summaryrefslogtreecommitdiff
path: root/builtin/blame.c
blob: 6fc7bff9a37e7b9acf24eb65546dfb8e3c87c8fb (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
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
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717
2718
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740
2741
2742
2743
2744
2745
2746
2747
2748
2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
2764
2765
2766
2767
2768
2769
2770
2771
2772
2773
2774
2775
2776
2777
2778
2779
2780
2781
2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
2807
2808
2809
2810
2811
2812
2813
2814
2815
2816
2817
2818
2819
2820
2821
2822
2823
2824
2825
2826
2827
2828
2829
2830
2831
/*
 * Blame
 *
 * Copyright (c) 2006, 2014 by its authors
 * See COPYING for licensing conditions
 */

#include "cache.h"
#include "refs.h"
#include "builtin.h"
#include "blob.h"
#include "commit.h"
#include "tag.h"
#include "tree-walk.h"
#include "diff.h"
#include "diffcore.h"
#include "revision.h"
#include "quote.h"
#include "xdiff-interface.h"
#include "cache-tree.h"
#include "string-list.h"
#include "mailmap.h"
#include "mergesort.h"
#include "parse-options.h"
#include "prio-queue.h"
#include "utf8.h"
#include "userdiff.h"
#include "line-range.h"
#include "line-log.h"
#include "dir.h"

static char blame_usage[] = N_("git blame [<options>] [<rev-opts>] [<rev>] [--] <file>");

static const char *blame_opt_usage[] = {
	blame_usage,
	"",
	N_("<rev-opts> are documented in git-rev-list(1)"),
	NULL
};

static int longest_file;
static int longest_author;
static int max_orig_digits;
static int max_digits;
static int max_score_digits;
static int show_root;
static int reverse;
static int blank_boundary;
static int incremental;
static int xdl_opts;
static int abbrev = -1;
static int no_whole_file_rename;

static struct date_mode blame_date_mode = { DATE_ISO8601 };
static size_t blame_date_width;

static struct string_list mailmap;

#ifndef DEBUG
#define DEBUG 0
#endif

/* stats */
static int num_read_blob;
static int num_get_patch;
static int num_commits;

#define PICKAXE_BLAME_MOVE		01
#define PICKAXE_BLAME_COPY		02
#define PICKAXE_BLAME_COPY_HARDER	04
#define PICKAXE_BLAME_COPY_HARDEST	010

/*
 * blame for a blame_entry with score lower than these thresholds
 * is not passed to the parent using move/copy logic.
 */
static unsigned blame_move_score;
static unsigned blame_copy_score;
#define BLAME_DEFAULT_MOVE_SCORE	20
#define BLAME_DEFAULT_COPY_SCORE	40

/* Remember to update object flag allocation in object.h */
#define METAINFO_SHOWN		(1u<<12)
#define MORE_THAN_ONE_PATH	(1u<<13)

/*
 * One blob in a commit that is being suspected
 */
struct origin {
	int refcnt;
	/* Record preceding blame record for this blob */
	struct origin *previous;
	/* origins are put in a list linked via `next' hanging off the
	 * corresponding commit's util field in order to make finding
	 * them fast.  The presence in this chain does not count
	 * towards the origin's reference count.  It is tempting to
	 * let it count as long as the commit is pending examination,
	 * but even under circumstances where the commit will be
	 * present multiple times in the priority queue of unexamined
	 * commits, processing the first instance will not leave any
	 * work requiring the origin data for the second instance.  An
	 * interspersed commit changing that would have to be
	 * preexisting with a different ancestry and with the same
	 * commit date in order to wedge itself between two instances
	 * of the same commit in the priority queue _and_ produce
	 * blame entries relevant for it.  While we don't want to let
	 * us get tripped up by this case, it certainly does not seem
	 * worth optimizing for.
	 */
	struct origin *next;
	struct commit *commit;
	/* `suspects' contains blame entries that may be attributed to
	 * this origin's commit or to parent commits.  When a commit
	 * is being processed, all suspects will be moved, either by
	 * assigning them to an origin in a different commit, or by
	 * shipping them to the scoreboard's ent list because they
	 * cannot be attributed to a different commit.
	 */
	struct blame_entry *suspects;
	mmfile_t file;
	unsigned char blob_sha1[20];
	unsigned mode;
	/* guilty gets set when shipping any suspects to the final
	 * blame list instead of other commits
	 */
	char guilty;
	char path[FLEX_ARRAY];
};

static int diff_hunks(mmfile_t *file_a, mmfile_t *file_b, long ctxlen,
		      xdl_emit_hunk_consume_func_t hunk_func, void *cb_data)
{
	xpparam_t xpp = {0};
	xdemitconf_t xecfg = {0};
	xdemitcb_t ecb = {NULL};

	xpp.flags = xdl_opts;
	xecfg.ctxlen = ctxlen;
	xecfg.hunk_func = hunk_func;
	ecb.priv = cb_data;
	return xdi_diff(file_a, file_b, &xpp, &xecfg, &ecb);
}

/*
 * Prepare diff_filespec and convert it using diff textconv API
 * if the textconv driver exists.
 * Return 1 if the conversion succeeds, 0 otherwise.
 */
int textconv_object(const char *path,
		    unsigned mode,
		    const unsigned char *sha1,
		    int sha1_valid,
		    char **buf,
		    unsigned long *buf_size)
{
	struct diff_filespec *df;
	struct userdiff_driver *textconv;

	df = alloc_filespec(path);
	fill_filespec(df, sha1, sha1_valid, mode);
	textconv = get_textconv(df);
	if (!textconv) {
		free_filespec(df);
		return 0;
	}

	*buf_size = fill_textconv(textconv, df, buf);
	free_filespec(df);
	return 1;
}

/*
 * Given an origin, prepare mmfile_t structure to be used by the
 * diff machinery
 */
static void fill_origin_blob(struct diff_options *opt,
			     struct origin *o, mmfile_t *file)
{
	if (!o->file.ptr) {
		enum object_type type;
		unsigned long file_size;

		num_read_blob++;
		if (DIFF_OPT_TST(opt, ALLOW_TEXTCONV) &&
		    textconv_object(o->path, o->mode, o->blob_sha1, 1, &file->ptr, &file_size))
			;
		else
			file->ptr = read_sha1_file(o->blob_sha1, &type, &file_size);
		file->size = file_size;

		if (!file->ptr)
			die("Cannot read blob %s for path %s",
			    sha1_to_hex(o->blob_sha1),
			    o->path);
		o->file = *file;
	}
	else
		*file = o->file;
}

/*
 * Origin is refcounted and usually we keep the blob contents to be
 * reused.
 */
static inline struct origin *origin_incref(struct origin *o)
{
	if (o)
		o->refcnt++;
	return o;
}

static void origin_decref(struct origin *o)
{
	if (o && --o->refcnt <= 0) {
		struct origin *p, *l = NULL;
		if (o->previous)
			origin_decref(o->previous);
		free(o->file.ptr);
		/* Should be present exactly once in commit chain */
		for (p = o->commit->util; p; l = p, p = p->next) {
			if (p == o) {
				if (l)
					l->next = p->next;
				else
					o->commit->util = p->next;
				free(o);
				return;
			}
		}
		die("internal error in blame::origin_decref");
	}
}

static void drop_origin_blob(struct origin *o)
{
	if (o->file.ptr) {
		free(o->file.ptr);
		o->file.ptr = NULL;
	}
}

/*
 * Each group of lines is described by a blame_entry; it can be split
 * as we pass blame to the parents.  They are arranged in linked lists
 * kept as `suspects' of some unprocessed origin, or entered (when the
 * blame origin has been finalized) into the scoreboard structure.
 * While the scoreboard structure is only sorted at the end of
 * processing (according to final image line number), the lists
 * attached to an origin are sorted by the target line number.
 */
struct blame_entry {
	struct blame_entry *next;

	/* the first line of this group in the final image;
	 * internally all line numbers are 0 based.
	 */
	int lno;

	/* how many lines this group has */
	int num_lines;

	/* the commit that introduced this group into the final image */
	struct origin *suspect;

	/* the line number of the first line of this group in the
	 * suspect's file; internally all line numbers are 0 based.
	 */
	int s_lno;

	/* how significant this entry is -- cached to avoid
	 * scanning the lines over and over.
	 */
	unsigned score;
};

/*
 * Any merge of blames happens on lists of blames that arrived via
 * different parents in a single suspect.  In this case, we want to
 * sort according to the suspect line numbers as opposed to the final
 * image line numbers.  The function body is somewhat longish because
 * it avoids unnecessary writes.
 */

static struct blame_entry *blame_merge(struct blame_entry *list1,
				       struct blame_entry *list2)
{
	struct blame_entry *p1 = list1, *p2 = list2,
		**tail = &list1;

	if (!p1)
		return p2;
	if (!p2)
		return p1;

	if (p1->s_lno <= p2->s_lno) {
		do {
			tail = &p1->next;
			if ((p1 = *tail) == NULL) {
				*tail = p2;
				return list1;
			}
		} while (p1->s_lno <= p2->s_lno);
	}
	for (;;) {
		*tail = p2;
		do {
			tail = &p2->next;
			if ((p2 = *tail) == NULL)  {
				*tail = p1;
				return list1;
			}
		} while (p1->s_lno > p2->s_lno);
		*tail = p1;
		do {
			tail = &p1->next;
			if ((p1 = *tail) == NULL) {
				*tail = p2;
				return list1;
			}
		} while (p1->s_lno <= p2->s_lno);
	}
}

static void *get_next_blame(const void *p)
{
	return ((struct blame_entry *)p)->next;
}

static void set_next_blame(void *p1, void *p2)
{
	((struct blame_entry *)p1)->next = p2;
}

/*
 * Final image line numbers are all different, so we don't need a
 * three-way comparison here.
 */

static int compare_blame_final(const void *p1, const void *p2)
{
	return ((struct blame_entry *)p1)->lno > ((struct blame_entry *)p2)->lno
		? 1 : -1;
}

static int compare_blame_suspect(const void *p1, const void *p2)
{
	const struct blame_entry *s1 = p1, *s2 = p2;
	/*
	 * to allow for collating suspects, we sort according to the
	 * respective pointer value as the primary sorting criterion.
	 * The actual relation is pretty unimportant as long as it
	 * establishes a total order.  Comparing as integers gives us
	 * that.
	 */
	if (s1->suspect != s2->suspect)
		return (intptr_t)s1->suspect > (intptr_t)s2->suspect ? 1 : -1;
	if (s1->s_lno == s2->s_lno)
		return 0;
	return s1->s_lno > s2->s_lno ? 1 : -1;
}

static struct blame_entry *blame_sort(struct blame_entry *head,
				      int (*compare_fn)(const void *, const void *))
{
	return llist_mergesort (head, get_next_blame, set_next_blame, compare_fn);
}

static int compare_commits_by_reverse_commit_date(const void *a,
						  const void *b,
						  void *c)
{
	return -compare_commits_by_commit_date(a, b, c);
}

/*
 * The current state of the blame assignment.
 */
struct scoreboard {
	/* the final commit (i.e. where we started digging from) */
	struct commit *final;
	/* Priority queue for commits with unassigned blame records */
	struct prio_queue commits;
	struct rev_info *revs;
	const char *path;

	/*
	 * The contents in the final image.
	 * Used by many functions to obtain contents of the nth line,
	 * indexed with scoreboard.lineno[blame_entry.lno].
	 */
	const char *final_buf;
	unsigned long final_buf_size;

	/* linked list of blames */
	struct blame_entry *ent;

	/* look-up a line in the final buffer */
	int num_lines;
	int *lineno;
};

static void sanity_check_refcnt(struct scoreboard *);

/*
 * If two blame entries that are next to each other came from
 * contiguous lines in the same origin (i.e. <commit, path> pair),
 * merge them together.
 */
static void coalesce(struct scoreboard *sb)
{
	struct blame_entry *ent, *next;

	for (ent = sb->ent; ent && (next = ent->next); ent = next) {
		if (ent->suspect == next->suspect &&
		    ent->s_lno + ent->num_lines == next->s_lno) {
			ent->num_lines += next->num_lines;
			ent->next = next->next;
			origin_decref(next->suspect);
			free(next);
			ent->score = 0;
			next = ent; /* again */
		}
	}

	if (DEBUG) /* sanity */
		sanity_check_refcnt(sb);
}

/*
 * Merge the given sorted list of blames into a preexisting origin.
 * If there were no previous blames to that commit, it is entered into
 * the commit priority queue of the score board.
 */

static void queue_blames(struct scoreboard *sb, struct origin *porigin,
			 struct blame_entry *sorted)
{
	if (porigin->suspects)
		porigin->suspects = blame_merge(porigin->suspects, sorted);
	else {
		struct origin *o;
		for (o = porigin->commit->util; o; o = o->next) {
			if (o->suspects) {
				porigin->suspects = sorted;
				return;
			}
		}
		porigin->suspects = sorted;
		prio_queue_put(&sb->commits, porigin->commit);
	}
}

/*
 * Given a commit and a path in it, create a new origin structure.
 * The callers that add blame to the scoreboard should use
 * get_origin() to obtain shared, refcounted copy instead of calling
 * this function directly.
 */
static struct origin *make_origin(struct commit *commit, const char *path)
{
	struct origin *o;
	size_t pathlen = strlen(path) + 1;
	o = xcalloc(1, sizeof(*o) + pathlen);
	o->commit = commit;
	o->refcnt = 1;
	o->next = commit->util;
	commit->util = o;
	memcpy(o->path, path, pathlen); /* includes NUL */
	return o;
}

/*
 * Locate an existing origin or create a new one.
 * This moves the origin to front position in the commit util list.
 */
static struct origin *get_origin(struct scoreboard *sb,
				 struct commit *commit,
				 const char *path)
{
	struct origin *o, *l;

	for (o = commit->util, l = NULL; o; l = o, o = o->next) {
		if (!strcmp(o->path, path)) {
			/* bump to front */
			if (l) {
				l->next = o->next;
				o->next = commit->util;
				commit->util = o;
			}
			return origin_incref(o);
		}
	}
	return make_origin(commit, path);
}

/*
 * Fill the blob_sha1 field of an origin if it hasn't, so that later
 * call to fill_origin_blob() can use it to locate the data.  blob_sha1
 * for an origin is also used to pass the blame for the entire file to
 * the parent to detect the case where a child's blob is identical to
 * that of its parent's.
 *
 * This also fills origin->mode for corresponding tree path.
 */
static int fill_blob_sha1_and_mode(struct origin *origin)
{
	if (!is_null_sha1(origin->blob_sha1))
		return 0;
	if (get_tree_entry(origin->commit->object.sha1,
			   origin->path,
			   origin->blob_sha1, &origin->mode))
		goto error_out;
	if (sha1_object_info(origin->blob_sha1, NULL) != OBJ_BLOB)
		goto error_out;
	return 0;
 error_out:
	hashclr(origin->blob_sha1);
	origin->mode = S_IFINVALID;
	return -1;
}

/*
 * We have an origin -- check if the same path exists in the
 * parent and return an origin structure to represent it.
 */
static struct origin *find_origin(struct scoreboard *sb,
				  struct commit *parent,
				  struct origin *origin)
{
	struct origin *porigin;
	struct diff_options diff_opts;
	const char *paths[2];

	/* First check any existing origins */
	for (porigin = parent->util; porigin; porigin = porigin->next)
		if (!strcmp(porigin->path, origin->path)) {
			/*
			 * The same path between origin and its parent
			 * without renaming -- the most common case.
			 */
			return origin_incref (porigin);
		}

	/* See if the origin->path is different between parent
	 * and origin first.  Most of the time they are the
	 * same and diff-tree is fairly efficient about this.
	 */
	diff_setup(&diff_opts);
	DIFF_OPT_SET(&diff_opts, RECURSIVE);
	diff_opts.detect_rename = 0;
	diff_opts.output_format = DIFF_FORMAT_NO_OUTPUT;
	paths[0] = origin->path;
	paths[1] = NULL;

	parse_pathspec(&diff_opts.pathspec,
		       PATHSPEC_ALL_MAGIC & ~PATHSPEC_LITERAL,
		       PATHSPEC_LITERAL_PATH, "", paths);
	diff_setup_done(&diff_opts);

	if (is_null_sha1(origin->commit->object.sha1))
		do_diff_cache(parent->tree->object.sha1, &diff_opts);
	else
		diff_tree_sha1(parent->tree->object.sha1,
			       origin->commit->tree->object.sha1,
			       "", &diff_opts);
	diffcore_std(&diff_opts);

	if (!diff_queued_diff.nr) {
		/* The path is the same as parent */
		porigin = get_origin(sb, parent, origin->path);
		hashcpy(porigin->blob_sha1, origin->blob_sha1);
		porigin->mode = origin->mode;
	} else {
		/*
		 * Since origin->path is a pathspec, if the parent
		 * commit had it as a directory, we will see a whole
		 * bunch of deletion of files in the directory that we
		 * do not care about.
		 */
		int i;
		struct diff_filepair *p = NULL;
		for (i = 0; i < diff_queued_diff.nr; i++) {
			const char *name;
			p = diff_queued_diff.queue[i];
			name = p->one->path ? p->one->path : p->two->path;
			if (!strcmp(name, origin->path))
				break;
		}
		if (!p)
			die("internal error in blame::find_origin");
		switch (p->status) {
		default:
			die("internal error in blame::find_origin (%c)",
			    p->status);
		case 'M':
			porigin = get_origin(sb, parent, origin->path);
			hashcpy(porigin->blob_sha1, p->one->sha1);
			porigin->mode = p->one->mode;
			break;
		case 'A':
		case 'T':
			/* Did not exist in parent, or type changed */
			break;
		}
	}
	diff_flush(&diff_opts);
	free_pathspec(&diff_opts.pathspec);
	return porigin;
}

/*
 * We have an origin -- find the path that corresponds to it in its
 * parent and return an origin structure to represent it.
 */
static struct origin *find_rename(struct scoreboard *sb,
				  struct commit *parent,
				  struct origin *origin)
{
	struct origin *porigin = NULL;
	struct diff_options diff_opts;
	int i;

	diff_setup(&diff_opts);
	DIFF_OPT_SET(&diff_opts, RECURSIVE);
	diff_opts.detect_rename = DIFF_DETECT_RENAME;
	diff_opts.output_format = DIFF_FORMAT_NO_OUTPUT;
	diff_opts.single_follow = origin->path;
	diff_setup_done(&diff_opts);

	if (is_null_sha1(origin->commit->object.sha1))
		do_diff_cache(parent->tree->object.sha1, &diff_opts);
	else
		diff_tree_sha1(parent->tree->object.sha1,
			       origin->commit->tree->object.sha1,
			       "", &diff_opts);
	diffcore_std(&diff_opts);

	for (i = 0; i < diff_queued_diff.nr; i++) {
		struct diff_filepair *p = diff_queued_diff.queue[i];
		if ((p->status == 'R' || p->status == 'C') &&
		    !strcmp(p->two->path, origin->path)) {
			porigin = get_origin(sb, parent, p->one->path);
			hashcpy(porigin->blob_sha1, p->one->sha1);
			porigin->mode = p->one->mode;
			break;
		}
	}
	diff_flush(&diff_opts);
	free_pathspec(&diff_opts.pathspec);
	return porigin;
}

/*
 * Append a new blame entry to a given output queue.
 */
static void add_blame_entry(struct blame_entry ***queue, struct blame_entry *e)
{
	origin_incref(e->suspect);

	e->next = **queue;
	**queue = e;
	*queue = &e->next;
}

/*
 * src typically is on-stack; we want to copy the information in it to
 * a malloced blame_entry that gets added to the given queue.  The
 * origin of dst loses a refcnt.
 */
static void dup_entry(struct blame_entry ***queue,
		      struct blame_entry *dst, struct blame_entry *src)
{
	origin_incref(src->suspect);
	origin_decref(dst->suspect);
	memcpy(dst, src, sizeof(*src));
	dst->next = **queue;
	**queue = dst;
	*queue = &dst->next;
}

static const char *nth_line(struct scoreboard *sb, long lno)
{
	return sb->final_buf + sb->lineno[lno];
}

static const char *nth_line_cb(void *data, long lno)
{
	return nth_line((struct scoreboard *)data, lno);
}

/*
 * It is known that lines between tlno to same came from parent, and e
 * has an overlap with that range.  it also is known that parent's
 * line plno corresponds to e's line tlno.
 *
 *                <---- e ----->
 *                   <------>
 *                   <------------>
 *             <------------>
 *             <------------------>
 *
 * Split e into potentially three parts; before this chunk, the chunk
 * to be blamed for the parent, and after that portion.
 */
static void split_overlap(struct blame_entry *split,
			  struct blame_entry *e,
			  int tlno, int plno, int same,
			  struct origin *parent)
{
	int chunk_end_lno;
	memset(split, 0, sizeof(struct blame_entry [3]));

	if (e->s_lno < tlno) {
		/* there is a pre-chunk part not blamed on parent */
		split[0].suspect = origin_incref(e->suspect);
		split[0].lno = e->lno;
		split[0].s_lno = e->s_lno;
		split[0].num_lines = tlno - e->s_lno;
		split[1].lno = e->lno + tlno - e->s_lno;
		split[1].s_lno = plno;
	}
	else {
		split[1].lno = e->lno;
		split[1].s_lno = plno + (e->s_lno - tlno);
	}

	if (same < e->s_lno + e->num_lines) {
		/* there is a post-chunk part not blamed on parent */
		split[2].suspect = origin_incref(e->suspect);
		split[2].lno = e->lno + (same - e->s_lno);
		split[2].s_lno = e->s_lno + (same - e->s_lno);
		split[2].num_lines = e->s_lno + e->num_lines - same;
		chunk_end_lno = split[2].lno;
	}
	else
		chunk_end_lno = e->lno + e->num_lines;
	split[1].num_lines = chunk_end_lno - split[1].lno;

	/*
	 * if it turns out there is nothing to blame the parent for,
	 * forget about the splitting.  !split[1].suspect signals this.
	 */
	if (split[1].num_lines < 1)
		return;
	split[1].suspect = origin_incref(parent);
}

/*
 * split_overlap() divided an existing blame e into up to three parts
 * in split.  Any assigned blame is moved to queue to
 * reflect the split.
 */
static void split_blame(struct blame_entry ***blamed,
			struct blame_entry ***unblamed,
			struct blame_entry *split,
			struct blame_entry *e)
{
	struct blame_entry *new_entry;

	if (split[0].suspect && split[2].suspect) {
		/* The first part (reuse storage for the existing entry e) */
		dup_entry(unblamed, e, &split[0]);

		/* The last part -- me */
		new_entry = xmalloc(sizeof(*new_entry));
		memcpy(new_entry, &(split[2]), sizeof(struct blame_entry));
		add_blame_entry(unblamed, new_entry);

		/* ... and the middle part -- parent */
		new_entry = xmalloc(sizeof(*new_entry));
		memcpy(new_entry, &(split[1]), sizeof(struct blame_entry));
		add_blame_entry(blamed, new_entry);
	}
	else if (!split[0].suspect && !split[2].suspect)
		/*
		 * The parent covers the entire area; reuse storage for
		 * e and replace it with the parent.
		 */
		dup_entry(blamed, e, &split[1]);
	else if (split[0].suspect) {
		/* me and then parent */
		dup_entry(unblamed, e, &split[0]);

		new_entry = xmalloc(sizeof(*new_entry));
		memcpy(new_entry, &(split[1]), sizeof(struct blame_entry));
		add_blame_entry(blamed, new_entry);
	}
	else {
		/* parent and then me */
		dup_entry(blamed, e, &split[1]);

		new_entry = xmalloc(sizeof(*new_entry));
		memcpy(new_entry, &(split[2]), sizeof(struct blame_entry));
		add_blame_entry(unblamed, new_entry);
	}
}

/*
 * After splitting the blame, the origins used by the
 * on-stack blame_entry should lose one refcnt each.
 */
static void decref_split(struct blame_entry *split)
{
	int i;

	for (i = 0; i < 3; i++)
		origin_decref(split[i].suspect);
}

/*
 * reverse_blame reverses the list given in head, appending tail.
 * That allows us to build lists in reverse order, then reverse them
 * afterwards.  This can be faster than building the list in proper
 * order right away.  The reason is that building in proper order
 * requires writing a link in the _previous_ element, while building
 * in reverse order just requires placing the list head into the
 * _current_ element.
 */

static struct blame_entry *reverse_blame(struct blame_entry *head,
					 struct blame_entry *tail)
{
	while (head) {
		struct blame_entry *next = head->next;
		head->next = tail;
		tail = head;
		head = next;
	}
	return tail;
}

/*
 * Process one hunk from the patch between the current suspect for
 * blame_entry e and its parent.  This first blames any unfinished
 * entries before the chunk (which is where target and parent start
 * differing) on the parent, and then splits blame entries at the
 * start and at the end of the difference region.  Since use of -M and
 * -C options may lead to overlapping/duplicate source line number
 * ranges, all we can rely on from sorting/merging is the order of the
 * first suspect line number.
 */
static void blame_chunk(struct blame_entry ***dstq, struct blame_entry ***srcq,
			int tlno, int offset, int same,
			struct origin *parent)
{
	struct blame_entry *e = **srcq;
	struct blame_entry *samep = NULL, *diffp = NULL;

	while (e && e->s_lno < tlno) {
		struct blame_entry *next = e->next;
		/*
		 * current record starts before differing portion.  If
		 * it reaches into it, we need to split it up and
		 * examine the second part separately.
		 */
		if (e->s_lno + e->num_lines > tlno) {
			/* Move second half to a new record */
			int len = tlno - e->s_lno;
			struct blame_entry *n = xcalloc(1, sizeof (struct blame_entry));
			n->suspect = e->suspect;
			n->lno = e->lno + len;
			n->s_lno = e->s_lno + len;
			n->num_lines = e->num_lines - len;
			e->num_lines = len;
			e->score = 0;
			/* Push new record to diffp */
			n->next = diffp;
			diffp = n;
		} else
			origin_decref(e->suspect);
		/* Pass blame for everything before the differing
		 * chunk to the parent */
		e->suspect = origin_incref(parent);
		e->s_lno += offset;
		e->next = samep;
		samep = e;
		e = next;
	}
	/*
	 * As we don't know how much of a common stretch after this
	 * diff will occur, the currently blamed parts are all that we
	 * can assign to the parent for now.
	 */

	if (samep) {
		**dstq = reverse_blame(samep, **dstq);
		*dstq = &samep->next;
	}
	/*
	 * Prepend the split off portions: everything after e starts
	 * after the blameable portion.
	 */
	e = reverse_blame(diffp, e);

	/*
	 * Now retain records on the target while parts are different
	 * from the parent.
	 */
	samep = NULL;
	diffp = NULL;
	while (e && e->s_lno < same) {
		struct blame_entry *next = e->next;

		/*
		 * If current record extends into sameness, need to split.
		 */
		if (e->s_lno + e->num_lines > same) {
			/*
			 * Move second half to a new record to be
			 * processed by later chunks
			 */
			int len = same - e->s_lno;
			struct blame_entry *n = xcalloc(1, sizeof (struct blame_entry));
			n->suspect = origin_incref(e->suspect);
			n->lno = e->lno + len;
			n->s_lno = e->s_lno + len;
			n->num_lines = e->num_lines - len;
			e->num_lines = len;
			e->score = 0;
			/* Push new record to samep */
			n->next = samep;
			samep = n;
		}
		e->next = diffp;
		diffp = e;
		e = next;
	}
	**srcq = reverse_blame(diffp, reverse_blame(samep, e));
	/* Move across elements that are in the unblamable portion */
	if (diffp)
		*srcq = &diffp->next;
}

struct blame_chunk_cb_data {
	struct origin *parent;
	long offset;
	struct blame_entry **dstq;
	struct blame_entry **srcq;
};

/* diff chunks are from parent to target */
static int blame_chunk_cb(long start_a, long count_a,
			  long start_b, long count_b, void *data)
{
	struct blame_chunk_cb_data *d = data;
	if (start_a - start_b != d->offset)
		die("internal error in blame::blame_chunk_cb");
	blame_chunk(&d->dstq, &d->srcq, start_b, start_a - start_b,
		    start_b + count_b, d->parent);
	d->offset = start_a + count_a - (start_b + count_b);
	return 0;
}

/*
 * We are looking at the origin 'target' and aiming to pass blame
 * for the lines it is suspected to its parent.  Run diff to find
 * which lines came from parent and pass blame for them.
 */
static void pass_blame_to_parent(struct scoreboard *sb,
				 struct origin *target,
				 struct origin *parent)
{
	mmfile_t file_p, file_o;
	struct blame_chunk_cb_data d;
	struct blame_entry *newdest = NULL;

	if (!target->suspects)
		return; /* nothing remains for this target */

	d.parent = parent;
	d.offset = 0;
	d.dstq = &newdest; d.srcq = &target->suspects;

	fill_origin_blob(&sb->revs->diffopt, parent, &file_p);
	fill_origin_blob(&sb->revs->diffopt, target, &file_o);
	num_get_patch++;

	if (diff_hunks(&file_p, &file_o, 0, blame_chunk_cb, &d))
		die("unable to generate diff (%s -> %s)",
		    sha1_to_hex(parent->commit->object.sha1),
		    sha1_to_hex(target->commit->object.sha1));
	/* The rest are the same as the parent */
	blame_chunk(&d.dstq, &d.srcq, INT_MAX, d.offset, INT_MAX, parent);
	*d.dstq = NULL;
	queue_blames(sb, parent, newdest);

	return;
}

/*
 * The lines in blame_entry after splitting blames many times can become
 * very small and trivial, and at some point it becomes pointless to
 * blame the parents.  E.g. "\t\t}\n\t}\n\n" appears everywhere in any
 * ordinary C program, and it is not worth to say it was copied from
 * totally unrelated file in the parent.
 *
 * Compute how trivial the lines in the blame_entry are.
 */
static unsigned ent_score(struct scoreboard *sb, struct blame_entry *e)
{
	unsigned score;
	const char *cp, *ep;

	if (e->score)
		return e->score;

	score = 1;
	cp = nth_line(sb, e->lno);
	ep = nth_line(sb, e->lno + e->num_lines);
	while (cp < ep) {
		unsigned ch = *((unsigned char *)cp);
		if (isalnum(ch))
			score++;
		cp++;
	}
	e->score = score;
	return score;
}

/*
 * best_so_far[] and this[] are both a split of an existing blame_entry
 * that passes blame to the parent.  Maintain best_so_far the best split
 * so far, by comparing this and best_so_far and copying this into
 * bst_so_far as needed.
 */
static void copy_split_if_better(struct scoreboard *sb,
				 struct blame_entry *best_so_far,
				 struct blame_entry *this)
{
	int i;

	if (!this[1].suspect)
		return;
	if (best_so_far[1].suspect) {
		if (ent_score(sb, &this[1]) < ent_score(sb, &best_so_far[1]))
			return;
	}

	for (i = 0; i < 3; i++)
		origin_incref(this[i].suspect);
	decref_split(best_so_far);
	memcpy(best_so_far, this, sizeof(struct blame_entry [3]));
}

/*
 * We are looking at a part of the final image represented by
 * ent (tlno and same are offset by ent->s_lno).
 * tlno is where we are looking at in the final image.
 * up to (but not including) same match preimage.
 * plno is where we are looking at in the preimage.
 *
 * <-------------- final image ---------------------->
 *       <------ent------>
 *         ^tlno ^same
 *    <---------preimage----->
 *         ^plno
 *
 * All line numbers are 0-based.
 */
static void handle_split(struct scoreboard *sb,
			 struct blame_entry *ent,
			 int tlno, int plno, int same,
			 struct origin *parent,
			 struct blame_entry *split)
{
	if (ent->num_lines <= tlno)
		return;
	if (tlno < same) {
		struct blame_entry this[3];
		tlno += ent->s_lno;
		same += ent->s_lno;
		split_overlap(this, ent, tlno, plno, same, parent);
		copy_split_if_better(sb, split, this);
		decref_split(this);
	}
}

struct handle_split_cb_data {
	struct scoreboard *sb;
	struct blame_entry *ent;
	struct origin *parent;
	struct blame_entry *split;
	long plno;
	long tlno;
};

static int handle_split_cb(long start_a, long count_a,
			   long start_b, long count_b, void *data)
{
	struct handle_split_cb_data *d = data;
	handle_split(d->sb, d->ent, d->tlno, d->plno, start_b, d->parent,
		     d->split);
	d->plno = start_a + count_a;
	d->tlno = start_b + count_b;
	return 0;
}

/*
 * Find the lines from parent that are the same as ent so that
 * we can pass blames to it.  file_p has the blob contents for
 * the parent.
 */
static void find_copy_in_blob(struct scoreboard *sb,
			      struct blame_entry *ent,
			      struct origin *parent,
			      struct blame_entry *split,
			      mmfile_t *file_p)
{
	const char *cp;
	mmfile_t file_o;
	struct handle_split_cb_data d;

	memset(&d, 0, sizeof(d));
	d.sb = sb; d.ent = ent; d.parent = parent; d.split = split;
	/*
	 * Prepare mmfile that contains only the lines in ent.
	 */
	cp = nth_line(sb, ent->lno);
	file_o.ptr = (char *) cp;
	file_o.size = nth_line(sb, ent->lno + ent->num_lines) - cp;

	/*
	 * file_o is a part of final image we are annotating.
	 * file_p partially may match that image.
	 */
	memset(split, 0, sizeof(struct blame_entry [3]));
	if (diff_hunks(file_p, &file_o, 1, handle_split_cb, &d))
		die("unable to generate diff (%s)",
		    sha1_to_hex(parent->commit->object.sha1));
	/* remainder, if any, all match the preimage */
	handle_split(sb, ent, d.tlno, d.plno, ent->num_lines, parent, split);
}

/* Move all blame entries from list *source that have a score smaller
 * than score_min to the front of list *small.
 * Returns a pointer to the link pointing to the old head of the small list.
 */

static struct blame_entry **filter_small(struct scoreboard *sb,
					 struct blame_entry **small,
					 struct blame_entry **source,
					 unsigned score_min)
{
	struct blame_entry *p = *source;
	struct blame_entry *oldsmall = *small;
	while (p) {
		if (ent_score(sb, p) <= score_min) {
			*small = p;
			small = &p->next;
			p = *small;
		} else {
			*source = p;
			source = &p->next;
			p = *source;
		}
	}
	*small = oldsmall;
	*source = NULL;
	return small;
}

/*
 * See if lines currently target is suspected for can be attributed to
 * parent.
 */
static void find_move_in_parent(struct scoreboard *sb,
				struct blame_entry ***blamed,
				struct blame_entry **toosmall,
				struct origin *target,
				struct origin *parent)
{
	struct blame_entry *e, split[3];
	struct blame_entry *unblamed = target->suspects;
	struct blame_entry *leftover = NULL;
	mmfile_t file_p;

	if (!unblamed)
		return; /* nothing remains for this target */

	fill_origin_blob(&sb->revs->diffopt, parent, &file_p);
	if (!file_p.ptr)
		return;

	/* At each iteration, unblamed has a NULL-terminated list of
	 * entries that have not yet been tested for blame.  leftover
	 * contains the reversed list of entries that have been tested
	 * without being assignable to the parent.
	 */
	do {
		struct blame_entry **unblamedtail = &unblamed;
		struct blame_entry *next;
		for (e = unblamed; e; e = next) {
			next = e->next;
			find_copy_in_blob(sb, e, parent, split, &file_p);
			if (split[1].suspect &&
			    blame_move_score < ent_score(sb, &split[1])) {
				split_blame(blamed, &unblamedtail, split, e);
			} else {
				e->next = leftover;
				leftover = e;
			}
			decref_split(split);
		}
		*unblamedtail = NULL;
		toosmall = filter_small(sb, toosmall, &unblamed, blame_move_score);
	} while (unblamed);
	target->suspects = reverse_blame(leftover, NULL);
}

struct blame_list {
	struct blame_entry *ent;
	struct blame_entry split[3];
};

/*
 * Count the number of entries the target is suspected for,
 * and prepare a list of entry and the best split.
 */
static struct blame_list *setup_blame_list(struct blame_entry *unblamed,
					   int *num_ents_p)
{
	struct blame_entry *e;
	int num_ents, i;
	struct blame_list *blame_list = NULL;

	for (e = unblamed, num_ents = 0; e; e = e->next)
		num_ents++;
	if (num_ents) {
		blame_list = xcalloc(num_ents, sizeof(struct blame_list));
		for (e = unblamed, i = 0; e; e = e->next)
			blame_list[i++].ent = e;
	}
	*num_ents_p = num_ents;
	return blame_list;
}

/*
 * For lines target is suspected for, see if we can find code movement
 * across file boundary from the parent commit.  porigin is the path
 * in the parent we already tried.
 */
static void find_copy_in_parent(struct scoreboard *sb,
				struct blame_entry ***blamed,
				struct blame_entry **toosmall,
				struct origin *target,
				struct commit *parent,
				struct origin *porigin,
				int opt)
{
	struct diff_options diff_opts;
	int i, j;
	struct blame_list *blame_list;
	int num_ents;
	struct blame_entry *unblamed = target->suspects;
	struct blame_entry *leftover = NULL;

	if (!unblamed)
		return; /* nothing remains for this target */

	diff_setup(&diff_opts);
	DIFF_OPT_SET(&diff_opts, RECURSIVE);
	diff_opts.output_format = DIFF_FORMAT_NO_OUTPUT;

	diff_setup_done(&diff_opts);

	/* Try "find copies harder" on new path if requested;
	 * we do not want to use diffcore_rename() actually to
	 * match things up; find_copies_harder is set only to
	 * force diff_tree_sha1() to feed all filepairs to diff_queue,
	 * and this code needs to be after diff_setup_done(), which
	 * usually makes find-copies-harder imply copy detection.
	 */
	if ((opt & PICKAXE_BLAME_COPY_HARDEST)
	    || ((opt & PICKAXE_BLAME_COPY_HARDER)
		&& (!porigin || strcmp(target->path, porigin->path))))
		DIFF_OPT_SET(&diff_opts, FIND_COPIES_HARDER);

	if (is_null_sha1(target->commit->object.sha1))
		do_diff_cache(parent->tree->object.sha1, &diff_opts);
	else
		diff_tree_sha1(parent->tree->object.sha1,
			       target->commit->tree->object.sha1,
			       "", &diff_opts);

	if (!DIFF_OPT_TST(&diff_opts, FIND_COPIES_HARDER))
		diffcore_std(&diff_opts);

	do {
		struct blame_entry **unblamedtail = &unblamed;
		blame_list = setup_blame_list(unblamed, &num_ents);

		for (i = 0; i < diff_queued_diff.nr; i++) {
			struct diff_filepair *p = diff_queued_diff.queue[i];
			struct origin *norigin;
			mmfile_t file_p;
			struct blame_entry this[3];

			if (!DIFF_FILE_VALID(p->one))
				continue; /* does not exist in parent */
			if (S_ISGITLINK(p->one->mode))
				continue; /* ignore git links */
			if (porigin && !strcmp(p->one->path, porigin->path))
				/* find_move already dealt with this path */
				continue;

			norigin = get_origin(sb, parent, p->one->path);
			hashcpy(norigin->blob_sha1, p->one->sha1);
			norigin->mode = p->one->mode;
			fill_origin_blob(&sb->revs->diffopt, norigin, &file_p);
			if (!file_p.ptr)
				continue;

			for (j = 0; j < num_ents; j++) {
				find_copy_in_blob(sb, blame_list[j].ent,
						  norigin, this, &file_p);
				copy_split_if_better(sb, blame_list[j].split,
						     this);
				decref_split(this);
			}
			origin_decref(norigin);
		}

		for (j = 0; j < num_ents; j++) {
			struct blame_entry *split = blame_list[j].split;
			if (split[1].suspect &&
			    blame_copy_score < ent_score(sb, &split[1])) {
				split_blame(blamed, &unblamedtail, split,
					    blame_list[j].ent);
			} else {
				blame_list[j].ent->next = leftover;
				leftover = blame_list[j].ent;
			}
			decref_split(split);
		}
		free(blame_list);
		*unblamedtail = NULL;
		toosmall = filter_small(sb, toosmall, &unblamed, blame_copy_score);
	} while (unblamed);
	target->suspects = reverse_blame(leftover, NULL);
	diff_flush(&diff_opts);
	free_pathspec(&diff_opts.pathspec);
}

/*
 * The blobs of origin and porigin exactly match, so everything
 * origin is suspected for can be blamed on the parent.
 */
static void pass_whole_blame(struct scoreboard *sb,
			     struct origin *origin, struct origin *porigin)
{
	struct blame_entry *e, *suspects;

	if (!porigin->file.ptr && origin->file.ptr) {
		/* Steal its file */
		porigin->file = origin->file;
		origin->file.ptr = NULL;
	}
	suspects = origin->suspects;
	origin->suspects = NULL;
	for (e = suspects; e; e = e->next) {
		origin_incref(porigin);
		origin_decref(e->suspect);
		e->suspect = porigin;
	}
	queue_blames(sb, porigin, suspects);
}

/*
 * We pass blame from the current commit to its parents.  We keep saying
 * "parent" (and "porigin"), but what we mean is to find scapegoat to
 * exonerate ourselves.
 */
static struct commit_list *first_scapegoat(struct rev_info *revs, struct commit *commit)
{
	if (!reverse) {
		if (revs->first_parent_only &&
		    commit->parents &&
		    commit->parents->next) {
			free_commit_list(commit->parents->next);
			commit->parents->next = NULL;
		}
		return commit->parents;
	}
	return lookup_decoration(&revs->children, &commit->object);
}

static int num_scapegoats(struct rev_info *revs, struct commit *commit)
{
	struct commit_list *l = first_scapegoat(revs, commit);
	return commit_list_count(l);
}

/* Distribute collected unsorted blames to the respected sorted lists
 * in the various origins.
 */
static void distribute_blame(struct scoreboard *sb, struct blame_entry *blamed)
{
	blamed = blame_sort(blamed, compare_blame_suspect);
	while (blamed)
	{
		struct origin *porigin = blamed->suspect;
		struct blame_entry *suspects = NULL;
		do {
			struct blame_entry *next = blamed->next;
			blamed->next = suspects;
			suspects = blamed;
			blamed = next;
		} while (blamed && blamed->suspect == porigin);
		suspects = reverse_blame(suspects, NULL);
		queue_blames(sb, porigin, suspects);
	}
}

#define MAXSG 16

static void pass_blame(struct scoreboard *sb, struct origin *origin, int opt)
{
	struct rev_info *revs = sb->revs;
	int i, pass, num_sg;
	struct commit *commit = origin->commit;
	struct commit_list *sg;
	struct origin *sg_buf[MAXSG];
	struct origin *porigin, **sg_origin = sg_buf;
	struct blame_entry *toosmall = NULL;
	struct blame_entry *blames, **blametail = &blames;

	num_sg = num_scapegoats(revs, commit);
	if (!num_sg)
		goto finish;
	else if (num_sg < ARRAY_SIZE(sg_buf))
		memset(sg_buf, 0, sizeof(sg_buf));
	else
		sg_origin = xcalloc(num_sg, sizeof(*sg_origin));

	/*
	 * The first pass looks for unrenamed path to optimize for
	 * common cases, then we look for renames in the second pass.
	 */
	for (pass = 0; pass < 2 - no_whole_file_rename; pass++) {
		struct origin *(*find)(struct scoreboard *,
				       struct commit *, struct origin *);
		find = pass ? find_rename : find_origin;

		for (i = 0, sg = first_scapegoat(revs, commit);
		     i < num_sg && sg;
		     sg = sg->next, i++) {
			struct commit *p = sg->item;
			int j, same;

			if (sg_origin[i])
				continue;
			if (parse_commit(p))
				continue;
			porigin = find(sb, p, origin);
			if (!porigin)
				continue;
			if (!hashcmp(porigin->blob_sha1, origin->blob_sha1)) {
				pass_whole_blame(sb, origin, porigin);
				origin_decref(porigin);
				goto finish;
			}
			for (j = same = 0; j < i; j++)
				if (sg_origin[j] &&
				    !hashcmp(sg_origin[j]->blob_sha1,
					     porigin->blob_sha1)) {
					same = 1;
					break;
				}
			if (!same)
				sg_origin[i] = porigin;
			else
				origin_decref(porigin);
		}
	}

	num_commits++;
	for (i = 0, sg = first_scapegoat(revs, commit);
	     i < num_sg && sg;
	     sg = sg->next, i++) {
		struct origin *porigin = sg_origin[i];
		if (!porigin)
			continue;
		if (!origin->previous) {
			origin_incref(porigin);
			origin->previous = porigin;
		}
		pass_blame_to_parent(sb, origin, porigin);
		if (!origin->suspects)
			goto finish;
	}

	/*
	 * Optionally find moves in parents' files.
	 */
	if (opt & PICKAXE_BLAME_MOVE) {
		filter_small(sb, &toosmall, &origin->suspects, blame_move_score);
		if (origin->suspects) {
			for (i = 0, sg = first_scapegoat(revs, commit);
			     i < num_sg && sg;
			     sg = sg->next, i++) {
				struct origin *porigin = sg_origin[i];
				if (!porigin)
					continue;
				find_move_in_parent(sb, &blametail, &toosmall, origin, porigin);
				if (!origin->suspects)
					break;
			}
		}
	}

	/*
	 * Optionally find copies from parents' files.
	 */
	if (opt & PICKAXE_BLAME_COPY) {
		if (blame_copy_score > blame_move_score)
			filter_small(sb, &toosmall, &origin->suspects, blame_copy_score);
		else if (blame_copy_score < blame_move_score) {
			origin->suspects = blame_merge(origin->suspects, toosmall);
			toosmall = NULL;
			filter_small(sb, &toosmall, &origin->suspects, blame_copy_score);
		}
		if (!origin->suspects)
			goto finish;

		for (i = 0, sg = first_scapegoat(revs, commit);
		     i < num_sg && sg;
		     sg = sg->next, i++) {
			struct origin *porigin = sg_origin[i];
			find_copy_in_parent(sb, &blametail, &toosmall,
					    origin, sg->item, porigin, opt);
			if (!origin->suspects)
				goto finish;
		}
	}

finish:
	*blametail = NULL;
	distribute_blame(sb, blames);
	/*
	 * prepend toosmall to origin->suspects
	 *
	 * There is no point in sorting: this ends up on a big
	 * unsorted list in the caller anyway.
	 */
	if (toosmall) {
		struct blame_entry **tail = &toosmall;
		while (*tail)
			tail = &(*tail)->next;
		*tail = origin->suspects;
		origin->suspects = toosmall;
	}
	for (i = 0; i < num_sg; i++) {
		if (sg_origin[i]) {
			drop_origin_blob(sg_origin[i]);
			origin_decref(sg_origin[i]);
		}
	}
	drop_origin_blob(origin);
	if (sg_buf != sg_origin)
		free(sg_origin);
}

/*
 * Information on commits, used for output.
 */
struct commit_info {
	struct strbuf author;
	struct strbuf author_mail;
	unsigned long author_time;
	struct strbuf author_tz;

	/* filled only when asked for details */
	struct strbuf committer;
	struct strbuf committer_mail;
	unsigned long committer_time;
	struct strbuf committer_tz;

	struct strbuf summary;
};

/*
 * Parse author/committer line in the commit object buffer
 */
static void get_ac_line(const char *inbuf, const char *what,
	struct strbuf *name, struct strbuf *mail,
	unsigned long *time, struct strbuf *tz)
{
	struct ident_split ident;
	size_t len, maillen, namelen;
	char *tmp, *endp;
	const char *namebuf, *mailbuf;

	tmp = strstr(inbuf, what);
	if (!tmp)
		goto error_out;
	tmp += strlen(what);
	endp = strchr(tmp, '\n');
	if (!endp)
		len = strlen(tmp);
	else
		len = endp - tmp;

	if (split_ident_line(&ident, tmp, len)) {
	error_out:
		/* Ugh */
		tmp = "(unknown)";
		strbuf_addstr(name, tmp);
		strbuf_addstr(mail, tmp);
		strbuf_addstr(tz, tmp);
		*time = 0;
		return;
	}

	namelen = ident.name_end - ident.name_begin;
	namebuf = ident.name_begin;

	maillen = ident.mail_end - ident.mail_begin;
	mailbuf = ident.mail_begin;

	if (ident.date_begin && ident.date_end)
		*time = strtoul(ident.date_begin, NULL, 10);
	else
		*time = 0;

	if (ident.tz_begin && ident.tz_end)
		strbuf_add(tz, ident.tz_begin, ident.tz_end - ident.tz_begin);
	else
		strbuf_addstr(tz, "(unknown)");

	/*
	 * Now, convert both name and e-mail using mailmap
	 */
	map_user(&mailmap, &mailbuf, &maillen,
		 &namebuf, &namelen);

	strbuf_addf(mail, "<%.*s>", (int)maillen, mailbuf);
	strbuf_add(name, namebuf, namelen);
}

static void commit_info_init(struct commit_info *ci)
{

	strbuf_init(&ci->author, 0);
	strbuf_init(&ci->author_mail, 0);
	strbuf_init(&ci->author_tz, 0);
	strbuf_init(&ci->committer, 0);
	strbuf_init(&ci->committer_mail, 0);
	strbuf_init(&ci->committer_tz, 0);
	strbuf_init(&ci->summary, 0);
}

static void commit_info_destroy(struct commit_info *ci)
{

	strbuf_release(&ci->author);
	strbuf_release(&ci->author_mail);
	strbuf_release(&ci->author_tz);
	strbuf_release(&ci->committer);
	strbuf_release(&ci->committer_mail);
	strbuf_release(&ci->committer_tz);
	strbuf_release(&ci->summary);
}

static void get_commit_info(struct commit *commit,
			    struct commit_info *ret,
			    int detailed)
{
	int len;
	const char *subject, *encoding;
	const char *message;

	commit_info_init(ret);

	encoding = get_log_output_encoding();
	message = logmsg_reencode(commit, NULL, encoding);
	get_ac_line(message, "\nauthor ",
		    &ret->author, &ret->author_mail,
		    &ret->author_time, &ret->author_tz);

	if (!detailed) {
		unuse_commit_buffer(commit, message);
		return;
	}

	get_ac_line(message, "\ncommitter ",
		    &ret->committer, &ret->committer_mail,
		    &ret->committer_time, &ret->committer_tz);

	len = find_commit_subject(message, &subject);
	if (len)
		strbuf_add(&ret->summary, subject, len);
	else
		strbuf_addf(&ret->summary, "(%s)", sha1_to_hex(commit->object.sha1));

	unuse_commit_buffer(commit, message);
}

/*
 * To allow LF and other nonportable characters in pathnames,
 * they are c-style quoted as needed.
 */
static void write_filename_info(const char *path)
{
	printf("filename ");
	write_name_quoted(path, stdout, '\n');
}

/*
 * Porcelain/Incremental format wants to show a lot of details per
 * commit.  Instead of repeating this every line, emit it only once,
 * the first time each commit appears in the output (unless the
 * user has specifically asked for us to repeat).
 */
static int emit_one_suspect_detail(struct origin *suspect, int repeat)
{
	struct commit_info ci;

	if (!repeat && (suspect->commit->object.flags & METAINFO_SHOWN))
		return 0;

	suspect->commit->object.flags |= METAINFO_SHOWN;
	get_commit_info(suspect->commit, &ci, 1);
	printf("author %s\n", ci.author.buf);
	printf("author-mail %s\n", ci.author_mail.buf);
	printf("author-time %lu\n", ci.author_time);
	printf("author-tz %s\n", ci.author_tz.buf);
	printf("committer %s\n", ci.committer.buf);
	printf("committer-mail %s\n", ci.committer_mail.buf);
	printf("committer-time %lu\n", ci.committer_time);
	printf("committer-tz %s\n", ci.committer_tz.buf);
	printf("summary %s\n", ci.summary.buf);
	if (suspect->commit->object.flags & UNINTERESTING)
		printf("boundary\n");
	if (suspect->previous) {
		struct origin *prev = suspect->previous;
		printf("previous %s ", sha1_to_hex(prev->commit->object.sha1));
		write_name_quoted(prev->path, stdout, '\n');
	}

	commit_info_destroy(&ci);

	return 1;
}

/*
 * The blame_entry is found to be guilty for the range.
 * Show it in incremental output.
 */
static void found_guilty_entry(struct blame_entry *ent)
{
	if (incremental) {
		struct origin *suspect = ent->suspect;

		printf("%s %d %d %d\n",
		       sha1_to_hex(suspect->commit->object.sha1),
		       ent->s_lno + 1, ent->lno + 1, ent->num_lines);
		emit_one_suspect_detail(suspect, 0);
		write_filename_info(suspect->path);
		maybe_flush_or_die(stdout, "stdout");
	}
}

/*
 * The main loop -- while we have blobs with lines whose true origin
 * is still unknown, pick one blob, and allow its lines to pass blames
 * to its parents. */
static void assign_blame(struct scoreboard *sb, int opt)
{
	struct rev_info *revs = sb->revs;
	struct commit *commit = prio_queue_get(&sb->commits);

	while (commit) {
		struct blame_entry *ent;
		struct origin *suspect = commit->util;

		/* find one suspect to break down */
		while (suspect && !suspect->suspects)
			suspect = suspect->next;

		if (!suspect) {
			commit = prio_queue_get(&sb->commits);
			continue;
		}

		assert(commit == suspect->commit);

		/*
		 * We will use this suspect later in the loop,
		 * so hold onto it in the meantime.
		 */
		origin_incref(suspect);
		parse_commit(commit);
		if (reverse ||
		    (!(commit->object.flags & UNINTERESTING) &&
		     !(revs->max_age != -1 && commit->date < revs->max_age)))
			pass_blame(sb, suspect, opt);
		else {
			commit->object.flags |= UNINTERESTING;
			if (commit->object.parsed)
				mark_parents_uninteresting(commit);
		}
		/* treat root commit as boundary */
		if (!commit->parents && !show_root)
			commit->object.flags |= UNINTERESTING;

		/* Take responsibility for the remaining entries */
		ent = suspect->suspects;
		if (ent) {
			suspect->guilty = 1;
			for (;;) {
				struct blame_entry *next = ent->next;
				found_guilty_entry(ent);
				if (next) {
					ent = next;
					continue;
				}
				ent->next = sb->ent;
				sb->ent = suspect->suspects;
				suspect->suspects = NULL;
				break;
			}
		}
		origin_decref(suspect);

		if (DEBUG) /* sanity */
			sanity_check_refcnt(sb);
	}
}

static const char *format_time(unsigned long time, const char *tz_str,
			       int show_raw_time)
{
	static struct strbuf time_buf = STRBUF_INIT;

	strbuf_reset(&time_buf);
	if (show_raw_time) {
		strbuf_addf(&time_buf, "%lu %s", time, tz_str);
	}
	else {
		const char *time_str;
		size_t time_width;
		int tz;
		tz = atoi(tz_str);
		time_str = show_date(time, tz, &blame_date_mode);
		strbuf_addstr(&time_buf, time_str);
		/*
		 * Add space paddings to time_buf to display a fixed width
		 * string, and use time_width for display width calibration.
		 */
		for (time_width = utf8_strwidth(time_str);
		     time_width < blame_date_width;
		     time_width++)
			strbuf_addch(&time_buf, ' ');
	}
	return time_buf.buf;
}

#define OUTPUT_ANNOTATE_COMPAT	001
#define OUTPUT_LONG_OBJECT_NAME	002
#define OUTPUT_RAW_TIMESTAMP	004
#define OUTPUT_PORCELAIN	010
#define OUTPUT_SHOW_NAME	020
#define OUTPUT_SHOW_NUMBER	040
#define OUTPUT_SHOW_SCORE      0100
#define OUTPUT_NO_AUTHOR       0200
#define OUTPUT_SHOW_EMAIL	0400
#define OUTPUT_LINE_PORCELAIN 01000

static void emit_porcelain_details(struct origin *suspect, int repeat)
{
	if (emit_one_suspect_detail(suspect, repeat) ||
	    (suspect->commit->object.flags & MORE_THAN_ONE_PATH))
		write_filename_info(suspect->path);
}

static void emit_porcelain(struct scoreboard *sb, struct blame_entry *ent,
			   int opt)
{
	int repeat = opt & OUTPUT_LINE_PORCELAIN;
	int cnt;
	const char *cp;
	struct origin *suspect = ent->suspect;
	char hex[GIT_SHA1_HEXSZ + 1];

	sha1_to_hex_r(hex, suspect->commit->object.sha1);
	printf("%s %d %d %d\n",
	       hex,
	       ent->s_lno + 1,
	       ent->lno + 1,
	       ent->num_lines);
	emit_porcelain_details(suspect, repeat);

	cp = nth_line(sb, ent->lno);
	for (cnt = 0; cnt < ent->num_lines; cnt++) {
		char ch;
		if (cnt) {
			printf("%s %d %d\n", hex,
			       ent->s_lno + 1 + cnt,
			       ent->lno + 1 + cnt);
			if (repeat)
				emit_porcelain_details(suspect, 1);
		}
		putchar('\t');
		do {
			ch = *cp++;
			putchar(ch);
		} while (ch != '\n' &&
			 cp < sb->final_buf + sb->final_buf_size);
	}

	if (sb->final_buf_size && cp[-1] != '\n')
		putchar('\n');
}

static void emit_other(struct scoreboard *sb, struct blame_entry *ent, int opt)
{
	int cnt;
	const char *cp;
	struct origin *suspect = ent->suspect;
	struct commit_info ci;
	char hex[GIT_SHA1_HEXSZ + 1];
	int show_raw_time = !!(opt & OUTPUT_RAW_TIMESTAMP);

	get_commit_info(suspect->commit, &ci, 1);
	sha1_to_hex_r(hex, suspect->commit->object.sha1);

	cp = nth_line(sb, ent->lno);
	for (cnt = 0; cnt < ent->num_lines; cnt++) {
		char ch;
		int length = (opt & OUTPUT_LONG_OBJECT_NAME) ? 40 : abbrev;

		if (suspect->commit->object.flags & UNINTERESTING) {
			if (blank_boundary)
				memset(hex, ' ', length);
			else if (!(opt & OUTPUT_ANNOTATE_COMPAT)) {
				length--;
				putchar('^');
			}
		}

		printf("%.*s", length, hex);
		if (opt & OUTPUT_ANNOTATE_COMPAT) {
			const char *name;
			if (opt & OUTPUT_SHOW_EMAIL)
				name = ci.author_mail.buf;
			else
				name = ci.author.buf;
			printf("\t(%10s\t%10s\t%d)", name,
			       format_time(ci.author_time, ci.author_tz.buf,
					   show_raw_time),
			       ent->lno + 1 + cnt);
		} else {
			if (opt & OUTPUT_SHOW_SCORE)
				printf(" %*d %02d",
				       max_score_digits, ent->score,
				       ent->suspect->refcnt);
			if (opt & OUTPUT_SHOW_NAME)
				printf(" %-*.*s", longest_file, longest_file,
				       suspect->path);
			if (opt & OUTPUT_SHOW_NUMBER)
				printf(" %*d", max_orig_digits,
				       ent->s_lno + 1 + cnt);

			if (!(opt & OUTPUT_NO_AUTHOR)) {
				const char *name;
				int pad;
				if (opt & OUTPUT_SHOW_EMAIL)
					name = ci.author_mail.buf;
				else
					name = ci.author.buf;
				pad = longest_author - utf8_strwidth(name);
				printf(" (%s%*s %10s",
				       name, pad, "",
				       format_time(ci.author_time,
						   ci.author_tz.buf,
						   show_raw_time));
			}
			printf(" %*d) ",
			       max_digits, ent->lno + 1 + cnt);
		}
		do {
			ch = *cp++;
			putchar(ch);
		} while (ch != '\n' &&
			 cp < sb->final_buf + sb->final_buf_size);
	}

	if (sb->final_buf_size && cp[-1] != '\n')
		putchar('\n');

	commit_info_destroy(&ci);
}

static void output(struct scoreboard *sb, int option)
{
	struct blame_entry *ent;

	if (option & OUTPUT_PORCELAIN) {
		for (ent = sb->ent; ent; ent = ent->next) {
			int count = 0;
			struct origin *suspect;
			struct commit *commit = ent->suspect->commit;
			if (commit->object.flags & MORE_THAN_ONE_PATH)
				continue;
			for (suspect = commit->util; suspect; suspect = suspect->next) {
				if (suspect->guilty && count++) {
					commit->object.flags |= MORE_THAN_ONE_PATH;
					break;
				}
			}
		}
	}

	for (ent = sb->ent; ent; ent = ent->next) {
		if (option & OUTPUT_PORCELAIN)
			emit_porcelain(sb, ent, option);
		else {
			emit_other(sb, ent, option);
		}
	}
}

static const char *get_next_line(const char *start, const char *end)
{
	const char *nl = memchr(start, '\n', end - start);
	return nl ? nl + 1 : end;
}

/*
 * To allow quick access to the contents of nth line in the
 * final image, prepare an index in the scoreboard.
 */
static int prepare_lines(struct scoreboard *sb)
{
	const char *buf = sb->final_buf;
	unsigned long len = sb->final_buf_size;
	const char *end = buf + len;
	const char *p;
	int *lineno;
	int num = 0;

	for (p = buf; p < end; p = get_next_line(p, end))
		num++;

	sb->lineno = lineno = xmalloc(sizeof(*sb->lineno) * (num + 1));

	for (p = buf; p < end; p = get_next_line(p, end))
		*lineno++ = p - buf;

	*lineno = len;

	sb->num_lines = num;
	return sb->num_lines;
}

/*
 * Add phony grafts for use with -S; this is primarily to
 * support git's cvsserver that wants to give a linear history
 * to its clients.
 */
static int read_ancestry(const char *graft_file)
{
	FILE *fp = fopen(graft_file, "r");
	struct strbuf buf = STRBUF_INIT;
	if (!fp)
		return -1;
	while (!strbuf_getwholeline(&buf, fp, '\n')) {
		/* The format is just "Commit Parent1 Parent2 ...\n" */
		struct commit_graft *graft = read_graft_line(buf.buf, buf.len);
		if (graft)
			register_commit_graft(graft, 0);
	}
	fclose(fp);
	strbuf_release(&buf);
	return 0;
}

static int update_auto_abbrev(int auto_abbrev, struct origin *suspect)
{
	const char *uniq = find_unique_abbrev(suspect->commit->object.sha1,
					      auto_abbrev);
	int len = strlen(uniq);
	if (auto_abbrev < len)
		return len;
	return auto_abbrev;
}

/*
 * How many columns do we need to show line numbers, authors,
 * and filenames?
 */
static void find_alignment(struct scoreboard *sb, int *option)
{
	int longest_src_lines = 0;
	int longest_dst_lines = 0;
	unsigned largest_score = 0;
	struct blame_entry *e;
	int compute_auto_abbrev = (abbrev < 0);
	int auto_abbrev = default_abbrev;

	for (e = sb->ent; e; e = e->next) {
		struct origin *suspect = e->suspect;
		int num;

		if (compute_auto_abbrev)
			auto_abbrev = update_auto_abbrev(auto_abbrev, suspect);
		if (strcmp(suspect->path, sb->path))
			*option |= OUTPUT_SHOW_NAME;
		num = strlen(suspect->path);
		if (longest_file < num)
			longest_file = num;
		if (!(suspect->commit->object.flags & METAINFO_SHOWN)) {
			struct commit_info ci;
			suspect->commit->object.flags |= METAINFO_SHOWN;
			get_commit_info(suspect->commit, &ci, 1);
			if (*option & OUTPUT_SHOW_EMAIL)
				num = utf8_strwidth(ci.author_mail.buf);
			else
				num = utf8_strwidth(ci.author.buf);
			if (longest_author < num)
				longest_author = num;
			commit_info_destroy(&ci);
		}
		num = e->s_lno + e->num_lines;
		if (longest_src_lines < num)
			longest_src_lines = num;
		num = e->lno + e->num_lines;
		if (longest_dst_lines < num)
			longest_dst_lines = num;
		if (largest_score < ent_score(sb, e))
			largest_score = ent_score(sb, e);
	}
	max_orig_digits = decimal_width(longest_src_lines);
	max_digits = decimal_width(longest_dst_lines);
	max_score_digits = decimal_width(largest_score);

	if (compute_auto_abbrev)
		/* one more abbrev length is needed for the boundary commit */
		abbrev = auto_abbrev + 1;
}

/*
 * For debugging -- origin is refcounted, and this asserts that
 * we do not underflow.
 */
static void sanity_check_refcnt(struct scoreboard *sb)
{
	int baa = 0;
	struct blame_entry *ent;

	for (ent = sb->ent; ent; ent = ent->next) {
		/* Nobody should have zero or negative refcnt */
		if (ent->suspect->refcnt <= 0) {
			fprintf(stderr, "%s in %s has negative refcnt %d\n",
				ent->suspect->path,
				sha1_to_hex(ent->suspect->commit->object.sha1),
				ent->suspect->refcnt);
			baa = 1;
		}
	}
	if (baa) {
		int opt = 0160;
		find_alignment(sb, &opt);
		output(sb, opt);
		die("Baa %d!", baa);
	}
}

static unsigned parse_score(const char *arg)
{
	char *end;
	unsigned long score = strtoul(arg, &end, 10);
	if (*end)
		return 0;
	return score;
}

static const char *add_prefix(const char *prefix, const char *path)
{
	return prefix_path(prefix, prefix ? strlen(prefix) : 0, path);
}

static int git_blame_config(const char *var, const char *value, void *cb)
{
	if (!strcmp(var, "blame.showroot")) {
		show_root = git_config_bool(var, value);
		return 0;
	}
	if (!strcmp(var, "blame.blankboundary")) {
		blank_boundary = git_config_bool(var, value);
		return 0;
	}
	if (!strcmp(var, "blame.showemail")) {
		int *output_option = cb;
		if (git_config_bool(var, value))
			*output_option |= OUTPUT_SHOW_EMAIL;
		else
			*output_option &= ~OUTPUT_SHOW_EMAIL;
		return 0;
	}
	if (!strcmp(var, "blame.date")) {
		if (!value)
			return config_error_nonbool(var);
		parse_date_format(value, &blame_date_mode);
		return 0;
	}

	if (userdiff_config(var, value) < 0)
		return -1;

	return git_default_config(var, value, cb);
}

static void verify_working_tree_path(struct commit *work_tree, const char *path)
{
	struct commit_list *parents;

	for (parents = work_tree->parents; parents; parents = parents->next) {
		const unsigned char *commit_sha1 = parents->item->object.sha1;
		unsigned char blob_sha1[20];
		unsigned mode;

		if (!get_tree_entry(commit_sha1, path, blob_sha1, &mode) &&
		    sha1_object_info(blob_sha1, NULL) == OBJ_BLOB)
			return;
	}
	die("no such path '%s' in HEAD", path);
}

static struct commit_list **append_parent(struct commit_list **tail, const unsigned char *sha1)
{
	struct commit *parent;

	parent = lookup_commit_reference(sha1);
	if (!parent)
		die("no such commit %s", sha1_to_hex(sha1));
	return &commit_list_insert(parent, tail)->next;
}

static void append_merge_parents(struct commit_list **tail)
{
	int merge_head;
	struct strbuf line = STRBUF_INIT;

	merge_head = open(git_path_merge_head(), O_RDONLY);
	if (merge_head < 0) {
		if (errno == ENOENT)
			return;
		die("cannot open '%s' for reading", git_path_merge_head());
	}

	while (!strbuf_getwholeline_fd(&line, merge_head, '\n')) {
		unsigned char sha1[20];
		if (line.len < 40 || get_sha1_hex(line.buf, sha1))
			die("unknown line in '%s': %s", git_path_merge_head(), line.buf);
		tail = append_parent(tail, sha1);
	}
	close(merge_head);
	strbuf_release(&line);
}

/*
 * This isn't as simple as passing sb->buf and sb->len, because we
 * want to transfer ownership of the buffer to the commit (so we
 * must use detach).
 */
static void set_commit_buffer_from_strbuf(struct commit *c, struct strbuf *sb)
{
	size_t len;
	void *buf = strbuf_detach(sb, &len);
	set_commit_buffer(c, buf, len);
}

/*
 * Prepare a dummy commit that represents the work tree (or staged) item.
 * Note that annotating work tree item never works in the reverse.
 */
static struct commit *fake_working_tree_commit(struct diff_options *opt,
					       const char *path,
					       const char *contents_from)
{
	struct commit *commit;
	struct origin *origin;
	struct commit_list **parent_tail, *parent;
	unsigned char head_sha1[20];
	struct strbuf buf = STRBUF_INIT;
	const char *ident;
	time_t now;
	int size, len;
	struct cache_entry *ce;
	unsigned mode;
	struct strbuf msg = STRBUF_INIT;

	time(&now);
	commit = alloc_commit_node();
	commit->object.parsed = 1;
	commit->date = now;
	parent_tail = &commit->parents;

	if (!resolve_ref_unsafe("HEAD", RESOLVE_REF_READING, head_sha1, NULL))
		die("no such ref: HEAD");

	parent_tail = append_parent(parent_tail, head_sha1);
	append_merge_parents(parent_tail);
	verify_working_tree_path(commit, path);

	origin = make_origin(commit, path);

	ident = fmt_ident("Not Committed Yet", "not.committed.yet", NULL, 0);
	strbuf_addstr(&msg, "tree 0000000000000000000000000000000000000000\n");
	for (parent = commit->parents; parent; parent = parent->next)
		strbuf_addf(&msg, "parent %s\n",
			    sha1_to_hex(parent->item->object.sha1));
	strbuf_addf(&msg,
		    "author %s\n"
		    "committer %s\n\n"
		    "Version of %s from %s\n",
		    ident, ident, path,
		    (!contents_from ? path :
		     (!strcmp(contents_from, "-") ? "standard input" : contents_from)));
	set_commit_buffer_from_strbuf(commit, &msg);

	if (!contents_from || strcmp("-", contents_from)) {
		struct stat st;
		const char *read_from;
		char *buf_ptr;
		unsigned long buf_len;

		if (contents_from) {
			if (stat(contents_from, &st) < 0)
				die_errno("Cannot stat '%s'", contents_from);
			read_from = contents_from;
		}
		else {
			if (lstat(path, &st) < 0)
				die_errno("Cannot lstat '%s'", path);
			read_from = path;
		}
		mode = canon_mode(st.st_mode);

		switch (st.st_mode & S_IFMT) {
		case S_IFREG:
			if (DIFF_OPT_TST(opt, ALLOW_TEXTCONV) &&
			    textconv_object(read_from, mode, null_sha1, 0, &buf_ptr, &buf_len))
				strbuf_attach(&buf, buf_ptr, buf_len, buf_len + 1);
			else if (strbuf_read_file(&buf, read_from, st.st_size) != st.st_size)
				die_errno("cannot open or read '%s'", read_from);
			break;
		case S_IFLNK:
			if (strbuf_readlink(&buf, read_from, st.st_size) < 0)
				die_errno("cannot readlink '%s'", read_from);
			break;
		default:
			die("unsupported file type %s", read_from);
		}
	}
	else {
		/* Reading from stdin */
		mode = 0;
		if (strbuf_read(&buf, 0, 0) < 0)
			die_errno("failed to read from stdin");
	}
	convert_to_git(path, buf.buf, buf.len, &buf, 0);
	origin->file.ptr = buf.buf;
	origin->file.size = buf.len;
	pretend_sha1_file(buf.buf, buf.len, OBJ_BLOB, origin->blob_sha1);

	/*
	 * Read the current index, replace the path entry with
	 * origin->blob_sha1 without mucking with its mode or type
	 * bits; we are not going to write this index out -- we just
	 * want to run "diff-index --cached".
	 */
	discard_cache();
	read_cache();

	len = strlen(path);
	if (!mode) {
		int pos = cache_name_pos(path, len);
		if (0 <= pos)
			mode = active_cache[pos]->ce_mode;
		else
			/* Let's not bother reading from HEAD tree */
			mode = S_IFREG | 0644;
	}
	size = cache_entry_size(len);
	ce = xcalloc(1, size);
	hashcpy(ce->sha1, origin->blob_sha1);
	memcpy(ce->name, path, len);
	ce->ce_flags = create_ce_flags(0);
	ce->ce_namelen = len;
	ce->ce_mode = create_ce_mode(mode);
	add_cache_entry(ce, ADD_CACHE_OK_TO_ADD|ADD_CACHE_OK_TO_REPLACE);

	/*
	 * We are not going to write this out, so this does not matter
	 * right now, but someday we might optimize diff-index --cached
	 * with cache-tree information.
	 */
	cache_tree_invalidate_path(&the_index, path);

	return commit;
}

static char *prepare_final(struct scoreboard *sb)
{
	int i;
	const char *final_commit_name = NULL;
	struct rev_info *revs = sb->revs;

	/*
	 * There must be one and only one positive commit in the
	 * revs->pending array.
	 */
	for (i = 0; i < revs->pending.nr; i++) {
		struct object *obj = revs->pending.objects[i].item;
		if (obj->flags & UNINTERESTING)
			continue;
		while (obj->type == OBJ_TAG)
			obj = deref_tag(obj, NULL, 0);
		if (obj->type != OBJ_COMMIT)
			die("Non commit %s?", revs->pending.objects[i].name);
		if (sb->final)
			die("More than one commit to dig from %s and %s?",
			    revs->pending.objects[i].name,
			    final_commit_name);
		sb->final = (struct commit *) obj;
		final_commit_name = revs->pending.objects[i].name;
	}
	return xstrdup_or_null(final_commit_name);
}

static char *prepare_initial(struct scoreboard *sb)
{
	int i;
	const char *final_commit_name = NULL;
	struct rev_info *revs = sb->revs;

	/*
	 * There must be one and only one negative commit, and it must be
	 * the boundary.
	 */
	for (i = 0; i < revs->pending.nr; i++) {
		struct object *obj = revs->pending.objects[i].item;
		if (!(obj->flags & UNINTERESTING))
			continue;
		while (obj->type == OBJ_TAG)
			obj = deref_tag(obj, NULL, 0);
		if (obj->type != OBJ_COMMIT)
			die("Non commit %s?", revs->pending.objects[i].name);
		if (sb->final)
			die("More than one commit to dig down to %s and %s?",
			    revs->pending.objects[i].name,
			    final_commit_name);
		sb->final = (struct commit *) obj;
		final_commit_name = revs->pending.objects[i].name;
	}
	if (!final_commit_name)
		die("No commit to dig down to?");
	return xstrdup(final_commit_name);
}

static int blame_copy_callback(const struct option *option, const char *arg, int unset)
{
	int *opt = option->value;

	/*
	 * -C enables copy from removed files;
	 * -C -C enables copy from existing files, but only
	 *       when blaming a new file;
	 * -C -C -C enables copy from existing files for
	 *          everybody
	 */
	if (*opt & PICKAXE_BLAME_COPY_HARDER)
		*opt |= PICKAXE_BLAME_COPY_HARDEST;
	if (*opt & PICKAXE_BLAME_COPY)
		*opt |= PICKAXE_BLAME_COPY_HARDER;
	*opt |= PICKAXE_BLAME_COPY | PICKAXE_BLAME_MOVE;

	if (arg)
		blame_copy_score = parse_score(arg);
	return 0;
}

static int blame_move_callback(const struct option *option, const char *arg, int unset)
{
	int *opt = option->value;

	*opt |= PICKAXE_BLAME_MOVE;

	if (arg)
		blame_move_score = parse_score(arg);
	return 0;
}

int cmd_blame(int argc, const char **argv, const char *prefix)
{
	struct rev_info revs;
	const char *path;
	struct scoreboard sb;
	struct origin *o;
	struct blame_entry *ent = NULL;
	long dashdash_pos, lno;
	char *final_commit_name = NULL;
	enum object_type type;

	static struct string_list range_list;
	static int output_option = 0, opt = 0;
	static int show_stats = 0;
	static const char *revs_file = NULL;
	static const char *contents_from = NULL;
	static const struct option options[] = {
		OPT_BOOL(0, "incremental", &incremental, N_("Show blame entries as we find them, incrementally")),
		OPT_BOOL('b', NULL, &blank_boundary, N_("Show blank SHA-1 for boundary commits (Default: off)")),
		OPT_BOOL(0, "root", &show_root, N_("Do not treat root commits as boundaries (Default: off)")),
		OPT_BOOL(0, "show-stats", &show_stats, N_("Show work cost statistics")),
		OPT_BIT(0, "score-debug", &output_option, N_("Show output score for blame entries"), OUTPUT_SHOW_SCORE),
		OPT_BIT('f', "show-name", &output_option, N_("Show original filename (Default: auto)"), OUTPUT_SHOW_NAME),
		OPT_BIT('n', "show-number", &output_option, N_("Show original linenumber (Default: off)"), OUTPUT_SHOW_NUMBER),
		OPT_BIT('p', "porcelain", &output_option, N_("Show in a format designed for machine consumption"), OUTPUT_PORCELAIN),
		OPT_BIT(0, "line-porcelain", &output_option, N_("Show porcelain format with per-line commit information"), OUTPUT_PORCELAIN|OUTPUT_LINE_PORCELAIN),
		OPT_BIT('c', NULL, &output_option, N_("Use the same output mode as git-annotate (Default: off)"), OUTPUT_ANNOTATE_COMPAT),
		OPT_BIT('t', NULL, &output_option, N_("Show raw timestamp (Default: off)"), OUTPUT_RAW_TIMESTAMP),
		OPT_BIT('l', NULL, &output_option, N_("Show long commit SHA1 (Default: off)"), OUTPUT_LONG_OBJECT_NAME),
		OPT_BIT('s', NULL, &output_option, N_("Suppress author name and timestamp (Default: off)"), OUTPUT_NO_AUTHOR),
		OPT_BIT('e', "show-email", &output_option, N_("Show author email instead of name (Default: off)"), OUTPUT_SHOW_EMAIL),
		OPT_BIT('w', NULL, &xdl_opts, N_("Ignore whitespace differences"), XDF_IGNORE_WHITESPACE),
		OPT_BIT(0, "minimal", &xdl_opts, N_("Spend extra cycles to find better match"), XDF_NEED_MINIMAL),
		OPT_STRING('S', NULL, &revs_file, N_("file"), N_("Use revisions from <file> instead of calling git-rev-list")),
		OPT_STRING(0, "contents", &contents_from, N_("file"), N_("Use <file>'s contents as the final image")),
		{ OPTION_CALLBACK, 'C', NULL, &opt, N_("score"), N_("Find line copies within and across files"), PARSE_OPT_OPTARG, blame_copy_callback },
		{ OPTION_CALLBACK, 'M', NULL, &opt, N_("score"), N_("Find line movements within and across files"), PARSE_OPT_OPTARG, blame_move_callback },
		OPT_STRING_LIST('L', NULL, &range_list, N_("n,m"), N_("Process only line range n,m, counting from 1")),
		OPT__ABBREV(&abbrev),
		OPT_END()
	};

	struct parse_opt_ctx_t ctx;
	int cmd_is_annotate = !strcmp(argv[0], "annotate");
	struct range_set ranges;
	unsigned int range_i;
	long anchor;

	git_config(git_blame_config, &output_option);
	init_revisions(&revs, NULL);
	revs.date_mode = blame_date_mode;
	DIFF_OPT_SET(&revs.diffopt, ALLOW_TEXTCONV);
	DIFF_OPT_SET(&revs.diffopt, FOLLOW_RENAMES);

	save_commit_buffer = 0;
	dashdash_pos = 0;

	parse_options_start(&ctx, argc, argv, prefix, options,
			    PARSE_OPT_KEEP_DASHDASH | PARSE_OPT_KEEP_ARGV0);
	for (;;) {
		switch (parse_options_step(&ctx, options, blame_opt_usage)) {
		case PARSE_OPT_HELP:
			exit(129);
		case PARSE_OPT_DONE:
			if (ctx.argv[0])
				dashdash_pos = ctx.cpidx;
			goto parse_done;
		}

		if (!strcmp(ctx.argv[0], "--reverse")) {
			ctx.argv[0] = "--children";
			reverse = 1;
		}
		parse_revision_opt(&revs, &ctx, options, blame_opt_usage);
	}
parse_done:
	no_whole_file_rename = !DIFF_OPT_TST(&revs.diffopt, FOLLOW_RENAMES);
	DIFF_OPT_CLR(&revs.diffopt, FOLLOW_RENAMES);
	argc = parse_options_end(&ctx);

	if (0 < abbrev)
		/* one more abbrev length is needed for the boundary commit */
		abbrev++;

	if (revs_file && read_ancestry(revs_file))
		die_errno("reading graft file '%s' failed", revs_file);

	if (cmd_is_annotate) {
		output_option |= OUTPUT_ANNOTATE_COMPAT;
		blame_date_mode.type = DATE_ISO8601;
	} else {
		blame_date_mode = revs.date_mode;
	}

	/* The maximum width used to show the dates */
	switch (blame_date_mode.type) {
	case DATE_RFC2822:
		blame_date_width = sizeof("Thu, 19 Oct 2006 16:00:04 -0700");
		break;
	case DATE_ISO8601_STRICT:
		blame_date_width = sizeof("2006-10-19T16:00:04-07:00");
		break;
	case DATE_ISO8601:
		blame_date_width = sizeof("2006-10-19 16:00:04 -0700");
		break;
	case DATE_RAW:
		blame_date_width = sizeof("1161298804 -0700");
		break;
	case DATE_SHORT:
		blame_date_width = sizeof("2006-10-19");
		break;
	case DATE_RELATIVE:
		/* TRANSLATORS: This string is used to tell us the maximum
		   display width for a relative timestamp in "git blame"
		   output.  For C locale, "4 years, 11 months ago", which
		   takes 22 places, is the longest among various forms of
		   relative timestamps, but your language may need more or
		   fewer display columns. */
		blame_date_width = utf8_strwidth(_("4 years, 11 months ago")) + 1; /* add the null */
		break;
	case DATE_NORMAL:
		blame_date_width = sizeof("Thu Oct 19 16:00:04 2006 -0700");
		break;
	case DATE_STRFTIME:
		blame_date_width = strlen(show_date(0, 0, &blame_date_mode)) + 1; /* add the null */
		break;
	}
	blame_date_width -= 1; /* strip the null */

	if (DIFF_OPT_TST(&revs.diffopt, FIND_COPIES_HARDER))
		opt |= (PICKAXE_BLAME_COPY | PICKAXE_BLAME_MOVE |
			PICKAXE_BLAME_COPY_HARDER);

	if (!blame_move_score)
		blame_move_score = BLAME_DEFAULT_MOVE_SCORE;
	if (!blame_copy_score)
		blame_copy_score = BLAME_DEFAULT_COPY_SCORE;

	/*
	 * We have collected options unknown to us in argv[1..unk]
	 * which are to be passed to revision machinery if we are
	 * going to do the "bottom" processing.
	 *
	 * The remaining are:
	 *
	 * (1) if dashdash_pos != 0, it is either
	 *     "blame [revisions] -- <path>" or
	 *     "blame -- <path> <rev>"
	 *
	 * (2) otherwise, it is one of the two:
	 *     "blame [revisions] <path>"
	 *     "blame <path> <rev>"
	 *
	 * Note that we must strip out <path> from the arguments: we do not
	 * want the path pruning but we may want "bottom" processing.
	 */
	if (dashdash_pos) {
		switch (argc - dashdash_pos - 1) {
		case 2: /* (1b) */
			if (argc != 4)
				usage_with_options(blame_opt_usage, options);
			/* reorder for the new way: <rev> -- <path> */
			argv[1] = argv[3];
			argv[3] = argv[2];
			argv[2] = "--";
			/* FALLTHROUGH */
		case 1: /* (1a) */
			path = add_prefix(prefix, argv[--argc]);
			argv[argc] = NULL;
			break;
		default:
			usage_with_options(blame_opt_usage, options);
		}
	} else {
		if (argc < 2)
			usage_with_options(blame_opt_usage, options);
		path = add_prefix(prefix, argv[argc - 1]);
		if (argc == 3 && !file_exists(path)) { /* (2b) */
			path = add_prefix(prefix, argv[1]);
			argv[1] = argv[2];
		}
		argv[argc - 1] = "--";

		setup_work_tree();
		if (!file_exists(path))
			die_errno("cannot stat path '%s'", path);
	}

	revs.disable_stdin = 1;
	setup_revisions(argc, argv, &revs, NULL);
	memset(&sb, 0, sizeof(sb));

	sb.revs = &revs;
	if (!reverse) {
		final_commit_name = prepare_final(&sb);
		sb.commits.compare = compare_commits_by_commit_date;
	}
	else if (contents_from)
		die("--contents and --children do not blend well.");
	else if (revs.first_parent_only)
		die("combining --first-parent and --reverse is not supported");
	else {
		final_commit_name = prepare_initial(&sb);
		sb.commits.compare = compare_commits_by_reverse_commit_date;
	}

	if (!sb.final) {
		/*
		 * "--not A B -- path" without anything positive;
		 * do not default to HEAD, but use the working tree
		 * or "--contents".
		 */
		setup_work_tree();
		sb.final = fake_working_tree_commit(&sb.revs->diffopt,
						    path, contents_from);
		add_pending_object(&revs, &(sb.final->object), ":");
	}
	else if (contents_from)
		die("Cannot use --contents with final commit object name");

	/*
	 * If we have bottom, this will mark the ancestors of the
	 * bottom commits we would reach while traversing as
	 * uninteresting.
	 */
	if (prepare_revision_walk(&revs))
		die(_("revision walk setup failed"));

	if (is_null_sha1(sb.final->object.sha1)) {
		o = sb.final->util;
		sb.final_buf = xmemdupz(o->file.ptr, o->file.size);
		sb.final_buf_size = o->file.size;
	}
	else {
		o = get_origin(&sb, sb.final, path);
		if (fill_blob_sha1_and_mode(o))
			die("no such path %s in %s", path, final_commit_name);

		if (DIFF_OPT_TST(&sb.revs->diffopt, ALLOW_TEXTCONV) &&
		    textconv_object(path, o->mode, o->blob_sha1, 1, (char **) &sb.final_buf,
				    &sb.final_buf_size))
			;
		else
			sb.final_buf = read_sha1_file(o->blob_sha1, &type,
						      &sb.final_buf_size);

		if (!sb.final_buf)
			die("Cannot read blob %s for path %s",
			    sha1_to_hex(o->blob_sha1),
			    path);
	}
	num_read_blob++;
	lno = prepare_lines(&sb);

	if (lno && !range_list.nr)
		string_list_append(&range_list, xstrdup("1"));

	anchor = 1;
	range_set_init(&ranges, range_list.nr);
	for (range_i = 0; range_i < range_list.nr; ++range_i) {
		long bottom, top;
		if (parse_range_arg(range_list.items[range_i].string,
				    nth_line_cb, &sb, lno, anchor,
				    &bottom, &top, sb.path))
			usage(blame_usage);
		if (lno < top || ((lno || bottom) && lno < bottom))
			die("file %s has only %lu lines", path, lno);
		if (bottom < 1)
			bottom = 1;
		if (top < 1)
			top = lno;
		bottom--;
		range_set_append_unsafe(&ranges, bottom, top);
		anchor = top + 1;
	}
	sort_and_merge_range_set(&ranges);

	for (range_i = ranges.nr; range_i > 0; --range_i) {
		const struct range *r = &ranges.ranges[range_i - 1];
		long bottom = r->start;
		long top = r->end;
		struct blame_entry *next = ent;
		ent = xcalloc(1, sizeof(*ent));
		ent->lno = bottom;
		ent->num_lines = top - bottom;
		ent->suspect = o;
		ent->s_lno = bottom;
		ent->next = next;
		origin_incref(o);
	}

	o->suspects = ent;
	prio_queue_put(&sb.commits, o->commit);

	origin_decref(o);

	range_set_release(&ranges);
	string_list_clear(&range_list, 0);

	sb.ent = NULL;
	sb.path = path;

	read_mailmap(&mailmap, NULL);

	if (!incremental)
		setup_pager();

	assign_blame(&sb, opt);

	free(final_commit_name);

	if (incremental)
		return 0;

	sb.ent = blame_sort(sb.ent, compare_blame_final);

	coalesce(&sb);

	if (!(output_option & OUTPUT_PORCELAIN))
		find_alignment(&sb, &output_option);

	output(&sb, output_option);
	free((void *)sb.final_buf);
	for (ent = sb.ent; ent; ) {
		struct blame_entry *e = ent->next;
		free(ent);
		ent = e;
	}

	if (show_stats) {
		printf("num read blob: %d\n", num_read_blob);
		printf("num get patch: %d\n", num_get_patch);
		printf("num commits: %d\n", num_commits);
	}
	return 0;
}