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 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229 3230 3231 3232 3233 3234 3235 3236 3237 3238 3239 3240 3241 3242 3243 3244 3245 3246 3247 3248 3249 3250 3251 3252 3253 3254 3255 3256 3257 3258 3259 3260 3261 3262 3263 3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275 3276 3277 3278 3279 3280 3281 3282 3283 3284 3285 3286 3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297 3298 3299 3300 3301 3302 3303 3304 3305 3306 3307 3308 3309 3310 3311 3312 3313 3314 3315 3316 3317 3318 3319 3320 3321 3322 3323 3324 3325 3326 3327 3328 3329 3330 3331 3332 3333 3334 3335 3336 3337 3338 3339 3340 3341 3342 3343 3344 3345 3346 3347 3348 3349 3350 3351 3352 3353 3354 3355 3356 3357 3358 3359 3360 3361 3362 3363 3364 3365 3366 3367 3368 3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381 3382 3383 3384 3385 3386 3387 3388 3389 3390 3391 3392 3393 3394 3395 3396 3397 3398 3399 3400 3401 3402 3403 3404 3405 3406 3407 3408 3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420 3421 3422 3423 3424 3425 3426 3427 3428 3429 3430 3431 3432 3433 3434 3435 3436 3437 3438 3439 3440 3441 3442 3443 3444 3445 3446 3447 3448 3449 3450 3451 3452 3453 3454 3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465 3466 3467 3468 3469 3470 3471 3472 3473 3474 3475 3476 3477 3478 3479 3480 3481 3482 3483 3484 3485 3486 3487 3488 3489 3490 3491 3492 3493 3494 3495 3496 3497 3498 3499 3500 3501 3502 3503 3504 3505 3506 3507 3508 3509 3510 3511 3512 3513 3514 3515 3516 3517 3518 3519 3520 3521 3522 3523 3524 3525 3526 3527 3528 3529 3530 3531 3532 3533 3534 3535 3536 3537 3538 3539 3540 3541 3542 3543 3544 3545 3546 3547 3548 3549 3550 3551 3552 3553 3554 3555 3556 3557 3558 3559 3560 3561 3562 3563 3564 3565 3566 3567 3568 3569 3570 3571 3572 3573 3574 3575 3576 3577 3578 3579 3580 3581 3582 3583 3584 3585 3586 3587 3588 3589 3590 3591 3592 3593 3594 3595 3596 3597 3598 3599 3600 3601 3602 3603 3604 3605 3606 3607 3608 3609 3610 3611 3612 3613 3614 3615 3616 3617 3618 3619 3620 3621 3622 3623 3624 3625 3626
|
% CLHEP/ZOOM Vector Package Formulas and Definitions
\documentclass[twoside,12pt]{article}
\flushbottom
\pagestyle{headings}
\setlength{\topmargin}{0.0in}
\setlength{\textwidth}{5.5in}
\setlength{\oddsidemargin}{.5in}
\setlength{\evensidemargin}{.5in}
\setlength{\textheight}{8.5in}
\addtolength{\parskip}{2pt}
\newcommand{\fpcl}{{\sc fpcl}}
\def \Point {{\tt Point3D}}
\def \Line {{\tt Line3D}}
\def \Direction {{\tt Direction}}
\def \UnitVector {{\tt UnitVector}}
\def \Plane {{\tt Plane3D}}
\def \SpaceVector {{\tt SpaceVector}}
\def \SV {{\tt Hep3Vector}}
\def \SVz {{\tt SpaceVector}}
\def \UV {{\tt UnitVector}}
\def \TangentVector {{\tt TangentVector}}
\def \TV {{\tt TangentVector}}
\def \Ro {{\tt HepRotation}}
\def \Ros {{\tt Rotation}}
\def \Rotation {{\tt Rotation}}
\def \RotationZ {{\tt HepRotationZ}}
\def \Transformation {{\tt Transformation}}
\def \Euclidean {{\tt EuclideanTransformation}}
\def \Angle {{\tt Angle}}
\def \LorentzVector {{\tt LorentzVector}}
\def \LorentzTransformation {{\tt LorentzTransformation}}
\def \LV {{\tt HepLorentzVector}}
\def \LVz {{\tt LorentzVector}}
\def \LT {{\tt HepLorentzRotation}}
\def \LTs {{\tt LorentzTransformation}}
\def \LB {{\tt HepBoost}}
\def \LBs {{\tt LorentzBoost}}
\def \PolarAngle {{\tt PolarAngle}}
\def \PAngle {{\tt PAngle}}
\def \AzimuthalAngle {{\tt AzimuthalAngle}}
\def \AAngle {{\tt AAngle}}
\def \EB {{\tt EBvector}}
\def \AV {{\tt Adjoint3Vector}}
\def \ALV {{\tt Adjoint4Vector}}
\def \Scalar {{\tt Scalar}}
\def \Ax {{\tt HepAxisAngle}}
\def \Es {{\tt HepEulerAngles}}
\newcommand {\see}[1] {\hfill$\triangleright$ see eqn.~#1}
\newenvironment{shortlist}{%
\begin{itemize}
\setlength{\itemsep}{0pt}
\setlength{\parskip}{0pt}
}{%
\end{itemize}
}
\setcounter{secnumdepth}{2} % Number sub- but not subsubsections
\setcounter{tocdepth}{2} % Only numbered section headings go in contents
\begin{document}
\title{CLHEP {\bf Vector} Package \\
and \\
ZOOM {\bf PhysicsVector} Package \\
Space and Lorentz Vector and Transformation Package \\
Formulas and Definitions}
\author{
Fermilab ``ZOOM'' Physics Class Library Task Force}
\date{Version 2.3, September 2, 2003}
\maketitle
The CLHEP Vector package implements 3-vectors, 4-vectors, rotations,
Lorentz transfomations and related concepts.
This includes all functionallity in the original CLHEP package, and in
the ZOOM PhysicsVectors package.
The latter is now implemented as wrapper headers, so that classes from
the ZOOM package can be used wherever the corresponding CLHEP class are
expected.
This document briefly lists the methods available, then presents relevant
mathematical definitions.
\tableofcontents
\section{Available Classes and Methods}
In this document we somewhat abbreviate the signatures of methods,
since coders will look at the header files anyway.
Any arguments specified without an explicit data type
are of the scalar type {\tt HepDouble}, which is just {\tt double} on
virtually every system.
Unless otherwise indicated,
parameter passage for scalars is by value,
and parameter passage for other types is by constant reference.
This document also does not discuss namespace issues.
\subsection{\protect\SV\ Class --- Vector of real quantities in 3-space}
Throughout this section,
arguments named v, v1, v2, etc., are of type \SV .
\subsubsection{Constructors and Accessors}
\begin{shortlist}
\item Hep3Vector(~)
\item Hep3Vector( x, y, z )
\end{shortlist}
\noindent
The following accessor methods are used
to obtain the named coordinate components,
as in \verb|double s = v.r();|.
\begin{shortlist}
\item x(~) \/\/\/ y(~) \/\/\/ z(~)
\item r(~) \see{\ref{eq:polar}}
\item theta(~) \see{\ref{eq:polar}}
\item eta(~) \see{\ref{eq:spherical}}
\item phi(~) \see{\ref{eq:polar}, \ref{eq:cylindrical}}
\item rho(~) \see{\ref{eq:cylindrical}}
\item getX(~) \/\/\/ getY(~) \/\/\/ getZ(~)
\item getR(~) \/\/\/ getTheta(~) \/\/\/ getPhi(~)
\item getEta(~) \/\/\/ getRho(~)
\end{shortlist}
\noindent
The Cartesian components may also be accessed by the index syntax,
using either square braces or parentheses.
In either case, the meaning of the index is 0-based,
and an enum is provided to help clarify this: {\tt Hep3Vector::X=0},
{\tt Hep3Vector::Y=1}, {\tt Hep3Vector::Z=2}.
\noindent
There is a family of methods of the form
{\tt set}{\it Component}{\tt ()} which may be used to set
one component in Cartesian or Polar coordinates, keeping the other
two constant.
Also available is {\tt setCylTheta()}, which modifies $\theta$ (the angle
against the Z axis) while keeing $\phi$ and the radial distance from the
Z axis $\rho$ constant.
\noindent
Finally, there is also a family of {\tt set()} methods
that may be used to update all of a \SV 's three coordinates at once.
These {\tt set()} methods' signatures are in one-to-one correspondence
with the non-default constructors listed above.
\subsubsection{Operators}
For \SV s v, v1, and v2, and for a scalar c,
the following arithmetic operations are provided,
in each case resulting in a \SV .
\begin{shortlist}
\item v1 + v2 and v1 $-$ v2
\item v $*$ c and c $*$ v
\item v / c
\item $-$ v
\end{shortlist}
\noindent
In addition,
the following arithmetic modify-assignment operations are provided.
\begin{shortlist}
\item v1 += v2 and v1 $-$= v2
\item v $*$= c and v /= c
\end{shortlist}
\noindent
The usual six relational operations
(==, !=, $<$, $<=$, $>$, $>=$) are provided (see eqn. \ref{eq:compSV}).
Further, the following member functions are useful to check for equality
within a relative tolerance.
\begin{shortlist}
\item bool isNear( v, epsilon ) \see{\ref{eq:isNear}}
\item Scalar howNear( v ) \see{\ref{eq:howNear}, \ref{eq:howNear:2}, \ref{eq:howNear:3}}
\item Scalar deltaR( v ) \see{\ref{eq:deltaR}}
\end{shortlist}
\noindent
The tolerance epsilon may be omitted.
The following class-wide (static) functions are used
to obtain and set the default tolerance for nearness.
\begin{shortlist}
\item HepDouble setTolerance( tol ) \see{\ref{eq:epsildef}}
\item HepDouble getTolerance(~)
\end{shortlist}
\subsubsection{Methods}
\begin{shortlist}
\item ostream \& operator$<<$( ostream \& os, v )
\item istream \& operator$>>$( istream \& is, Hep3Vector \& v )
\end{shortlist}
\begin{shortlist}
\item HepDouble dot( v ) \see{\ref{eq:dot}}
\item Hep3Vector cross( v ) \see{\ref{eq:cross}}
\item HepDouble diff2( v ) \see{\ref{eq:diff2}}
\item bool isParallel( v, epsilon ) \see{\ref{eq:isPar}}
\item bool isOrthogonal( v, epsilon ) \see{\ref{eq:isOrtho}}
\item HepDouble howParallel( v ) \see{\ref{eq:howPar}, \ref{eq:howPar:2}}
\item HepDouble howOrthogonal( v ) \see{\ref{eq:howOrtho}, \ref{eq:howOrtho:2}}
\end{shortlist}
\begin{shortlist}
\item HepDouble mag2(~) \see{\ref{eq:mag2}}
\item HepDouble mag(~) \see{\ref{eq:mag}, \ref{eq:mag:2}}
\item HepDouble beta(~) \see{\ref{eq:beta}}
\item HepDouble gamma(~) \see{\ref{eq:gamma}}
\item HepDouble pseudoRapidity(~) \see{\ref{eq:spherical}}
\item HepDouble coLinearRapidity(~) \see{\ref{eq:coLinRap}}
\item Hep3Vector unit(~) \see{\ref{eq:svunit}}
\item Hep3Vector orthogonal(~) \see{\ref{eq:orthogonal}}
\end{shortlist}
\noindent
The following methods depend on a reference direction
(specified by a \SV\ u). Signatures omitting
the reference direction are also supplied---$\hat{z}$ is implied in these
cases and the methods take advantage of the simpler form.
\begin{shortlist}
\item HepDouble perp( u ) \see{\ref{eq:perp}}
\item HepDouble perp2( u ) \see{\ref{eq:perp2}}
\item SpaceVector perpPart( u ) \see{\ref{eq:perpPart}}
\item SpaceVector project( u ) \see{\ref{eq:project}}
\item HepDouble angle( u ) \see{\ref{eq:angle}}
\item HepDouble theta( u ) \see{\ref{eq:theta}}
\item HepDouble cosTheta( u ) \see{\ref{eq:cosTheta}}
\item HepDouble cos2Theta( u ) \see{\ref{eq:cos2Theta}}
\item HepDouble eta( u ) \see{\ref{eq:eta}, \ref{eq:eta:2}, \ref{eq:eta:3}, \ref{eq:eta:4}}
\item HepDouble polarAngle( v2, u ) \see{\ref{eq:polarA}, \ref{eq:polarA:2}}
\item HepDouble deltaPhi( v2 ) \see{\ref{eq:azim}}
\item HepDouble azimAngle( v2, u ) \see{\ref{eq:azim}, \ref{eq:azim:2}}
\item HepDouble rapidity( u ) \see{\ref{eq:rap}, \ref{eq:rap:2}}
\end{shortlist}
\subsubsection{Rotations}
\noindent
These methods change the vector:
\begin{shortlist}
\item Hep3Vector \& rotateX( delta ) \see{\ref{eq:rotX}}
\item Hep3Vector \& rotateY( delta ) \see{\ref{eq:rotY}}
\item Hep3Vector \& rotateZ( delta ) \see{\ref{eq:rotZ}, \ref{eq:rotZ:2}}
\item Hep3Vector \& rotateUZ( Hep3Vector ) \see{\ref{eq:rotUz}}
\end{shortlist}
\begin{shortlist}
\item Hep3Vector \& rotate( axis, delta ) \see{\ref{eq:axisrot}}
\item Hep3Vector \& rotate( const AxisAngle \& ax ) \see{\ref{eq:axisrot}}
\item Hep3Vector \& rotate( phi, theta, psi ) \see{\ref{eq:eulerrot}}
\item Hep3Vector \& rotate( const EulerAngles \& e ) \see{\ref{eq:eulerrot}}
\end{shortlist}
\noindent
The following methods both do $\vec{v} \Longleftarrow R v$.
(Notice the order of multiplication for {\tt v *= R}.)
\begin{shortlist}
\item Hep3Vector transform (const HepRotation \& R) \see{\ref{eq:opstareq}}
\item Hep3Vector operator *= (const HepRotation \& R) \see{\ref{eq:opstareq}}
\end{shortlist}
\noindent
These global functions do not change the \SV\ v:
\begin{shortlist}
\item Hep3Vector rotationXOf( v, delta ) \see{\ref{eq:rotX}}
\item Hep3Vector rotationYOf( v, delta ) \see{\ref{eq:rotY}}
\item Hep3Vector rotationZOf( v, delta ) \see{\ref{eq:rotZ}, \ref{eq:rotZ:2}}
\end{shortlist}
\begin{shortlist}
\item Hep3Vector rotationOf( v, axis, delta ) \see{\ref{eq:axisrot}}
\item Hep3Vector rotationOf( v, const AxisAngle \& ax ) \see{\ref{eq:axisrot}}
\item Hep3Vector rotationOf( v, phi, theta, psi ) \see{\ref{eq:eulerrot}}
\item Hep3Vector rotationOf( v, const EulerAngles \& e ) \see{\ref{eq:eulerrot}}
\end{shortlist}
\subsection{\protect\SVz\ Class --- Derived from \SV }
The \SVz\ class provides backward compatibility with the original ZOOM
PhysicsVectors package.
It is publicly derived from \SV .
It would simply be a typedef off \SV , in the appropriate namespace,
but for one set of features which were felt to be overkill in the
CLHEP context.
This is the ability to construct a \SVz\ providing spherical or cylindrical
coordinates.
Associated with these constructors are a set of defined keywords which
allow disambiguation of the various forms of constructors.
These keywords are {\tt RADIANS}, {\tt DEGREES}, and {\tt ETA},
Throughout this section, we will illustrate using {\tt RADIANS},
but this may be replaced by {\tt DEGREES}
to indicate the corresponding angle's units.
For theta, this may also be replaced by {\tt ETA}, to indicate that the
pseudorapidity is being supplied.
\begin{shortlist}
\item SpaceVector(~)
\item SpaceVector( x, y, z )
\item SpaceVector( r, theta, RADIANS, phi, RADIANS ) \see{\ref{eq:polar}}
\item SpaceVector( rho, phi, RADIANS, z ) \see{\ref{eq:cylindrical}}
\end{shortlist}
\subsection{\protect\UV\ Class}
It was found useful in the ZOOM package to express the concept of a
\UV, that is, a vector known to be inherently of unit length.
CLHEP neither has such a class, and it is felt (at this time) that
it should not.
In the merged package,
\UV\ is provided as a header file which appears only in the PhysicsVectors
area.
Since \UV\ depends on no non-header implementation code, issues of where
to place the library containing non-CLHEP code do not arise.
Although the \UV\ class is not derived from the \SVz\ class,
all const methods (except for three relativistic kinematic methods
which only make sense for a vector of length less than one)
are provided for \UV .
Thos non-const methods of \SV\ which do not risk violating the unit-length
property, such as rotation, are also provided.
In this section, therefore,
only the differences in the classes are described.
\subsubsection{Constructors and Accessors}
\UV\ constructors (and \verb|set()| methods)
have the same signatures as do the corresponding \SVz\ methods,
but normalize before returning.
The default \UV\ constructor yields $\hat{z}$.
In addition, we have the conversion constructor
\begin{shortlist}
\item UnitVector( Hep3Vector v )
\end{shortlist}
\noindent
Unlike the case for \SV , \UV\ disallows setting a
single Cartesian coordinate. Also, of course,
modifying the radius of a \UV\ is forbidden.
\subsubsection{Operators}
\UV s are treated as \SV s for purposes of arithmetic and comparisons,
except that unary minus returns a \UV .
Modify-assignment (e.g., +=) is forbidden on \UV s.
\subsubsection{Methods}
\label{unitMethods}
Most \UV\ methods match those of the \SV\ class; some are modified,
however, to take advantage of the r~=~1 property.
Forbidden methods include
\verb|beta()|, \verb|gamma()|, and \verb|rapidity()|.
\subsubsection{Rotations}
All methods and functions in this category apply equally to
\UV s as they do to \SV s.
\subsection{\protect\LV\ Class --- Vector of real quantities in 4-space}
Throughout this section,
arguments named p, p1, p2, etc., are of type \LV ,
arguments named v, v1, v2, etc., are of type \SV ,
and an argument named t will refer to a scalar meant as a time component.
\subsubsection{Constructors and Accessors}
\begin{shortlist}
\item HepLorentzVector(~)
\item HepLorentzVector( x, y, z, t )
\item HepLorentzVector( v, t )
\item HepLorentzVector( t, v )
\item HepLorentzVector( t )
\item HepLorentzVector( x, y, z )
\item HepLorentzVector( v )
\end{shortlist}
\noindent
The following accessor methods are used
to obtain the named coordinate components,
as in \verb|double s = p.t();|.
\begin{shortlist}
\item x(~) \/\/\/ y(~) \/\/\/ z(~) \/\/\/ t(~)
\item px(~) \/\/\/ py(~) \/\/\/ pz(~) \/\/\/ e(~)
\item getX(~) \/\/\/ getY(~) \/\/\/ getZ(~) \/\/\/ getT(~)
\item v(~) or getV(~) or vect(~)
\end{shortlist}
\noindent
The Cartesian components may also be accessed by the index syntax,
using either square braces or parentheses.
In either case, the meaning of the index is 0-based, {\em with the
time component last}.
An enum is nested in the {\tt HepLorentzVector} class to help clarify this:
{\tt X=0}, {\tt Y=1}, {\tt Z=2}, {\tt T=3}.
The spatial components can also be accessed in spherical coordinates:
\begin{shortlist}
\item HepDouble theta() const;
\item HepDouble cosTheta() const;
\item HepDouble phi() const;
\item HepDouble rho() const;
\end{shortlist}
\noindent
There is a family of methods of the form
{\tt set}{\it Component}{\tt ()} which may be used to set
one component in Cartesian or spherical coordinates, or the $\rho$
cylindrical coordinate,
keeping the other components in that system constant.
\begin{shortlist}
\item setX(HepDouble); \/\/\/ setPx(HepDouble);
\item setY(HepDouble); \/\/\/ setPy(HepDouble);
\item setZ(HepDouble); \/\/\/ setPz(HepDouble);
\item setT(HepDouble); \/\/\/ setE(HepDouble);
\item setTheta(HepDouble); \see{\ref{eq:polar}}
\item setPhi(HepDouble); \see{\ref{eq:polar}}
\item setRho(HepDouble); \/\/\/ setPerp(HepDouble);
\see{\ref{eq:cylindrical}}
\end{shortlist}
The entire spatial component can be set at once, keeping the time component
constant.
\begin{shortlist}
\item setVect(HepDouble); \/\/\/ setV(HepDouble);
\end{shortlist}
\noindent
And there is a family of {\tt set()} methods, corresponding to the
constructors,
that may be used to update all four of a \LV 's coordinates at once.
\begin{shortlist}
\item set (x, y, z, t);
\item set (v, t);
\item set (t, x);
\end{shortlist}
\noindent
Assignment from a \SV\ is supported, as in \verb|p = v;|.
\noindent
Finally, there are conversion operators to const and non-const Hep3Vector;
these were present in the original CLHEP classes.
They ignore the time component.
\subsubsection{Operators}
For \LV s p, p1, and p2, and for a scalar c,
the following arithmetic operations are provided,
in each case resulting in a \LV .
\begin{shortlist}
\item p1 + p2 and p1 $-$ p2
\item p $*$ c and c $*$ p
\item p / c
\item $-$ p
\end{shortlist}
\noindent
In addition, the following arithmetic modify-assignment operations are provided.
\begin{shortlist}
\item p1 += p2 and p1 $-$= p2
\item p $*$= c and p /= c
\end{shortlist}
The usual six relational operations
(==, !=, $<$, $<=$, $>$, $>=$) are provided (see eqn. \ref{eq:wcomp}).
Further, the following member functions are useful
to check for equality, etc., within a relative tolerance.
\begin{shortlist}
\item bool isNear( p, epsilon ) \see{\ref{eq:wisNear}}
\item HepDouble howNear( p ) \see{\ref{eq:whowNear}, \ref{eq:whowNear:2}}
\item bool isNearCM( p, epsilon ) \see{\ref{eq:wisNearCM}, \ref{eq:wisNearCM:2}}
\item HepDouble howNearCM( p ) \see{\ref{eq:whowNearCM}, \ref{eq:whowNearCM:2}}
\item bool isParallel( p, epsilon ) \see{\ref{eq:wisPar}, \ref{eq:wisPar:2}}
\item HepDouble howParallel( p) \see{\ref{eq:whowPar}, \ref{eq:whowPar:2}}
\item HepDouble deltaR( p ) \see{\ref{eq:deltaR}}
\end{shortlist}
\noindent
The tolerance epsilon may be omitted.
The following class-wide (static) functions are used
to obtain and set the default tolerance for nearness.
\begin{shortlist}
\item HepDouble setTolerance( HepDouble tol )
\item HepDouble getTolerance(~)
\end{shortlist}
\subsubsection{Methods}
\noindent
Establishing a metric convention (static):
\begin{shortlist}
\item ZMpvMetric\_t setMetric( ZMpvMetric\_t m )
\item ZMpvMetric\_t getMetric(~)
\end{shortlist}
\noindent
Metric-independent properties and methods of this 4-vector
(the results of these methods do not change sign if you go from the
TimePositive to the TimeNegative metric):
\begin{shortlist}
\item ostream \& operator$<<$( ostream \& os, p )
\item istream \& operator$>>$( istream \& is, LorentzVector \& p )
\item bool isSpacelike(~) \see{\ref{eq:wisSl}}
\item bool isTimelike(~) \see{\ref{eq:wisTl}}
\item bool isLightlike( epsilon ) \see{\ref{eq:wisLl}}
\item HepDouble howLightlike(~) \see{\ref{eq:whowLl}, \ref{eq:whowLl:2}}
\item HepDouble plus(~) \see{\ref{eq:wplus}}
\item HepDouble minus(~) \see{\ref{eq:wminus}}
\item HepDouble euclideanNorm2(~) \see{\ref{eq:wENorm2}}
\item HepDouble euclideanNorm(~) \see{\ref{eq:wENorm}}
\item HepDouble restMass2(~) \see{\ref{eq:wrestM2}}
\item HepDouble restMass(~) \see{\ref{eq:wrestM}}
\item HepDouble m2(~) \see{\ref{eq:winvMass2}}
\item HepDouble invariantMass2(~) \see{\ref{eq:winvMass2}}
\item HepDouble m(~) \/\/\/ mag(~) \see{\ref{eq:wmag}}
\item HepDouble invariantMass(~) \see{\ref{eq:winvMass}}
\item HepDouble mt2(~) \see{\ref{eq:wmt2}}
\item HepDouble mt(~) \see{\ref{eq:wmt}}
\item HepDouble et2(~) \see{\ref{eq:wet2}}
\item HepDouble et(~) \see{\ref{eq:wet}}
\item LorentzVector rest4Vector(~) \see{\ref{eq:wrest4V}}
\item SpaceVector boostVector(~) \see{\ref{boostvector}}
\item HepDouble beta(~) \see{\ref{eq:wbeta}}
\item HepDouble gamma(~) \see{\ref{eq:wgamma}}
\item HepDouble eta(~) \see{\ref{eq:weta}, \ref{eq:weta:2}, \ref{eq:weta:3}}
\item HepDouble rapidity(~) \see{\ref{eq:wrapid}, \ref{eq:wrapid:2}}
\item HepDouble coLinearRapidity(~) \see{\ref{eq:wcoLinRap}, \ref{eq:wcoLinRap:2}}
\item SpaceVector findBoostToCM(~) \see{\ref{eq:wfindBoost}}
\end{shortlist}
\noindent
Metric-dependent properties of this 4-vector:
(the results of these methods change sign if you go from the
TimePositive to the TimeNegative metric):
\begin{shortlist}
\item HepDouble mag2(~) \see{\ref{eq:wmag2}}
\end{shortlist}
\noindent
Metric-independent methods combining two 4-vectors:
\begin{shortlist}
\item HepDouble delta2Euclidean( p ) \see{\ref{eq:wdelta2E}}
\item HepDouble plus( p ) \see{\ref{eq:wplus:2}}
\item HepDouble minus( p ) \see{\ref{eq:wminus:2}}
\item HepDouble eta( p ) \see{\ref{eq:weta}, \ref{eq:weta:2}, \ref{eq:weta:3}, \ref{eq:weta:4}}
\item HepDouble rapidity( p ) \see{\ref{eq:wrapid}, \ref{eq:wrapid:2}, \ref{eq:wrapid:3}, \ref{eq:wrapid:4}}
\item HepDouble invariantMass2( p ) \see{\ref{eq:winvMass2}}
\item HepDouble invariantMass( p ) \see{\ref{eq:winvMass}}
\item SpaceVector findBoostToCM( p ) \see{\ref{eq:wfindBoost}}
\end{shortlist}
\noindent
Metric-dependent methods combining two 4-vectors:
\begin{shortlist}
\item HepDouble dot( p ) \see{\ref{eq:wdot}, \ref{eq:wdot:2}}
\item HepDouble operator*( p ) \see{\ref{eq:wdot}, \ref{eq:wdot:2}}
\item HepDouble diff2( p ) \see{\ref{eq:wdiff2}}
\end{shortlist}
\noindent
Methods involving properties of the spatial part of the 4-vector
(these could be invoked as p.v().whatever() but the p.whatever()
syntax is shorter).
Most of these take a reference direction v; if v is omitted,
$\hat{z}$ is implied and the methods take advantage of the simpler form.
\begin{shortlist}
\item HepDouble perp( v ) \see{\ref{eq:perp}}
\item HepDouble perp2( v ) \see{\ref{eq:perp2}}
\item HepDouble angle(~) \see{\ref{eq:angle}}
\item setVectM(HepDouble); \/\/\/ setVectMag(HepDouble);
\see{\ref{eq:polar}}
\item setRho(HepDouble); \/\/\/ setPerp(HepDouble);
\see{\ref{eq:cylindrical}}
\item HepDouble pseudoRapidity(~) \see{\ref{eq:spherical}}
\end{shortlist}
\subsubsection{Rotations and Boosts}
These methods change the 4-vector.
\begin{shortlist}
\item LorentzVector \& rotateX( delta ) \see{\ref{eq:rotX}}
\item LorentzVector \& rotateY( delta ) \see{\ref{eq:rotY}}
\item LorentzVector \& rotateZ( delta ) \see{\ref{eq:rotZ}}
\item LorentzVector \& rotate( v, delta ) \see{\ref{eq:axisrot}}
\item LorentzVector \& rotate( delta, v ) \see{\ref{eq:axisrot}}
\item LorentzVector \& rotate( phi, theta, psi ) \see{\ref{eq:eulerrot}}
\item LorentzVector \& rotate( EulerAngles \& e ) \see{\ref{eq:eulerrot}}
\item LorentzVector \& rotateUz( v ) \see{\ref{eq:rotUz}}
\item LorentzVector \& boostX( beta ) \see{\ref{eq:wbX}}
\item LorentzVector \& boostY( beta ) \see{\ref{eq:wbY}}
\item LorentzVector \& boostZ( beta ) \see{\ref{eq:wbZ}}
\item LorentzVector \& boost( v ) \see{\ref{eq:wboostvec}}
\item LorentzVector \& boost( v, beta ) \see{\ref{eq:pureboost}}
\end{shortlist}
\noindent
The following methods all do $\vec{p} \Longleftarrow R p$ where $R$ is
either a \Ro or a \LT .
(Notice the order of multiplication for {\tt p *= R}.)
\begin{shortlist}
\item HepLorentzVector transform (const HepRotation \& R)
\see{\ref{eq:wopstareq}}
\item HepLorentzVector operator *= (const HepRotation \& R)
\see{\ref{eq:wopstareq}}
\item HepLorentzVector transform (const HepLorentzRotation \& R)
\see{\ref{eq:wopstareq}}
\item HepLorentzVector operator *= (const HepLorentzRotation \& R)
\see{\ref{eq:wopstareq}}
\end{shortlist}
\noindent
These functions return a new 4-vector.
\begin{shortlist}
\item LorentzVector rotationXOf( p, delta ) \see{\ref{eq:rotX}}
\item LorentzVector rotationYOf( p, delta ) \see{\ref{eq:rotY}}
\item LorentzVector rotationZOf( p, delta ) \see{\ref{eq:rotZ}}
\item LorentzVector rotationOf( p, v, delta ) \see{\ref{eq:axisrot}}
\item LorentzVector rotationOf( p, phi, theta, psi ) \see{\ref{eq:eulerrot}}
\item LorentzVector rotationOf( p, const EulerAngles \& e ) \see{\ref{eq:eulerrot}}
\item LorentzVector boostXOf( p, beta ) \see{\ref{eq:wbX}}
\item LorentzVector boostYOf( p, beta ) \see{\ref{eq:wbY}}
\item LorentzVector boostZOf( p, beta ) \see{\ref{eq:wbZ}}
\item LorentzVector boostOf( p, betaVector ) \see{\ref{eq:wboostvec}}
\item LorentzVector boostOf( p, v, beta ) \see{\ref{eq:pureboost}}
\end{shortlist}
\subsection{\protect\LVz\ Class --- Typedefed from \LV }
The \LVz\ class provides backward compatibility with the original ZOOM
PhysicsVectors package.
It is simply a typedef off \LV , in the appropriate namespace.
This is because there were no features or constructors in the ZOOM
product which were felt to be
overkill in the CLHEP context.
\subsection{{\tt Hep2Vector} Class}
The {\tt Hep2Vector} class is a simple plane vector.
Throughout this section, arguments named s, s1, s2, etc., are of type
{\tt Hep2Vector}.
\subsubsection{Constructors and Accessors}
\begin{shortlist}
\item Hep2Vector(~)
\item Hep2Vector( x, y )
\item Hep2Vector( Hep3Vector v )
\end{shortlist}
\noindent
That last constructor will supress the Z component of $\vec{v}$.
Cartesian and polar coordinates may be accessed; these are identical to those
of a \SV, with the Z component fixed at zero.
\begin{shortlist}
\item x(~) \/\/\/ y(~)
\item r(~) \see{\ref{eq:polar}}
\item phi(~) \see{\ref{eq:polar}}
\end{shortlist}
\noindent
The Cartesian components may also be accessed by the index syntax,
using either square braces or parentheses.
In either case, the meaning of the index is 0-based,
and an enum is provided to help clarify this: {\tt Hep2Vector::X=0},
{\tt Hep2Vector::Y=1}.
\noindent
There is a family of methods of the form
{\tt set}{\it Component}{\tt ()} which may be used to set
one component in Cartesian or Polar coordinates, keeping the other
component constant.
\begin{shortlist}
\item setX(x) \/\/\/ setY(y)
\item setR(r) \/\/\/ setMag(r) \see{\ref{eq:polar}}
\item setPhi(phi) \see{\ref{eq:polar}}
\end{shortlist}
\noindent
Finally, there is also a family of {\tt set()} methods
that may be used to update all of a \SV 's three coordinates at once.
These {\tt set()} methods' signatures are in one-to-one correspondence
with the non-default constructors listed above.
\subsubsection{Operators}
For {\tt Hep2Vector}s s, s1, and s2, and for a scalar c,
the following arithmetic operations are provided,
in each case resulting in a {\tt Hep2Vector}.
\begin{shortlist}
\item s1 + s2 and s1 $-$ s2
\item s $*$ c and c $*$ s
\item s / c
\item $-$ s
\end{shortlist}
\noindent
In addition,
the following arithmetic modify-assignment operations are provided.
\begin{shortlist}
\item s1 += s2 and v1 $-$= s2
\item s $*$= c and v /= c
\end{shortlist}
\noindent
The usual six relational operations
(==, !=, $<$, $<=$, $>$, $>=$) are provided (see eqn. \ref{eq:compSV}).
Further, the following member functions are useful to check for equality
within a relative tolerance. (Again, these match the corresponding methods
for \SV, with Z-component pinned at zero.)
\begin{shortlist}
\item bool isNear( v, epsilon ) \see{\ref{eq:isNear}}
\item Scalar howNear( v ) \see{\ref{eq:howNear}, \ref{eq:howNear:2}, \ref{eq:howNear:3}}
\item Scalar deltaR( v ) \see{\ref{eq:deltaR}}
\end{shortlist}
\noindent
The tolerance epsilon may be omitted.
The following class-wide (static) functions are used
to obtain and set the default tolerance for nearness.
\begin{shortlist}
\item HepDouble setTolerance( tol ) \see{\ref{eq:epsildef}}
\item HepDouble getTolerance(~)
\end{shortlist}
\subsubsection{Methods}
The set of methods is somewhat simpler that for \SV, but in all applicable
cases the definitions match those for \SV:
\begin{shortlist}
\item ostream \& operator$<<$( ostream \& os, s )
\end{shortlist}
\begin{shortlist}
\item HepDouble dot( v ) \see{\ref{eq:dot}}
\item bool isParallel( v, epsilon ) \see{\ref{eq:isPar}}
\item bool isOrthogonal( v, epsilon ) \see{\ref{eq:isOrtho}}
\item HepDouble howParallel( v ) \see{\ref{eq:howPar}, \ref{eq:howPar:2}}
\item HepDouble howOrthogonal( v ) \see{\ref{eq:howOrtho}, \ref{eq:howOrtho:2}}
\item HepDouble angle( s2 ) \see{\ref{eq:angle}}
\end{shortlist}
\begin{shortlist}
\item HepDouble mag2(~) \see{\ref{eq:mag2}}
\item HepDouble mag(~) \see{\ref{eq:mag}, \ref{eq:mag:2}}
\item Hep2Vector unit(~) \see{\ref{eq:svunit}}
\end{shortlist}
\noindent
There is one method which is applicable for {\tt Hep2Vector} but which
does not match the definition for \SV:
\begin{shortlist}
\item Hep2Vector orthogonal(~)
\end{shortlist}
\noindent
\verb$s.orthogonal()$ is $s$ rotated clockwise by $90^\circ$ if
$|s_x| < |s_y|$, and counterclockwise by $90^\circ$ if
$|s_x| \geq |s_y|$, and counterclockwise.
\vspace{.16 in}
In the ZOOM area, the class {\tt PlaneVector} is typedefed to
{\tt Hep2Vector}.
\subsection{\protect\Ro\ Classes}
Throughout this section, arguments named r, r1, r2, etc., are of type \Ro.
Also, rowX, rowY, rowZ, colX, colY, colZ will represent \SV\ arguments.
\subsubsection{Constructors and Accessors}
\begin{shortlist}
% // construct an identity rotation by default:
\item Rotation(~)
% // copy/assign any rotation:
\item Rotation( r )
\item Rotation \& operator=( r )
\item Rotation \& set( r )
% // supply three Euler angles(in radians):
\item Rotation( phi, theta, psi ) \see{\ref{eq:eulerrot}}
\item Rotation \& set( phi, theta, psi ) \see{\ref{eq:eulerrot}}
% // supply EulerAngles structure:
\item Rotation( const EulerAngles \& e ) \see{\ref{eq:eulerrot}}
\item Rotation \& set( const EulerAngles \& e ) \see{\ref{eq:eulerrot}}
% // supply axis and angle:
\item Rotation( const Hep3Vector \& axis, delta ) \see{\ref{eq:axisrot}}
\item Rotation \& set( const Hep3Vector \& axis, delta ) \see{\ref{eq:axisrot}}
% // supply AxisAngle structure:
\item Rotation( const AxisAngle \& ax ) \see{\ref{eq:axisrot}}
\item Rotation \& set( const AxisAngle \& ax ) \see{\ref{eq:axisrot}}
% // supply three *orthogonal* UnitVectors for the columns:
\item Rotation( colX, colY, colZ )
\item Rotation \& set( colX, colY, colZ )
\item Rotation \& setRows( rowX, rowY, rowZ )
% // NOTE:
% // This constructor and set method will check that the
% // cols form an orthonormal matrix, and adjust so that
% // relation is as exact as possible.
% // supply ZMpvRep3x3 structure:
\item Rotation( const ZMpvRep3x3 \& rep )
\item Rotation \& set( const ZMpvRep3x3 \& rep )
% // WARNING:
% // This constructor and set method will assume the
% // ZMpvRep3x3 supplied is infact an orthogonal matrix.
% // No checking or correction is done. If you are
% // not certain the matrix is orthogonal, break it
% // into three SpaceVector cols and use the form
% // Rotation(UnitVector, UnitVector, UnitVector)
\end{shortlist}
\begin{shortlist}
\item phi(~) \/\/\/ theta(~) \/\/\/ psi(~)
\see{\ref{eq:eulerrot},\ref{eq:phipsiconv},\ref{eq:thetaconv}}
\item eulerAngles(~)
\see{\ref{eq:eulerrot},\ref{eq:phipsiconv},\ref{eq:thetaconv}}
\item axis(~) \/\/\/ delta(~)
\item axisAngle(~) \see{\ref{eq:axisrot},\ref{eq:deltaconv}}
%// for orthogonal unit vectors
\item colX(~) \/\/\/ colY(~) \/\/\/ colZ(~)
\item rowX(~) \/\/\/ rowY(~) \/\/\/ rowZ(~)
% // for individual elements:
\item xx(~) \/\/\/ xy(~) \/\/\/ xz(~)
\item yx(~) \/\/\/ yy(~) \/\/\/ yz(~)
\item zx(~) \/\/\/ zy(~) \/\/\/ zz(~)
% // for all 9 elements:
\item HepRep3x3 rep3x3(~)
% // for a 16-element xyzt matrix:
\item HepRep4x4 rep4x4(~)
\end{shortlist}
And the matrix elements of a \Ro\ may be accessed using two integer indices
in parentheses or square brackets, with indices running from 0 to 2.
The following methods alter one component of a \Ro, in some way of viewing
that rotation:
\begin{shortlist}
\item setPhi(HepDouble) \/\/\/ setTheta(HepDouble) \/\/\/ setPsi(HepDouble)
\see{\ref{eq:eulerrotR}}
\item setAxis(Hep3Vector) \/\/\/ setDelta(Hep3Vector)
\see{\ref{eq:axisrotR}}
\end{shortlist}
\subsubsection{Use of \protect\Ro\ as a 4-Rotation}
The \Ro\ may be considered to be a 4-rotation. Thus, other accessors
applicable to \LT\ (see section (\ref{ltmethods}))
may be applied to \Ro\ as well.
\subsubsection{Operators and Methods}
% // arithmetic:
\begin{shortlist}
\item Rotation \& operator\verb$*=$( r )
\item friend Rotation operator$*$( r1, r2 )
\end{shortlist}
% // relative comparison:
\begin{shortlist}
\item static HepDouble getTolerance(~) \see{\ref{eq:epsildefR}}
\item static HepDouble setTolerance( tol )
\item bool isNear( r, epsilon ) \see{\ref{eq:nearrot}}
\item HepDouble howNear( r ) \see{\ref{eq:nearrot}}
\item HepDouble distance2( r ) \see{\ref{eq:dist2rot}}
\end{shortlist}
% // ordering comparison:
\begin{shortlist}
\item r1 == r2 \/\/\/ r1 != r2
\item r1 $>$ r2 \/\/\/ r1 $>=$ r2 \/\/\/ r1 $<$ r2 \/\/\/ r1 $<=$ r2
\see{\ref{eq:ordrot}}
\end{shortlist}
% // ---------- Apply rotations:
\begin{shortlist}
\item Hep3Vector operator(~)~( const Hep3Vector \& v ) \see{\ref{eq:Rv}}
\item Hep3Vector operator* ~ ( const Hep3Vector \& v ) \see{\ref{eq:Rv}}
\item LorentzVector operator(~)~( const LorentzVector \& p ) \see{\ref{eq:Rv}}
\item LorentzVector operator* ~ ( const LorentzVector \& p ) \see{\ref{eq:Rv}}
\end{shortlist}
\begin{shortlist}
\item HepDouble norm2(~) \see{\ref{eq:norm2rot}}
\item HepDouble isIdentity(~)
\item ostream \& print ( ostream \& os )
\item ostream \& operator$<<$( ostream \& os, const RotationInterface \& r )
\item rectify(~) \see{\ref{eq:rectRot}}
\end{shortlist}
\subsubsection{The Rotation Group}
The following methods use \Ro s as a group, that is, they invert, multiply,
and so forth:
\begin{shortlist}
% // multiplication
\item Hep3Rotation operator* ~ ( const Hep3Rotation \& r )
\see{\ref{eq:opmulrot}}
\item operator*= ~ ( const Hep3Rotation \& r )
\see{\ref{eq:opstrot}}
\item transform ~ ( const Hep3Rotation \& r )
\see{\ref{eq:transrot}}
\item RotateX (delta) \/\/\/ RotateY (delta) \/\/\/ RotateZ (delta)
\see{\ref{eq:xyzrot}}
\item RotateAxes (newX, newY, newZ)
\see{\ref{eq:rotaxes}}
% // inversion:
\item invert(~)
\item HepRotation inverse(~)
\item HepRotation inverseOf( r )
\end{shortlist}
\subsubsection{Axial Rotations}
There are threee specialized rotation classes, {\tt HepRotationX},
{\tt HepRotationZ}, and {\tt HepRotationZ}.
These use substantially less storage than the general \Ro,
and for some methods it is quicker to work with a specialized axial rotation
rather than the general case.
All information which can be obtained from a general \Ro\ can be obtained
from any of these specialized axial rotations.
However, the set of methods which modify the rotations are very restricted
for the specialized cases:
\begin{shortlist}
% // supply angle of rotation(an identity rotation by default):
\item RotationX(~)
\item RotationX( delta ) \see{\ref{eq:rotX}}
\item RotationX \& set( delta ) \see{\ref{eq:rotX}}
\item RotationY(~)
\item RotationY( delta ) \see{\ref{eq:rotY}}
\item RotationY \& set( delta ) \see{\ref{eq:rotY}}
\item RotationZ(~)
\item RotationZ( delta ) \see{\ref{eq:rotZ}}
\item RotationZ \& set( delta ) \see{\ref{eq:rotZ}}
\end{shortlist}
\subsection{\protect\Ros\ Class --- Derived from \Ro }
The \Ros\ classes (which also include {\tt RotationX},
{\tt RotationY}, and {\tt RotationZ}) provide backward compatibility
with the original ZOOM PhysicsVectors package.
These are simply typedefs the corresponding CLHEP classes,
in the appropriate namespace.
This is because there were no features or constructors in the ZOOM
product which were felt to be overkill in the CLHEP context.
The {\tt Rotation.h} header in the ZOOM area also defines
{\tt ZMpvRep3x3}, {\tt ZMpvRep3x3}, and {\tt ZMpvRep3x3} which
are typedefs for the corresponding CLHEP structs.
\subsection{\protect\LT\ Classes}
\label{ltmethods}
There are three sorts of Lorentz transformation objects suppoprted:
General \LT s, pure Lorentz boosts {\tt HepBoost}, and pure boosts
along axes {\tt HepBoostX}, {\tt HepBoostY}, {\tt HepBoostZ}.
(Technically, ordinary \Ro s and axial rotations can also be considered
as Lorentz transformations, and indeed one of those classes can be used
wherever a general Lorentz transformation is called for. The class
{\tt Hep4RotationInterface} represents the abstract concept.
But the rotation classes have been defined above, so we will restrict this
section to discussing transformations involving the time component.)
Throughout this section,
arguments named lt, lt1, lt2, etc., are of type \LT,
arguments named b, b1, b2, etc., are of type \LB, and
arguments named r, r1, r2, etc., are of type \Ro .
Arguments named R can be any sort of {\tt Hep4RotationInterface} including
\LT s, boosts, rotations, and axial boosts and rotations.
Also, row1, row2, row3, row4, col1, col2, col3, col4
will represent \LV\ arguments.
\subsubsection{Constructors and Accessors---HepLorentzRotation}
\begin{shortlist}
% // construct an identity transformation by default:
\item HepLorentzRotation(~)
% // copy/assign
\item HepLorentzRotation ( lt )
\item HepLorentzRotation ( r )
\item HepLorentzRotation ( R )
\item HepLorentzRotation \& operator=( lt )
\item HepLorentzRotation \& operator=( r )
\item HepLorentzRotation \& operator=( R )
\item HepLorentzRotation \& set( lt )
\item HepLorentzRotation \& set( r )
\item HepLorentzRotation \& set( R )
% // supply boost
\item HepLorentzRotation ( Hep3Vector boostVector )
\item HepLorentzRotation ( HepBoost boost )
\item HepLorentzRotation ( boostX, boostY, boostZ )
\item HepLorentzRotation \& set ( Hep3Vector boostVector )
\item HepLorentzRotation \& set ( HepBoost boost )
\item HepLorentzRotation \& set ( boostX, boostY, boostZ )
% // supply R and B
\item HepLorentzRotation ( b, r ) \see{\ref{eq:decomBR}}
\item HepLorentzRotation \& set( b, r )
\item HepLorentzRotation ( r, b ) \see{\ref{eq:decomRB}}
\item HepLorentzRotation \& set( b, r )
% // supply LorentzVectors for the cols:
\item HepLorentzRotation( col1, col2, col3, col4 )
\item HepLorentzRotation \& set( col1, col2, col3, col4 )
\item HepLorentzRotation \& setRows( row1, row2, row3, row4 )
% // NOTE:
% // This constructor and set method will check that the
% // cols form an orthonormal matrix, and adjust so that
% // relation is as exact as possible.
% // supply HepRep4x4 structure:
\item HepLorentzRotation( const HepRep4x4 \& rep )
\item HepLorentzRotation \& set( const HepRep4x4 \& rep )
% // output
\item print ( ostream \& os )
\end{shortlist}
\subsubsection{Constructors and Accessors---HepBoost}
\begin{shortlist}
\item HepBoost(~)
\item HepBoost( b )
\item HepBoost \& set( b )
% // supply axis and beta;
\item HepBoost ( const Hep3Vector \& direction, beta )
\see{\ref{eq:boostform}}
\item HepBoost \& set( const Hep3Vector \& direction, beta )
\item HepBoost ( const Hep3Vector \& betaVector )
\see{\ref{eq:boostform}}
\item HepBoost \& set( const Hep3Vector \& betaVector )
\item HepBoost ( betaX, betaY, betaZ ) \see{\ref{eq:boostform}}
\item HepBoost \& set( betaX, betaY, betaZ )
\item HepBoost( const HepRep4x4Symmetric \& rep )
\item HepBoost \& set( const HepRep4x4Symmetric \& rep )
\subsubsection{Constructors and Accessors---Axial Boosts}
% // supply beta
\item HepBoostX(~)
\item HepBoostX( beta ) \see{\ref{eq:boostx}}
\item HepBoostX \& set( beta )
\item HepBoostY(~)
\item HepBoostY( beta )
\item HepBoostY \& set( beta )
\item HepBoostZ(~)
\item HepBoostZ( beta )
\item HepBoostZ \& set( beta )
\end{shortlist}
\subsubsection{Components and Decomposition}
\begin{shortlist}
\item decompose(HepBoost \& b, HepRotation \& r) \see{\ref{eq:decomBR}}
\item decompose(HepRotation \& r, HepBoost \& b) \see{\ref{eq:decomRB}}
%// for orthosymplectic Lorentz vectors
\item col1(~) \/\/\/ col2(~) \/\/\/ col3(~) \/\/\/ col4(~)
\item row1(~) \/\/\/ row2(~) \/\/\/ row3(~) \/\/\/ row4(~)
% // for individual elements:
\item xx(~) \/\/\/ xy(~) \/\/\/ xz(~) \/\/\/ xt(~)
\item yx(~) \/\/\/ yy(~) \/\/\/ yz(~) \/\/\/ yt(~)
\item zx(~) \/\/\/ zy(~) \/\/\/ zz(~) \/\/\/ zt(~)
\item tx(~) \/\/\/ ty(~) \/\/\/ tz(~) \/\/\/ tt(~)
% // for all 16 elements:
\item HepRep4x4 rep4x4(~)
\end{shortlist}
\begin{shortlist}
\item ostream \& operator$<<$( ostream \& os, const Hep4RotationInterface \& lt )
\end{shortlist}
\noindent
Also, components can be accessed by C-style and array-style subscripting:
\begin{shortlist}
\item lt[i][j] \see{\ref{eq:ltsubscript}}
\item lt(i,j) \see{\ref{eq:ltsubscript}}
\end{shortlist}
The following are applicable to \LB\ but not to \LT:
\begin{shortlist}
\item Hep3vector direction(~) \/\/\/ beta(~) \/\/\/ gamma(~)
\item Hep3Vector boostVector(~)
\item HepRep4x4Symmetric rep4x4Symmetric(~)
\end{shortlist}
\subsubsection{Application to 4-vectors}
% // ---------- Apply
\begin{shortlist}
\item LorentzVector operator*~( const LorentzVector \& w )
\item LorentzVector operator(~)~( const LorentzVector \& w )
\end{shortlist}
\subsubsection{Comparisons and Nearness}
% // intolerant comparison:
\begin{shortlist}
\item lt1 == lt2 \/\/\/ lt1 != lt2 \see{\ref{eq:ltexact}}
\item lt1 $>$ lt2 \/\/\/ lt1 $>=$ lt2 \/\/\/ lt1 $<$ lt2 \/\/\/ lt1 $<=$ lt2
\see{\ref{eq:ltorder}}
\item isIdentity()
\end{shortlist}
% // tolerant comparison:
\begin{shortlist}
\item getTolerance(~) \see{\ref{eq:epsildefR}}
\item setTolerance( tol )
\item int compare(lt) \see{\ref{eq:ltorder}}
\item bool isNear( lt, epsilon ) \see{\ref{eq:nearboost},
\ref{eq:isnearLT}}
\item double distance2( lt ) \see{\ref{eq:hownearboost},
\ref{eq:isnearLT}}
\item double howNear( lt ) \see{\ref{eq:hownearboost},
\ref{eq:isnearLT}}
\item double norm2() \see{\ref{eq:boostnorm2},
\ref{eq:ltnorm2}}
\end{shortlist}
\subsubsection{Arithmetic in the Lorentz Group}
% // arithmetic:
\begin{shortlist}
\item lt1 * lt2
\item HepLorentzRotation \& operator\verb$*=$( lt )
\item HepLorentzRotation \& operator\verb$*=$( b )
\item HepLorentzRotation \& operator\verb$*=$( r )
\item HepLorentzRotation \& transform ( lt ) \see{\ref{eq:lttrans}}
\end{shortlist}
% // inversion:
\begin{shortlist}
\item invert(~)
\item HepLorentzRotation inverseOf( lt )
\end{shortlist}
% // boosts and rotations
\begin{shortlist}
\item HepLorentzRotation \& rotate ( delta, axis ) \see{\ref{eq:ltrot}}
\item HepLorentzRotation \& rotateX ( delta )
\item HepLorentzRotation \& rotateY ( delta )
\item HepLorentzRotation \& rotateZ ( delta )
\item HepLorentzRotation \& boost ( betaX, betaY, betaZ) \see{\ref{eq:ltboost}}
\item HepLorentzRotation \& boost ( v ) \see{\ref{eq:ltboost}}
\item HepLorentzRotation \& boostX ( beta )
\item HepLorentzRotation \& boostY ( beta )
\item HepLorentzRotation \& boostZ ( beta )
\end{shortlist}
% // rectify
\begin{shortlist}
\item rectify(~) \see{\ref{eq:rectLT}}
\end{shortlist}
\subsubsection{Arithmetic on Boosts}
The pure boosts do not form a group, since the product of two
boosts is a Lorentz transformation which in general involves a rotation.
The pure boosts along an axial direction do form groups.
\begin{shortlist}
\item LorentzBoost inverseOf( b )
\item LorentzBoostX inverseOf( LorentzBoostX bx )
\item LorentzBoostY inverseOf( LorentzBoostY by )
\item LorentzBoostZ inverseOf( LorentzBoostZ bz )
\end{shortlist}
\begin{shortlist}
\item bx = bx1 * bx2
\item by = by1 * by2
\item bz = bz1 * bz2
\end{shortlist}
\subsection{\protect\LTs\ Class --- Derived from \LT }
The \LT\ classes provides backward compatibility
with the original ZOOM PhysicsVectors package.
It is defined in {\tt LorentzTransformation.h} in the ZOOM area, and
is a typedef for \LT, in the appropriate namespace.
This is because there were no features or constructors in the ZOOM
product which were felt to be overkill in the CLHEP context.
Similarly, that files establishes typedefs
{\tt LorentzBoost} for {\tt HepBoost},
{\tt LorentzBoostX} for {\tt HepBoostX},
{\tt LorentzBoostY} for {\tt HepBoostY}, and
{\tt LorentzBoostZ} for {\tt HepBoostZ}.
Also, for any ZOOM users using {\tt LorentzTransformationInterface},
this is typedefed in {\tt LorentzTransformation.h} as
{\tt Hep4RotationInterface}.
\newpage
\section{\protect\SV\ and \protect\SVz\ Classes}
\SV s may be expressed as Cartesian coordinates, Spherical coordinates,
or Cylindrical coordinates.
\begin{eqnarray}
( x, y, z ) \label{eq:cartesian}\\
( r, \theta, \phi ) \nonumber \\
\left\{
\begin{array}{r}
x = r \sin \theta \cos \phi \\
y = r \sin \theta \sin \phi \\
z = r \cos \theta \label{eq:polar}
\end{array}
\right. \\
( \rho, \phi, z ) \nonumber \\
\left\{
\begin{array}{r}
x = \rho \cos \phi \\
y = \rho \sin \phi \\
z = z \label{eq:cylindrical}
\end{array}
\right.
\end{eqnarray}
For Spherical coordinates one may optionally
specify the pseudorapidity $\eta$ instead of $\theta$.
\begin{eqnarray}
\left( \rho, \phi, \eta = - \ln { \tan { \frac {\theta}{2} } } \right)
\label{eq:spherical}
\end{eqnarray}
When accessing the angles in Sperical coordinates, the values obtained will
always be in the range $0 \leq \theta \leq \pi$ and $-\pi < \phi \leq +\pi$.
\subsection {Dot and Cross Products}
\noindent
Let $\vec{v}_{1}$ and $\vec{v}_{2}$ be \SV s. Then:
\begin{eqnarray}
\vec{v}_{1}.\mbox{dot} (\vec{v}_{2}) \equiv
\vec{v}_{1} \cdot \vec{v}_{2} = \sum_{i} \vec{v}_{1i} \vec{v}_{2i}
\label{eq:dot} \\
\vec{v}_{1}.\mbox{cross} (\vec{v}_{2}) \equiv
\vec{v}_{1} \times \vec{v}_{2} = \stackrel{\longrightarrow}
{ {\textstyle \left(
\sum_{jk} \epsilon_{ijk} \vec{v}_{1j} \vec{v}_{2k} \right) } }
\label{eq:cross} \\
\vec{v}_{1}.\mbox{diff2} (\vec{v}_{2}) =
\left| \vec{v}_{1} - \vec{v}_{2} \right| ^ 2 \label{eq:diff2}
\end{eqnarray}
\noindent
Here, $\epsilon_{ijk}$ is the three-index anti-symmetric symbol.
\subsection {Near Equality and isOrthogonal/isParallel}
\noindent
We structure the definitions of near equality and orthogonal/parallel such that
they are commutative (order of vectors makes no difference), rotationally
invariant, and scale invariant (multiplying both vectors by the same non-zero
constant makes no difference). They also match the definitions of relative
equality (or perpendicularity or parallelism) within $\epsilon$ for the case
where each vector is along or near an axis. This fixes the definitions, up to
order $\epsilon$.
\noindent
Let the method {\tt isNear()} be represented by the symbol $\approx$,
and let $\vec{v}_{1}, \vec{v}_{2}$ be \SV s.
Then:
\begin{equation}
\label{vecisnear}
\vec{v}_{1} \approx \vec{v}_{2} \mbox{ if }
\left| \vec{v}_{1} - \vec{v}_{2} \right| ^ 2
\leq \epsilon^2 \vec{v}_{1} \cdot \vec{v}_{2}
\label{eq:isNear}
\end{equation}
\noindent
However, if $\vec{v}_{1}$ and/or $\vec{v}_2$ is known to be a UnitVector
(designated as $\hat{u}_{1}$), this sets a meaningful scale---if the
vectors are nearly equal, they are both of magnitude near unity.
In that case a simpler absolute formula, which is which is equivalent
to order $\epsilon$ to the relative criterion \ref{vecisnear}, is used:
\begin{equation}
\hat{u}_{1} \approx \vec{v}_{2} \mbox{ if }
\left| \hat{u}_{1} - \vec{v}_{2} \right| ^ 2 \leq \epsilon^2
\end{equation}
\noindent
Tests for {\tt v1.isParallel(v2)}
and {\tt v1.isOrthogonal(v2)} utilize the dot and
cross products:
\begin{eqnarray}
\vec{v}_{1} \parallel \vec{v}_{2} \mbox{ if }
\left| \frac{\vec{v}_{1} \times \vec{v}_{2}}
{\vec{v}_{1} \cdot \vec{v}_{2}} \right| ^2
\le \epsilon^2
\label{eq:isPar} \\
\vec{v}_{1} \perp \vec{v}_{2} \mbox{ if }
\left| \frac{\vec{v}_{1} \cdot \vec{v}_{2}}
{\vec{v}_{1} \times \vec{v}_{2}} \right| ^2
\le \epsilon^2
\label{eq:isOrtho}
\end{eqnarray}
\noindent
Care is taken to avoid taking products of more than two potentially large
quantities in evaluating these. Thus these tests can be done on any vectors
which could safely be squared.
\noindent
$\epsilon$ is assumed to be small; in some cases, short cuts are taken
to determine the result without potentially generating large quantities.
These techniques are not necessarily faithful to the above formulae when
$\epsilon \ge 1$.
\noindent
The definition of {\tt v1.isParallel(v2)} is equivalent, to order $\epsilon$,
to a simple (but computationally more expensive) definition involving
normalizing the vectors:
\begin{displaymath}
\vec{v}_{1} \parallel \vec{v}_{2} \mbox{ if }
\left| \hat{v}_{1} - \hat{v}_{2} \right| ^2 \le \epsilon^2
\end{displaymath}
\noindent
If one of the vectors is the zero vector, the above relations may be ambiguous
instead the following hold:
($\vec{v}$ here is any non-zero vector):
\begin{eqnarray}
\vec{v} \approx \vec{0} \mbox{ iff } \vec{v} = \vec{0} \\
\vec{v} \perp \vec{0} \mbox{ for all } \vec{v} \\
\vec{v} \parallel \vec{0} \mbox{ iff } \vec{v} = \vec{0}
\end{eqnarray}
\noindent
The default tolerance (which may be modified by the class static method
{\tt setTolerance()}) for vectors and Lorentz vectors
is 100 times the double precision epsilon.
\begin{equation}
\epsilon_{\mbox{default}} \approx 2.2 \cdot 10^{-14}
\label{eq:epsildef}
\end{equation}
\subsection {Measures of Near-ness}
Each boolean tolerance comparison method is accompanied by a method returning
the measure used to compare to the tolerance $\epsilon$.
The formulae can be deduced from those above:
\begin{eqnarray}
\vec{v}_{1} \mbox{.howNear} (\vec{v}_{2}) =
\max \left( \sqrt{ \frac {\left| \vec{v}_{1} - \vec{v}_{2} \right| ^ 2}
{\vec{v}_{1} \cdot \vec{v}_{2}} } , 1 \right)
\label{eq:howNear} \\
\vec{v}_{1} \mbox{.howParallel} (\vec{v}_{2}) =
\max \left( \left| \frac{\vec{v}_{1} \times \vec{v}_{2}}
{\vec{v}_{1} \cdot \vec{v}_{2}} \right| , 1 \right)
\label{eq:howPar} \\
\vec{v}_{1} \mbox{.howOrthogonal} (\vec{v}_{2}) =
\max \left( \left| \frac{\vec{v}_{1} \cdot \vec{v}_{2}}
{\vec{v}_{1} \times \vec{v}_{2}} \right| , 1 \right)
\label{eq:howOrtho}
\end{eqnarray}
\noindent
The above relative measures are limited to a maximum of 1; cases where
the denominator would be zero may safely be done without overflow.
(Of course, {\tt howParallel()} and {\tt howOrthogonal()} are close to
zero for nearly parallel and perpendicular vectors, respectively).
Note that since a UnitVector has a natural absolute scale, the
{\tt Unitvector::howNear()} method is absolute, not relative.
When a comparison method involves absolute tolerance, the measure returned
is not truncated at 1:
\begin{eqnarray}
\hat{u}_{1} \mbox{.howNear} (\vec{v}_{2}) =
\vec{v}_{2} \mbox{.howNear} (\hat{u}_{1}) =
\left| \hat{u}_{1} - \vec{v}_{2} \right|
\label{eq:howNear:2}
\end{eqnarray}
In addition, we provide \verb$deltaR()$,
a measure of nearness useful in collider physics analysis.
It is defined by
\begin{equation}
\vec{v}_{1} \mbox{deltaR} (\vec{v}_{2}) =
\sqrt{ (\Delta \phi)^2 + (\Delta \eta)^2) } =
\sqrt{ \left( \vec{v}_{1}\mbox{.phi} - \vec{v}_{2}\mbox{.phi} \right)^2 +
\left( \vec{v}_{1}\mbox{.eta} - \vec{v}_{1}\mbox{.eta} \right)^2 }
\label{eq:deltaR}
\end{equation}
\noindent
where of course the angular $\phi$ difference is corrected to lie in the range
$(-\pi, \pi]$ (the {\tt deltaPhi()} method is used).
When one vector or the other is zero,
\begin{eqnarray}
\vec{v} \mbox{.howNear} (\vec{0}) =
\vec{0} \mbox{.howNear} (\vec{v}) = 1
\label{eq:howNear:3}
\\
\vec{v} \mbox{.howParallel} (\vec{0}) =
\vec{0} \mbox{.howParallel} (\vec{v}) = 1
\label{eq:howPar:2}
\\
\vec{v} \mbox{.howOrthogonal} (\vec{0}) =
\vec{0} \mbox{.howOrthogonal} (\vec{v}) = 0
\label{eq:howOrtho:2}
\\
\vec{v} \mbox{.deltaR} (\vec{0}) = \vec{0} \mbox{.deltaR} (\vec{v}) =
\left| \Delta \eta \right| =
\left| \vec{v}\mbox{.eta()} \right|
\label{eq:deltaR:2}
\end{eqnarray}
When both vectors are zero, all these measures will return zero.
\subsubsection{Ordering Comparisons for \protect\SV s}
The comparison operators \verb$ ( >, >=, <, <= )$ for \SV\ (and for
\UV) use a ``dictionary ordering'',
comparing first the Z, then the Y, then the
X components:
\begin{eqnarray}
\vec{v}_1 > \vec{v}_2 \mbox{ if } \nonumber \\
z_1 > z_2 \mbox { or } \nonumber \\
\left[
z_1 = z_2 \mbox { and }
y_1 > y_2 \right] \mbox { or } \nonumber \\
\left[
z_1 = z_2 \mbox { and }
y_1 = y_2 \mbox { and }
x_1 > x_2 \right]
\label{eq:compSV}
\end{eqnarray}
\subsection{Intrinsic Properties and Relativistic Quantities}
A \SV\ does not have much in the way of intrinsic properties---just its
magnitude.
We can also talk about the pseudorapidity, which depends only on the angle
against the Z axis.
\begin{eqnarray}
\vec{v1}\mbox{.mag() } = \sqrt { \vec{v}_{1} \cdot \vec{v}_{1} }
\label{eq:mag} \\
\vec{v1}\mbox{.mag2() } = \vec{v}_{1} \cdot \vec{v}_{1}
\label{eq:mag2} \\
\vec{v1}\mbox{.eta() } = -\ln \tan \frac {\theta_{\vec{v}_1,\hat{z}}}{2}
= -\ln \tan \frac {\hat{v}_1 \cdot \hat{z}}{2}
\label{eq:eta}
\end{eqnarray}
Another intrinsic is the unit vector in the direction of $\vec{v}$:
\begin{eqnarray}
\vec{v}\mbox{.unit() } = \frac{\vec{v}}{|v|}
\label{eq:svunit}
\end{eqnarray}
A somewhat contrived intrinsic property, useful for Geant4, is a special
vector orthognal to $\vec{v}$, lying in a plane defined by two coordinate
axes.
The plane is chosen so as to supress the smallest component of $\vec{v}$.
\begin{eqnarray}
\min( v_x, v_y, v_z ) = v_z \Longrightarrow
\vec{v}\mbox{.orthogonal() } = \left\{ v_y, -v_x, 0 \right\}
\label{eq:orthogonal} \\
\min( v_x, v_y, v_z ) = v_y \Longrightarrow
\vec{v}\mbox{.orthogonal() } = \left\{ -v_z, 0, v_x \right\}
\nonumber \\
\min( v_x, v_y, v_z ) = v_z \Longrightarrow
\vec{v}\mbox{.orthogonal() } = \left\{ 0, v_z, -v_y \right\}
\nonumber
\end{eqnarray}
(In the case of equal components, $v_z$ is considered smallest, then $v_y$.)
\noindent
A \SV\ of length less than 1 may be considered as defining a Lorentz boost;
in that sense, we can discuss $\beta$ and $\gamma$ of the \SV, and the rapidity
associated with that boost.
\begin{eqnarray}
\vec{v1}\mbox{.beta() } = \vec{v1}\mbox{ .mag() } = \beta
\label{eq:mag:2} \label{eq:beta} \\
\vec{v1}\mbox{.gamma() } = \frac{1} {\sqrt {1-\beta^2 } } = \gamma
\label{eq:gamma} \\
\vec{v1}\mbox{.rapidity() } = \tanh^{-1} (\vec{v}_{1} \cdot \hat{z})
\label{eq:rap} \\
\vec{v1}\mbox{.coLinearRapidity() } = \tanh^{-1} \beta
= \tanh^{-1} (\vec{v}_{1} \cdot \hat{v}_1)
\label{eq:coLinRap}
\end{eqnarray}
The rapidity and pseudorapidity are discussed further in
section \S{\ref{rapidity}}.
\subsection{Properties Involving Vectors and Directions}
Let a reference direction be represented by a unit vector $\hat{u}$ in that
direction.
In all the following methods, the reference direction may be
omitted, defaulting to $\hat{\bf z}$.
And in these methods, a {\em non-zero} second vector may be
substituted for $\hat{u}$; the unit vector in the direction of $\vec{v}_2$
will be used in that case.
Then the following definitions of methods relative to the reference direction
apply:
\begin{eqnarray}
\vec{v}\mbox{.angle} (\hat{u}) =
\theta_{\vec{v},\hat{u}} =
\cos^{-1} \left( \frac{ \vec{v} \cdot \hat{u} }
{ \left| \vec{v} \right| } \right)
\label{eq:angle} \label{eq:theta} \\
\vec{v}\mbox{.cosTheta} (\hat{u}) =
\cos \theta_{\vec{v},\hat{u}}
\label{eq:cosTheta} \\
\vec{v}\mbox{.cos2Theta} (\hat{u}) =
\cos^2 \theta_{\vec{v},\hat{u}}
\label{eq:cos2Theta} \\
\vec{v}\mbox{.eta} (\hat{u}) =
-\ln \tan \frac {\theta_{\vec{v},\hat{u}}}{2}
\label{eq:eta:2} \\
\vec{v}\mbox{.rapidity} (\hat{u}) =
\tanh^{-1} (\vec{v}_{1} \cdot \hat{u})
\label{eq:rap:2} \\
\vec{v}\mbox{.perp} (\hat{u}) =
\left| \vec{v} - (\vec{v} \cdot \hat{u}) \hat{u} \right|
\label{eq:perp} \\
\vec{v}\mbox{.perp2} (\hat{u}) =
\left| \vec{v} - (\vec{v} \cdot \hat{u}) \hat{u} \right| ^2
\label{eq:perp2} \\
\vec{v}\mbox{.perpPart} (\hat{u}) =
\vec{v} -
\hat{u} \cos \theta_{\vec{v},\hat{u}}
\left| \vec{v} \right| =
\vec{v} - (\vec{v} \cdot \hat{u}) \hat{u}
\label{eq:perpPart} \\
\vec{v}\mbox{.project} (\hat{u}) =
\hat{u} \cos \theta_{\vec{v},\hat{u}}
\left| \vec{v} \right| =
(\vec{v} \cdot \hat{u}) \hat{u}
\label{eq:project} \\
\vec{v}\mbox{.polarAngle} (\vec{v_2})
\equiv \vec{v}\mbox{.polarAngle} (\vec{v_2}, \hat{z}) =
\vec{v_2}\mbox{.theta()} - \vec{v}\mbox{.theta()}
\label{eq:polarA} \\
\vec{v}\mbox{.polarEta} (\vec{v_2})
\equiv \vec{v}\mbox{.polarEta} (\vec{v_2}, \hat{z}) =
\vec{v_2}\mbox{.eta()} - \vec{v}\mbox{.eta()}
\label{eq:polarEta} \\
\vec{v}\mbox{.polarAngle} (\vec{v_2}, \hat{u}) =
\vec{v_2}\mbox{.eta}(\hat{u}) - \vec{v}\mbox{.eta}(\hat{u})
\label{eq:polarA:2} \\
\vec{v}\mbox{.azimAngle} (\vec{v_2})
\equiv \vec{v}\mbox{.azimAngle} (\vec{v_2}, \hat{z}) =
\vec{v_2}\mbox{.phi()} - \vec{v}\mbox{.phi()} ^*
\label{eq:azim} \\
\vec{v}\mbox{.deltaPhi} (\vec{v_2})
\equiv \vec{v}\mbox{.azimAngle} (\vec{v_2})
\label{eq:deltaPhi} \\
\vec{v}\mbox{.azimAngle} (\vec{v_2}, \hat{u}) =
\theta_{\vec{v_2}\mbox{.perpPart}(\hat{u}),
\vec{v}\mbox{.perpPart}(\hat{u})}
\mbox{ sign} \left( \hat{u} \cdot (\vec{v} \times \vec{v_2}) \right)
\label{eq:azim:2}
\end{eqnarray}
\noindent
* Equation (\ref{eq:azim}) is not quite definitive:
The azimuthal angle between two vectors (or delta phi)
will always be translated into an
equivalent angle in the range $(-\pi,\pi]$.
The azimuthal angle between two vectors with respect to a reference direction,
as shown in equation (\ref{eq:azim:2}) above, is found by
projecting both vectors into the plane defined by the reference direction,
and taking the angle between those projections, in the clockwise sense about
the reference axis. Again, this will be in the range $(-\pi,\pi]$.
\subsection{Direct Vector Rotations}
\label{rotations}
Direct vector rotations are methods of \SV\ or \LV\ which
modify the vector being acted
upon---{\tt v.rotate}$(\phi, \theta, \psi)$---or
global functions which form a new
vector---{\tt rotationOf}$(\vec{v}, \phi, \theta, \psi)$.
\noindent
Rotations about the X, Y, or Z axis are defined in the counter-clockwise
sense. Thus for rotations about the Z axis, if $\vec{v}$ has representation
in polar coordinates $( v_r, v_\theta, v_\phi )$
\begin{equation}
\vec{v}.\mbox{rotateZ} ( \delta ) \mbox{ is equivalent to }
\vec{v}_\phi \Longrightarrow \vec{v}_\phi + \delta
\label{eq:rotZ:2}
\end{equation}
\noindent
The axis rotations are implemented taking advantage of their
simple form. When $\vec{v} = (x, y, z)$ is rotated by angle $\delta$,
\begin{eqnarray}
\vec{v}.\mbox{rotateZ} (\delta) \Longrightarrow
( x \cos\delta - y \sin\delta, x \sin\delta + y \cos\delta, z )
\label{eq:rotZ} \\
\vec{v}.\mbox{rotateY} (\delta) \Longrightarrow
( z \sin\delta + x \cos\delta, y, z \cos\delta - x \sin\delta )
\label{eq:rotY} \\
\vec{v}.\mbox{rotateX} (\delta) \Longrightarrow
( x, y \cos\delta - z \sin\delta, y \sin\delta + z \cos\delta )
\label{eq:rotX}
\end{eqnarray}
\noindent
More general rotations may be expressed in terms of
an angle $\delta$ (counter-clockwise) about an axis given
as a \UV\ $\hat{u}$, or in terms of Euler Angles
$(\phi, \theta, \psi)$:
\[ \vec{v}.\mbox{rotate}(\hat{u},\delta) \Longrightarrow \]
\begin{equation}
\label{eq:axisrot}
\left(
\begin{array}{ccc}
\cos \delta + (1 - \cos \delta ) u_x^2 &
(1 - \cos \delta ) u_x u_y - \sin \delta u_z &
(1 - \cos \delta ) u_x u_z + \sin \delta u_y \\
(1 - \cos \delta ) u_y u_x + \sin \delta u_z &
\cos \delta + (1 - \cos \delta ) u_y^2 &
(1 - \cos \delta ) u_y u_z - \sin \delta u_x \\
(1 - \cos \delta ) u_z u_x - \sin \delta u_y &
(1 - \cos \delta ) u_z u_y + \sin \delta u_x &
\cos \delta + (1 - \cos \delta ) u_z^2
\end{array}
\right)
\left(
\begin{array}{c}
v_x\\
v_y\\
v_z
\end{array}
\right)
\end{equation}
\[ \vec{v}.\mbox{rotate}(\phi, \theta, \psi) \Longrightarrow \]
\begin{equation}
\label{eq:eulerrot}
\left(
\begin{array}{ccc}
\cos \psi \cos \phi - \sin \psi \cos \theta \sin \phi &
\cos \psi \sin \phi + \sin \psi \cos \theta \cos \phi &
\sin \psi \sin \theta \\
- \sin \psi \cos \phi - \cos \psi \cos \theta \sin \phi &
- \sin \psi \sin \phi + \cos \psi \cos \theta \cos \phi &
\cos \psi \sin \theta \\
\sin \theta \sin \phi &
- \sin \theta \cos \phi &
\cos \theta
\end{array}
\right)
\left(
\begin{array}{c}
v_x\\
v_y\\
v_z
\end{array}
\right)
\end{equation}
\noindent
The Euler angles definition matches that found in found in
{\em Classical Mechanics} (Goldstein), page 109.
This treats the Euler angles as a sequence of counter-clockwise {\bf passive}
rotations;
that is, the vector remains fixed while the coordinate axes are rotated---new
vector components are computed in new coordinate frame.
It is unnatural (though possible) to view an Euler angles transformation as
as sequence of active rotations.
HEP computations ordinarily use the active rotation viewpoint.
Therefore, rotations about an axis imply {\bf active} counter-clockwise
rotation in this package.
Consequently, a rotation by angle $\delta$ around the X axis is
equivalent to a rotation with Euler angles
$(\phi=\psi=0, \mbox{ } \theta = - \delta)$
and a rotation about the Z axis is
equivalent to a rotation with Euler angles
$(\theta = 0, \mbox{ } \phi=\psi= - \delta/2)$.
\subsubsection{RotateUz}
Another way to specify a direct rotation is via {\tt v.rotateUz(u)}
where u is required to be a unit \SV .
This rotates the reference frame such that the original Z-axis will lie
in the direction of $\hat{u}$. Many rotations would accomplish this; the
one selected uses $u$ as its third column and is given by:
\[ \vec{v}.\mbox{rotateUz}(u_x, u_y, u_z)\Longrightarrow \]
\begin{equation}
\label{eq:rotUz}
\left(
\begin{array}{ccc}
u_x u_z / u_\perp & - u_y / u_\perp & u_x \\
u_y u_z / u_\perp & u_x / u_\perp & u_y \\
- u_\perp & 0 & u_z
\end{array}
\right)
\left(
\begin{array}{c}
v_x\\
v_y\\
v_z
\end{array}
\right)
\end{equation}
\noindent
Here, $u_\perp \equiv \sqrt{u_x^2 + u_y^2}$. Using $u$ as the third column of
the rotation removes the ambiguity except if $u$ is parallel to $\hat{z}$.
In that case, if $u = \hat{z}$ the vector $\vec{v}$ is left untouched,
while if
$u = - \hat{z}$,
If u is $-\hat{z}$, $\vec{v}$ is rotated by 180 degrees about the Y axis.
\subsubsection{Applying \protect\Ro s to \protect\SV s}
In CLHEP, the \SV\ class is aware of the existence of the \Ro\ class.
This is reflected in two routines to apply a \Ro\ to the \SV :
\begin{equation}
\label{eq:opstareq}
\left.
\begin{array} {r}
\vec{v} \mbox{ *= } R \\
\vec{v} \mbox{.transform} (R)
\end{array}
\right\}
\Longleftrightarrow
\vec{v} \leftarrow R \vec{v}
\end{equation}
\noindent
Notice that these are identical.
In contrast to the usual {\tt operator *=} semantics,
{\tt v *= R } {\it left} multiplies the matrix representing {\tt v}
by the matrix representing {\tt R}.
\subsection{Overflow for Large Vectors}
When vector components are near the limits of floating point representation,
there are cases where operations have mathematical results which can be
represented, but intermediate steps give too large a result. An extreme
example is that testing whether two vectors are nearly orthogonal should give
a result of type \verb$bool$, yet the intermediate steps may involve a dot
product adding three huge terms.
\noindent
Although each of the vector operations described above can be performed
protecting against overflow (by judicious re-scaling) this generally leads
to unacceptable inefficiency in the overwhelming majority of cases where the
vectors are not so large. The compromise used in this package is:
\begin{itemize}
\item All vectors are assumed to be ``squarable,'' that is, the algorithms
may freely take a dot product of a vector with itself. For \verb$double$
numbers, this implies that components are limited to about $10^{153}$
in magnitude.
\item Care is taken that any operation, applied to squarable vectors, gives
the proper result. This implies that we never take products of more than two
powers of components without checking for size and potentially re-scaling.
In particular, {\tt isParallel()} and {\tt isOrthogonal()} do extra work
to avoid trouble when vectors have components on the order of $10^{76}$ to
$10^{152}$.
\end{itemize}
\section{\protect\LV\ Class}
We always take the time component to be the `4' index, and $c = 1$.
For these equations, let the sense of the metric be $ {\cal M} = \pm 1$ and
the metric be represented as $g_{ij}$ where $g_{44} = {\cal M}$,
$g_{ii} = -{\cal M}$, and all other $g_{ij} = 0$.
\noindent
Let ${\bf w}_{1}$ and ${\bf w}_{2}$ be \LV s.
Then ${\bf w}_{1}$.dot(${\bf w}_{2}$) is defined by:
\begin{equation}
\label{eq:wdot}
{\bf w}_{1} \cdot {\bf w}_{2} = \sum_{ij} g{_ij}
{\bf w}_{1}^{i} {\bf w}_{2}^{j} \\
\end{equation}
\noindent
The sign of this dot product is dependent on $\cal M$.
For the remainder of these definitions we will take the metric to be
(-- -- -- +) and point out any definitions which change sign when the
(+ + + --) metric is chosen.
\subsection{Combinations and Properties of \protect\LV s}
The dot product, and Lorentz-invariant magnitude squared and
squared norm of the difference between two 4-vectors, are metric-dependent.
Let $w_i = ( \vec{v}_i, t_i )$:
\begin{eqnarray}
\label{eq:wdot:2}
w_1\mbox{.dot}(w_2) = w_1 \cdot w_2 = t_1 t_2 - \vec{v}_1 \cdot \vec{v}_2 \\
\label{eq:wmag2}
w_1\mbox{.mag2}() = w_1 \cdot w_1 = t_1^2 - |\vec{v}_1|^2 \\
\label{eq:wdiff2}
w_1\mbox{.diff2}(w_2) = (w_1-w_2) \cdot (w_1-w_2) =
(t_1 - t_2)^2 - \left| \vec{v}_1 - \vec{v}_2 \right| ^2 \\
\label{eq:wmag}
w_1\mbox{.mag(~)} = w_1\mbox{.m(~)} =
\mbox{sign}(t_1^2 - \vec{v}_1^2)
\sqrt{\left|t_1^2 - \vec{v}_1^2\right|}
\end{eqnarray}
The Euclidean-norm, and Euclidean-norm difference squared, are given by
\begin{eqnarray}
\label{eq:wENorm2}
w_1\mbox{.EuclideanNorm2}() = t_1^2 + |\vec{v}_1|^2 \\
\label{eq:wENorm}
w_1\mbox{.EuclideanNorm}() = \sqrt{t_1^2 + |\vec{v}_1|^2} \\
\label{eq:wdelta2E}
w_1\mbox{.delta2Euclidean}(w_2) = t_1 t_2 + \vec{v}_1 \cdot \vec{v}_2
\end{eqnarray}
It is convenient to have methods returning $t \pm z$; for completeness
we also provide the plus() and minus() methods relative to an arbitrary
direction $\hat{u}$:
\begin{eqnarray}
\label{eq:wplus}
w_1\mbox{.plus}() = t_1 + z_1 \\
\label{eq:wminus}
w_1\mbox{.minus}() = t_1 - z_1 \\
\label{eq:wplus:2}
w_1\mbox{.plus}(\hat{u}) = t_1 + \vec{v}_1 \cdot \hat{u} \\
\label{eq:wminus:2}
w_1\mbox{.minus}(\hat{u}) = t_1 - \vec{v}_1 \cdot \hat{u}
\end{eqnarray}
\subsection{Kinematics of \protect\LV s}
The rest mass and its square are independent of metric.
Note that {\tt restMass2()} differs from {\tt mag2()} in that no matter
which metric is selected, {\tt restMass2()} remains $t^2-v^2$.
The sign of the
rest mass is set to match the time component; thus the rest mass of
$(0,0,0,-m)$ will return as $-m$.
The method {\tt rest4Vector()} will return a 4-vector equal to this vector
in its rest frame.
\begin{eqnarray}
\label{eq:wrestM2}
p\mbox{.restMass2}() = p\mbox{.invariantMass2}() = E^2 - |\vec{p}|^2 \\
\label{eq:wrestM}
p\mbox{.restMass}() = p\mbox{.invariantMass}() =
\sqrt {E^2 - |\vec{p}|^2} \times \mbox{sign}(t_1) \\
\label{eq:wrest4V}
p\mbox{.rest4Vector}() = \left( 0, 0, 0, p\mbox{.restMass}() \right)
\end{eqnarray}
\noindent
Taking the rest mass of a spacelike 4-vector will ZMthrow the exception
{\tt ZMxpvSpacelike}.
\noindent
The boost which if applied to {\tt w.rest4Vector()} would give the result
{\tt w} is {\tt w.boostVector()}. A pure boost can be expressed as a \SV:
\begin{eqnarray}
\label{boostvector}
w_1\mbox{.boostVector}() = \frac{\vec{v}_1}{t_1} \\
\left(w_1\mbox{.rest4Vector}()\right)\mbox{.boosted}
\left(w_1.\mbox{.boostVector}() \right) \equiv w_1 \nonumber
\end{eqnarray}
\noindent
The boostVector for a zero 4-vector will return as zero.
{\tt boostVector()}
will return $v/t$ even if the 4-vector is spacelike, but will ZMthrow a
{\tt ZMxpvTachyonic} error.
If $ t=0 $ it will throw {\tt ZMxpvInfiniteVector} and do the divisions,
returning an infinite vector if ignored.
Beta and gamma refer to this boost vector:
\begin{eqnarray}
\label{eq:wbeta}
w_1\mbox{.beta}() = \beta = \left| w_1\mbox{.boostVector}() \right |
= \frac{\left| \vec{v}_1 \right|} {| t_1 |} \\
\label{eq:wgamma}
w_1\mbox{.gamma}() = \gamma = \frac{1}{\sqrt{1-\beta^2}}
\end{eqnarray}
Psuedorapidity and rapidity are discussed below (\S\ref{rapidity}):
\begin{eqnarray}
\label{eq:weta}
w_1\mbox{.eta()} = - \ln \tan \frac{\theta}{2} \\
\label{eq:weta:2}
w_1\mbox{.eta(u)} = - \ln \tan \frac{\hat{p} \cdot \hat{u}}{2} \\
\label{eq:wrapid}
w_1\mbox{.rapidity() }
= \frac{1}{2} \ln \left( \frac{E + p_z}{E - p_z} \right)
= \tanh^{-1} \frac {z}{t} \\
\label{eq:wrapid:2}
w_1\mbox{.rapidity(u) }
= \frac{1}{2} \ln \left( \frac{E + \vec{p} \cdot \hat{u}}
{E - \vec{p} \cdot \hat{u}} \right)
= \tanh^{-1} \frac {\vec{v} \cdot \hat{u}} {t} \\
\label{eq:wcoLinRap}
w_1\mbox{.coLinearRapidity() }
= \frac{1}{2} \ln \left( \frac{E + |p|}{E - |p|} \right)
= \tanh^{-1} \frac {|v|}{t}
\end{eqnarray}
\noindent
The ``transverse mass'' is found by neglecting the x- and
y-components of the 4-vector. This is the square root of the
sum of the rest mass squared and the transverse momentum squared:
\begin{eqnarray}
\label{eq:wmt2}
p\mbox{.mt2}() = E^2 - p_z^2 \\
\label{eq:wmt}
p\mbox{.mt}() = \sqrt { \left| E^2 - |\vec{p}|^2 \right| }
\times \mbox{sign}(E^2 - |\vec{p}|^2)
\end{eqnarray}
\noindent
Note that $m_t \ge m$. {\it Warning:} Although this definition for $m_t$
matches that of equation (34.36) in the Review of Particle Physics and this
definition has always been in CLHEP, some experimenters have indicated that
various experiments may use different definitions, which are of more use to
their applications.
\noindent
The ``transverse energy'' is defined as $E \sin \theta$.
This quantity is invariant, to order $m/E$, under boosts in the Z direction.
It is most easily expressed in terms of the energy, momentum, and
transverse momentum:
\begin{eqnarray}
\label{eq:wet2}
p\mbox{.et2}() = \frac {E^2 p_\perp^2}{|p|^2} \\
\label{eq:wet}
p\mbox{.et}() = \sqrt { \left| \frac {E^2 p_\perp^2}{|p|^2} \right| }
\times \mbox{sign}(E)
\end{eqnarray}
\noindent
For completeness, the transverse mass and energy are also defined with respect
to a direction specified by a \SV $\vec{v}$:
\begin{eqnarray}
p\mbox{.mt2}(\vec{v}) = E^2 - ( \vec{p} \cdot \hat{v} )^2 \\
p\mbox{.mt}(\vec{v}) = \sqrt {\left| E^2 - (\vec{p} \cdot \hat{v})^2\right|}
\times \mbox{sign}(E^2 - ( \vec{p} \cdot \hat{v} )^2 )
\\
p\mbox{.et2}(\vec{v}) = \frac {E^2 p.\mbox{perp}(\hat{v})^2} {|p|^2} \\
p\mbox{.et}(\vec{v}) = \sqrt { \left| \frac {E^2 p.\mbox{perp}(\hat{v})^2}
{|p|^2} \right| }
\times \mbox{sign}(E)
\end{eqnarray}
\subsection{Invariant Mass and the Center-of-Mass Frame}
The invariant mass of a pair of \LV s is given by:
\begin{eqnarray}
\label{eq:winvMass2}
w_1\mbox{.invariantMass2}(w_2) =
\left(t_1+t_2\right)^2 - \left| \vec{v}_1 - \vec{v}_2 \right| ^2 \\
\label{eq:winvMass}
w_1\mbox{.invariantMass}(w_2) =
\left(t_1+t_2\right)^2 - \left| \vec{v}_1 - \vec{v}_2 \right| ^2
\times \mbox{sign}(t_1 + t_2)
\end{eqnarray}
The boost necessary top bring a pair of vectors into their Center-of-Mass
frame is given by
\begin{eqnarray}
\label{eq:wfindBoost}
w_1\mbox{.findBoostToCM}(w_2) = - \frac{\vec{v}_1 + \vec{v}_2}{t_1+t_2}
\end{eqnarray}
\noindent
If the sum of the two 4-vectors is spacelike, this makes analytic sense but
is physically meaningless; a {\tt ZMxpvTachyonic} error will be ZMthrown.
If the sum of the time components is zero, a
{\tt ZMxpvInfiniteVector} error will be ZMthrown.
\subsection{Various Forms of Masses and Magnitudes}
The \LV\ class combines the features of the CLHEP and ZOOM 4-vectors.
Each of these deal with several concepts related to the magnitude or
``mass'' associated with the 4-vector.
This is further complicated by the possibility that the metric,
normally taken to be (-- -- -- +) which we designate as $\cal M$ $= +1$,
can be set to (+ + + --) which we will designate as $\cal M$ $= -1$.
So some of the definitions below will involve $\cal M$.
In the CLHEP original package, of course, $\cal M$ would always be $+1$.
These methods are defined above, but perhaps it will be helpful to
provide them all in one place.
In the below definitions, since we are conceptually dealing with
energy-momentum 4-vectors, we will use $(\vec{p},e)$ for the 4-vector
components.
% orig PV CLHEP Merge
%mag2(~) M * (t^2-v^2) (t^2-v^2) PV
%m2(~) --- (t^2-v^2) CLHEP
%mag(~) --- sign(m2)*sqrt|m2| CLHEP
%m(~) --- sign(m2)*sqrt|m2| CLHEP
%mt2(~) --- e^2 - pz^2 CLHEP
%mt(~) --- sign(mt2)*sqrt|mt2| CLHEP
%w.dot(w) M * (t^2-v^2) (t^2-v^2) PV
%w.invariantMass2(~) --- --- (t^2-v^2)
%w.invariantMass(~) --- --- sign(t)*sqrt(t^2-v^2)
%restMass2(~) (t^2-v^2) --- (t^2-v^2)
%restMass(~) sign(t)*sqrt(t^2-v^2),# ---
% sign(t)*sign(t^2-v^2)*sqrt|t^2-v^2|
\begin{eqnarray}
w\mbox{.mag2()} = w\mbox{.dot(w)} = {\cal M} (E^2 - \vec{p}^2) \\
w\mbox{.m2(~)} = E^2 - \vec{p}^2 \\
w\mbox{.mag(~)} = w\mbox{.m(~)} =
\mbox{sign}(E^2 - \vec{p}^2) \sqrt{\left|E^2 - |\vec{p}|^2\right|} \\
w\mbox{.invariantMass2(~)} = w\mbox{.restMass2(~)} = E^2 - |\vec{p}|^2 \\
w\mbox{.w.invariantMass(~)} = w\mbox{.restMass(~)} =
\mbox{sign}(E) \sqrt{E^2 - |\vec{p}|^2} \\
w\mbox{.mt2(~)} = E^2 - p_z^2 \\
w\mbox{.mt(~)} =
\mbox{sign}(E^2 - p_z^2) \sqrt{\left|E^2 - p_z^2\right|} \\
w\mbox{.et2}() = \frac {E^2 p_\perp^2}{|p|^2} \\
w\mbox{.et}() = \sqrt { \left| \frac {E^2 p_\perp^2}{|p|^2} \right| }
\times \mbox{sign}(E)
\end{eqnarray}
Thus a tachyonic particle (for which $t^2-v^2 < 0$) is assigned a
negative mass m(), but a positive restMass().
\subsection{Direct \protect\LV\ Boosts and Rotations}
Direct boosts and rotations are methods of \LV\ which
modify the 4-vector being acted
upon--- {\it e.g.}, {\tt w.boost}$(\hat{u}, \beta)$---or
global functions which form a new
vector--- {\it e.g.}, {\tt boostOf}$(w, \hat{u}, \beta)$.
Rotations act in the obvious manner, affecting only the $\vec{v}$ component of
the 4-vector---see \S\ref{rotations}.
In analogy with our ``active'' rotation viewpoint, boosts are treated as
``active'' transformations rather than transformations of the coordinate system.
That is, if you take a 4-vector at rest, with positive mass ($t$),
and boost it by a positive amount in the X direction, the resulting 4-vector
will have positive $x$.
Boosts along the X, Y, or Z axis are simpler than the general case.
Let $w = (\vec{v}, t) = (x, y, z, t)$:
\begin{eqnarray}
\label{eq:wbX}
w.\mbox{boostX} (\beta) \Longrightarrow
( \gamma x + \beta \gamma t, y, z, \gamma t + \beta \gamma x ) \\
\label{eq:wbY}
w.\mbox{boostY} (\beta) \Longrightarrow
( x, \gamma y + \beta \gamma t, z, \gamma t + \beta \gamma y ) \\
\label{eq:wbZ}
w.\mbox{boostZ} (\beta) \Longrightarrow
( x, y, \gamma z + \beta \gamma t, \gamma t + \beta \gamma z ) \\
\gamma \equiv \frac{1}{\sqrt{1-\beta^2}} \nonumber
\end{eqnarray}
More general rotations boosts may be expressed in terms of
$\beta$ along an axis given as a \SV\ $\hat{u}$ (which will be normalized),
or in terms of a \SV\ boost $\vec{\beta}$, which must obey $|\vec{\beta}|<1$.
(Boosts beyond the speed of light ZMthrow a
{\tt ZMxpvTachyonic} error, and leave the 4-vector unchanged
if this is ignored.)
For the axis $(\hat{u}, \beta)$ form,
\begin{eqnarray}
\label{eq:pureboost}
\left\{
\begin{array}{lcl}
t & \longleftarrow & \gamma t + \beta \gamma \vec{v} \cdot \hat{u} \\
\vec{v} & \longleftarrow & \vec{v} + \left[
\frac{\gamma-1}{\beta^2} \beta \vec{v} \cdot \hat{u} +
\beta \gamma t \right] \hat{u}
\end{array}
\right. \\
\gamma \equiv \frac{1}{\sqrt{1-\beta^2}} \nonumber
\end{eqnarray}
For the boost vector $(\vec{\beta})$ form,
\begin{eqnarray}
\label{eq:wboostvec}
\left\{
\begin{array}{lcl}
t & \longleftarrow & \gamma t + \gamma \vec{v} \cdot \vec{\beta} \\
\vec{v} & \longleftarrow & \vec{v} + \left[
\frac{\gamma-1}{|\vec{\beta}|^2} \vec{v} \cdot \vec{\beta} +
\gamma t \right] \vec{\beta}
\end{array}
\right. \\
\gamma \equiv \frac{1}{\sqrt{1-|\vec{\beta}|^2}} \nonumber
\end{eqnarray}
\subsubsection{Applying \protect\Ro s and \protect\LT s to \protect\SV s}
In CLHEP, the \LV\ class is aware of the existence of the \Ro\ and
\LT\ classes.
This is reflected in routines to apply these the \LV :
\begin{equation}
\label{eq:wopstareq}
\left.
\begin{array} {r}
p \mbox{ *= } R \\
p \mbox{.transform} (R)
\end{array}
\right\}
\Longleftrightarrow
p \leftarrow R p
\end{equation}
\noindent
Notice that these are identical.
In contrast to the usual {\tt operator *=} semantics,
{\tt p *= R } {\it left} multiplies the matrix representing {\tt p}
by the matrix representing {\tt R}.
\subsection{Near-equality of \protect\LV s}
We keep in mind that the prime utility of {\tt isNear()} and related methods
is to see whether two vectors, which have been created along two computational
paths or from
two sets of fuzzy quantities, ought mathematically to be taken as equal.
{\em Relative} tolerance is needed,
and this is more involved than just
checking for approximate equality in the time and space sectors respectively.
For example, though no non-zero vector can be near the zero vector,
$(\vec{\epsilon},t=1)$ with $\vec{\epsilon}$ a very small vector
should be considered close to $(\vec{0},t=1)$.
The temptation is to use, as the normaliztion to determine relative nearness,
the length $|\vec{v}|^2 + t^2$. Let us call this the Euclidean norm, since it
is the norm in complex 4-space $(\vec{v}, i t)$.
This is not Lorentz invariant, but for
many purposes is a good criteria for calling two 4-vectors close.
We use the Euclidean norm to define {\tt isNear()} for {\tt LorentzVectors};
it has the virtue of simplicity and can be applied to any 4-vectors.
Let the method {\tt isNear()} be represented by the symbol $\approx$:
\begin{equation}
\label{eq:wisNear}
w_1 \approx w_2 \Longleftrightarrow
\left|
\vec{v}_1 - \vec{v}_2
\right| ^2
+ (t_1 - t_2)^2 \leq
\epsilon^2
\left[
\left|
\vec{v}_1 \cdot \vec{v}_2
\right|
+
\left(
\frac{t_1 + t_2}{2}
\right)^2
\right]
\end{equation}
This definition not Lorentz invariant, but it
is (to order $\epsilon \times \gamma$) independent of frame
within the space of {\em small} Lorentz transformations.
It turns out to be impossible to find a definition which is
Lorentz invariant for all possible 4-vectors, under arbitrarily large
boosts, and still behaves like a measure of near-ness.
A second useful definition, which is Lorentz invariant, but is
sensibly applicable only to timelike 4-vectors,
is to look at the Euclidean norm of the difference of two vectors
{\em in their Center-of-Mass frame}.
This is intuitively appealing to HEP practitioners;
we make this method available as well, calling it {\tt isNearCM()}.
Let the method {\tt isNear()} be represented by the symbol $\approx$,
and the method {\tt isNearCM()} be represented by the symbol
$\stackrel{\mbox{\tiny CM}}{\approx}$.
Let ${\bf w}_{1}$, ${\bf w}_{2}$ be LorentzVectors.
Also, let
$\vec{v}_i, t_i$ be the space vector and time components of ${\bf w}_{i}$.
Then using the boost vector to the joint center of mass frame
\begin{equation}
\vec{b} = - \frac { \left| \vec{v}_1 + \vec{v}_2 \right| } { t_1 + t_2 }
\end{equation}
\noindent
and assuming that $|\vec{b}| < 1$ so we can take
\begin{eqnarray}
\beta = |\vec{b}| \\
\gamma = \frac{1} {\sqrt {1-\beta^2 } }
\end{eqnarray}
\noindent
we define the condition
\begin{eqnarray}
\label{eq:wisNearCM}
w_1 \stackrel{\mbox{\tiny CM}}{\approx} w_2 \Longleftrightarrow
\left( w_1\mbox{.boost}(\vec{b}) \right) \approx
\left( w_2\mbox{.boost}(\vec{b}) \right)
\end{eqnarray}
As mentioned earlier, this criterion makes sense only for timelike 4-vectors.
If applied to 4-vectors whose sum is not timelike, the {\tt isNearCM()}
method will return a test for exact equality.
\begin{eqnarray}
\label{eq:wisNearCM:2}
\mbox{if } | \vec{v}_1 + \vec{v}_1 | \geq | t_1 + t_2 | \mbox{ then }
w_1 \stackrel{\mbox{\tiny CM}}{\approx} w|2 \Longleftrightarrow w_1 = v_2
\end{eqnarray}
\subsubsection{DeltaR for \protect\LV s}
Another method to compare two \LV s is {\tt w1.deltaR(w2)}, which acts only
on the 3-vector components of the \LV s and applies {\tt deltaR} as defined
by equation (\ref{eq:deltaR}).
\subsubsection{Ordering Comparisons for \protect\LV s}
The comparison operators \verb$ ( >, >=, <, <= )$ for \LV\ act by comparing
first the time component, then the \SV\ part. The latter comparison is done
using definition (\ref{eq:compSV}).
\begin{equation}
\label{eq:wcomp}
w_1 > w_2 \mbox{ if }
t_1 > t_2 \mbox { or } \left[
t_1 = t_2 \mbox { and }
\vec{v}_1 > \vec{v}_2 \right]
\end{equation}
\subsection{Other Boolean Methods for \protect\LV s}
The {\tt w1.isParallel(w2)} method works with a relative tolerance.
If the difference of the normalized 4-vectors is small,
then those 4-vectors are considered nearly parallel.
For this purpose, we use the Euclidean norm to define the normalization and
size of difference.
Let
\begin{eqnarray}
\overline{w}_1 \equiv \frac{w_1}{|\vec{v}_1|^2 + t_1^2} \nonumber \\
\overline{w}_2 \equiv \frac{w_1}{|\vec{v}_2|^2 + t_2^2} \nonumber
\end{eqnarray}
\noindent
and
\begin{eqnarray}
\overline{w}_1 - \overline{w_2} \equiv
( \overline{v}_{12} , \overline{t}_{12} ) \nonumber
\end{eqnarray}
\noindent
then
\begin{equation}
\label{eq:wisPar}
w_1 \parallel w_2 \mbox{ iff }
\left| \overline{v}_{12} \right| ^2 + \overline{t}_{12}^2 \leq \epsilon^2
\end{equation}
As in the case of \SV s, only the zero 4-vector is considered parallel to
the zero 4-vector.
\begin{equation}
\label{eq:wisPar:2}
w \parallel (0, 0, 0, 0) \mbox{ iff } w = (0, 0, 0, 0)
\end{equation}
The {\tt w1.howParallel(w2)} method, applied to two non-zero \LV s,
returns the Euclidean norm of the differnce. This can range from zero
(if {\tt w2} is a positive multiple of {\tt w1}) to
2 (if {\tt w2} is a negative multiple of {\tt w1}). If both \LV s are
zero, {\tt w1.howParallel(w2)} returns zero; if one is zero, it returns 1.
\vspace{.25 in}
The boolean tests {\tt isSpacelike()}, {\tt isTimelike()},
and {\tt isLightlike()},
work with the {\tt restMass2()} function, which returns $t^2 - |\vec{v}|^2$.
\begin{eqnarray}
\label{eq:wisSl}
w\mbox{.isSpacelike}() \Longleftrightarrow t^2 - |\vec{v}|^2 < 0 \\
\label{eq:wisTl}
w\mbox{.isTimelike}() \Longleftrightarrow t^2 - |\vec{v}|^2 > 0
\end{eqnarray}
The test for {\tt isLightlike()}
uses a tolerance relative to the time component of the vector.
It determines if, starting from an exactly lightlike vector, you can perturb
$t$ and $v$ by a small relative amount to reach the actual
vector.
\begin{equation}
\label{eq:wisLl}
w\mbox{.isLightlike}() \Longleftrightarrow
\left| t^2 - |\vec{v}|^2 \right| <= 2 \epsilon t^2
\end{equation}
\noindent
The $2 \epsilon t^2$ limit is chosen such that $(0, 0, (1\pm\epsilon)t, t)$
is just on the boundary of being considered lightlike.
Since the {\tt isLightlike()} method is tolerant of small perturbations,
these three methods are not mutually exclusive: A 4-vector can test true for
{\tt isLightlike()}
and either {\tt isSpacelike()} or {\tt isTimelike()} as well.
By these definitions, the zero 4-vector is considered lightlike.
\subsection{Nearness measures for \protect\LV s}
Since both {\tt isNear()} and {\tt isNearCM()} for \LV s use relative tolerance,
the corresponding nearness measures are truncated at a maximum of 1:
\begin{eqnarray}
\label{eq:whowNear}
w_1 \mbox{.howNear} (w_2) = \max \left( \sqrt { \frac
{\left| \vec{v}_1 - \vec{v}_2 \right| ^2 + (t_1 - t_2)^2 }
{ \left| \vec{v}_1 \cdot \vec{v}_2 \right|
+ \left( \frac{t_1 + t_2}{2} \right)^2 }
} \;, \; 1 \right)
\\
\label{eq:whowNearCM}
w_1 \mbox{.howNearCM} (w_2) =
\left( w_1\mbox{.boost}(\vec{b}) \right)\mbox{.howNear}
\left( w_1\mbox{.boost}(\vec{b}) \right)
\end{eqnarray}
\noindent
with, as before, the boost vector $\vec{b}$ given by
\[
\vec{b} = - \frac { \left| \vec{v}_1 + \vec{v}_2 \right| } { t_1 + t_2 }
\]
\noindent
For two unequal \LV s,
{\tt w1.howNearCM(w2)} will return 1
if the boost to the rest frame is tachyonic.
The {\tt w1.isParallel(w2)}
and {\tt w.isLightlike()}
methods also work with a relative tolerance.
The corresponding measures are defined by:
\begin{eqnarray}
\label{eq:whowPar}
w_1\mbox{.howParallel}(w_2) = \max \left( \sqrt {
\left| \overline{v}_{12} \right| ^2 +
\overline{t}_{12}^2 } \; , \; 1 \right)
\end{eqnarray}
\noindent
(where the notation used is that use for equation \ref{eq:wisPar}).
\noindent
If $w = (v,t)$
\begin{eqnarray}
\label{eq:whowLl}
w\mbox{.howLightlike}() = \max \left(
\frac { \left| t^2 - |\vec{v}|^2 \right| } { 2 t^2 }
\; , \; 1 \right)
\end{eqnarray}
\noindent
As before, if this measure is very nearly zero, the 4-vectors are nearly
parallel or the 4-vector is nearly lightlike.
When one of the 4-vectors (but not the other) is zero,
\begin{eqnarray}
\label{eq:whowNear:2}
w \mbox{.howNear}(\mbox{{\it 0}}) =
\mbox{{\it 0}} \mbox{.howNear} (w) = 1
\\
\label{eq:whowNearCM:2}
w \mbox{.howNearCM}(\mbox{{\it 0}}) =
\mbox{{\it 0}} \mbox{.howNearCM} (w) = 1
\\
\label{eq:whowPar:2}
w \mbox{.howParallel}(\mbox{{\it 0}}) =
\mbox{{\it 0}} \mbox{.howParallel} (w) = 1
\\
\label{eq:whowLl:2}
\mbox{{\it 0}} \mbox{.howLightlike()} = 0
\end{eqnarray}
When both 4-vectors are zero, the nearness measures will all return zero.
\section {Pseudorapidity, Rapidity and CoLinearRapidity}
\label{rapidity}
\noindent
Pseudorapidity (conventionally labelled $\eta$) and rapidity are properties
which apply to both \SV s and \LV s.
Pseudorapidity and rapidity are defined relative to the $z$ direction.
The pseudorapidity {\tt eta} of either a \LV\ or a \SV\ is defined in terms of
the angle the 3-vector part forms with the Z axis:
This can be found knowing nothing of the mass of a particle, and
in the limit of large momentum is approximately the same as the true rapidity.
Unlike the case for rapidity(), eta() makes mathematical sense for any
vector---it a simple function of the tangent of the angle between the vector and
the Z axis.
The pseudorapidity of a zero vector will be assigned the value zero.
Let $\vec{v_1}$ be a \SV, and $w_1$ be a \LV\ with decomposition
$(t, \vec{v})$.
And let the angle formed between the Z axis and
$\vec{v_1}$ or $\vec{v}$ be $\theta$. Then:
\begin{eqnarray}
\label{eq:eta:3}
v_1 \mbox{.eta()} = - \ln \tan \frac{\theta}{2} \\
\label{eq:weta:3}
w_1 \mbox{.eta()} = - \ln \tan \frac{\theta}{2}
\end{eqnarray}
\noindent
The true rapidity of a LorentzVector is defined (see the Kinematics section of
the Review of Particle Properties, page 177 in the 1997 version) such that
the rapidity transforms under a boost {\em along the Z axis}
by adding the rapidity of the boost:
This treats the LorentzVector as a timelike 4-momentum
(using the $t$ and $z$ components as $E$ and $p_z$).
In analogy with the familiar definition for \LV, the rapidity of a \SV\ is
defined with respect to the $z$ direction.
Mathematically, this is the same as the rapidity of a 4-vector
$(\vec{v}, 1)$.
Letting $w_1$ be $(x, y, z, t)$ or $(p_x, p_y, p_z, E)$,
\begin{eqnarray}
\label{eq:wrapid:3}
w_1 \mbox{.rapidity() }
= \frac{1}{2} \ln \left( \frac{E + p_z}{E - p_z} \right)
= \tanh^{-1} \frac {z}{t} \\
v_1 \mbox{.rapidity() }
= \tanh^{-1} (\vec{v}_1 \cdot \hat{z})
\end{eqnarray}
\noindent
The shape of a rapidity distribution is a invariant under boosts in the $z$
direction.
Pseudorapidity can always be determined from direction information alone,
and when $ p^2 \gg m^2 $ {\em and the time component is positive}
pseudorapidity matches rapidity to order $m^2$/$p^2$.
\vspace{.2 in}
Although the $z$ direction is special for many HEP uses,
rapidity and pseudorapidity can be defined with respect to an arbitrary
direction $\hat{u}$. Letting $w_1$ be $(\vec{v}, t)$:
\begin{eqnarray}
\label{eq:weta:4}
w_1 \mbox{.eta(u)} = - \ln \tan \frac{\hat{p} \cdot {u}}{2} \\
\label{eq:eta:4}
v_1 \mbox{.eta(u)} = - \ln \tan \frac{\hat{v}_1 \cdot \hat{u}} {2} \\
\label{eq:wrapid:4}
w_1 \mbox{.rapidity(u) }
= \frac{1}{2} \ln \left( \frac{E + \vec{p} \cdot \hat{u}}
{E - \vec{p} \cdot \hat{u}} \right)
= \tanh^{-1} \frac {\vec{v} \cdot \hat{u}} {t} \\
v_1 \mbox{.rapidity(u) } = \tanh^{-1} (\vec{v}_{1} \cdot \hat{u})
\end{eqnarray}
\noindent
Relativity texts discuss rapidity along the direction of the vector.
This concept can apply to \SV s or \LV s.
This function adds when you compbine two boosts in the same direction.
Such a method may not be needed for typical HEP calculations but
is provided for completeness.
To distinguish this from the rapidity with respect to a specific
direction---or a default rapidity, which is with respect to $\hat{z}$---the
package names the rapidity along the direction of the vector
{\tt coLinearRapidity()}.
\begin{eqnarray}
v_1 \mbox{.coLinearRapidity() } = \tanh^{-1} \beta
= \tanh^{-1} \left|\vec{v_1}\right| \\
\label{eq:wcoLinRap:2}
w_1 \mbox{.coLinearRapidity() }
= \frac{1}{2} \ln \left( \frac{E + |p|}{E - |p|} \right)
= \tanh^{-1} \frac {|v|}{t}
\end{eqnarray}
\noindent
The co-linear rapidity of a 3-vector is inherently non-negative; for a
4-vector it will have the same sign as the time component of the 4-vector.
\section{\protect\Ro\ Class}
\Ro s may be expressed in terms of an axis $\hat{u}$
and angle $\delta$ of counter-clockwise
rotation, or as a set of three Euler Angles $(\phi, \theta, \psi)$.
Definitions and conventions for these classes
match those described in \S \ref{rotations}
for directly rotating a \SV:
\[ \mbox{HepRotation}(\hat{u},\delta) \Longrightarrow \]
\begin{equation}
\label{eq:axisrotR}
\left(
\begin{array}{ccc}
\cos \delta + (1 - \cos \delta ) u_x^2 &
(1 - \cos \delta ) u_x u_y - \sin \delta u_z &
(1 - \cos \delta ) u_x u_z + \sin \delta u_y \\
(1 - \cos \delta ) u_y u_x + \sin \delta u_z &
\cos \delta + (1 - \cos \delta ) u_y^2 &
(1 - \cos \delta ) u_y u_z - \sin \delta u_x \\
(1 - \cos \delta ) u_z u_x - \sin \delta u_y &
(1 - \cos \delta ) u_z u_y + \sin \delta u_x &
\cos \delta + (1 - \cos \delta ) u_z^2
\end{array}
\right)
\end{equation}
\[ \mbox{HepRotation}(\phi, \theta, \psi) \Longrightarrow \]
\begin{equation}
\label{eq:eulerrotR}
\left(
\begin{array}{ccc}
\cos \psi \cos \phi - \sin \psi \cos \theta \sin \phi &
\cos \psi \sin \phi + \sin \psi \cos \theta \cos \phi &
\sin \psi \sin \theta \\
- \sin \psi \cos \phi - \cos \psi \cos \theta \sin \phi &
- \sin \psi \sin \phi + \cos \psi \cos \theta \cos \phi &
\cos \psi \sin \theta \\
\sin \theta \sin \phi &
- \sin \theta \cos \phi &
\cos \theta
\end{array}
\right)
\end{equation}
\noindent
The Euler angles definition matches that found in found in
{\em Classical Mechanics} (Goldstein), page 109.
This treats the Euler angles as a sequence of counter-clockwise {\bf passive}
rotations;
that is, the vector remains fixed while the coordinate axes are rotated---new
vector components are computed in new coordinate frame.
HEP computations ordinarily use the active rotation viewpoint.
Therefore, rotations about an axis imply {\bf active} counter-clockwise
rotation in this package.
Consequently, a rotation by angle $\delta$ around the X axis is
equivalent to a rotation with Euler angles
$(\phi=\psi=0, \mbox{ } \theta = - \delta)$
and a rotation about the Z axis is
equivalent to a rotation with Euler angles
$(\theta = 0, \mbox{ } \phi=\psi= - \delta/2)$.
\subsection{Applying Rotations to Vectors and 4-Vectors}
A \Ro may be applied to a \SV using either of two notations:
\begin{equation}
\label{eq:Rv}
\mbox{R * v} \equiv \mbox{R(v)} \equiv {\boldmath R} \vec{v}
\end{equation}
\noindent
where ${\boldmath R}$ is the matrix representing R, as in equation
\ref{eq:eulerrotR} or \ref{eq:axisrotR}. The same syntaxes may be
used to apply a \Ro to a \LV ---the rotation matrix acts on the space
components of the 4-vector.
Note that {\tt R *= v} is meaningless and not supported.
Also note the warning given earlier:
$ \vec{v} \mbox{*=} R \Longleftrightarrow \vec{v} = R \mbox{*} \vec{v} $
\subsection{Axial Rotations}
The special case rotations along the X, Y, and Z axes will often be specified
by just the rotation angle $\delta$. Thus a
{\tt HepRotationX}, {\tt HepRotationY}, or {\tt HepRotationZ}
with angle $\delta$ would be represented by the matrix
\begin{eqnarray}
\label{eq:rotx}
\mbox{HepRotationX}(\delta) =
\left(
\begin{array}{ccc}
1 & 0 & 0 \\
0 & \cos \delta & - \sin \delta \\
0 & \sin \delta & \cos \delta
\end{array}
\right) \\
\label{eq:roty}
\mbox{HepRotationY}(\delta) =
\left(
\begin{array}{ccc}
\cos \delta & 0 & \sin \delta \\
0 & 1 & 0 \\
- \sin \delta & 0 & \cos \delta
\end{array}
\right) \\
\label{eq:rotz}
\mbox{HepRotationZ}(\delta) =
\left(
\begin{array}{ccc}
\cos \delta & - \sin \delta & 0 \\
\sin \delta & \cos \delta & 0 \\
0 & 0 & 1
\end{array}
\right) \\
\end{eqnarray}
A \Ro\ is an object in its own right; two \Ro s may be multiplied,
tested for near equality, and so forth.
Multiplication is done by multiplying the matrix representations of two
\Ro s (though for matching axis rotations, simplifications are
done to effciently compute this product).
The matrix representing a \Ro\ is orthonormal: Each row, considered
as a vector, has length 1, and the dot product of any two distinct rows
is zero. This leads to a trivial inversion method---simply transpose the
matrix.
When dealing with structures holding an axis and angle, or Euler angles,
we do not provide direct analogues of all rotation methods. For example,
we do not provide for multiplication of two \Ax objects to form a third
\Ax. We do provide comparison and nearness methods for these classes,
and this document specifies whether the criteria match the corresponding
\Rotation\ criteria.
\subsection{Expressing a \protect\Ro\ as \protect\Ax\ or \protect\Es\ }
It is worth noting that equations (\ref{eq:axisrot}) and (\ref{eq:eulerrot})
are not single-valued when extracting $ ( \hat{u}, \delta) $ or
$ ( \phi, \theta, \psi ) $ from a \Rotation\ matrix. We adhere to the
following conventions to resolve that ambiguity whenever forming an
\Ax\ from a \Rotation\ $R$:
\begin{eqnarray}
R = {\bf I} \Longrightarrow \delta = 0, \; \hat{u} = \hat{z}
\\
\label{eq:deltaconv}
R \longrightarrow ( \hat{u}, \delta ) \Longrightarrow 0 \leq \delta \leq \pi
\end{eqnarray}
And we adhere to the
following conventions to resolve that ambiguity whenever forming an
\Es\ from a \Rotation\ $R$:
\begin{eqnarray}
\label{eq:thetaconv}
R \longrightarrow ( \phi, \theta, \psi )
\Longrightarrow 0 \leq \theta \leq \pi
\\
\label{eq:tphiconv}
R \longrightarrow ( \phi, \theta, \psi )
\Longrightarrow -\pi < \phi \leq \pi
\\
\label{eq:psiconv}
R \longrightarrow ( \phi, \theta, \psi )
\Longrightarrow -\pi < \psi \leq \pi
\\
\label{eq:phipsiconv}
R \longrightarrow ( \phi, \; \theta = 0, \; \psi )
\Longrightarrow -\pi/2 < \phi = \psi \leq \pi/2
\\
R \longrightarrow ( \phi, \; \theta = \pi, \; \psi )
\Longrightarrow -\pi/2 < \phi = -\psi \leq \pi/2
\end{eqnarray}
\noindent
Thus when supplying values for Euler angles,
we return $(0, 0, 0)$ whenever the rotation is
equivalent to the identity, and return $(0, \pi, 0)$ in preference to
the equivalent $(\pi, \pi, \pi)$.
In particular, special case rotations such as
{\tt HepRotationX} with $\delta$ supplied as zero or $\pi$
obey these conventions: even though for $\delta = \pi - |\epsilon|$
a {\tt RotationX} would have Euler angles $( \pi, \delta, \pi, )$,
when $\delta = \pi$ exactly, the Euler angles are $(0, \delta=\pi, 0)$.
These conventions for how methods return values for Euler angles
do not affect the user's right to supply explicit values for
or $ ( \phi, \theta, \psi ) $ which may not obey the conventions,
when defining an \Es\ structure or a \Rotation--it is just that when
such a \Rotation\ is read back as Euler angles, the values will not be
the ones originally supplied.
Similarly, the conventions for reading axis and angle do not affect the
user's ability to supply arbitrary $ ( \hat{u}, \delta) $ values.
Obeying the above rules, Euler angles returned for
rotations about coordinate axes behave as follows:
\begin{equation}
\mbox{HepRotationX}(\delta) \longrightarrow
\left\{
\begin{array}{ccc}
( 0, -\delta, 0 ) & \mbox{ if } & \delta \leq 0 \\
( \pi, \delta, \pi ) & \mbox{ if } & 0 < \delta < \pi \\
( 0, \pi, 0 ) & \mbox{ if } & \delta = \pi
\end{array}
\right.
\label{eq:rotXconv}
\end{equation}
\begin{equation}
\mbox{HepRotationY}(\delta) \longrightarrow
\left\{
\begin{array}{ccc}
( +\pi/2, -\delta, -\pi/2 ) & \mbox{ if } & \delta < 0 \\
( 0, 0, 0) & \mbox{ if } & \delta = 0 \\
( -\pi/2, \delta, +\pi/2 ) & \mbox{ if } & 0 < \delta < \pi \\
( \pi/2, \delta, -\pi/2 ) & \mbox{ if } & \delta = \pi
\end{array}
\right.
\label{eq:rotYconv}
\end{equation}
\begin{equation}
\mbox{HepRotationZ}(\delta) \longrightarrow ( -\delta/2, 0, -\delta/2 )
\label{eq:rotZconv}
\end{equation}
\noindent always assuming that $\delta$ represents an angle of active rotation
between $-\pi$ and $\pi$.
\subsection{Nearness Measure for \protect\Ro s, \protect\Ax s, and \protect\Es\ }
The definition used for {\tt isNear()} and {\tt howNear()} on \Rotation s
has the following properties:
\begin{enumerate}
\item {\tt isNear()} and {\tt howNear()} use the same measure: Two rotations
are considered near if their {\tt howNear()} measure is less that $\epsilon$.
\item Transitivity: $r_1 \approx r_2$ and {\tt r1.howNear(r2)}
are exactly equivalent to $r_2 \approx r_1$ and {\tt r2.howNear(r1)}.
\item Rotational invariance to order $\epsilon$:
If $x = ${\tt r1.howNear(r2)} is small, then for any third rotation {\tt r3}
{\tt (r3*r1).howNear(r3*r2)} $ x + O(x^2)$.
\item For the special case of rotations that happen to be around the same axis,
the measure agrees with a natural definition for measure of rotation about an
axis (\S\ref{rotsame}).
\end{enumerate}
Although it is possible to formulate a measure definition with exact rotational
in\-var\-i\-ance---based on $
\sup_{\hat{u}} \left\{
\left| R_1(\hat{u}) - R_2(\hat{u}) \right|
\right\}
$---this definition would require finding the 2-norm (or the largest eigenvalue)
of the matrix $r_1 r_2^{-1}$, which is computationally difficult.
The measure we use is the same for small answers, and has the above desirable
properties.
\begin{eqnarray}
\label{eq:nearrot}
r_1 \mbox{.howNear}(r_2) = \sqrt{ 3 - \mbox{Tr}(r_1 r_2^{-1}) }
= \sqrt {3 - \sum_{ij} r_{1_{ij}} r_{2_{ij}} }
\\
\label{eq:dist2rot}
r_1 \mbox{.distance2}(r_2) = 3 - \mbox{Tr}(r_1 r_2^{-1})
= 3 - \sum_{ij} r_{1_{ij}} r_{2_{ij}}
\\
r_1 \approx r_2 \Longleftrightarrow 3 - \mbox{Tr}(r_1 r_2^{-1})
\leq \epsilon^2
\end{eqnarray}
And a norm is provided:
\begin{eqnarray}
\label{eq:norm2rot}
r_1 \mbox{.norm2}(~) = 3 - \mbox{Tr}(r_1)
\end{eqnarray}
\Ax s have the property that two apparently unequal forms may be equivalent,
for example if the axes are in opposite directions and one angle is the
negative of the other.
Similarly, two different-looking \Es\ can represent the same
rotation. To avoid a whole spectrum of special cases, we adapt the rule that,
letting $\Upsilon$ be an \Ax and $\Xi$ be an \Es\ structure, and
$R(\Upsilon), R(\Xi)$ be the corresponding \Ro s,
\begin{eqnarray}
\Upsilon_1 \mbox{.howNear} (\Upsilon_2) \equiv
R \left( \Upsilon_1 \right) \mbox{.howNear}
\left( R \left( \Upsilon_2 \right) \right)
\\
\Xi_1 \mbox{.howNear} (\Xi_2) \equiv
R \left( \Xi_1 \right) \mbox{.howNear}
\left( R \left( \Xi_2 \right) \right)
\end{eqnarray}
Since for double precision computations these nearness measures cannot
count on precision to better than $10^{-8}$,
the default tolerance for Rotations
(which may be modified by the class static method {\tt setTolerance()})
is set to 100 times that.
\begin{equation}
\epsilon_{\mbox{default}} = 10^{-6}
\label{eq:epsildefR}
\end{equation}
\subsubsection{Nearness for \protect\Ro s About the Same Axis}
\label{rotsame}
The above definition of {\tt howNear()} for two general \Rotation s reduces,
when both \Rotation s are around the same axis by angles $\delta_1$ and
$\delta_2$, to
\begin{equation}
r_1 \mbox{.howNear} (r_2) = \sqrt{ 2 - 2 \cos (\delta_1 - \delta_2) }
= | \delta_1 - \delta_2 | + O \left( (\delta_1 - \delta_2)^3 \right)
\end{equation}
\noindent
(Note that when the two rotations are special-case coordinate axis rotations,
computing the cosine of that angle difference is trivial, given
that the structures already hold the sine and cosine of $\delta_i$.)
Although equivalent for most small angular differences to the simpler concept
of $ | \delta_1 - \delta_2 | $, the definition above also properly handles the
case where one $\delta$ is near $\pi$ and the other near $-\pi$.
\subsection{Comparison for \protect\Ro s, \protect\Ax s, and \protect\Es\ }
It is useful to have definitions of the various comparison operators
so that \Ro s, \Ax s, and \Es\ can be placed into {\tt std::} containers.
Of the three classes, \Ax\ has a natural meaning for ordering comparisons,
taking advantage of the ordering relation already available for the \UV\
axes:
\begin{equation}
\label{eq:ordrot}
( \hat{u}_1 , \delta_1 ) > ( \hat{u}_2 , \delta_2 ) \mbox{ if }
\hat{u}_1 > \hat{u}_2 \mbox { or } \left[
\hat{u}_1 = \hat{u}_2 \mbox { and }
\delta_1 > \delta_2 \right]
\end{equation}
For \Rotation, we could use the ordering induced by its \Ax\ expression;
extracting the \Ax\ corresponding to a \Ro\ is a fairly simple task.
But that is unnecessarily complex--instead, we use dictionary ordering,
starting with $zz, zy, zx, yz, \ldots, xx$. This agrees with the definition
used in the orignal CLHEP Vector package.
For \Es, rather than laboriously going over to \Ro s and then using the
induced dictionary ordering comparison, we adapt simple dictionary ordering:
\begin{eqnarray}
( \phi_1 , \theta_1, \psi_1 ) > ( \phi_2 , \theta_2, \psi_2 ) \mbox{ if }
\nonumber \\
\phi_1 > \phi_2 \mbox { or }
\nonumber \\
\left[
\phi_1 = \phi_2 \mbox { and } \theta_1 > \theta_2 \right]
\mbox { or }
\nonumber \\
\left[
\phi_1 = \phi_2 \mbox { and } \theta_1 = \theta_2
\mbox { and } \psi_1 > \psi_2
\right]
\end{eqnarray}
Because we use dictionary ordering comparisons, one \Ro\ may be considered
greater than another, but if you take may Euler angles (or for that matter
their \Ax\ structures), the comparison may have the opposite sense.
\subsection{The Rotation Group}
Inversion of a \Ro\ is supported. Since the matrix is orthogonal,
the inverse matches the transpose:
\[
R\mbox{.inverse()} \equiv R^{-1} = R^T
\]
Multiplication is available in three syntaxes:
\begin{eqnarray}
\label{eq:opmulrot}
\mbox{ R = R1 * R2} \Longrightarrow {\boldmath R = R_1 R_2}
\label{eq:opstrot}
\\
\mbox{ R *= R1 } \Longrightarrow {\boldmath R = R R_1 }
\\
\label{eq:transrot}
\mbox{ R.transform(R1) } \Longrightarrow {\boldmath R = R_1 R }
\end{eqnarray}
To complete the group concept, the identity \Ro\ can easily be obtained; the
default constructor for \Ro\ gives the identity.
Two sorts of specialized transformations are available. The first transforms
by a rotation around an axis:
\begin{eqnarray}
\label{eq:xyzrot}
R\mbox{.rotateX}(\delta) \Longrightarrow R = \mbox{RotationX}(\delta) R \\
\nonumber
R\mbox{.rotateY}(\delta) \Longrightarrow R = \mbox{RotationY}(\delta) R \\
\nonumber
R\mbox{.rotateZ}(\delta) \Longrightarrow R = \mbox{RotationZ}(\delta) R
\end{eqnarray}
\noindent
The other specialized transformation rotates such that the original X axis
becomes a specified new X axis, and similarly for the Y and Z axes.
In the specified new axes are labeled $\vec{X}^\prime$, $\vec{Y}^\prime$,
$\vec{Z}^\prime$, this transformation is equivalent to:
\begin{equation}
\label{eq:rotaxes}
\mbox{R.rotateAxes}(X^\prime, Y^\prime, Z^\prime) \Longrightarrow
{\boldmath R} =
\left(
\begin{array}{ccc}
X^\prime_x & Y^\prime_x & Z^\prime_x \\
X^\prime_y & Y^\prime_y & Z^\prime_y \\
X^\prime_z & Y^\prime_z & Z^\prime_z
\end{array}
\right)
{\boldmath R}
\end{equation}
\noindent
The supplied
$\vec{X}^\prime$, $\vec{Y}^\prime$ and $\vec{Z}^\prime$ must be orthonormal;
no checking is done, and if the supplied new axes are not orthonormal, the
result will be an ill-formed (non-orthogonal) rotation matrix.
\subsection{Rectifying Rotations}
The operations on \Ro s are such that mathematically, the orthonormality of
the representation is always preserved. And methods take advantage of this
property.
However, a long series of operations could, due to round-off, produce a
\Ro\ object with a representation that slightly deviates from the
mathematical
ideal. This deviation can be repaired by extracting the axis and delta
for the \Ro, and freshly setting the axis and delta to those values.
\begin{equation}
\label{eq:rectRot}
\mbox{R.rectify()} \rightarrow \mbox{R.set (R.axis(), R.delta())}
\end{equation}
A technical point: If the rotation has strayed significantly from a true
orthonormal matrix, then extracting the axis is not necessarily an accurate
process. To minimze such effects, before performing the formal algorithm
to extract the axis, the rectify() method averages the purported rotation
with the transpose of its inverse. (A true rotaion is identical to the
transpose of its inverse). This in principle eliminates errors to lowest
order.
\section{\protect\LT\ Class}
A \LT\ may be
expressed in terms of a \Rotation\ in the space sector, followed by
a pure \LB\ along some direction.
Alternatively, it may be expressed as a pure boost followed by a rotation.
In any event, just as for rotations,
we use the convention of active transformations changing 4-vectors
(rather than transformations of a reference frame). So a boost by
$\beta$ along the in the X direction, for example, would be represented by
the matrix
\begin{equation}
\label{eq:boostx}
\mbox{LorentzBoost}(\beta) =
\left(
\begin{array}{cccc}
\gamma & 0 & 0 & \beta \gamma \\
0 & 1 & 0 & 0 \\
0 & 0 & 1 & 0 \\
\beta \gamma & 0 & 0 & \gamma
\end{array}
\right)
\end{equation}
A \LT\ is an object in its own right; two \LT s may be multiplied,
tested for near equality, and so forth.
Multiplication is done by multiplying the matrix representations of two
\LT s.
The matrix representing a \LT\ is orthosymplectic: The last (T) row,
constdered as a 4-vector, has $ t^2 - |\vec{v}|^2 = 1 $, each of rows
X, Y, and Z have $ t^2 - |\vec{v}|^2 = -1$, and the Minkowski space
dot product of any two distinct rows
is zero. This leads to a trivial inversion method---simply transpose the
matrix and negate any elment with just one of its two indices refering to
a space direction.
\subsection{Pure Lorentz Boosts}
Definitions and
and conventions for a pure boost classes match those described in equation
(\ref{eq:pureboost}).
Here we will write out the matrix in detail, for a boost specified by
$\vec{b}$, a \SV\ with magnitude $0 < \beta < 1$ (here we write
$\vec{b} = \beta \hat{u} $ and
$\gamma \equiv \frac{1}{\sqrt{1-\beta^2}}$):
\begin{eqnarray}
\label{eq:boostform}
\mbox{LorentzBoost}(\beta \hat{u}) =
\left(
\begin{array}{cccc}
(\gamma - 1) u_x^2 + 1 & (\gamma - 1) u_x u_y &
(\gamma - 1) u_x u_z & \beta \gamma u_x
\\
(\gamma - 1) u_y u_x & (\gamma - 1) u_y^2 + 1 &
(\gamma - 1) u_y u_z & \beta \gamma u_y
\\
(\gamma - 1) u_z u_x & (\gamma - 1) u_z u_y &
(\gamma - 1) u_z^2 + 1 & \beta \gamma u_z
\\
\beta \gamma u_x & \beta \gamma u_y &
\beta \gamma u_z & \gamma
\end{array}
\right)
\end{eqnarray}
\subsubsection{isNear() and howNear() for LorentzBoost}
Though a pure boosts may be specified by a \SV\ of magnitude less than one,
we do not define a nearness measure as the (relative) measure that induced
by those \SV s. This is because two \LB s in the same direction
with large but quite different
values of $\gamma$ should be viewed as quite different transformations.
The \SV s representing two such boosts would both be very close to the same
unit vector. That is, (.9999, 0, 0) and (.999999, 0, 0) are not very similar:
The former would boost a muon to 10 GeV, while the latter would boost it to
200 GeV; yet as \SV s, (.9999, 0, 0) is equal to (.999999, 0, 0) within
a relative tolerance $10^{-4}$.
LorentzBoosts have a natural scale, set by the speed of light, so absolute
rather than relative tolerances are appropriate.
The correct way to measure nearness is to find the difference between the
values of $\gamma b$. If $\vec{\beta}_1$ and
$\vec{\beta}_2$ specify two pure boosts
$B_1$ and $B_2$
and $\gamma_i \equiv \frac{1}{\sqrt{1 - |\vec{\beta}_i|^2}}$ then
\begin{eqnarray}
\label{eq:nearboost}
B_1 \mbox{.isNear} (B_2) \Longleftrightarrow
\left| \gamma_1 \vec{\beta}_1 - \gamma_2 \vec{\beta}_2 \right|^2 \leq
\epsilon^2
\\
\label{eq:hownearboost}
B_1 \mbox{.howNear} (B_2) =
\left| \gamma_1 \vec{\beta}_1 - \gamma_2 \vec{\beta}_2 \right|
\\
B_1 \mbox{.distance2} (B_2) =
\left| \gamma_1 \vec{\beta}_1 - \gamma_2 \vec{\beta}_2 \right|^2
\\
\label{eq:boostnorm2}
B\mbox{.norm2}( ) \equiv B\mbox{.distance2} (I)
= \gamma^2 \beta^2 = 1 - \gamma^2
\end{eqnarray}
\subsubsection{Ordering Comparisons of LorentzBoosts}
Any pure boost can be viewed as having a non-negative $\beta$, and s
direction.
For pure boosts in the same direction both with positive $\beta$,
we wish to order LorentzBoosts according $\beta$:
\begin{equation}
\label{ordboost}
\mbox{LorentzBoost}( \beta_1 > 0, \hat{u} ) >
\mbox{LorentzBoost}( \beta_2 > 0, \hat{u} ) \Longleftrightarrow
\beta_1 > \beta_2
\end{equation}
For pure boosts which may be in different directions,
we use a comparison
condition induced by the \SV s specifying the two boosts.
But in order to match the above ordering for identical directions, we must
reverse the sense of the naive induced ordering when both vectors are
negative. That is, we want to say that the boost (-.2, -.2, -.2) is
``more than'' the boost (-.1, -.1, -.1).
\begin{eqnarray}
\mbox{LorentzBoost}( \vec{\beta}_1 ) > \mbox{LorentzBoost}( \vec{\beta}_2 )
\mbox{ if }
\left\{
\begin{array}{ccc}
( \vec{\beta}_1 \geq \vec{0} & \mbox{ and } &
\vec{\beta}_1 > \vec{\beta}_2 ) \\
& \mbox{or} \\
( \vec{\beta}_2 < \vec{0} & \mbox{ and } &
\vec{\beta}_1 < \vec{\beta}_2 )
\end{array}
\right.
\end{eqnarray}
\subsection{Components of HepLorentzRotations}
Beyond the obvious methods returning single components, such as
{\tt lt.yt()} there is also the method {\tt rep4x4()},
which returns a struct of type {\tt HepRep4x4}---this has publicly
visible data members {\tt xx\_, xy\_, $\ldots$, tz\_, tt\_}. These
methods may also be used for pure {\tt HepBoost} boosts. For those
classes, there is also a method {\tt rep4x4Symmetric()}, which returns
a ten-element struct {\tt HepRep4x4Symmetric}.
For \LT, there are also indexing methods with syntax
{\tt lt[i][j]} and {\tt lt(i,j)}.
In this syntax, the indices range from 0 to 3, with the time component last.
That is, 0 refers to an X component, 1 to Y, 2 to Z, and 3 to T.
If R is the four by four representation of a \LT\ L, then:
\begin{eqnarray}
\label{eq:ltsubscript}
\begin{array}{llll}
L [0] [0] = R_{xx} & L [0] [1] = R_{xy} &
L [0] [2] = R_{xz} & L [0] [3] = R_{xt} \\
\nonumber
L [1] [0] = R_{yx} & L [1] [1] = R_{yy} &
L [1] [2] = R_{yz} & L [1] [3] = R_{yt} \\
\nonumber
L [2] [0] = R_{zx} & L [2] [1] = R_{zy} &
L [2] [2] = R_{zz} & L [2] [3] = R_{zt} \\
\nonumber
L [3] [0] = R_{tx} & L [3] [1] = R_{ty} &
L [3] [2] = R_{tz} & L [3] [3] = R_{tt}
\end{array}
\end{eqnarray}
\subsection{Decomposition of Transformations into Boost and Rotation}
A \LT\ $T$ may be decomposed either into the form $ B R $ or the form
$ R B $. Here we will refer to the
first form as $T = \acute{B}(T) \grave{R}(T)$
and the second as $T = \acute{R}(T) \grave{B}(T)$.
When decomposing into the product $ B R $, the boost
will have the same last column as the \LT\ $T$. Using the value of
$\gamma$ read off $T_{tt}$, one can apply equation (\ref{eq:boostform}) to
find the matrix $\acute{B}(T)$ for that boost. Then
\begin{equation}
\label{eq:decomBR}
\grave{R}(T) = \left[ \acute{B}(T) \right] ^{-1} T
\end{equation}
When decomposing into the product $ R B $, the boost
will have the same last row as the \LT\ $T$. Again
one can apply equation (\ref{eq:boostform}) to
find the matrix (this time $\grave{B}(T)$) for that boost. Then
\begin{equation}
\label{eq:decomRB}
\acute{R}(T) = T \left[ \grave{B}(T) \right] ^{-1}
\end{equation}
Naively applying the above equations leads to non-neglible round-off errors
in the components of the Rotation---of order $3 \cdot 10^{-14}$ if the
boost has $\beta = .95$.
Since a Rotation representation which is non-orthogonal on that scale would
lead to errors in distance measures of more than one part in $10^{-7}$,
the decompose() method rectifies the Rotation before returning.
\subsection{isNear() and howNear() for \protect\LT s}
For \LT s, the analog of the the \Rotation measure (\ref{eq:nearrot}) would be
\[
T_1 \mbox{.howNear}(T_2) = \sqrt{ 4 - \mbox{Tr}(T_1 T_2^{-1}) }
\]
This, however, would be a horrible definitition of distance since that trace
can equal 4 for very different $T_1$ and $T_2$. Instead, we use a definition
which matches the definitions (\ref{eq:nearrot}) and (\ref{eq:hownearboost})
for \LT s which happen to be pure boosts or pure rotations:
\begin{eqnarray}
\label{eq:isnearLT}
\mbox{if } T_1 = B_1 R_1 \mbox{ and } T_2 = B_2 R_2 \nonumber \\
\mbox{then } T_1\mbox{.distance2}(T_2) =
B_1\mbox{.distance2}(B_2) + R_1\mbox{.distance2}(R_2)
\\
\nonumber
T_1\mbox{.howNear}(T_2) = \sqrt {T_1\mbox{.distance2}(T_2)}
\end{eqnarray}
\noindent
The {\tt isNear()} relationship uses this same measure, but when the
transformations have boosts that differ by more that $\epsilon$, {\tt isNear()}
will be quite a bit quicker since the boost portion of this formula is trivial
to compute.
And, as usual, {\tt norm2()} is the squared distance from the identity:
\begin{equation}
\label{eq:ltnorm2}
T = B R \Longrightarrow T\mbox{.norm2}() =
B.\mbox{.norm2}() + R\mbox{.norm2}()
\end{equation}
\subsection{Ordering Comparisons of \protect\LT s}
The \verb$ ( >, >=, <, <= )$ comparisons of \LT s are defined as those
induced by the decomposition $ T = \acute{B} \grave{R} $,
comparing the pure boost part
(using (\ref{ordboost})) and
then, if those are equal, comparing the rotation part (using (\ref{eq:ordrot})).
\begin{eqnarray}
\label{eq:ltorder}
T_1\mbox{.compare}(T_2) = 1 \mbox{ if }
\acute{B}_1 > \acute{B}_2 \mbox { or } \left[
\acute{B}_1 = \acute{B}_2 \mbox { and }
\grave{R}_1 > \grave{R}_2 \right] \\
\nonumber
T_1\mbox{.compare}(T_2) = -1 \mbox{ if }
\acute{B}_1 < \acute{B}_2 \mbox { or } \left[
\acute{B}_1 = \acute{B}_2 \mbox { and }
\grave{R}_1 < \grave{R}_2 \right] \\
\nonumber
T_1\mbox{.compare}(T_2) = 0 \mbox{ if }
\acute{B}_1 = \acute{B}_2 \mbox { and }
\grave{R}_1 = \grave{R}_2
\\
T_1 > T_2 \mbox{ if } T_1\mbox{.compare}(T_2) = 1 \\
\nonumber
T_1 < T_2 \mbox{ if } T_1\mbox{.compare}(T_2) = -1
\end{eqnarray}
Although \verb$==$ and \verb$!=$ comparisons could use this same comparison
algorithm, decomposition is expensive and unnecessary. Instead,
\begin{eqnarray}
\label{eq:ltexact}
T == S \Longleftrightarrow
\left\{
\begin{array}{cccc}
T_{xx} = S_{xx} \wedge & T_{xy} = S_{xy} \wedge &
T_{xz} = S_{xz} \wedge & T_{xt} = S_{xt} \wedge \\
T_{yx} = S_{yx} \wedge & T_{yy} = S_{yy} \wedge &
T_{yz} = S_{yz} \wedge & T_{yt} = S_{yt} \wedge \\
T_{zx} = S_{zx} \wedge & T_{zy} = S_{zy} \wedge &
T_{zz} = S_{zz} \wedge & T_{zt} = S_{zt} \wedge \\
T_{tx} = S_{tx} \wedge & T_{ty} = S_{ty} \wedge &
T_{tz} = S_{tz} \wedge & T_{tt} = S_{tt}
\end{array}
\right.
\end{eqnarray}
\subsection{The Lorentz Group}
Inversion of a \LT\ is supported. Since the matrix is orthosymplectic,
the inverse matches the transpose, but with the signs of all components
with mixed space and time indices reversed.
Multiplication is available in three syntaxes:
\begin{eqnarray}
\mbox{ T = T1 * T2} \Longrightarrow {\boldmath T = T_1 T_2}
\\
\mbox{ T *= T1 } \Longrightarrow {\boldmath T = T T_1 }
\\
\label{eq:lttrans}
\mbox{ T.transform(T1) } \Longrightarrow {\boldmath T = T_1 T }
\end{eqnarray}
To complete the group concept, the identity \LT\ can easily be obtained; the
default constructor for \LT\ gives the identity.
Specialized transformations based on pure rotations and pure boosts
are available.
\begin{eqnarray}
\label{eq:ltrot}
T\mbox{.rotate}(\delta, \vec{v}) \Longrightarrow T =
\mbox{Rotation}(\delta, \vec{v}) T \\
\nonumber
T\mbox{.rotateX}(\delta) \Longrightarrow T = \mbox{RotationX}(\delta) T \\
\nonumber
T\mbox{.rotateY}(\delta) \Longrightarrow T = \mbox{RotationY}(\delta) T \\
\nonumber
T\mbox{.rotateZ}(\delta) \Longrightarrow T = \mbox{RotationZ}(\delta) T
\end{eqnarray}
\begin{eqnarray}
\label{eq:ltboost}
T\mbox{.boost}(\vec{\beta}) \Longrightarrow T =
\mbox{HepBoost}(\vec{\beta}) T \\
\nonumber
T\mbox{.boost}(\beta_x, \beta_y, \beta_z) \Longrightarrow T =
\mbox{HepBoost}(\beta_x, \beta_y, \beta_z) T \\
\nonumber
T\mbox{.boostX}(\beta) \Longrightarrow T = \mbox{HepBoostX}(\beta) T \\
\nonumber
T\mbox{.boostY}(\beta) \Longrightarrow T = \mbox{HepBoostY}(\beta) T \\
\nonumber
T\mbox{.boostZ}(\beta) \Longrightarrow T = \mbox{HepBoostZ}(\beta) T
\end{eqnarray}
\subsection{Rectifying HepLorentzRotations}
The operations on \LT s are such that mathematically, the ortosymplectic
property of
the representation is always preserved. And methods take advantage of this
property.
However, a long series of operations could, due to round-off, produce a
\LT\ object with a representation that slightly deviates from the mathematical
ideal. This deviation can be repaired by extracting the boost vector based on
row 4, multiplying by the inverse of that boost to form what should ideally be
a rotation, rectifying that rotation, and setting the \LT\ ,based on that boost
and rotation.
\begin{eqnarray}
\label{eq:rectLT}
\mbox{T.rectify()} \rightarrow \nonumber \\
\vec{\beta} = \mbox {T.row4()} \nonumber \\
R = T \mbox{ Boost}(-\vec{\beta}) \nonumber \\
\mbox {drop time components of R} \nonumber \\
\mbox {R.rectify()} \nonumber \\
T = R \mbox{ Boost}(\beta)
\end{eqnarray}
\newpage
\section{When Exceptions Occur}
Although this section is not about mathematical definitions, it may be useful
to understand what will happen if the user code sends a method
data which does not make physical or mathematical sense. For example,
a user may supply a vector of magnitude greater than one, to a method
which will use that data to form a Boost. The resulting tachyonic boost
is unlikely to be the result the user had intended.
\subsection{How Problems Are Dealt With}
In general, the package is configurable in one of three ways:
\begin{enumerate}
\item
If {\tt ENABLE\_ZOOM\_EXCEPTIONS} is defined, then the Zoom Exceptions
package is used. This allows the user code to control the behaviour to
a considerable extent, including ignoring versus handling specific types
of exceptional conditions and checking a stack of prior potential throws.
Ultimately, an unhandled exception which is deemed or serious severity will
throw a C++ exception, which if un-caught will cause the program to terminate.
\item
If {\tt ENABLE\_ZOOM\_EXCEPTIONS} is defined, but C++ exceptions are not
enabled (some compilers have a switch to do this; some experiments choose
to use this switch in an effort to improve performance) then the behavior
for an unhandled serios exception is to explicitly abort.
\item
If {\tt ENABLE\_ZOOM\_EXCEPTIONS} is defined, then this compilation is designed
for CLHEP, which currently does not use the Zoom Exception package.
In that case, each problem will cause a message to be emitted on {\em cerr}.
In addition, the various problems which were deemed too severe to
ignore will throw actual C++ exceptions. These are all derived from
a class {\tt CLHEP\_vector\_exception} which is in turn derived from
the {\tt std::exception} class - see section 14.10 in Stroustrup.
User code catching a general {\tt CLHEP\_vector\_exception} can call the
virtual method {\em what()} to get a complete text message, or {\em name()}
to just get the exception name.
\end{enumerate}
\subsection{Possible Exceptions}
The following sorts of problems may be detected (and reported to {\em cerr}
by the Vector package,
and may in some circumstances be non-ignorable.
If the problem can't be ignored, then the corresponding exception
(derived from {\tt CLHEP\_vector\_exception} is thrown.
\begin{description}
\item [ZMxPhysicsVectors]
Parent exception of all ZMexceptions
particular to classes in the package.
\item [ZMxpvInfiniteVector]
Mathematical operation will lead
to infinity or NAN in a component
of a result vector.
\item [ZMxpvZeroVector]
A zero vector was used to specify
a direction based on vector.unit().
\item [ZMxpvTachyonic]
A relativistic kinematic function was
taken, involving a vector representing
a speed at or beyond that of light (=1).
\item [ZMxpvSpacelike]
A spacelike 4-vector was used in a
context where its restMass or gamma
needs to be computed: The result is
formally imaginary (a zero result is
supplied).
\item [ZMxpvInfinity]
Mathematical operation will lead
to infinity as a Scalar result.
\item [ZMxpvNegativeMass]
Kinematic operation, e.g. invariant
mass, rendered meaningless by an input
with negative time component.
\item [ZMxpvVectorInputFails]
Input to a SpaceVector or Lorentz
Vector failed due to bad format or EOF.
\item [ZMxpvParallelCols]
Purportedly orthogonal col's supplied
to form a Rotation are exactly
parallel instead.
\item [ZMxpvImproperRotation]
Orthogonal col's supplied form a
refection (determinant -1) more
nearly than rather than a rotation.
\item [ZMxpvImproperTransformation]
Orthogonalized rows supplied form a
tachyonic boost, a reflection, or
a combination of those flaws,
more nearly than a proper Lorentz
transformation.
\item [ZMxpvFixedAxis]
Attempt to change a RotationX,
RotationY, or RotationZ in such a way
that the axis might no longer be X,
Y, or Z respectively.
\item [ZMxpvIndexRange]
When using the syntax of v(i) to get
a vector component, i is out of range.
\end{description}
The following sorts of problems may be detected (and reported to {\em cerr}
by the Vector package, as warnings.
In these cases, the methdos have sensible behaviors avaialble, and will continue
processing after reporting the warning.
\begin{description}
\item [ZMxpvNotOrthogonal]
Purportedly orthogonal col's supplied
to form a Rotation or LT are not
orthogonal within the tolerance.
\item [ZMxpvNotSymplectic]
A row supplied to form a Lorentz
transformation has a value of restmass
incorrect by more than the tolerance:
It should be -1 for rows 1-3,
+1 for row 4.
\item [ZMxpvAmbiguousAngle]
Method involves taking an angle against
a reference vector of zero length, or
phi in polar coordinates of a vector
along the Z axis.
\item [ZMxpvNegativeR]
R of a supplied vector is negative.
The mathematical operation done is
still formally valid.
\item [ZMxpvUnusualTheta]
Theta supplied to construct or set
a vector is outside the range [0,PI].
The mathematical operation done is
still formally valid. But note that
when sin(theta) < 0, phi becomes an
angle against the -X axis.
\end{description}
\newpage
\section{Names and Keywords}
\subsection {Symbols in the CLHEP Vector Package}
Here we list the keywords that the user may need to be aware are used by the
CLHEP Vectors package. Some symbols not listed here are:
\begin{itemize}
\item
All symbols particular to the ZOOM PhysicsVectors package are listed in
the next section. Pure CLHEP users who do not include files fro a ZOOM
area need not be concerned with these.
\item
Method names in class scope, and names accessible only in the scope of a
class (such as Rotation::IDENTITY) cannot clash with user names.
\end{itemize}
Names of functions defined at global scope do not actually pollute the user
namespace, as long as one or more of their arguments involves a class
defined in this package. These are therefore listed separately here.
\begin{verbatim}
File Defines Class Names
------------ --------------------------------------------
HEP_THREEVEVCTOR_H Hep3Vector
HEP_LORENTZVECTOR_H HepLorentzVector
HEP_ROTATION_H HepRotation HepRotationX
HepRotationY HepRotationZ
HEP_LORENTZ_ROTATION_H HepLorentzRotation HepBoost
HepBoostX HepBoostY HepBoostZ
HEP_ROTATION_INTERFACES_H Hep4RotationInterface Hep3RotationInterface
HepRep3x3 HepRep4x4 HepRep4x4Symmetric
HEP_EULERANGLES_H HepEulerAngles
HEP_AXISANGLE_H HepAxisAngle
\end{verbatim}
\begin{verbatim}
Keywords and Types Constants Global Functions
------------------ --------- ----------------
Tcomponent X_HAT4 rotationOf
TimePositive Y_HAT4 rotationXOf
TimeNegative Z_HAT4 rotationYOf
T_HAT4 rotationZOf
X_HAT2 boostOf
Y_HAT2 boostXOf
HepXHat boostYOf
HepYHat boostZOf
HepZHat inverseOf
\end{verbatim}
\subsection {Further Symbols Defined in ZOOM Headers}
For backward compatibility, ZOOM headers are made available.
These typedef classes to the corresponding CLHEP versions,
and in the two cases of features which were in ZOOM but are not in
CLHEP---{\tt UnitVector} and construction of vectors from
spherical coordinates---provide the appropriate features.
These ZOOM features are implemented purely at the header level; there
is no separate ZOOM PhysicsVectors library beyond the CLHEP library.
Note that for users including these headers, all symbols starting with
{\tt ZMpv} are to be considered as reserved. These therefore do not
appear in this list of keywords.
\begin{verbatim}
File Defines Class Names
------------ --------------------------------------------
PHYSICSVECTORS_H
SPACEVECTOR_H SpaceVector
UNITVECTOR_H UnitVector
LORENTZVECTOR_H LorentzVector
ROTATION_H Rotation RotationX RotationY RotationZ
LORENTZ_TRANSFORMATION_H LorentzTransformation LorentzBoost
LorentzBoostX LorentzBoostY LorentzBoostZ
EULERANGLES_H EulerAngles
AXISANGLE_H AxisAngle
\end{verbatim}
\begin{verbatim}
Keywords and Types Constants
------------------ ---------
DEGREES X_HAT
RADIANS Y_HAT
ETA Z_HAT
\end{verbatim}
\end{document}
|