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
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
|
/* File object implementation */
#define PY_SSIZE_T_CLEAN
#include "Python.h"
#include "structmember.h"
#ifdef HAVE_SYS_TYPES_H
#include <sys/types.h>
#endif /* HAVE_SYS_TYPES_H */
#ifdef MS_WINDOWS
#define fileno _fileno
/* can simulate truncate with Win32 API functions; see file_truncate */
#define HAVE_FTRUNCATE
#define WIN32_LEAN_AND_MEAN
#include <windows.h>
#endif
#ifdef _MSC_VER
/* Need GetVersion to see if on NT so safe to use _wfopen */
#define WIN32_LEAN_AND_MEAN
#include <windows.h>
#endif /* _MSC_VER */
#if defined(PYOS_OS2) && defined(PYCC_GCC)
#include <io.h>
#endif
#define BUF(v) PyString_AS_STRING((PyStringObject *)v)
#ifndef DONT_HAVE_ERRNO_H
#include <errno.h>
#endif
#ifdef HAVE_GETC_UNLOCKED
#define GETC(f) getc_unlocked(f)
#define FLOCKFILE(f) flockfile(f)
#define FUNLOCKFILE(f) funlockfile(f)
#else
#define GETC(f) getc(f)
#define FLOCKFILE(f)
#define FUNLOCKFILE(f)
#endif
/* Bits in f_newlinetypes */
#define NEWLINE_UNKNOWN 0 /* No newline seen, yet */
#define NEWLINE_CR 1 /* \r newline seen */
#define NEWLINE_LF 2 /* \n newline seen */
#define NEWLINE_CRLF 4 /* \r\n newline seen */
/*
* These macros release the GIL while preventing the f_close() function being
* called in the interval between them. For that purpose, a running total of
* the number of currently running unlocked code sections is kept in
* the unlocked_count field of the PyFileObject. The close() method raises
* an IOError if that field is non-zero. See issue #815646, #595601.
*/
#define FILE_BEGIN_ALLOW_THREADS(fobj) \
{ \
fobj->unlocked_count++; \
Py_BEGIN_ALLOW_THREADS
#define FILE_END_ALLOW_THREADS(fobj) \
Py_END_ALLOW_THREADS \
fobj->unlocked_count--; \
assert(fobj->unlocked_count >= 0); \
}
#define FILE_ABORT_ALLOW_THREADS(fobj) \
Py_BLOCK_THREADS \
fobj->unlocked_count--; \
assert(fobj->unlocked_count >= 0);
#ifdef __cplusplus
extern "C" {
#endif
FILE *
PyFile_AsFile(PyObject *f)
{
if (f == NULL || !PyFile_Check(f))
return NULL;
else
return ((PyFileObject *)f)->f_fp;
}
void PyFile_IncUseCount(PyFileObject *fobj)
{
fobj->unlocked_count++;
}
void PyFile_DecUseCount(PyFileObject *fobj)
{
fobj->unlocked_count--;
assert(fobj->unlocked_count >= 0);
}
PyObject *
PyFile_Name(PyObject *f)
{
if (f == NULL || !PyFile_Check(f))
return NULL;
else
return ((PyFileObject *)f)->f_name;
}
/* This is a safe wrapper around PyObject_Print to print to the FILE
of a PyFileObject. PyObject_Print releases the GIL but knows nothing
about PyFileObject. */
static int
file_PyObject_Print(PyObject *op, PyFileObject *f, int flags)
{
int result;
PyFile_IncUseCount(f);
result = PyObject_Print(op, f->f_fp, flags);
PyFile_DecUseCount(f);
return result;
}
/* On Unix, fopen will succeed for directories.
In Python, there should be no file objects referring to
directories, so we need a check. */
static PyFileObject*
dircheck(PyFileObject* f)
{
#if defined(HAVE_FSTAT) && defined(S_IFDIR) && defined(EISDIR)
struct stat buf;
if (f->f_fp == NULL)
return f;
if (fstat(fileno(f->f_fp), &buf) == 0 &&
S_ISDIR(buf.st_mode)) {
char *msg = strerror(EISDIR);
PyObject *exc = PyObject_CallFunction(PyExc_IOError, "(is)",
EISDIR, msg);
PyErr_SetObject(PyExc_IOError, exc);
Py_XDECREF(exc);
return NULL;
}
#endif
return f;
}
static PyObject *
fill_file_fields(PyFileObject *f, FILE *fp, PyObject *name, char *mode,
int (*close)(FILE *))
{
assert(name != NULL);
assert(f != NULL);
assert(PyFile_Check(f));
assert(f->f_fp == NULL);
Py_DECREF(f->f_name);
Py_DECREF(f->f_mode);
Py_DECREF(f->f_encoding);
Py_DECREF(f->f_errors);
Py_INCREF(name);
f->f_name = name;
f->f_mode = PyString_FromString(mode);
f->f_close = close;
f->f_softspace = 0;
f->f_binary = strchr(mode,'b') != NULL;
f->f_buf = NULL;
f->f_univ_newline = (strchr(mode, 'U') != NULL);
f->f_newlinetypes = NEWLINE_UNKNOWN;
f->f_skipnextlf = 0;
Py_INCREF(Py_None);
f->f_encoding = Py_None;
Py_INCREF(Py_None);
f->f_errors = Py_None;
if (f->f_mode == NULL)
return NULL;
f->f_fp = fp;
f = dircheck(f);
return (PyObject *) f;
}
/* check for known incorrect mode strings - problem is, platforms are
free to accept any mode characters they like and are supposed to
ignore stuff they don't understand... write or append mode with
universal newline support is expressly forbidden by PEP 278.
Additionally, remove the 'U' from the mode string as platforms
won't know what it is. Non-zero return signals an exception */
int
_PyFile_SanitizeMode(char *mode)
{
char *upos;
size_t len = strlen(mode);
if (!len) {
PyErr_SetString(PyExc_ValueError, "empty mode string");
return -1;
}
upos = strchr(mode, 'U');
if (upos) {
memmove(upos, upos+1, len-(upos-mode)); /* incl null char */
if (mode[0] == 'w' || mode[0] == 'a') {
PyErr_Format(PyExc_ValueError, "universal newline "
"mode can only be used with modes "
"starting with 'r'");
return -1;
}
if (mode[0] != 'r') {
memmove(mode+1, mode, strlen(mode)+1);
mode[0] = 'r';
}
if (!strchr(mode, 'b')) {
memmove(mode+2, mode+1, strlen(mode));
mode[1] = 'b';
}
} else if (mode[0] != 'r' && mode[0] != 'w' && mode[0] != 'a') {
PyErr_Format(PyExc_ValueError, "mode string must begin with "
"one of 'r', 'w', 'a' or 'U', not '%.200s'", mode);
return -1;
}
return 0;
}
static PyObject *
open_the_file(PyFileObject *f, char *name, char *mode)
{
char *newmode;
assert(f != NULL);
assert(PyFile_Check(f));
#ifdef MS_WINDOWS
/* windows ignores the passed name in order to support Unicode */
assert(f->f_name != NULL);
#else
assert(name != NULL);
#endif
assert(mode != NULL);
assert(f->f_fp == NULL);
/* probably need to replace 'U' by 'rb' */
newmode = PyMem_MALLOC(strlen(mode) + 3);
if (!newmode) {
PyErr_NoMemory();
return NULL;
}
strcpy(newmode, mode);
if (_PyFile_SanitizeMode(newmode)) {
f = NULL;
goto cleanup;
}
/* rexec.py can't stop a user from getting the file() constructor --
all they have to do is get *any* file object f, and then do
type(f). Here we prevent them from doing damage with it. */
if (PyEval_GetRestricted()) {
PyErr_SetString(PyExc_IOError,
"file() constructor not accessible in restricted mode");
f = NULL;
goto cleanup;
}
errno = 0;
#ifdef MS_WINDOWS
if (PyUnicode_Check(f->f_name)) {
PyObject *wmode;
wmode = PyUnicode_DecodeASCII(newmode, strlen(newmode), NULL);
if (f->f_name && wmode) {
FILE_BEGIN_ALLOW_THREADS(f)
/* PyUnicode_AS_UNICODE OK without thread
lock as it is a simple dereference. */
f->f_fp = _wfopen(PyUnicode_AS_UNICODE(f->f_name),
PyUnicode_AS_UNICODE(wmode));
FILE_END_ALLOW_THREADS(f)
}
Py_XDECREF(wmode);
}
#endif
if (NULL == f->f_fp && NULL != name) {
FILE_BEGIN_ALLOW_THREADS(f)
f->f_fp = fopen(name, newmode);
FILE_END_ALLOW_THREADS(f)
}
if (f->f_fp == NULL) {
#if defined _MSC_VER && (_MSC_VER < 1400 || !defined(__STDC_SECURE_LIB__))
/* MSVC 6 (Microsoft) leaves errno at 0 for bad mode strings,
* across all Windows flavors. When it sets EINVAL varies
* across Windows flavors, the exact conditions aren't
* documented, and the answer lies in the OS's implementation
* of Win32's CreateFile function (whose source is secret).
* Seems the best we can do is map EINVAL to ENOENT.
* Starting with Visual Studio .NET 2005, EINVAL is correctly
* set by our CRT error handler (set in exceptions.c.)
*/
if (errno == 0) /* bad mode string */
errno = EINVAL;
else if (errno == EINVAL) /* unknown, but not a mode string */
errno = ENOENT;
#endif
/* EINVAL is returned when an invalid filename or
* an invalid mode is supplied. */
if (errno == EINVAL)
PyErr_Format(PyExc_IOError,
"invalid filename: %s or mode: %s",
name, mode);
else
PyErr_SetFromErrnoWithFilenameObject(PyExc_IOError, f->f_name);
f = NULL;
}
if (f != NULL)
f = dircheck(f);
cleanup:
PyMem_FREE(newmode);
return (PyObject *)f;
}
static PyObject *
close_the_file(PyFileObject *f)
{
int sts = 0;
int (*local_close)(FILE *);
FILE *local_fp = f->f_fp;
if (local_fp != NULL) {
local_close = f->f_close;
if (local_close != NULL && f->unlocked_count > 0) {
if (f->ob_refcnt > 0) {
PyErr_SetString(PyExc_IOError,
"close() called during concurrent "
"operation on the same file object.");
} else {
/* This should not happen unless someone is
* carelessly playing with the PyFileObject
* struct fields and/or its associated FILE
* pointer. */
PyErr_SetString(PyExc_SystemError,
"PyFileObject locking error in "
"destructor (refcnt <= 0 at close).");
}
return NULL;
}
/* NULL out the FILE pointer before releasing the GIL, because
* it will not be valid anymore after the close() function is
* called. */
f->f_fp = NULL;
if (local_close != NULL) {
Py_BEGIN_ALLOW_THREADS
errno = 0;
sts = (*local_close)(local_fp);
Py_END_ALLOW_THREADS
if (sts == EOF)
return PyErr_SetFromErrno(PyExc_IOError);
if (sts != 0)
return PyInt_FromLong((long)sts);
}
}
Py_RETURN_NONE;
}
PyObject *
PyFile_FromFile(FILE *fp, char *name, char *mode, int (*close)(FILE *))
{
PyFileObject *f = (PyFileObject *)PyFile_Type.tp_new(&PyFile_Type,
NULL, NULL);
if (f != NULL) {
PyObject *o_name = PyString_FromString(name);
if (o_name == NULL)
return NULL;
if (fill_file_fields(f, fp, o_name, mode, close) == NULL) {
Py_DECREF(f);
f = NULL;
}
Py_DECREF(o_name);
}
return (PyObject *) f;
}
PyObject *
PyFile_FromString(char *name, char *mode)
{
extern int fclose(FILE *);
PyFileObject *f;
f = (PyFileObject *)PyFile_FromFile((FILE *)NULL, name, mode, fclose);
if (f != NULL) {
if (open_the_file(f, name, mode) == NULL) {
Py_DECREF(f);
f = NULL;
}
}
return (PyObject *)f;
}
void
PyFile_SetBufSize(PyObject *f, int bufsize)
{
PyFileObject *file = (PyFileObject *)f;
if (bufsize >= 0) {
int type;
switch (bufsize) {
case 0:
type = _IONBF;
break;
#ifdef HAVE_SETVBUF
case 1:
type = _IOLBF;
bufsize = BUFSIZ;
break;
#endif
default:
type = _IOFBF;
#ifndef HAVE_SETVBUF
bufsize = BUFSIZ;
#endif
break;
}
fflush(file->f_fp);
if (type == _IONBF) {
PyMem_Free(file->f_setbuf);
file->f_setbuf = NULL;
} else {
file->f_setbuf = (char *)PyMem_Realloc(file->f_setbuf,
bufsize);
}
#ifdef HAVE_SETVBUF
setvbuf(file->f_fp, file->f_setbuf, type, bufsize);
#else /* !HAVE_SETVBUF */
setbuf(file->f_fp, file->f_setbuf);
#endif /* !HAVE_SETVBUF */
}
}
/* Set the encoding used to output Unicode strings.
Return 1 on success, 0 on failure. */
int
PyFile_SetEncoding(PyObject *f, const char *enc)
{
return PyFile_SetEncodingAndErrors(f, enc, NULL);
}
int
PyFile_SetEncodingAndErrors(PyObject *f, const char *enc, char* errors)
{
PyFileObject *file = (PyFileObject*)f;
PyObject *str, *oerrors;
assert(PyFile_Check(f));
str = PyString_FromString(enc);
if (!str)
return 0;
if (errors) {
oerrors = PyString_FromString(errors);
if (!oerrors) {
Py_DECREF(str);
return 0;
}
} else {
oerrors = Py_None;
Py_INCREF(Py_None);
}
Py_DECREF(file->f_encoding);
file->f_encoding = str;
Py_DECREF(file->f_errors);
file->f_errors = oerrors;
return 1;
}
static PyObject *
err_closed(void)
{
PyErr_SetString(PyExc_ValueError, "I/O operation on closed file");
return NULL;
}
/* Refuse regular file I/O if there's data in the iteration-buffer.
* Mixing them would cause data to arrive out of order, as the read*
* methods don't use the iteration buffer. */
static PyObject *
err_iterbuffered(void)
{
PyErr_SetString(PyExc_ValueError,
"Mixing iteration and read methods would lose data");
return NULL;
}
static void drop_readahead(PyFileObject *);
/* Methods */
static void
file_dealloc(PyFileObject *f)
{
PyObject *ret;
if (f->weakreflist != NULL)
PyObject_ClearWeakRefs((PyObject *) f);
ret = close_the_file(f);
if (!ret) {
PySys_WriteStderr("close failed in file object destructor:\n");
PyErr_Print();
}
else {
Py_DECREF(ret);
}
PyMem_Free(f->f_setbuf);
Py_XDECREF(f->f_name);
Py_XDECREF(f->f_mode);
Py_XDECREF(f->f_encoding);
Py_XDECREF(f->f_errors);
drop_readahead(f);
Py_TYPE(f)->tp_free((PyObject *)f);
}
static PyObject *
file_repr(PyFileObject *f)
{
if (PyUnicode_Check(f->f_name)) {
#ifdef Py_USING_UNICODE
PyObject *ret = NULL;
PyObject *name = PyUnicode_AsUnicodeEscapeString(f->f_name);
const char *name_str = name ? PyString_AsString(name) : "?";
ret = PyString_FromFormat("<%s file u'%s', mode '%s' at %p>",
f->f_fp == NULL ? "closed" : "open",
name_str,
PyString_AsString(f->f_mode),
f);
Py_XDECREF(name);
return ret;
#endif
} else {
return PyString_FromFormat("<%s file '%s', mode '%s' at %p>",
f->f_fp == NULL ? "closed" : "open",
PyString_AsString(f->f_name),
PyString_AsString(f->f_mode),
f);
}
}
static PyObject *
file_close(PyFileObject *f)
{
PyObject *sts = close_the_file(f);
PyMem_Free(f->f_setbuf);
f->f_setbuf = NULL;
return sts;
}
/* Our very own off_t-like type, 64-bit if possible */
#if !defined(HAVE_LARGEFILE_SUPPORT)
typedef off_t Py_off_t;
#elif SIZEOF_OFF_T >= 8
typedef off_t Py_off_t;
#elif SIZEOF_FPOS_T >= 8
typedef fpos_t Py_off_t;
#else
#error "Large file support, but neither off_t nor fpos_t is large enough."
#endif
/* a portable fseek() function
return 0 on success, non-zero on failure (with errno set) */
static int
_portable_fseek(FILE *fp, Py_off_t offset, int whence)
{
#if !defined(HAVE_LARGEFILE_SUPPORT)
return fseek(fp, offset, whence);
#elif defined(HAVE_FSEEKO) && SIZEOF_OFF_T >= 8
return fseeko(fp, offset, whence);
#elif defined(HAVE_FSEEK64)
return fseek64(fp, offset, whence);
#elif defined(__BEOS__)
return _fseek(fp, offset, whence);
#elif SIZEOF_FPOS_T >= 8
/* lacking a 64-bit capable fseek(), use a 64-bit capable fsetpos()
and fgetpos() to implement fseek()*/
fpos_t pos;
switch (whence) {
case SEEK_END:
#ifdef MS_WINDOWS
fflush(fp);
if (_lseeki64(fileno(fp), 0, 2) == -1)
return -1;
#else
if (fseek(fp, 0, SEEK_END) != 0)
return -1;
#endif
/* fall through */
case SEEK_CUR:
if (fgetpos(fp, &pos) != 0)
return -1;
offset += pos;
break;
/* case SEEK_SET: break; */
}
return fsetpos(fp, &offset);
#else
#error "Large file support, but no way to fseek."
#endif
}
/* a portable ftell() function
Return -1 on failure with errno set appropriately, current file
position on success */
static Py_off_t
_portable_ftell(FILE* fp)
{
#if !defined(HAVE_LARGEFILE_SUPPORT)
return ftell(fp);
#elif defined(HAVE_FTELLO) && SIZEOF_OFF_T >= 8
return ftello(fp);
#elif defined(HAVE_FTELL64)
return ftell64(fp);
#elif SIZEOF_FPOS_T >= 8
fpos_t pos;
if (fgetpos(fp, &pos) != 0)
return -1;
return pos;
#else
#error "Large file support, but no way to ftell."
#endif
}
static PyObject *
file_seek(PyFileObject *f, PyObject *args)
{
int whence;
int ret;
Py_off_t offset;
PyObject *offobj, *off_index;
if (f->f_fp == NULL)
return err_closed();
drop_readahead(f);
whence = 0;
if (!PyArg_ParseTuple(args, "O|i:seek", &offobj, &whence))
return NULL;
off_index = PyNumber_Index(offobj);
if (!off_index) {
if (!PyFloat_Check(offobj))
return NULL;
/* Deprecated in 2.6 */
PyErr_Clear();
if (PyErr_WarnEx(PyExc_DeprecationWarning,
"integer argument expected, got float",
1) < 0)
return NULL;
off_index = offobj;
Py_INCREF(offobj);
}
#if !defined(HAVE_LARGEFILE_SUPPORT)
offset = PyInt_AsLong(off_index);
#else
offset = PyLong_Check(off_index) ?
PyLong_AsLongLong(off_index) : PyInt_AsLong(off_index);
#endif
Py_DECREF(off_index);
if (PyErr_Occurred())
return NULL;
FILE_BEGIN_ALLOW_THREADS(f)
errno = 0;
ret = _portable_fseek(f->f_fp, offset, whence);
FILE_END_ALLOW_THREADS(f)
if (ret != 0) {
PyErr_SetFromErrno(PyExc_IOError);
clearerr(f->f_fp);
return NULL;
}
f->f_skipnextlf = 0;
Py_INCREF(Py_None);
return Py_None;
}
#ifdef HAVE_FTRUNCATE
static PyObject *
file_truncate(PyFileObject *f, PyObject *args)
{
Py_off_t newsize;
PyObject *newsizeobj = NULL;
Py_off_t initialpos;
int ret;
if (f->f_fp == NULL)
return err_closed();
if (!PyArg_UnpackTuple(args, "truncate", 0, 1, &newsizeobj))
return NULL;
/* Get current file position. If the file happens to be open for
* update and the last operation was an input operation, C doesn't
* define what the later fflush() will do, but we promise truncate()
* won't change the current position (and fflush() *does* change it
* then at least on Windows). The easiest thing is to capture
* current pos now and seek back to it at the end.
*/
FILE_BEGIN_ALLOW_THREADS(f)
errno = 0;
initialpos = _portable_ftell(f->f_fp);
FILE_END_ALLOW_THREADS(f)
if (initialpos == -1)
goto onioerror;
/* Set newsize to current postion if newsizeobj NULL, else to the
* specified value.
*/
if (newsizeobj != NULL) {
#if !defined(HAVE_LARGEFILE_SUPPORT)
newsize = PyInt_AsLong(newsizeobj);
#else
newsize = PyLong_Check(newsizeobj) ?
PyLong_AsLongLong(newsizeobj) :
PyInt_AsLong(newsizeobj);
#endif
if (PyErr_Occurred())
return NULL;
}
else /* default to current position */
newsize = initialpos;
/* Flush the stream. We're mixing stream-level I/O with lower-level
* I/O, and a flush may be necessary to synch both platform views
* of the current file state.
*/
FILE_BEGIN_ALLOW_THREADS(f)
errno = 0;
ret = fflush(f->f_fp);
FILE_END_ALLOW_THREADS(f)
if (ret != 0)
goto onioerror;
#ifdef MS_WINDOWS
/* MS _chsize doesn't work if newsize doesn't fit in 32 bits,
so don't even try using it. */
{
HANDLE hFile;
/* Have to move current pos to desired endpoint on Windows. */
FILE_BEGIN_ALLOW_THREADS(f)
errno = 0;
ret = _portable_fseek(f->f_fp, newsize, SEEK_SET) != 0;
FILE_END_ALLOW_THREADS(f)
if (ret)
goto onioerror;
/* Truncate. Note that this may grow the file! */
FILE_BEGIN_ALLOW_THREADS(f)
errno = 0;
hFile = (HANDLE)_get_osfhandle(fileno(f->f_fp));
ret = hFile == (HANDLE)-1;
if (ret == 0) {
ret = SetEndOfFile(hFile) == 0;
if (ret)
errno = EACCES;
}
FILE_END_ALLOW_THREADS(f)
if (ret)
goto onioerror;
}
#else
FILE_BEGIN_ALLOW_THREADS(f)
errno = 0;
ret = ftruncate(fileno(f->f_fp), newsize);
FILE_END_ALLOW_THREADS(f)
if (ret != 0)
goto onioerror;
#endif /* !MS_WINDOWS */
/* Restore original file position. */
FILE_BEGIN_ALLOW_THREADS(f)
errno = 0;
ret = _portable_fseek(f->f_fp, initialpos, SEEK_SET) != 0;
FILE_END_ALLOW_THREADS(f)
if (ret)
goto onioerror;
Py_INCREF(Py_None);
return Py_None;
onioerror:
PyErr_SetFromErrno(PyExc_IOError);
clearerr(f->f_fp);
return NULL;
}
#endif /* HAVE_FTRUNCATE */
static PyObject *
file_tell(PyFileObject *f)
{
Py_off_t pos;
if (f->f_fp == NULL)
return err_closed();
FILE_BEGIN_ALLOW_THREADS(f)
errno = 0;
pos = _portable_ftell(f->f_fp);
FILE_END_ALLOW_THREADS(f)
if (pos == -1) {
PyErr_SetFromErrno(PyExc_IOError);
clearerr(f->f_fp);
return NULL;
}
if (f->f_skipnextlf) {
int c;
c = GETC(f->f_fp);
if (c == '\n') {
f->f_newlinetypes |= NEWLINE_CRLF;
pos++;
f->f_skipnextlf = 0;
} else if (c != EOF) ungetc(c, f->f_fp);
}
#if !defined(HAVE_LARGEFILE_SUPPORT)
return PyInt_FromLong(pos);
#else
return PyLong_FromLongLong(pos);
#endif
}
static PyObject *
file_fileno(PyFileObject *f)
{
if (f->f_fp == NULL)
return err_closed();
return PyInt_FromLong((long) fileno(f->f_fp));
}
static PyObject *
file_flush(PyFileObject *f)
{
int res;
if (f->f_fp == NULL)
return err_closed();
FILE_BEGIN_ALLOW_THREADS(f)
errno = 0;
res = fflush(f->f_fp);
FILE_END_ALLOW_THREADS(f)
if (res != 0) {
PyErr_SetFromErrno(PyExc_IOError);
clearerr(f->f_fp);
return NULL;
}
Py_INCREF(Py_None);
return Py_None;
}
static PyObject *
file_isatty(PyFileObject *f)
{
long res;
if (f->f_fp == NULL)
return err_closed();
FILE_BEGIN_ALLOW_THREADS(f)
res = isatty((int)fileno(f->f_fp));
FILE_END_ALLOW_THREADS(f)
return PyBool_FromLong(res);
}
#if BUFSIZ < 8192
#define SMALLCHUNK 8192
#else
#define SMALLCHUNK BUFSIZ
#endif
#if SIZEOF_INT < 4
#define BIGCHUNK (512 * 32)
#else
#define BIGCHUNK (512 * 1024)
#endif
static size_t
new_buffersize(PyFileObject *f, size_t currentsize)
{
#ifdef HAVE_FSTAT
off_t pos, end;
struct stat st;
if (fstat(fileno(f->f_fp), &st) == 0) {
end = st.st_size;
/* The following is not a bug: we really need to call lseek()
*and* ftell(). The reason is that some stdio libraries
mistakenly flush their buffer when ftell() is called and
the lseek() call it makes fails, thereby throwing away
data that cannot be recovered in any way. To avoid this,
we first test lseek(), and only call ftell() if lseek()
works. We can't use the lseek() value either, because we
need to take the amount of buffered data into account.
(Yet another reason why stdio stinks. :-) */
pos = lseek(fileno(f->f_fp), 0L, SEEK_CUR);
if (pos >= 0) {
pos = ftell(f->f_fp);
}
if (pos < 0)
clearerr(f->f_fp);
if (end > pos && pos >= 0)
return currentsize + end - pos + 1;
/* Add 1 so if the file were to grow we'd notice. */
}
#endif
if (currentsize > SMALLCHUNK) {
/* Keep doubling until we reach BIGCHUNK;
then keep adding BIGCHUNK. */
if (currentsize <= BIGCHUNK)
return currentsize + currentsize;
else
return currentsize + BIGCHUNK;
}
return currentsize + SMALLCHUNK;
}
#if defined(EWOULDBLOCK) && defined(EAGAIN) && EWOULDBLOCK != EAGAIN
#define BLOCKED_ERRNO(x) ((x) == EWOULDBLOCK || (x) == EAGAIN)
#else
#ifdef EWOULDBLOCK
#define BLOCKED_ERRNO(x) ((x) == EWOULDBLOCK)
#else
#ifdef EAGAIN
#define BLOCKED_ERRNO(x) ((x) == EAGAIN)
#else
#define BLOCKED_ERRNO(x) 0
#endif
#endif
#endif
static PyObject *
file_read(PyFileObject *f, PyObject *args)
{
long bytesrequested = -1;
size_t bytesread, buffersize, chunksize;
PyObject *v;
if (f->f_fp == NULL)
return err_closed();
/* refuse to mix with f.next() */
if (f->f_buf != NULL &&
(f->f_bufend - f->f_bufptr) > 0 &&
f->f_buf[0] != '\0')
return err_iterbuffered();
if (!PyArg_ParseTuple(args, "|l:read", &bytesrequested))
return NULL;
if (bytesrequested < 0)
buffersize = new_buffersize(f, (size_t)0);
else
buffersize = bytesrequested;
if (buffersize > PY_SSIZE_T_MAX) {
PyErr_SetString(PyExc_OverflowError,
"requested number of bytes is more than a Python string can hold");
return NULL;
}
v = PyString_FromStringAndSize((char *)NULL, buffersize);
if (v == NULL)
return NULL;
bytesread = 0;
for (;;) {
FILE_BEGIN_ALLOW_THREADS(f)
errno = 0;
chunksize = Py_UniversalNewlineFread(BUF(v) + bytesread,
buffersize - bytesread, f->f_fp, (PyObject *)f);
FILE_END_ALLOW_THREADS(f)
if (chunksize == 0) {
if (!ferror(f->f_fp))
break;
clearerr(f->f_fp);
/* When in non-blocking mode, data shouldn't
* be discarded if a blocking signal was
* received. That will also happen if
* chunksize != 0, but bytesread < buffersize. */
if (bytesread > 0 && BLOCKED_ERRNO(errno))
break;
PyErr_SetFromErrno(PyExc_IOError);
Py_DECREF(v);
return NULL;
}
bytesread += chunksize;
if (bytesread < buffersize) {
clearerr(f->f_fp);
break;
}
if (bytesrequested < 0) {
buffersize = new_buffersize(f, buffersize);
if (_PyString_Resize(&v, buffersize) < 0)
return NULL;
} else {
/* Got what was requested. */
break;
}
}
if (bytesread != buffersize)
_PyString_Resize(&v, bytesread);
return v;
}
static PyObject *
file_readinto(PyFileObject *f, PyObject *args)
{
char *ptr;
Py_ssize_t ntodo;
Py_ssize_t ndone, nnow;
Py_buffer pbuf;
if (f->f_fp == NULL)
return err_closed();
/* refuse to mix with f.next() */
if (f->f_buf != NULL &&
(f->f_bufend - f->f_bufptr) > 0 &&
f->f_buf[0] != '\0')
return err_iterbuffered();
if (!PyArg_ParseTuple(args, "w*", &pbuf))
return NULL;
ptr = pbuf.buf;
ntodo = pbuf.len;
ndone = 0;
while (ntodo > 0) {
FILE_BEGIN_ALLOW_THREADS(f)
errno = 0;
nnow = Py_UniversalNewlineFread(ptr+ndone, ntodo, f->f_fp,
(PyObject *)f);
FILE_END_ALLOW_THREADS(f)
if (nnow == 0) {
if (!ferror(f->f_fp))
break;
PyErr_SetFromErrno(PyExc_IOError);
clearerr(f->f_fp);
PyBuffer_Release(&pbuf);
return NULL;
}
ndone += nnow;
ntodo -= nnow;
}
PyBuffer_Release(&pbuf);
return PyInt_FromSsize_t(ndone);
}
/**************************************************************************
Routine to get next line using platform fgets().
Under MSVC 6:
+ MS threadsafe getc is very slow (multiple layers of function calls before+
after each character, to lock+unlock the stream).
+ The stream-locking functions are MS-internal -- can't access them from user
code.
+ There's nothing Tim could find in the MS C or platform SDK libraries that
can worm around this.
+ MS fgets locks/unlocks only once per line; it's the only hook we have.
So we use fgets for speed(!), despite that it's painful.
MS realloc is also slow.
Reports from other platforms on this method vs getc_unlocked (which MS doesn't
have):
Linux a wash
Solaris a wash
Tru64 Unix getline_via_fgets significantly faster
CAUTION: The C std isn't clear about this: in those cases where fgets
writes something into the buffer, can it write into any position beyond the
required trailing null byte? MSVC 6 fgets does not, and no platform is (yet)
known on which it does; and it would be a strange way to code fgets. Still,
getline_via_fgets may not work correctly if it does. The std test
test_bufio.py should fail if platform fgets() routinely writes beyond the
trailing null byte. #define DONT_USE_FGETS_IN_GETLINE to disable this code.
**************************************************************************/
/* Use this routine if told to, or by default on non-get_unlocked()
* platforms unless told not to. Yikes! Let's spell that out:
* On a platform with getc_unlocked():
* By default, use getc_unlocked().
* If you want to use fgets() instead, #define USE_FGETS_IN_GETLINE.
* On a platform without getc_unlocked():
* By default, use fgets().
* If you don't want to use fgets(), #define DONT_USE_FGETS_IN_GETLINE.
*/
#if !defined(USE_FGETS_IN_GETLINE) && !defined(HAVE_GETC_UNLOCKED)
#define USE_FGETS_IN_GETLINE
#endif
#if defined(DONT_USE_FGETS_IN_GETLINE) && defined(USE_FGETS_IN_GETLINE)
#undef USE_FGETS_IN_GETLINE
#endif
#ifdef USE_FGETS_IN_GETLINE
static PyObject*
getline_via_fgets(PyFileObject *f, FILE *fp)
{
/* INITBUFSIZE is the maximum line length that lets us get away with the fast
* no-realloc, one-fgets()-call path. Boosting it isn't free, because we have
* to fill this much of the buffer with a known value in order to figure out
* how much of the buffer fgets() overwrites. So if INITBUFSIZE is larger
* than "most" lines, we waste time filling unused buffer slots. 100 is
* surely adequate for most peoples' email archives, chewing over source code,
* etc -- "regular old text files".
* MAXBUFSIZE is the maximum line length that lets us get away with the less
* fast (but still zippy) no-realloc, two-fgets()-call path. See above for
* cautions about boosting that. 300 was chosen because the worst real-life
* text-crunching job reported on Python-Dev was a mail-log crawler where over
* half the lines were 254 chars.
*/
#define INITBUFSIZE 100
#define MAXBUFSIZE 300
char* p; /* temp */
char buf[MAXBUFSIZE];
PyObject* v; /* the string object result */
char* pvfree; /* address of next free slot */
char* pvend; /* address one beyond last free slot */
size_t nfree; /* # of free buffer slots; pvend-pvfree */
size_t total_v_size; /* total # of slots in buffer */
size_t increment; /* amount to increment the buffer */
size_t prev_v_size;
/* Optimize for normal case: avoid _PyString_Resize if at all
* possible via first reading into stack buffer "buf".
*/
total_v_size = INITBUFSIZE; /* start small and pray */
pvfree = buf;
for (;;) {
FILE_BEGIN_ALLOW_THREADS(f)
pvend = buf + total_v_size;
nfree = pvend - pvfree;
memset(pvfree, '\n', nfree);
assert(nfree < INT_MAX); /* Should be atmost MAXBUFSIZE */
p = fgets(pvfree, (int)nfree, fp);
FILE_END_ALLOW_THREADS(f)
if (p == NULL) {
clearerr(fp);
if (PyErr_CheckSignals())
return NULL;
v = PyString_FromStringAndSize(buf, pvfree - buf);
return v;
}
/* fgets read *something* */
p = memchr(pvfree, '\n', nfree);
if (p != NULL) {
/* Did the \n come from fgets or from us?
* Since fgets stops at the first \n, and then writes
* \0, if it's from fgets a \0 must be next. But if
* that's so, it could not have come from us, since
* the \n's we filled the buffer with have only more
* \n's to the right.
*/
if (p+1 < pvend && *(p+1) == '\0') {
/* It's from fgets: we win! In particular,
* we haven't done any mallocs yet, and can
* build the final result on the first try.
*/
++p; /* include \n from fgets */
}
else {
/* Must be from us: fgets didn't fill the
* buffer and didn't find a newline, so it
* must be the last and newline-free line of
* the file.
*/
assert(p > pvfree && *(p-1) == '\0');
--p; /* don't include \0 from fgets */
}
v = PyString_FromStringAndSize(buf, p - buf);
return v;
}
/* yuck: fgets overwrote all the newlines, i.e. the entire
* buffer. So this line isn't over yet, or maybe it is but
* we're exactly at EOF. If we haven't already, try using the
* rest of the stack buffer.
*/
assert(*(pvend-1) == '\0');
if (pvfree == buf) {
pvfree = pvend - 1; /* overwrite trailing null */
total_v_size = MAXBUFSIZE;
}
else
break;
}
/* The stack buffer isn't big enough; malloc a string object and read
* into its buffer.
*/
total_v_size = MAXBUFSIZE << 1;
v = PyString_FromStringAndSize((char*)NULL, (int)total_v_size);
if (v == NULL)
return v;
/* copy over everything except the last null byte */
memcpy(BUF(v), buf, MAXBUFSIZE-1);
pvfree = BUF(v) + MAXBUFSIZE - 1;
/* Keep reading stuff into v; if it ever ends successfully, break
* after setting p one beyond the end of the line. The code here is
* very much like the code above, except reads into v's buffer; see
* the code above for detailed comments about the logic.
*/
for (;;) {
FILE_BEGIN_ALLOW_THREADS(f)
pvend = BUF(v) + total_v_size;
nfree = pvend - pvfree;
memset(pvfree, '\n', nfree);
assert(nfree < INT_MAX);
p = fgets(pvfree, (int)nfree, fp);
FILE_END_ALLOW_THREADS(f)
if (p == NULL) {
clearerr(fp);
if (PyErr_CheckSignals()) {
Py_DECREF(v);
return NULL;
}
p = pvfree;
break;
}
p = memchr(pvfree, '\n', nfree);
if (p != NULL) {
if (p+1 < pvend && *(p+1) == '\0') {
/* \n came from fgets */
++p;
break;
}
/* \n came from us; last line of file, no newline */
assert(p > pvfree && *(p-1) == '\0');
--p;
break;
}
/* expand buffer and try again */
assert(*(pvend-1) == '\0');
increment = total_v_size >> 2; /* mild exponential growth */
prev_v_size = total_v_size;
total_v_size += increment;
/* check for overflow */
if (total_v_size <= prev_v_size ||
total_v_size > PY_SSIZE_T_MAX) {
PyErr_SetString(PyExc_OverflowError,
"line is longer than a Python string can hold");
Py_DECREF(v);
return NULL;
}
if (_PyString_Resize(&v, (int)total_v_size) < 0)
return NULL;
/* overwrite the trailing null byte */
pvfree = BUF(v) + (prev_v_size - 1);
}
if (BUF(v) + total_v_size != p)
_PyString_Resize(&v, p - BUF(v));
return v;
#undef INITBUFSIZE
#undef MAXBUFSIZE
}
#endif /* ifdef USE_FGETS_IN_GETLINE */
/* Internal routine to get a line.
Size argument interpretation:
> 0: max length;
<= 0: read arbitrary line
*/
static PyObject *
get_line(PyFileObject *f, int n)
{
FILE *fp = f->f_fp;
int c;
char *buf, *end;
size_t total_v_size; /* total # of slots in buffer */
size_t used_v_size; /* # used slots in buffer */
size_t increment; /* amount to increment the buffer */
PyObject *v;
int newlinetypes = f->f_newlinetypes;
int skipnextlf = f->f_skipnextlf;
int univ_newline = f->f_univ_newline;
#if defined(USE_FGETS_IN_GETLINE)
if (n <= 0 && !univ_newline )
return getline_via_fgets(f, fp);
#endif
total_v_size = n > 0 ? n : 100;
v = PyString_FromStringAndSize((char *)NULL, total_v_size);
if (v == NULL)
return NULL;
buf = BUF(v);
end = buf + total_v_size;
for (;;) {
FILE_BEGIN_ALLOW_THREADS(f)
FLOCKFILE(fp);
if (univ_newline) {
c = 'x'; /* Shut up gcc warning */
while ( buf != end && (c = GETC(fp)) != EOF ) {
if (skipnextlf ) {
skipnextlf = 0;
if (c == '\n') {
/* Seeing a \n here with
* skipnextlf true means we
* saw a \r before.
*/
newlinetypes |= NEWLINE_CRLF;
c = GETC(fp);
if (c == EOF) break;
} else {
newlinetypes |= NEWLINE_CR;
}
}
if (c == '\r') {
skipnextlf = 1;
c = '\n';
} else if ( c == '\n')
newlinetypes |= NEWLINE_LF;
*buf++ = c;
if (c == '\n') break;
}
if ( c == EOF && skipnextlf )
newlinetypes |= NEWLINE_CR;
} else /* If not universal newlines use the normal loop */
while ((c = GETC(fp)) != EOF &&
(*buf++ = c) != '\n' &&
buf != end)
;
FUNLOCKFILE(fp);
FILE_END_ALLOW_THREADS(f)
f->f_newlinetypes = newlinetypes;
f->f_skipnextlf = skipnextlf;
if (c == '\n')
break;
if (c == EOF) {
if (ferror(fp)) {
PyErr_SetFromErrno(PyExc_IOError);
clearerr(fp);
Py_DECREF(v);
return NULL;
}
clearerr(fp);
if (PyErr_CheckSignals()) {
Py_DECREF(v);
return NULL;
}
break;
}
/* Must be because buf == end */
if (n > 0)
break;
used_v_size = total_v_size;
increment = total_v_size >> 2; /* mild exponential growth */
total_v_size += increment;
if (total_v_size > PY_SSIZE_T_MAX) {
PyErr_SetString(PyExc_OverflowError,
"line is longer than a Python string can hold");
Py_DECREF(v);
return NULL;
}
if (_PyString_Resize(&v, total_v_size) < 0)
return NULL;
buf = BUF(v) + used_v_size;
end = BUF(v) + total_v_size;
}
used_v_size = buf - BUF(v);
if (used_v_size != total_v_size)
_PyString_Resize(&v, used_v_size);
return v;
}
/* External C interface */
PyObject *
PyFile_GetLine(PyObject *f, int n)
{
PyObject *result;
if (f == NULL) {
PyErr_BadInternalCall();
return NULL;
}
if (PyFile_Check(f)) {
PyFileObject *fo = (PyFileObject *)f;
if (fo->f_fp == NULL)
return err_closed();
/* refuse to mix with f.next() */
if (fo->f_buf != NULL &&
(fo->f_bufend - fo->f_bufptr) > 0 &&
fo->f_buf[0] != '\0')
return err_iterbuffered();
result = get_line(fo, n);
}
else {
PyObject *reader;
PyObject *args;
reader = PyObject_GetAttrString(f, "readline");
if (reader == NULL)
return NULL;
if (n <= 0)
args = PyTuple_New(0);
else
args = Py_BuildValue("(i)", n);
if (args == NULL) {
Py_DECREF(reader);
return NULL;
}
result = PyEval_CallObject(reader, args);
Py_DECREF(reader);
Py_DECREF(args);
if (result != NULL && !PyString_Check(result) &&
!PyUnicode_Check(result)) {
Py_DECREF(result);
result = NULL;
PyErr_SetString(PyExc_TypeError,
"object.readline() returned non-string");
}
}
if (n < 0 && result != NULL && PyString_Check(result)) {
char *s = PyString_AS_STRING(result);
Py_ssize_t len = PyString_GET_SIZE(result);
if (len == 0) {
Py_DECREF(result);
result = NULL;
PyErr_SetString(PyExc_EOFError,
"EOF when reading a line");
}
else if (s[len-1] == '\n') {
if (result->ob_refcnt == 1)
_PyString_Resize(&result, len-1);
else {
PyObject *v;
v = PyString_FromStringAndSize(s, len-1);
Py_DECREF(result);
result = v;
}
}
}
#ifdef Py_USING_UNICODE
if (n < 0 && result != NULL && PyUnicode_Check(result)) {
Py_UNICODE *s = PyUnicode_AS_UNICODE(result);
Py_ssize_t len = PyUnicode_GET_SIZE(result);
if (len == 0) {
Py_DECREF(result);
result = NULL;
PyErr_SetString(PyExc_EOFError,
"EOF when reading a line");
}
else if (s[len-1] == '\n') {
if (result->ob_refcnt == 1)
PyUnicode_Resize(&result, len-1);
else {
PyObject *v;
v = PyUnicode_FromUnicode(s, len-1);
Py_DECREF(result);
result = v;
}
}
}
#endif
return result;
}
/* Python method */
static PyObject *
file_readline(PyFileObject *f, PyObject *args)
{
int n = -1;
if (f->f_fp == NULL)
return err_closed();
/* refuse to mix with f.next() */
if (f->f_buf != NULL &&
(f->f_bufend - f->f_bufptr) > 0 &&
f->f_buf[0] != '\0')
return err_iterbuffered();
if (!PyArg_ParseTuple(args, "|i:readline", &n))
return NULL;
if (n == 0)
return PyString_FromString("");
if (n < 0)
n = 0;
return get_line(f, n);
}
static PyObject *
file_readlines(PyFileObject *f, PyObject *args)
{
long sizehint = 0;
PyObject *list = NULL;
PyObject *line;
char small_buffer[SMALLCHUNK];
char *buffer = small_buffer;
size_t buffersize = SMALLCHUNK;
PyObject *big_buffer = NULL;
size_t nfilled = 0;
size_t nread;
size_t totalread = 0;
char *p, *q, *end;
int err;
int shortread = 0;
if (f->f_fp == NULL)
return err_closed();
/* refuse to mix with f.next() */
if (f->f_buf != NULL &&
(f->f_bufend - f->f_bufptr) > 0 &&
f->f_buf[0] != '\0')
return err_iterbuffered();
if (!PyArg_ParseTuple(args, "|l:readlines", &sizehint))
return NULL;
if ((list = PyList_New(0)) == NULL)
return NULL;
for (;;) {
if (shortread)
nread = 0;
else {
FILE_BEGIN_ALLOW_THREADS(f)
errno = 0;
nread = Py_UniversalNewlineFread(buffer+nfilled,
buffersize-nfilled, f->f_fp, (PyObject *)f);
FILE_END_ALLOW_THREADS(f)
shortread = (nread < buffersize-nfilled);
}
if (nread == 0) {
sizehint = 0;
if (!ferror(f->f_fp))
break;
PyErr_SetFromErrno(PyExc_IOError);
clearerr(f->f_fp);
goto error;
}
totalread += nread;
p = (char *)memchr(buffer+nfilled, '\n', nread);
if (p == NULL) {
/* Need a larger buffer to fit this line */
nfilled += nread;
buffersize *= 2;
if (buffersize > PY_SSIZE_T_MAX) {
PyErr_SetString(PyExc_OverflowError,
"line is longer than a Python string can hold");
goto error;
}
if (big_buffer == NULL) {
/* Create the big buffer */
big_buffer = PyString_FromStringAndSize(
NULL, buffersize);
if (big_buffer == NULL)
goto error;
buffer = PyString_AS_STRING(big_buffer);
memcpy(buffer, small_buffer, nfilled);
}
else {
/* Grow the big buffer */
if ( _PyString_Resize(&big_buffer, buffersize) < 0 )
goto error;
buffer = PyString_AS_STRING(big_buffer);
}
continue;
}
end = buffer+nfilled+nread;
q = buffer;
do {
/* Process complete lines */
p++;
line = PyString_FromStringAndSize(q, p-q);
if (line == NULL)
goto error;
err = PyList_Append(list, line);
Py_DECREF(line);
if (err != 0)
goto error;
q = p;
p = (char *)memchr(q, '\n', end-q);
} while (p != NULL);
/* Move the remaining incomplete line to the start */
nfilled = end-q;
memmove(buffer, q, nfilled);
if (sizehint > 0)
if (totalread >= (size_t)sizehint)
break;
}
if (nfilled != 0) {
/* Partial last line */
line = PyString_FromStringAndSize(buffer, nfilled);
if (line == NULL)
goto error;
if (sizehint > 0) {
/* Need to complete the last line */
PyObject *rest = get_line(f, 0);
if (rest == NULL) {
Py_DECREF(line);
goto error;
}
PyString_Concat(&line, rest);
Py_DECREF(rest);
if (line == NULL)
goto error;
}
err = PyList_Append(list, line);
Py_DECREF(line);
if (err != 0)
goto error;
}
cleanup:
Py_XDECREF(big_buffer);
return list;
error:
Py_CLEAR(list);
goto cleanup;
}
static PyObject *
file_write(PyFileObject *f, PyObject *args)
{
Py_buffer pbuf;
char *s;
Py_ssize_t n, n2;
if (f->f_fp == NULL)
return err_closed();
if (f->f_binary) {
if (!PyArg_ParseTuple(args, "s*", &pbuf))
return NULL;
s = pbuf.buf;
n = pbuf.len;
} else
if (!PyArg_ParseTuple(args, "t#", &s, &n))
return NULL;
f->f_softspace = 0;
FILE_BEGIN_ALLOW_THREADS(f)
errno = 0;
n2 = fwrite(s, 1, n, f->f_fp);
FILE_END_ALLOW_THREADS(f)
if (f->f_binary)
PyBuffer_Release(&pbuf);
if (n2 != n) {
PyErr_SetFromErrno(PyExc_IOError);
clearerr(f->f_fp);
return NULL;
}
Py_INCREF(Py_None);
return Py_None;
}
static PyObject *
file_writelines(PyFileObject *f, PyObject *seq)
{
#define CHUNKSIZE 1000
PyObject *list, *line;
PyObject *it; /* iter(seq) */
PyObject *result;
int index, islist;
Py_ssize_t i, j, nwritten, len;
assert(seq != NULL);
if (f->f_fp == NULL)
return err_closed();
result = NULL;
list = NULL;
islist = PyList_Check(seq);
if (islist)
it = NULL;
else {
it = PyObject_GetIter(seq);
if (it == NULL) {
PyErr_SetString(PyExc_TypeError,
"writelines() requires an iterable argument");
return NULL;
}
/* From here on, fail by going to error, to reclaim "it". */
list = PyList_New(CHUNKSIZE);
if (list == NULL)
goto error;
}
/* Strategy: slurp CHUNKSIZE lines into a private list,
checking that they are all strings, then write that list
without holding the interpreter lock, then come back for more. */
for (index = 0; ; index += CHUNKSIZE) {
if (islist) {
Py_XDECREF(list);
list = PyList_GetSlice(seq, index, index+CHUNKSIZE);
if (list == NULL)
goto error;
j = PyList_GET_SIZE(list);
}
else {
for (j = 0; j < CHUNKSIZE; j++) {
line = PyIter_Next(it);
if (line == NULL) {
if (PyErr_Occurred())
goto error;
break;
}
PyList_SetItem(list, j, line);
}
}
if (j == 0)
break;
/* Check that all entries are indeed strings. If not,
apply the same rules as for file.write() and
convert the results to strings. This is slow, but
seems to be the only way since all conversion APIs
could potentially execute Python code. */
for (i = 0; i < j; i++) {
PyObject *v = PyList_GET_ITEM(list, i);
if (!PyString_Check(v)) {
const char *buffer;
if (((f->f_binary &&
PyObject_AsReadBuffer(v,
(const void**)&buffer,
&len)) ||
PyObject_AsCharBuffer(v,
&buffer,
&len))) {
PyErr_SetString(PyExc_TypeError,
"writelines() argument must be a sequence of strings");
goto error;
}
line = PyString_FromStringAndSize(buffer,
len);
if (line == NULL)
goto error;
Py_DECREF(v);
PyList_SET_ITEM(list, i, line);
}
}
/* Since we are releasing the global lock, the
following code may *not* execute Python code. */
f->f_softspace = 0;
FILE_BEGIN_ALLOW_THREADS(f)
errno = 0;
for (i = 0; i < j; i++) {
line = PyList_GET_ITEM(list, i);
len = PyString_GET_SIZE(line);
nwritten = fwrite(PyString_AS_STRING(line),
1, len, f->f_fp);
if (nwritten != len) {
FILE_ABORT_ALLOW_THREADS(f)
PyErr_SetFromErrno(PyExc_IOError);
clearerr(f->f_fp);
goto error;
}
}
FILE_END_ALLOW_THREADS(f)
if (j < CHUNKSIZE)
break;
}
Py_INCREF(Py_None);
result = Py_None;
error:
Py_XDECREF(list);
Py_XDECREF(it);
return result;
#undef CHUNKSIZE
}
static PyObject *
file_self(PyFileObject *f)
{
if (f->f_fp == NULL)
return err_closed();
Py_INCREF(f);
return (PyObject *)f;
}
static PyObject *
file_xreadlines(PyFileObject *f)
{
if (PyErr_WarnPy3k("f.xreadlines() not supported in 3.x, "
"try 'for line in f' instead", 1) < 0)
return NULL;
return file_self(f);
}
static PyObject *
file_exit(PyObject *f, PyObject *args)
{
PyObject *ret = PyObject_CallMethod(f, "close", NULL);
if (!ret)
/* If error occurred, pass through */
return NULL;
Py_DECREF(ret);
/* We cannot return the result of close since a true
* value will be interpreted as "yes, swallow the
* exception if one was raised inside the with block". */
Py_RETURN_NONE;
}
PyDoc_STRVAR(readline_doc,
"readline([size]) -> next line from the file, as a string.\n"
"\n"
"Retain newline. A non-negative size argument limits the maximum\n"
"number of bytes to return (an incomplete line may be returned then).\n"
"Return an empty string at EOF.");
PyDoc_STRVAR(read_doc,
"read([size]) -> read at most size bytes, returned as a string.\n"
"\n"
"If the size argument is negative or omitted, read until EOF is reached.\n"
"Notice that when in non-blocking mode, less data than what was requested\n"
"may be returned, even if no size parameter was given.");
PyDoc_STRVAR(write_doc,
"write(str) -> None. Write string str to file.\n"
"\n"
"Note that due to buffering, flush() or close() may be needed before\n"
"the file on disk reflects the data written.");
PyDoc_STRVAR(fileno_doc,
"fileno() -> integer \"file descriptor\".\n"
"\n"
"This is needed for lower-level file interfaces, such os.read().");
PyDoc_STRVAR(seek_doc,
"seek(offset[, whence]) -> None. Move to new file position.\n"
"\n"
"Argument offset is a byte count. Optional argument whence defaults to\n"
"0 (offset from start of file, offset should be >= 0); other values are 1\n"
"(move relative to current position, positive or negative), and 2 (move\n"
"relative to end of file, usually negative, although many platforms allow\n"
"seeking beyond the end of a file). If the file is opened in text mode,\n"
"only offsets returned by tell() are legal. Use of other offsets causes\n"
"undefined behavior."
"\n"
"Note that not all file objects are seekable.");
#ifdef HAVE_FTRUNCATE
PyDoc_STRVAR(truncate_doc,
"truncate([size]) -> None. Truncate the file to at most size bytes.\n"
"\n"
"Size defaults to the current file position, as returned by tell().");
#endif
PyDoc_STRVAR(tell_doc,
"tell() -> current file position, an integer (may be a long integer).");
PyDoc_STRVAR(readinto_doc,
"readinto() -> Undocumented. Don't use this; it may go away.");
PyDoc_STRVAR(readlines_doc,
"readlines([size]) -> list of strings, each a line from the file.\n"
"\n"
"Call readline() repeatedly and return a list of the lines so read.\n"
"The optional size argument, if given, is an approximate bound on the\n"
"total number of bytes in the lines returned.");
PyDoc_STRVAR(xreadlines_doc,
"xreadlines() -> returns self.\n"
"\n"
"For backward compatibility. File objects now include the performance\n"
"optimizations previously implemented in the xreadlines module.");
PyDoc_STRVAR(writelines_doc,
"writelines(sequence_of_strings) -> None. Write the strings to the file.\n"
"\n"
"Note that newlines are not added. The sequence can be any iterable object\n"
"producing strings. This is equivalent to calling write() for each string.");
PyDoc_STRVAR(flush_doc,
"flush() -> None. Flush the internal I/O buffer.");
PyDoc_STRVAR(close_doc,
"close() -> None or (perhaps) an integer. Close the file.\n"
"\n"
"Sets data attribute .closed to True. A closed file cannot be used for\n"
"further I/O operations. close() may be called more than once without\n"
"error. Some kinds of file objects (for example, opened by popen())\n"
"may return an exit status upon closing.");
PyDoc_STRVAR(isatty_doc,
"isatty() -> true or false. True if the file is connected to a tty device.");
PyDoc_STRVAR(enter_doc,
"__enter__() -> self.");
PyDoc_STRVAR(exit_doc,
"__exit__(*excinfo) -> None. Closes the file.");
static PyMethodDef file_methods[] = {
{"readline", (PyCFunction)file_readline, METH_VARARGS, readline_doc},
{"read", (PyCFunction)file_read, METH_VARARGS, read_doc},
{"write", (PyCFunction)file_write, METH_VARARGS, write_doc},
{"fileno", (PyCFunction)file_fileno, METH_NOARGS, fileno_doc},
{"seek", (PyCFunction)file_seek, METH_VARARGS, seek_doc},
#ifdef HAVE_FTRUNCATE
{"truncate", (PyCFunction)file_truncate, METH_VARARGS, truncate_doc},
#endif
{"tell", (PyCFunction)file_tell, METH_NOARGS, tell_doc},
{"readinto", (PyCFunction)file_readinto, METH_VARARGS, readinto_doc},
{"readlines", (PyCFunction)file_readlines, METH_VARARGS, readlines_doc},
{"xreadlines",(PyCFunction)file_xreadlines, METH_NOARGS, xreadlines_doc},
{"writelines",(PyCFunction)file_writelines, METH_O, writelines_doc},
{"flush", (PyCFunction)file_flush, METH_NOARGS, flush_doc},
{"close", (PyCFunction)file_close, METH_NOARGS, close_doc},
{"isatty", (PyCFunction)file_isatty, METH_NOARGS, isatty_doc},
{"__enter__", (PyCFunction)file_self, METH_NOARGS, enter_doc},
{"__exit__", (PyCFunction)file_exit, METH_VARARGS, exit_doc},
{NULL, NULL} /* sentinel */
};
#define OFF(x) offsetof(PyFileObject, x)
static PyMemberDef file_memberlist[] = {
{"mode", T_OBJECT, OFF(f_mode), RO,
"file mode ('r', 'U', 'w', 'a', possibly with 'b' or '+' added)"},
{"name", T_OBJECT, OFF(f_name), RO,
"file name"},
{"encoding", T_OBJECT, OFF(f_encoding), RO,
"file encoding"},
{"errors", T_OBJECT, OFF(f_errors), RO,
"Unicode error handler"},
/* getattr(f, "closed") is implemented without this table */
{NULL} /* Sentinel */
};
static PyObject *
get_closed(PyFileObject *f, void *closure)
{
return PyBool_FromLong((long)(f->f_fp == 0));
}
static PyObject *
get_newlines(PyFileObject *f, void *closure)
{
switch (f->f_newlinetypes) {
case NEWLINE_UNKNOWN:
Py_INCREF(Py_None);
return Py_None;
case NEWLINE_CR:
return PyString_FromString("\r");
case NEWLINE_LF:
return PyString_FromString("\n");
case NEWLINE_CR|NEWLINE_LF:
return Py_BuildValue("(ss)", "\r", "\n");
case NEWLINE_CRLF:
return PyString_FromString("\r\n");
case NEWLINE_CR|NEWLINE_CRLF:
return Py_BuildValue("(ss)", "\r", "\r\n");
case NEWLINE_LF|NEWLINE_CRLF:
return Py_BuildValue("(ss)", "\n", "\r\n");
case NEWLINE_CR|NEWLINE_LF|NEWLINE_CRLF:
return Py_BuildValue("(sss)", "\r", "\n", "\r\n");
default:
PyErr_Format(PyExc_SystemError,
"Unknown newlines value 0x%x\n",
f->f_newlinetypes);
return NULL;
}
}
static PyObject *
get_softspace(PyFileObject *f, void *closure)
{
if (PyErr_WarnPy3k("file.softspace not supported in 3.x", 1) < 0)
return NULL;
return PyInt_FromLong(f->f_softspace);
}
static int
set_softspace(PyFileObject *f, PyObject *value)
{
int new;
if (PyErr_WarnPy3k("file.softspace not supported in 3.x", 1) < 0)
return -1;
if (value == NULL) {
PyErr_SetString(PyExc_TypeError,
"can't delete softspace attribute");
return -1;
}
new = PyInt_AsLong(value);
if (new == -1 && PyErr_Occurred())
return -1;
f->f_softspace = new;
return 0;
}
static PyGetSetDef file_getsetlist[] = {
{"closed", (getter)get_closed, NULL, "True if the file is closed"},
{"newlines", (getter)get_newlines, NULL,
"end-of-line convention used in this file"},
{"softspace", (getter)get_softspace, (setter)set_softspace,
"flag indicating that a space needs to be printed; used by print"},
{0},
};
static void
drop_readahead(PyFileObject *f)
{
if (f->f_buf != NULL) {
PyMem_Free(f->f_buf);
f->f_buf = NULL;
}
}
/* Make sure that file has a readahead buffer with at least one byte
(unless at EOF) and no more than bufsize. Returns negative value on
error, will set MemoryError if bufsize bytes cannot be allocated. */
static int
readahead(PyFileObject *f, int bufsize)
{
Py_ssize_t chunksize;
if (f->f_buf != NULL) {
if( (f->f_bufend - f->f_bufptr) >= 1)
return 0;
else
drop_readahead(f);
}
if ((f->f_buf = (char *)PyMem_Malloc(bufsize)) == NULL) {
PyErr_NoMemory();
return -1;
}
FILE_BEGIN_ALLOW_THREADS(f)
errno = 0;
chunksize = Py_UniversalNewlineFread(
f->f_buf, bufsize, f->f_fp, (PyObject *)f);
FILE_END_ALLOW_THREADS(f)
if (chunksize == 0) {
if (ferror(f->f_fp)) {
PyErr_SetFromErrno(PyExc_IOError);
clearerr(f->f_fp);
drop_readahead(f);
return -1;
}
}
f->f_bufptr = f->f_buf;
f->f_bufend = f->f_buf + chunksize;
return 0;
}
/* Used by file_iternext. The returned string will start with 'skip'
uninitialized bytes followed by the remainder of the line. Don't be
horrified by the recursive call: maximum recursion depth is limited by
logarithmic buffer growth to about 50 even when reading a 1gb line. */
static PyStringObject *
readahead_get_line_skip(PyFileObject *f, int skip, int bufsize)
{
PyStringObject* s;
char *bufptr;
char *buf;
Py_ssize_t len;
if (f->f_buf == NULL)
if (readahead(f, bufsize) < 0)
return NULL;
len = f->f_bufend - f->f_bufptr;
if (len == 0)
return (PyStringObject *)
PyString_FromStringAndSize(NULL, skip);
bufptr = (char *)memchr(f->f_bufptr, '\n', len);
if (bufptr != NULL) {
bufptr++; /* Count the '\n' */
len = bufptr - f->f_bufptr;
s = (PyStringObject *)
PyString_FromStringAndSize(NULL, skip+len);
if (s == NULL)
return NULL;
memcpy(PyString_AS_STRING(s)+skip, f->f_bufptr, len);
f->f_bufptr = bufptr;
if (bufptr == f->f_bufend)
drop_readahead(f);
} else {
bufptr = f->f_bufptr;
buf = f->f_buf;
f->f_buf = NULL; /* Force new readahead buffer */
assert(skip+len < INT_MAX);
s = readahead_get_line_skip(
f, (int)(skip+len), bufsize + (bufsize>>2) );
if (s == NULL) {
PyMem_Free(buf);
return NULL;
}
memcpy(PyString_AS_STRING(s)+skip, bufptr, len);
PyMem_Free(buf);
}
return s;
}
/* A larger buffer size may actually decrease performance. */
#define READAHEAD_BUFSIZE 8192
static PyObject *
file_iternext(PyFileObject *f)
{
PyStringObject* l;
if (f->f_fp == NULL)
return err_closed();
l = readahead_get_line_skip(f, 0, READAHEAD_BUFSIZE);
if (l == NULL || PyString_GET_SIZE(l) == 0) {
Py_XDECREF(l);
return NULL;
}
return (PyObject *)l;
}
static PyObject *
file_new(PyTypeObject *type, PyObject *args, PyObject *kwds)
{
PyObject *self;
static PyObject *not_yet_string;
assert(type != NULL && type->tp_alloc != NULL);
if (not_yet_string == NULL) {
not_yet_string = PyString_InternFromString("<uninitialized file>");
if (not_yet_string == NULL)
return NULL;
}
self = type->tp_alloc(type, 0);
if (self != NULL) {
/* Always fill in the name and mode, so that nobody else
needs to special-case NULLs there. */
Py_INCREF(not_yet_string);
((PyFileObject *)self)->f_name = not_yet_string;
Py_INCREF(not_yet_string);
((PyFileObject *)self)->f_mode = not_yet_string;
Py_INCREF(Py_None);
((PyFileObject *)self)->f_encoding = Py_None;
Py_INCREF(Py_None);
((PyFileObject *)self)->f_errors = Py_None;
((PyFileObject *)self)->weakreflist = NULL;
((PyFileObject *)self)->unlocked_count = 0;
}
return self;
}
static int
file_init(PyObject *self, PyObject *args, PyObject *kwds)
{
PyFileObject *foself = (PyFileObject *)self;
int ret = 0;
static char *kwlist[] = {"name", "mode", "buffering", 0};
char *name = NULL;
char *mode = "r";
int bufsize = -1;
int wideargument = 0;
assert(PyFile_Check(self));
if (foself->f_fp != NULL) {
/* Have to close the existing file first. */
PyObject *closeresult = file_close(foself);
if (closeresult == NULL)
return -1;
Py_DECREF(closeresult);
}
#ifdef Py_WIN_WIDE_FILENAMES
if (GetVersion() < 0x80000000) { /* On NT, so wide API available */
PyObject *po;
if (PyArg_ParseTupleAndKeywords(args, kwds, "U|si:file",
kwlist, &po, &mode, &bufsize)) {
wideargument = 1;
if (fill_file_fields(foself, NULL, po, mode,
fclose) == NULL)
goto Error;
} else {
/* Drop the argument parsing error as narrow
strings are also valid. */
PyErr_Clear();
}
}
#endif
if (!wideargument) {
PyObject *o_name;
if (!PyArg_ParseTupleAndKeywords(args, kwds, "et|si:file", kwlist,
Py_FileSystemDefaultEncoding,
&name,
&mode, &bufsize))
return -1;
/* We parse again to get the name as a PyObject */
if (!PyArg_ParseTupleAndKeywords(args, kwds, "O|si:file",
kwlist, &o_name, &mode,
&bufsize))
goto Error;
if (fill_file_fields(foself, NULL, o_name, mode,
fclose) == NULL)
goto Error;
}
if (open_the_file(foself, name, mode) == NULL)
goto Error;
foself->f_setbuf = NULL;
PyFile_SetBufSize(self, bufsize);
goto Done;
Error:
ret = -1;
/* fall through */
Done:
PyMem_Free(name); /* free the encoded string */
return ret;
}
PyDoc_VAR(file_doc) =
PyDoc_STR(
"file(name[, mode[, buffering]]) -> file object\n"
"\n"
"Open a file. The mode can be 'r', 'w' or 'a' for reading (default),\n"
"writing or appending. The file will be created if it doesn't exist\n"
"when opened for writing or appending; it will be truncated when\n"
"opened for writing. Add a 'b' to the mode for binary files.\n"
"Add a '+' to the mode to allow simultaneous reading and writing.\n"
"If the buffering argument is given, 0 means unbuffered, 1 means line\n"
"buffered, and larger numbers specify the buffer size. The preferred way\n"
"to open a file is with the builtin open() function.\n"
)
PyDoc_STR(
"Add a 'U' to mode to open the file for input with universal newline\n"
"support. Any line ending in the input file will be seen as a '\\n'\n"
"in Python. Also, a file so opened gains the attribute 'newlines';\n"
"the value for this attribute is one of None (no newline read yet),\n"
"'\\r', '\\n', '\\r\\n' or a tuple containing all the newline types seen.\n"
"\n"
"'U' cannot be combined with 'w' or '+' mode.\n"
);
PyTypeObject PyFile_Type = {
PyVarObject_HEAD_INIT(&PyType_Type, 0)
"file",
sizeof(PyFileObject),
0,
(destructor)file_dealloc, /* tp_dealloc */
0, /* tp_print */
0, /* tp_getattr */
0, /* tp_setattr */
0, /* tp_compare */
(reprfunc)file_repr, /* tp_repr */
0, /* tp_as_number */
0, /* tp_as_sequence */
0, /* tp_as_mapping */
0, /* tp_hash */
0, /* tp_call */
0, /* tp_str */
PyObject_GenericGetAttr, /* tp_getattro */
/* softspace is writable: we must supply tp_setattro */
PyObject_GenericSetAttr, /* tp_setattro */
0, /* tp_as_buffer */
Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE | Py_TPFLAGS_HAVE_WEAKREFS, /* tp_flags */
file_doc, /* tp_doc */
0, /* tp_traverse */
0, /* tp_clear */
0, /* tp_richcompare */
offsetof(PyFileObject, weakreflist), /* tp_weaklistoffset */
(getiterfunc)file_self, /* tp_iter */
(iternextfunc)file_iternext, /* tp_iternext */
file_methods, /* tp_methods */
file_memberlist, /* tp_members */
file_getsetlist, /* tp_getset */
0, /* tp_base */
0, /* tp_dict */
0, /* tp_descr_get */
0, /* tp_descr_set */
0, /* tp_dictoffset */
file_init, /* tp_init */
PyType_GenericAlloc, /* tp_alloc */
file_new, /* tp_new */
PyObject_Del, /* tp_free */
};
/* Interface for the 'soft space' between print items. */
int
PyFile_SoftSpace(PyObject *f, int newflag)
{
long oldflag = 0;
if (f == NULL) {
/* Do nothing */
}
else if (PyFile_Check(f)) {
oldflag = ((PyFileObject *)f)->f_softspace;
((PyFileObject *)f)->f_softspace = newflag;
}
else {
PyObject *v;
v = PyObject_GetAttrString(f, "softspace");
if (v == NULL)
PyErr_Clear();
else {
if (PyInt_Check(v))
oldflag = PyInt_AsLong(v);
assert(oldflag < INT_MAX);
Py_DECREF(v);
}
v = PyInt_FromLong((long)newflag);
if (v == NULL)
PyErr_Clear();
else {
if (PyObject_SetAttrString(f, "softspace", v) != 0)
PyErr_Clear();
Py_DECREF(v);
}
}
return (int)oldflag;
}
/* Interfaces to write objects/strings to file-like objects */
int
PyFile_WriteObject(PyObject *v, PyObject *f, int flags)
{
PyObject *writer, *value, *args, *result;
if (f == NULL) {
PyErr_SetString(PyExc_TypeError, "writeobject with NULL file");
return -1;
}
else if (PyFile_Check(f)) {
PyFileObject *fobj = (PyFileObject *) f;
#ifdef Py_USING_UNICODE
PyObject *enc = fobj->f_encoding;
int result;
#endif
if (fobj->f_fp == NULL) {
err_closed();
return -1;
}
#ifdef Py_USING_UNICODE
if ((flags & Py_PRINT_RAW) &&
PyUnicode_Check(v) && enc != Py_None) {
char *cenc = PyString_AS_STRING(enc);
char *errors = fobj->f_errors == Py_None ?
"strict" : PyString_AS_STRING(fobj->f_errors);
value = PyUnicode_AsEncodedString(v, cenc, errors);
if (value == NULL)
return -1;
} else {
value = v;
Py_INCREF(value);
}
result = file_PyObject_Print(value, fobj, flags);
Py_DECREF(value);
return result;
#else
return file_PyObject_Print(v, fobj, flags);
#endif
}
writer = PyObject_GetAttrString(f, "write");
if (writer == NULL)
return -1;
if (flags & Py_PRINT_RAW) {
if (PyUnicode_Check(v)) {
value = v;
Py_INCREF(value);
} else
value = PyObject_Str(v);
}
else
value = PyObject_Repr(v);
if (value == NULL) {
Py_DECREF(writer);
return -1;
}
args = PyTuple_Pack(1, value);
if (args == NULL) {
Py_DECREF(value);
Py_DECREF(writer);
return -1;
}
result = PyEval_CallObject(writer, args);
Py_DECREF(args);
Py_DECREF(value);
Py_DECREF(writer);
if (result == NULL)
return -1;
Py_DECREF(result);
return 0;
}
int
PyFile_WriteString(const char *s, PyObject *f)
{
if (f == NULL) {
/* Should be caused by a pre-existing error */
if (!PyErr_Occurred())
PyErr_SetString(PyExc_SystemError,
"null file for PyFile_WriteString");
return -1;
}
else if (PyFile_Check(f)) {
PyFileObject *fobj = (PyFileObject *) f;
FILE *fp = PyFile_AsFile(f);
if (fp == NULL) {
err_closed();
return -1;
}
FILE_BEGIN_ALLOW_THREADS(fobj)
fputs(s, fp);
FILE_END_ALLOW_THREADS(fobj)
return 0;
}
else if (!PyErr_Occurred()) {
PyObject *v = PyString_FromString(s);
int err;
if (v == NULL)
return -1;
err = PyFile_WriteObject(v, f, Py_PRINT_RAW);
Py_DECREF(v);
return err;
}
else
return -1;
}
/* Try to get a file-descriptor from a Python object. If the object
is an integer or long integer, its value is returned. If not, the
object's fileno() method is called if it exists; the method must return
an integer or long integer, which is returned as the file descriptor value.
-1 is returned on failure.
*/
int PyObject_AsFileDescriptor(PyObject *o)
{
int fd;
PyObject *meth;
if (PyInt_Check(o)) {
fd = PyInt_AsLong(o);
}
else if (PyLong_Check(o)) {
fd = PyLong_AsLong(o);
}
else if ((meth = PyObject_GetAttrString(o, "fileno")) != NULL)
{
PyObject *fno = PyEval_CallObject(meth, NULL);
Py_DECREF(meth);
if (fno == NULL)
return -1;
if (PyInt_Check(fno)) {
fd = PyInt_AsLong(fno);
Py_DECREF(fno);
}
else if (PyLong_Check(fno)) {
fd = PyLong_AsLong(fno);
Py_DECREF(fno);
}
else {
PyErr_SetString(PyExc_TypeError,
"fileno() returned a non-integer");
Py_DECREF(fno);
return -1;
}
}
else {
PyErr_SetString(PyExc_TypeError,
"argument must be an int, or have a fileno() method.");
return -1;
}
if (fd < 0) {
PyErr_Format(PyExc_ValueError,
"file descriptor cannot be a negative integer (%i)",
fd);
return -1;
}
return fd;
}
/* From here on we need access to the real fgets and fread */
#undef fgets
#undef fread
/*
** Py_UniversalNewlineFgets is an fgets variation that understands
** all of \r, \n and \r\n conventions.
** The stream should be opened in binary mode.
** If fobj is NULL the routine always does newline conversion, and
** it may peek one char ahead to gobble the second char in \r\n.
** If fobj is non-NULL it must be a PyFileObject. In this case there
** is no readahead but in stead a flag is used to skip a following
** \n on the next read. Also, if the file is open in binary mode
** the whole conversion is skipped. Finally, the routine keeps track of
** the different types of newlines seen.
** Note that we need no error handling: fgets() treats error and eof
** identically.
*/
char *
Py_UniversalNewlineFgets(char *buf, int n, FILE *stream, PyObject *fobj)
{
char *p = buf;
int c;
int newlinetypes = 0;
int skipnextlf = 0;
int univ_newline = 1;
if (fobj) {
if (!PyFile_Check(fobj)) {
errno = ENXIO; /* What can you do... */
return NULL;
}
univ_newline = ((PyFileObject *)fobj)->f_univ_newline;
if ( !univ_newline )
return fgets(buf, n, stream);
newlinetypes = ((PyFileObject *)fobj)->f_newlinetypes;
skipnextlf = ((PyFileObject *)fobj)->f_skipnextlf;
}
FLOCKFILE(stream);
c = 'x'; /* Shut up gcc warning */
while (--n > 0 && (c = GETC(stream)) != EOF ) {
if (skipnextlf ) {
skipnextlf = 0;
if (c == '\n') {
/* Seeing a \n here with skipnextlf true
** means we saw a \r before.
*/
newlinetypes |= NEWLINE_CRLF;
c = GETC(stream);
if (c == EOF) break;
} else {
/*
** Note that c == EOF also brings us here,
** so we're okay if the last char in the file
** is a CR.
*/
newlinetypes |= NEWLINE_CR;
}
}
if (c == '\r') {
/* A \r is translated into a \n, and we skip
** an adjacent \n, if any. We don't set the
** newlinetypes flag until we've seen the next char.
*/
skipnextlf = 1;
c = '\n';
} else if ( c == '\n') {
newlinetypes |= NEWLINE_LF;
}
*p++ = c;
if (c == '\n') break;
}
if ( c == EOF && skipnextlf )
newlinetypes |= NEWLINE_CR;
FUNLOCKFILE(stream);
*p = '\0';
if (fobj) {
((PyFileObject *)fobj)->f_newlinetypes = newlinetypes;
((PyFileObject *)fobj)->f_skipnextlf = skipnextlf;
} else if ( skipnextlf ) {
/* If we have no file object we cannot save the
** skipnextlf flag. We have to readahead, which
** will cause a pause if we're reading from an
** interactive stream, but that is very unlikely
** unless we're doing something silly like
** execfile("/dev/tty").
*/
c = GETC(stream);
if ( c != '\n' )
ungetc(c, stream);
}
if (p == buf)
return NULL;
return buf;
}
/*
** Py_UniversalNewlineFread is an fread variation that understands
** all of \r, \n and \r\n conventions.
** The stream should be opened in binary mode.
** fobj must be a PyFileObject. In this case there
** is no readahead but in stead a flag is used to skip a following
** \n on the next read. Also, if the file is open in binary mode
** the whole conversion is skipped. Finally, the routine keeps track of
** the different types of newlines seen.
*/
size_t
Py_UniversalNewlineFread(char *buf, size_t n,
FILE *stream, PyObject *fobj)
{
char *dst = buf;
PyFileObject *f = (PyFileObject *)fobj;
int newlinetypes, skipnextlf;
assert(buf != NULL);
assert(stream != NULL);
if (!fobj || !PyFile_Check(fobj)) {
errno = ENXIO; /* What can you do... */
return 0;
}
if (!f->f_univ_newline)
return fread(buf, 1, n, stream);
newlinetypes = f->f_newlinetypes;
skipnextlf = f->f_skipnextlf;
/* Invariant: n is the number of bytes remaining to be filled
* in the buffer.
*/
while (n) {
size_t nread;
int shortread;
char *src = dst;
nread = fread(dst, 1, n, stream);
assert(nread <= n);
if (nread == 0)
break;
n -= nread; /* assuming 1 byte out for each in; will adjust */
shortread = n != 0; /* true iff EOF or error */
while (nread--) {
char c = *src++;
if (c == '\r') {
/* Save as LF and set flag to skip next LF. */
*dst++ = '\n';
skipnextlf = 1;
}
else if (skipnextlf && c == '\n') {
/* Skip LF, and remember we saw CR LF. */
skipnextlf = 0;
newlinetypes |= NEWLINE_CRLF;
++n;
}
else {
/* Normal char to be stored in buffer. Also
* update the newlinetypes flag if either this
* is an LF or the previous char was a CR.
*/
if (c == '\n')
newlinetypes |= NEWLINE_LF;
else if (skipnextlf)
newlinetypes |= NEWLINE_CR;
*dst++ = c;
skipnextlf = 0;
}
}
if (shortread) {
/* If this is EOF, update type flags. */
if (skipnextlf && feof(stream))
newlinetypes |= NEWLINE_CR;
break;
}
}
f->f_newlinetypes = newlinetypes;
f->f_skipnextlf = skipnextlf;
return dst - buf;
}
#ifdef __cplusplus
}
#endif
|