;
case 'c':
checkonly = 1;
break;
case 'k':
if (s[1])
++s;
else
{
if (i == argc - 1)
error (SORT_FAILURE, 0,
_("option `-k' requires an argument"));
else
s = argv[++i];
}
if (key)
insertkey (key);
key = (struct keyfield *)
xmalloc (sizeof (struct keyfield));
key_init (key);
/* Get POS1. */
if (!ISDIGIT (*s))
badfieldspec (argv[i]);
for (t = 0; ISDIGIT (*s); ++s)
t = 10 * t + *s - '0';
if (t == 0)
{
/* Provoke with `sort -k0' */
error (0, 0, _("the starting field number argument \
to the `-k' option must be positive"));
badfieldspec (argv[i]);
}
--t;
t2 = 0;
if (*s == '.')
{
if (!ISDIGIT (s[1]))
{
/* Provoke with `sort -k1.' */
error (0, 0, _("starting field spec has `.' but \
lacks following character offset"));
badfieldspec (argv[i]);
}
for (++s; ISDIGIT (*s); ++s)
t2 = 10 * t2 + *s - '0';
if (t2 == 0)
{
/* Provoke with `sort -k1.0' */
error (0, 0, _("starting field character offset \
argument to the `-k' option\nmust be positive"));
badfieldspec (argv[i]);
}
--t2;
}
if (t2 || t)
{
key->sword = t;
key->schar = t2;
}
else
key->sword = -1;
s = set_ordering (s, key, bl_start);
if (*s == 0)
{
key->eword = -1;
key->echar = 0;
}
else if (*s != ',')
badfieldspec (argv[i]);
else if (*s == ',')
{
/* Skip over comma. */
++s;
if (*s == 0)
{
/* Provoke with `sort -k1,' */
error (0, 0, _("field specification has `,' but \
lacks following field spec"));
badfieldspec (argv[i]);
}
/* Get POS2. */
for (t = 0; ISDIGIT (*s); ++s)
t = t * 10 + *s - '0';
if (t == 0)
{
/* Provoke with `sort -k1,0' */
error (0, 0, _("ending field number argument \
to the `-k' option must be positive"));
badfieldspec (argv[i]);
}
--t;
t2 = 0;
if (*s == '.')
{
if (!ISDIGIT (s[1]))
{
/* Provoke with `sort -k1,1.' */
error (0, 0, _("ending field spec has `.' \
but lacks following character offset"));
badfieldspec (argv[i]);
}
for (++s; ISDIGIT (*s); ++s)
t2 = t2 * 10 + *s - '0';
}
else
{
/* `-k 2,3' is equivalent to `+1 -3'. */
++t;
}
key->eword = t;
key->echar = t2;
s = set_ordering (s, key, bl_end);
if (*s)
badfieldspec (argv[i]);
}
insertkey (key);
key = NULL;
goto outer;
case 'm':
mergeonly = 1;
break;
case 'o':
if (s[1])
outfile = s + 1;
else
{
if (i == argc - 1)
error (SORT_FAILURE, 0,
_("option `-o' requires an argument"));
else
outfile = argv[++i];
}
close_stdout_set_file_name (outfile);
goto outer;
case 's':
stable = 1;
break;
case 't':
if (s[1])
tab = *++s;
else if (i < argc - 1)
{
tab = *argv[++i];
goto outer;
}
else
error (SORT_FAILURE, 0,
_("option `-t' requires an argument"));
break;
case 'T':
if (s[1])
add_temp_dir (++s);
else
{
if (i < argc - 1)
add_temp_dir (argv[++i]);
else
error (SORT_FAILURE, 0,
_("option `-T' requires an argument"));
}
goto outer;
/* break; */
case 'u':
unique = 1;
break;
case 'z':
eolchar = 0;
break;
case 'y':
/* Accept and ignore e.g. -y0 for compatibility with
Solaris 2. */
goto outer;
default:
fprintf (stderr, _("%s: unrecognized option `-%c'\n"),
argv[0], *s);
usage (SORT_FAILURE);
}
if (*s)
++s;
}
}
else /* Not an option. */
{
files[nfiles++] = argv[i];
}
outer:;
}
if (key)
insertkey (key);
/* Inheritance of global options to individual keys. */
for (key = keylist; key; key = key->next)
if (!key->ignore && !key->translate && !key->skipsblanks && !key->reverse
&& !key->skipeblanks && !key->month && !key->numeric
&& !key->general_numeric)
{
key->ignore = gkey.ignore;
key->translate = gkey.translate;
key->skipsblanks = gkey.skipsblanks;
key->skipeblanks = gkey.skipeblanks;
key->month = gkey.month;
key->numeric = gkey.numeric;
key->general_numeric = gkey.general_numeric;
key->reverse = gkey.reverse;
}
if (!keylist && (gkey.ignore || gkey.translate || gkey.skipsblanks
|| gkey.skipeblanks || gkey.month || gkey.numeric
|| gkey.general_numeric))
insertkey (&gkey);
reverse = gkey.reverse;
if (temp_dir_count == 0)
{
char const *tmp_dir = getenv ("TMPDIR");
add_temp_dir (tmp_dir ? tmp_dir : DEFAULT_TMPDIR);
}
if (nfiles == 0)
{
nfiles = 1;
files = −
}
if (checkonly)
{
if (nfiles > 1)
error (SORT_FAILURE, 0,
_("too many arguments; with -c, there may be at most\
one file argument"));
/* POSIX requires that sort return 1 IFF invoked with -c and the
input is not properly sorted. */
exit (check (files, nfiles) == 0 ? EXIT_SUCCESS : SORT_OUT_OF_ORDER);
}
if (!STREQ (outfile, "-"))
{
struct stat outstat;
if (stat (outfile, &outstat) == 0)
{
/* FIXME: warn about this */
/* The following code prevents a race condition when
people use the brain dead shell programming idiom:
cat file | sort -o file
This feature is provided for historical compatibility,
but we strongly discourage ever relying on this in
new shell programs. */
/* Temporarily copy each input file that might be another name
for the output file. When in doubt (e.g. a pipe), copy. */
for (i = 0; i < nfiles; ++i)
{
char buf[8192];
FILE *in_fp;
FILE *out_fp;
int cc;
if (S_ISREG (outstat.st_mode) && !STREQ (outfile, files[i]))
{
struct stat instat;
if ((STREQ (files[i], "-")
? fstat (STDIN_FILENO, &instat)
: stat (files[i], &instat)) != 0)
{
error (0, errno, "%s", files[i]);
cleanup ();
exit (SORT_FAILURE);
}
if (S_ISREG (instat.st_mode) && !SAME_INODE (instat, outstat))
{
/* We know the files are distinct. */
continue;
}
}
in_fp = xfopen (files[i], "r");
tmp = tempname ();
out_fp = xtmpfopen (tmp);
/* FIXME: maybe use copy.c(copy) here. */
while ((cc = fread (buf, 1, sizeof buf, in_fp)) > 0)
write_bytes (buf, cc, out_fp, tmp);
if (ferror (in_fp))
{
error (0, errno, "%s", files[i]);
cleanup ();
exit (SORT_FAILURE);
}
xfclose (out_fp);
xfclose (in_fp);
files[i] = tmp;
}
ofp = xfopen (outfile, "w");
}
else
{
/* A non-`-' outfile was specified, but the file doesn't yet exist.
Before opening it for writing (thus creating it), make sure all
of the input files exist. Otherwise, creating the output file
could create an otherwise missing input file, making sort succeed
when it should fail. */
for (i = 0; i < nfiles; ++i)
{
struct stat sb;
if (STREQ (files[i], "-"))
continue;
if (stat (files[i], &sb))
{
error (0, errno, "%s", files[i]);
cleanup ();
exit (SORT_FAILURE);
}
}
ofp = xfopen (outfile, "w");
}
}
else
{
ofp = stdout;
}
if (mergeonly)
merge (files, nfiles, ofp, outfile);
else
sort (files, nfiles, ofp, outfile);
cleanup ();
/* If we wait for the implicit flush on exit, and the parent process
has closed stdout (e.g., exec >&- in a shell), then the output file
winds up empty. I don't understand why. This is under SunOS,
Solaris, Ultrix, and Irix. This premature fflush makes the output
reappear. --karl@cs.umb.edu */
if (fflush (ofp) < 0)
error (SORT_FAILURE, errno, _("%s: write error"), outfile);
if (have_read_stdin && fclose (stdin) == EOF)
error (SORT_FAILURE, errno, "%s", outfile);
exit (EXIT_SUCCESS);
}
|