I have multiple measurements per ‘Subject’ in file 1. I only want to use the highest quality, singular measurement per Subject. In my second file I have the exact list of which measurement is the best for each Subject. This information is contained in the column ‘seriesnumber’. The number in the ‘seriesnumber’ column in file 2 corresponds to the best measurement for a Subject. I Need to extract only these rows from my file 1.
I have tried to use awk, join, and merge to try and accomplish this but came up with errors and strange incomplete files.
join code:
join -j2 file1 file2
awk code:
awk ' FILENAME=="file1" {arr[$2]=$0; next}
FILENAME=="file2" {print arr[$2]} ' file1 file2 > newfile
File 1 Example
Subject Seriesnumber 19-1-1001 2 8655 661 15250 60747 8005 3919 7393 2264 1479 1663 22968 4180 1712 689 781 4255 90 1260 7233 154 15643 63421 7361 4384 6932 2062 4526 1742 686 4575 100 1684 0 1194 0 0 5 0 0 147 699 315 305 317 565 1361200 1338210 1338690 304258 308180 612438 250614 255920 506534 66645 802424 1206450 1187010 1185180 1816840 1 1 21 17 38 1765590 19-1-1001 10 8992 507 15722 64032 8728 3929 7208 2075 1529 1529 22503 3993 1819 710 764 3870 87 1247 7361 65 16128 66226 8165 4384 6669 1805 4405 1752 779 4039 103 1705 0 1280 0 0 10 0 0 186 685 300 318 320 598 1370490 1347160 1347520 306588 307188 613775 251704 256521 508225 65808 808802 1208880 1189150 1187450 1827880 1 1 22 26 48 1778960 19-1-1103 2 3303 317 12146 57569 7008 3617 6910 2018 811 1593 18708 4708 1429 408 668 3279 14 1289 2351 85 13730 60206 6731 4137 7034 2038 4407 1483 749 3576 85 1668 0 948 0 0 7 0 0 129 602 288 291 285 748 1250030 1238540 1238820 301810 301062 602872 215029 218080 433108 61555 781150 1107360 1098510 1097220 1635560 1 1 32 47 79 1555850 19-1-1103 9 3236 286 12490 59477 7000 3558 6782 2113 894 1752 19338 4818 1724 387 649 3345 56 1314 2077 133 13885 60414 6628 4078 7063 2031 4269 1709 610 3707 112 1947 0 990 0 0 8 0 0 245 604 279 280 284 693 1269820 1258050 1258320 306856 309614 616469 215658 220876 436534 61859 796760 1124870 1115990 1114510 1630740 1 1 32 42 74 1556790 19-10-1010 2 3344 608 14744 59165 8389 4427 6962 2008 716 1496 21980 4008 1474 769 652 3715 61 1400 3049 1072 15767 61919 8325 4824 7117 1936 4001 1546 684 3935 103 1434 0 1624 0 0 3 0 0 316 834 413 520 517 833 1350760 1337040 1336840 311985 312592 624577 246800 251133 497933 65699 809736 1200320 1189410 1188280 1731270 1 1 17 13 30 1606700 19-10-1010 6 3242 616 15205 61330 8019 4520 6791 2093 735 1558 22824 3981 1546 653 614 3672 96 1227 2992 1070 16450 64189 8489 4407 6953 2099 4096 1668 680 4116 99 1449 0 2161 0 0 19 0 0 263 848 387 525 528 824 1339090 1325830 1325780 309464 311916 621380 239958 244616 484574 65493 810887 1183120 1172600 1171430 1720000 1 1 16 26 42 1587100
File 2 Example
Subject seriesnumber 19-10-1010 2 19-10-1166 2 19-102-10005 2 19-102-10006 2 19-103-10009 2 19-103-10010 2 19-104-10013 11 19-104-10014 2 19-105-10017 6 19-105-10018 6
The desired output would like something like this: Where I no longer have duplicate entries per subject. The second column will look different because the preferred series number will differ per subject.
19-10-1010 2 3344 608 14744 59165 8389 4427 6962 2008 716 1496 21980 4008 1474 769 652 3715 61 1400 3049 1072 15767 61919 8325 4824 7117 1936 4001 1546 684 3935 103 1434 0 1624 0 0 3 0 0 316 834 413 520 517 833 1350760 1337040 1336840 311985 312592 624577 246800 251133 497933 65699 809736 1200320 1189410 1188280 1731270 1 1 17 13 30 1606700 19-10-1166 2 3699 312 15373 61787 8026 4248 6385 1955 608 2194 21394 4260 1563 886 609 3420 25 1101 3415 417 16909 63040 7236 4264 5933 1852 4156 1213 654 4007 53 1336 5 1597 0 0 18 0 0 110 821 300 514 466 854 1193020 1179470 1179420 282241 273236 555477 204883 203228 408111 61343 740736 1036210 1026080 1024910 1563950 1 1 39 40 79 1415890 19-102-10005 2 8733 514 13024 50735 7729 3775 4955 1575 1045 1141 20415 3924 1537 990 651 3515 134 1259 8571 232 13487 51374 7150 4169 5192 1664 3760 1620 596 3919 189 1958 0 1479 0 0 36 0 0 203 837 459 409 439 1072 1224350 1200010 1200120 287659 290445 578104 216976 220545 437521 57457 737161 1095770 1074440 1073050 1637570 1 1 31 22 53 1618600 19-102-10006 2 8347 604 13735 42231 7266 3836 6473 2057 1099 1007 18478 3769 1351 978 639 3332 125 1197 8207 454 13774 43750 6758 4274 6148 1921 3732 1584 614 3521 180 1611 0 1241 0 0 25 0 0 254 813 410 352 372 833 1092800 1069450 1069190 244104 245787 489891 202201 205897 408098 59170 634640 978807 958350 957462 1485600 1 1 19 19 38 1472020 19-103-10009 2 4222 596 14702 52038 7428 4065 6598 2166 835 1854 22613 3397 1387 879 568 3729 93 1315 3414 222 14580 52639 7316 3997 6447 1986 4067 1529 596 3778 113 1689 0 2097 0 0 23 0 0 260 761 326 400 359 772 1204670 1190100 1189780 256560 260381 516941 237316 243326 480642 60653 681040 1070620 1059370 1058440 1605990 1 1 25 23 48 1593730 19-103-10010 2 5254 435 14688 47120 7772 3130 5414 1711 741 1912 20643 3594 1449 882 717 3663 41 999 6465 605 14820 49390 6361 3826 5527 1523 3513 1537 639 3596 80 1261 0 1475 0 0 18 0 0 283 827 383 414 297 627 1135490 1117320 1116990 243367 245896 489263 221809 227084 448893 55338 639719 1009370 994519 993639 1568140 1 1 14 11 25 1542210 19-104-10013 2 7276 341 11836 53018 7912 3942 6105 2334 795 2532 21239 4551 1258 1176 430 3636 83 1184 8811 396 12760 53092 7224 4361 6306 1853 4184 1278 543 3921 175 1814 0 2187 0 0 8 0 0 266 783 381 382 357 793 1011640 987712 987042 206633 228397 435031 170375 191222 361597 61814 601948 879229 859619 859103 1586150 1 1 224 162 386 1557120 19-104-10014 2 5964 355 13297 55439 8599 4081 5628 1730 970 1308 20196 4519 1363 992 697 3474 62 1232 6830 472 14729 59478 7006 4443 6156 1825 4492 1726 827 4017 122 1804 0 1412 0 0 17 0 0 259 672 299 305 319 779 1308470 1288970 1288910 284018 285985 570003 258525 257355 515880 62485 746108 1166160 1149700 1148340 1826660 1 1 33 24 57 1630580 19-105-10017 2 7018 307 13848 53855 8345 3734 6001 2095 899 1932 20712 4196 1349 645 823 4212 72 1475 3346 1119 13970 55202 7411 3975 5672 1737 3778 1490 657 4089 132 1689 0 1318 0 0 23 0 0 234 745 474 367 378 760 1122360 1104380 1104520 235806 233881 469687 217939 220736 438675 61471 639143 985718 970903 969619 1583800 1 1 51 51 102 1558470 19-105-10018 2 16454 1098 12569 52521 8215 3788 5858 1805 788 1147 21028 3496 1492 665 634 3796 39 1614 10700 617 12813 52098 8091 3901 5367 1646 3544 1388 723 3938 47 1819 0 1464 0 0 42 0 0 330 832 301 319 400 788 1148940 1114080 1113560 225179 227218 452397 237056 237295 474351 59172 614884 1019300 986820 986144 1607900 1 1 19 28 47 1591480 19-105-10020 2 4096 451 13042 48597 7601 3228 5665 1582 778 1670 19769 3612 1187 717 617 3672 103 962 2627 467 13208 48466 6619 3461 5217 1360 3575 1388 718 3783 90 1370 0 862 0 0 6 0 0 216 673 386 439 401 682 1081580 1068850 1068890 233290 235396 468686 209666 214472 424139 54781 619447 958522 948737 947554 1493740 1 1 16 11 27 1452900
Advertisement
Answer
For file1 containing (I removed long useless lines):
Subject Seriesnumber 19-1-1001 2 8655 661 15250 60747 800 19-1-1001 10 8992 507 15722 64032 872 19-1-1103 2 3303 317 12146 57569 700 19-1-1103 9 3236 286 12490 59477 700 19-10-1010 2 3344 608 14744 59165 838 19-10-1010 6 3242 616 15205 61330 801
and file2 containig:
Subject seriesnumber 19-10-1010 2 19-10-1166 2 19-102-10005 2 19-102-10006 2 19-103-10009 2 19-103-10010 2 19-104-10013 11 19-104-10014 2 19-105-10017 6 19-105-10018 6
The following awk will output:
$ awk 'NR==FNR{a[$1, $2];next} ($1, $2) in a' file2 file1
19-10-1010 2 3344 608 14744 59165 838
Note that the first file argument to awk is file2 not file1 (small optimization)! How it works:
NR == FNR– if line number is file line number. Ie. choose only first file passed to awk.a[$1, $2]– remember index$1,$2in associative arrayanext– do not parse rest of script and restart with next line($1, $2) in a– check if$1, $2is in associative arraya- because of
nextthis is run only for the second file as passed toawk - if this expression returns with
true, then the line will be printed (this is how awk works).
- because of
Alternatively you could do the follow, but it will store the whole file1 in memory, which is… memory consuming…, the code above only stores $1, $2 indexes in memory.
awk 'NR==FNR{arr[$1, $2]=$0} NR!=FNR{print arr[$1, $2]}' file1 file2