dash_mpd/fetch.rs
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 3627 3628 3629 3630 3631 3632 3633 3634 3635 3636 3637 3638 3639 3640 3641 3642 3643 3644 3645 3646 3647 3648 3649 3650 3651 3652 3653 3654 3655 3656 3657 3658 3659 3660 3661 3662 3663 3664 3665 3666 3667 3668 3669 3670 3671 3672 3673 3674 3675 3676 3677 3678 3679 3680 3681 3682 3683 3684 3685 3686 3687 3688 3689 3690 3691 3692 3693 3694 3695 3696 3697 3698 3699 3700 3701 3702 3703 3704 3705 3706 3707 3708 3709 3710 3711 3712 3713 3714 3715 3716 3717 3718 3719 3720 3721 3722 3723 3724 3725 3726 3727 3728 3729 3730 3731 3732 3733 3734 3735 3736 3737 3738 3739 3740 3741 3742 3743 3744 3745 3746 3747 3748 3749 3750 3751 3752 3753 3754 3755 3756 3757 3758 3759 3760 3761 3762 3763 3764 3765 3766 3767 3768 3769 3770 3771 3772 3773 3774 3775 3776 3777 3778 3779 3780 3781 3782 3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797 3798 3799 3800 3801 3802 3803 3804 3805 3806 3807 3808 3809 3810 3811 3812 3813 3814 3815 3816 3817 3818 3819 3820 3821 3822 3823 3824 3825 3826 3827 3828 3829 3830 3831 3832 3833 3834 3835 3836 3837 3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848 3849 3850 3851 3852 3853 3854 3855 3856 3857 3858 3859 3860 3861 3862 3863 3864 3865 3866 3867 3868 3869 3870 3871 3872 3873 3874 3875 3876 3877 3878 3879 3880 3881 3882 3883 3884 3885 3886 3887 3888 3889 3890 3891 3892 3893 3894 3895 3896 3897 3898 3899 3900 3901 3902 3903 3904 3905 3906 3907 3908 3909 3910 3911 3912 3913 3914 3915 3916 3917 3918 3919 3920 3921 3922 3923 3924 3925 3926 3927 3928 3929 3930 3931 3932 3933 3934 3935 3936 3937 3938 3939 3940 3941 3942 3943 3944 3945 3946 3947 3948 3949 3950 3951 3952 3953 3954 3955 3956 3957 3958 3959 3960 3961 3962 3963 3964 3965 3966 3967 3968 3969 3970 3971 3972 3973 3974 3975 3976 3977 3978 3979 3980 3981 3982 3983 3984 3985 3986 3987 3988 3989 3990 3991 3992 3993 3994 3995 3996 3997 3998 3999 4000 4001 4002 4003 4004 4005 4006 4007 4008 4009 4010 4011 4012 4013 4014 4015 4016 4017 4018 4019 4020 4021 4022 4023 4024 4025 4026 4027 4028 4029 4030 4031 4032 4033 4034 4035 4036 4037 4038 4039 4040 4041 4042 4043 4044 4045 4046 4047 4048 4049 4050 4051 4052 4053 4054 4055 4056 4057 4058 4059 4060 4061 4062 4063 4064 4065 4066 4067 4068 4069 4070 4071 4072 4073 4074 4075 4076 4077 4078 4079 4080 4081 4082 4083 4084 4085 4086 4087 4088 4089 4090 4091 4092 4093 4094 4095 4096 4097 4098 4099 4100 4101 4102 4103 4104 4105 4106 4107 4108 4109 4110 4111 4112 4113 4114 4115 4116 4117 4118 4119 4120 4121 4122 4123 4124 4125 4126 4127 4128 4129 4130 4131 4132 4133 4134 4135 4136 4137 4138 4139 4140 4141 4142 4143 4144 4145 4146 4147 4148 4149 4150 4151 4152 4153 4154 4155 4156 4157 4158 4159 4160 4161 4162 4163 4164 4165 4166 4167 4168 4169 4170 4171 4172 4173 4174 4175 4176 4177 4178 4179 4180 4181 4182 4183 4184 4185 4186 4187 4188 4189 4190 4191 4192 4193 4194 4195 4196 4197 4198 4199 4200 4201 4202 4203 4204 4205 4206 4207 4208 4209 4210 4211 4212 4213 4214 4215 4216 4217 4218 4219 4220 4221 4222 4223 4224 4225 4226 4227 4228 4229 4230 4231 4232 4233 4234 4235 4236 4237 4238 4239 4240 4241 4242 4243 4244 4245 4246 4247 4248 4249 4250 4251 4252 4253 4254 4255 4256 4257 4258 4259 4260 4261 4262 4263 4264 4265 4266 4267 4268 4269 4270 4271 4272 4273 4274 4275 4276 4277 4278 4279 4280 4281 4282 4283 4284 4285 4286 4287 4288 4289 4290 4291 4292 4293 4294 4295 4296 4297 4298 4299 4300 4301 4302 4303 4304 4305 4306 4307 4308 4309 4310 4311 4312 4313 4314 4315 4316 4317 4318 4319 4320 4321 4322 4323 4324 4325 4326 4327 4328 4329 4330 4331 4332 4333 4334 4335 4336 4337 4338 4339 4340 4341 4342 4343 4344 4345 4346 4347 4348 4349 4350 4351 4352 4353 4354 4355 4356 4357 4358 4359 4360 4361 4362 4363 4364 4365 4366 4367 4368 4369 4370 4371 4372 4373 4374 4375 4376 4377 4378 4379 4380 4381 4382 4383 4384 4385 4386 4387 4388 4389 4390 4391 4392 4393 4394 4395 4396 4397 4398 4399 4400 4401 4402 4403 4404 4405 4406 4407 4408 4409 4410 4411 4412 4413 4414 4415 4416 4417 4418 4419 4420 4421 4422 4423 4424 4425 4426 4427 4428 4429 4430 4431 4432 4433 4434 4435 4436 4437 4438 4439 4440 4441 4442 4443 4444 4445 4446 4447 4448 4449 4450 4451 4452 4453 4454 4455 4456 4457 4458 4459 4460 4461 4462 4463 4464 4465 4466 4467 4468 4469 4470 4471 4472 4473 4474 4475 4476 4477 4478 4479 4480 4481 4482 4483 4484 4485 4486 4487 4488 4489 4490 4491 4492 4493 4494 4495 4496 4497 4498 4499 4500 4501
//! Support for downloading content from DASH MPD media streams.
use std::io;
use std::env;
use fs_err as fs;
use fs::File;
use std::io::{Read, Write, Seek, BufReader, BufWriter};
use std::path::{Path, PathBuf};
use std::process::Command;
use std::time::Duration;
use tokio::time::Instant;
use std::sync::Arc;
use std::borrow::Cow;
use std::collections::HashMap;
use std::cmp::min;
use std::ffi::OsStr;
use std::num::NonZeroU32;
use tracing::{trace, info, warn, error};
use colored::*;
use regex::Regex;
use url::Url;
use bytes::Bytes;
use data_url::DataUrl;
use reqwest::header::{RANGE, CONTENT_TYPE};
use backoff::{future::retry_notify, ExponentialBackoff};
use governor::{Quota, RateLimiter};
use async_recursion::async_recursion;
use lazy_static::lazy_static;
use crate::{MPD, Period, Representation, AdaptationSet, SegmentBase, DashMpdError};
use crate::{parse, mux_audio_video, copy_video_to_container, copy_audio_to_container};
use crate::{is_audio_adaptation, is_video_adaptation, is_subtitle_adaptation};
use crate::{subtitle_type, content_protection_type, SubtitleType};
use crate::check_conformity;
#[cfg(not(feature = "libav"))]
use crate::ffmpeg::concat_output_files;
use crate::media::temporary_outpath;
#[allow(unused_imports)]
use crate::media::video_containers_concatable;
/// A `Client` from the `reqwest` crate, that we use to download content over HTTP.
pub type HttpClient = reqwest::Client;
type DirectRateLimiter = RateLimiter<governor::state::direct::NotKeyed,
governor::state::InMemoryState,
governor::clock::DefaultClock,
governor::middleware::NoOpMiddleware>;
// When reading stdout or stderr from an external commandline application to display for the user,
// this is the maximum number of octets read.
pub fn partial_process_output(output: &[u8]) -> Cow<'_, str> {
let len = min(output.len(), 4096);
#[allow(clippy::indexing_slicing)]
String::from_utf8_lossy(&output[0..len])
}
// This doesn't work correctly on modern Android, where there is no global location for temporary
// files (fix needed in the tempfile crate)
fn tmp_file_path(prefix: &str, extension: &OsStr) -> Result<PathBuf, DashMpdError> {
if let Some(ext) = extension.to_str() {
// suffix should include the "." separator
let fmt = format!(".{}", extension.to_string_lossy());
let suffix = if ext.starts_with('.') {
extension
} else {
OsStr::new(&fmt)
};
let file = tempfile::Builder::new()
.prefix(prefix)
.suffix(suffix)
.rand_bytes(7)
.tempfile()
.map_err(|e| DashMpdError::Io(e, String::from("creating temporary file")))?;
Ok(file.path().to_path_buf())
} else {
Err(DashMpdError::Other(String::from("converting filename extension")))
}
}
/// Receives updates concerning the progression of the download, and can display this information to
/// the user, for example using a progress bar.
pub trait ProgressObserver: Send + Sync {
fn update(&self, percent: u32, message: &str);
}
/// Preference for retrieving media representation with highest quality (and highest file size) or
/// lowest quality (and lowest file size).
#[derive(PartialEq, Eq, Clone, Copy, Default)]
pub enum QualityPreference { #[default] Lowest, Intermediate, Highest }
/// The DashDownloader allows the download of streaming media content from a DASH MPD manifest.
///
/// This involves:
/// - fetching the manifest file
/// - parsing its XML contents
/// - identifying the different Periods, potentially filtering out Periods that contain undesired
/// content such as advertising
/// - selecting for each Period the desired audio and video representations, according to user
/// preferences concerning the audio language, video dimensions and quality settings, and other
/// attributes such as label and role
/// - downloading all the audio and video segments for each Representation
/// - concatenating the audio segments and video segments into a stream
/// - potentially decrypting the audio and video content, if DRM is present
/// - muxing the audio and video streams to produce a single video file including audio
/// - concatenating the streams from each Period into a single media container.
///
/// This should work with both MPEG-DASH MPD manifests (where the media segments are typically
/// placed in fragmented MP4 or MPEG-2 TS containers) and for
/// [WebM-DASH](http://wiki.webmproject.org/adaptive-streaming/webm-dash-specification).
pub struct DashDownloader {
pub mpd_url: String,
pub redirected_url: Url,
referer: Option<String>,
auth_username: Option<String>,
auth_password: Option<String>,
auth_bearer_token: Option<String>,
pub output_path: Option<PathBuf>,
http_client: Option<HttpClient>,
quality_preference: QualityPreference,
language_preference: Option<String>,
role_preference: Vec<String>,
video_width_preference: Option<u64>,
video_height_preference: Option<u64>,
fetch_video: bool,
fetch_audio: bool,
fetch_subtitles: bool,
keep_video: Option<PathBuf>,
keep_audio: Option<PathBuf>,
concatenate_periods: bool,
fragment_path: Option<PathBuf>,
decryption_keys: HashMap<String, String>,
xslt_stylesheets: Vec<PathBuf>,
minimum_period_duration: Option<Duration>,
content_type_checks: bool,
conformity_checks: bool,
use_index_range: bool,
fragment_retry_count: u32,
max_error_count: u32,
progress_observers: Vec<Arc<dyn ProgressObserver>>,
sleep_between_requests: u8,
allow_live_streams: bool,
force_duration: Option<f64>,
rate_limit: u64,
bw_limiter: Option<DirectRateLimiter>,
pub verbosity: u8,
record_metainformation: bool,
pub muxer_preference: HashMap<String, String>,
pub concat_preference: HashMap<String, String>,
pub decryptor_preference: String,
pub ffmpeg_location: String,
pub vlc_location: String,
pub mkvmerge_location: String,
pub mp4box_location: String,
pub mp4decrypt_location: String,
pub shaka_packager_location: String,
}
// We don't want to test this code example on the CI infrastructure as it's too expensive
// and requires network access.
#[cfg(not(doctest))]
/// The DashDownloader follows the builder pattern to allow various optional arguments concerning
/// the download of DASH media content (preferences concerning bitrate/quality, specifying an HTTP
/// proxy, etc.).
///
/// # Example
///
/// ```rust
/// use dash_mpd::fetch::DashDownloader;
///
/// let url = "https://storage.googleapis.com/shaka-demo-assets/heliocentrism/heliocentrism.mpd";
/// match DashDownloader::new(url)
/// .worst_quality()
/// .download().await
/// {
/// Ok(path) => println!("Downloaded to {path:?}"),
/// Err(e) => eprintln!("Download failed: {e}"),
/// }
/// ```
impl DashDownloader {
/// Create a `DashDownloader` for the specified DASH manifest URL `mpd_url`.
pub fn new(mpd_url: &str) -> DashDownloader {
DashDownloader {
mpd_url: String::from(mpd_url),
redirected_url: Url::parse(mpd_url).unwrap(),
referer: None,
auth_username: None,
auth_password: None,
auth_bearer_token: None,
output_path: None,
http_client: None,
quality_preference: QualityPreference::Lowest,
language_preference: None,
role_preference: vec!["main".to_string(), "alternate".to_string()],
video_width_preference: None,
video_height_preference: None,
fetch_video: true,
fetch_audio: true,
fetch_subtitles: false,
keep_video: None,
keep_audio: None,
concatenate_periods: true,
fragment_path: None,
decryption_keys: HashMap::new(),
xslt_stylesheets: Vec::new(),
minimum_period_duration: None,
content_type_checks: true,
conformity_checks: true,
use_index_range: true,
fragment_retry_count: 10,
max_error_count: 30,
progress_observers: Vec::new(),
sleep_between_requests: 0,
allow_live_streams: false,
force_duration: None,
rate_limit: 0,
bw_limiter: None,
verbosity: 0,
record_metainformation: true,
muxer_preference: HashMap::new(),
concat_preference: HashMap::new(),
decryptor_preference: String::from("mp4decrypt"),
ffmpeg_location: String::from("ffmpeg"),
vlc_location: if cfg!(target_os = "windows") {
// The official VideoLan Windows installer doesn't seem to place its installation
// directory in the PATH, so we try with the default full path.
String::from("c:/Program Files/VideoLAN/VLC/vlc.exe")
} else {
String::from("vlc")
},
mkvmerge_location: String::from("mkvmerge"),
mp4box_location: if cfg!(target_os = "windows") {
String::from("MP4Box.exe")
} else if cfg!(target_os = "linux") || cfg!(target_os = "macos") {
String::from("MP4Box")
} else {
String::from("mp4box")
},
mp4decrypt_location: String::from("mp4decrypt"),
shaka_packager_location: String::from("shaka-packager"),
}
}
/// Specify the reqwest Client to be used for HTTP requests that download the DASH streaming
/// media content. Allows you to specify a proxy, the user agent, custom request headers,
/// request timeouts, additional root certificates to trust, client identity certificates, etc.
///
/// # Example
///
/// ```rust
/// use dash_mpd::fetch::DashDownloader;
///
/// let client = reqwest::Client::builder()
/// .user_agent("Mozilla/5.0")
/// .timeout(Duration::new(30, 0))
/// .build()
/// .expect("creating HTTP client");
/// let url = "https://cloudflarestream.com/31c9291ab41fac05471db4e73aa11717/manifest/video.mpd";
/// let out = PathBuf::from(env::temp_dir()).join("cloudflarestream.mp4");
/// DashDownloader::new(url)
/// .with_http_client(client)
/// .download_to(out)
// .await
/// ```
pub fn with_http_client(mut self, client: HttpClient) -> DashDownloader {
self.http_client = Some(client);
self
}
/// Specify the value for the Referer HTTP header used in network requests. This value is used
/// when retrieving the MPD manifest, when retrieving video and audio media segments, and when
/// retrieving subtitle data.
pub fn with_referer(mut self, referer: String) -> DashDownloader {
self.referer = Some(referer);
self
}
/// Specify the username and password to use to authenticate network requests for the manifest
/// and media segments.
pub fn with_authentication(mut self, username: String, password: String) -> DashDownloader {
self.auth_username = Some(username.clone());
self.auth_password = Some(password.clone());
self
}
/// Specify the Bearer token to use to authenticate network requests for the manifest and media
/// segments.
pub fn with_auth_bearer(mut self, token: String) -> DashDownloader {
self.auth_bearer_token = Some(token.clone());
self
}
/// Add a observer implementing the ProgressObserver trait, that will receive updates concerning
/// the progression of the download (allows implementation of a progress bar, for example).
pub fn add_progress_observer(mut self, observer: Arc<dyn ProgressObserver>) -> DashDownloader {
self.progress_observers.push(observer);
self
}
/// If the DASH manifest specifies several Adaptations with different bitrates (levels of
/// quality), prefer the Adaptation with the highest bitrate (largest output file).
pub fn best_quality(mut self) -> DashDownloader {
self.quality_preference = QualityPreference::Highest;
self
}
/// If the DASH manifest specifies several Adaptations with different bitrates (levels of
/// quality), prefer the Adaptation with an intermediate bitrate (closest to the median value).
pub fn intermediate_quality(mut self) -> DashDownloader {
self.quality_preference = QualityPreference::Intermediate;
self
}
/// If the DASH manifest specifies several Adaptations with different bitrates (levels of
/// quality), prefer the Adaptation with the lowest bitrate (smallest output file).
pub fn worst_quality(mut self) -> DashDownloader {
self.quality_preference = QualityPreference::Lowest;
self
}
/// Specify the preferred language when multiple audio streams with different languages are
/// available. Must be in RFC 5646 format (e.g. "fr" or "en-AU"). If a preference is not
/// specified and multiple audio streams are present, the first one listed in the DASH manifest
/// will be downloaded.
pub fn prefer_language(mut self, lang: String) -> DashDownloader {
self.language_preference = Some(lang);
self
}
/// Specify the preference ordering for Role annotations on AdaptationSet elements. Some DASH
/// streams include multiple AdaptationSets, one annotated "main" and another "alternate", for
/// example. If `role_preference` is ["main", "alternate"] and one of the AdaptationSets is
/// annotated "main", then we will only download that AdaptationSet. If no role annotations are
/// specified, this preference is ignored. This preference selection is applied before the
/// preferences related to stream quality and video height/width: for example an AdaptationSet
/// with role=alternate will be ignored when a role=main AdaptationSet is present, even if we
/// also specify a quality preference for highest and the role=alternate stream has a higher
/// quality.
pub fn prefer_roles(mut self, role_preference: Vec<String>) -> DashDownloader {
if role_preference.len() < u8::MAX.into() {
self.role_preference = role_preference;
} else {
warn!("Ignoring role_preference ordering due to excessive length");
}
self
}
/// If the DASH manifest specifies several video Adaptations with different resolutions, prefer
/// the Adaptation whose width is closest to the specified `width`.
pub fn prefer_video_width(mut self, width: u64) -> DashDownloader {
self.video_width_preference = Some(width);
self
}
/// If the DASH manifest specifies several video Adaptations with different resolutions, prefer
/// the Adaptation whose height is closest to the specified `height`.
pub fn prefer_video_height(mut self, height: u64) -> DashDownloader {
self.video_height_preference = Some(height);
self
}
/// If the media stream has separate audio and video streams, only download the video stream.
pub fn video_only(mut self) -> DashDownloader {
self.fetch_audio = false;
self.fetch_video = true;
self
}
/// If the media stream has separate audio and video streams, only download the audio stream.
pub fn audio_only(mut self) -> DashDownloader {
self.fetch_audio = true;
self.fetch_video = false;
self
}
/// Keep the file containing video at the specified path. If the path already exists, file
/// contents will be overwritten.
pub fn keep_video_as<P: Into<PathBuf>>(mut self, video_path: P) -> DashDownloader {
self.keep_video = Some(video_path.into());
self
}
/// Keep the file containing audio at the specified path. If the path already exists, file
/// contents will be overwritten.
pub fn keep_audio_as<P: Into<PathBuf>>(mut self, audio_path: P) -> DashDownloader {
self.keep_audio = Some(audio_path.into());
self
}
/// Save media fragments to the directory `fragment_path`. The directory will be created if it
/// does not exist.
pub fn save_fragments_to<P: Into<PathBuf>>(mut self, fragment_path: P) -> DashDownloader {
self.fragment_path = Some(fragment_path.into());
self
}
/// Add a key to be used to decrypt MPEG media streams that use Common Encryption (cenc). This
/// function may be called several times to specify multiple kid/key pairs. Decryption uses the
/// external commandline application specified by `with_decryptor_preference` (either mp4decrypt
/// from Bento4 or shaka-packager, defaulting to mp4decrypt), run as a subprocess.
///
/// # Arguments
///
/// * `id` - a track ID in decimal or, a 128-bit KID in hexadecimal format (32 hex characters).
/// Examples: "1" or "eb676abbcb345e96bbcf616630f1a3da".
///
/// * `key` - a 128-bit key in hexadecimal format.
pub fn add_decryption_key(mut self, id: String, key: String) -> DashDownloader {
self.decryption_keys.insert(id, key);
self
}
/// Register an XSLT stylesheet that will be applied to the MPD manifest after XLink processing
/// and before deserialization into Rust structs. The stylesheet will be applied to the manifest
/// using the xsltproc commandline tool, which supports XSLT 1.0. If multiple stylesheets are
/// registered, they will be called in sequence in the same order as their registration. If the
/// application of a stylesheet fails, the download will be aborted.
///
/// This is an experimental API which may change in future versions of the library.
///
/// # Arguments
///
/// * `stylesheet`: the path to an XSLT stylesheet.
pub fn with_xslt_stylesheet<P: Into<PathBuf>>(mut self, stylesheet: P) -> DashDownloader {
self.xslt_stylesheets.push(stylesheet.into());
self
}
/// Don't download (skip) Periods in the manifest whose duration is less than the specified
/// value.
pub fn minimum_period_duration(mut self, value: Duration) -> DashDownloader {
self.minimum_period_duration = Some(value);
self
}
/// Parameter `value` determines whether audio content is downloaded. If disabled, the output
/// media file will either contain only a video track (if fetch_video is true and the manifest
/// includes a video stream), or will be empty.
pub fn fetch_audio(mut self, value: bool) -> DashDownloader {
self.fetch_audio = value;
self
}
/// Parameter `value` determines whether video content is downloaded. If disabled, the output
/// media file will either contain only an audio track (if fetch_audio is true and the manifest
/// includes an audio stream which is separate from the video stream), or will be empty.
pub fn fetch_video(mut self, value: bool) -> DashDownloader {
self.fetch_video = value;
self
}
/// Specify whether subtitles should be fetched, if they are available. If subtitles are
/// requested and available, they will be downloaded to a file named with the same name as the
/// media output and an appropriate extension (".vtt", ".ttml", ".srt", etc.).
///
/// # Arguments
///
/// * `value`: enable or disable the retrieval of subtitles.
pub fn fetch_subtitles(mut self, value: bool) -> DashDownloader {
self.fetch_subtitles = value;
self
}
/// For multi-Period manifests, parameter `value` determines whether the content of multiple
/// Periods is concatenated into a single output file where their resolutions, frame rate and
/// aspect ratios are compatible, or kept in individual files.
pub fn concatenate_periods(mut self, value: bool) -> DashDownloader {
self.concatenate_periods = value;
self
}
/// Don't check that the content-type of downloaded segments corresponds to audio or video
/// content (may be necessary with poorly configured HTTP servers).
pub fn without_content_type_checks(mut self) -> DashDownloader {
self.content_type_checks = false;
self
}
/// Specify whether to check that the content-type of downloaded segments corresponds to audio
/// or video content (this may need to be set to false with poorly configured HTTP servers).
pub fn content_type_checks(mut self, value: bool) -> DashDownloader {
self.content_type_checks = value;
self
}
/// Specify whether to run various conformity checks on the content of the DASH manifest before
/// downloading media segments.
pub fn conformity_checks(mut self, value: bool) -> DashDownloader {
self.conformity_checks = value;
self
}
/// Specify whether the use the sidx/Cue index for SegmentBase@indexRange addressing.
///
/// If set to true (the default value), downloads of media whose manifest uses
/// SegmentBase@indexRange addressing will retrieve the index information (currently only sidx
/// information used in ISOBMFF/MP4 containers; Cue information for WebM containers is currently
/// not supported) with a byte range request, then retrieve and concatenate the different bytes
/// ranges indicated in the index. This is the download method used by most DASH players
/// (set-top box and browser-based). It avoids downloading the content identified by the
/// BaseURL as a very large chunk, which can fill up RAM and may be banned by certain content
/// servers.
///
/// If set to false, the BaseURL content will be downloaded as a single large chunk. This may be
/// more robust on certain content streams that have been encoded in a manner which is not
/// suitable for byte range retrieval.
pub fn use_index_range(mut self, value: bool) -> DashDownloader {
self.use_index_range = value;
self
}
/// The upper limit on the number of times to attempt to fetch a media segment, even in the
/// presence of network errors. Transient network errors (such as timeouts) do not count towards
/// this limit.
pub fn fragment_retry_count(mut self, count: u32) -> DashDownloader {
self.fragment_retry_count = count;
self
}
/// The upper limit on the number of non-transient network errors encountered for this download
/// before we abort the download. Transient network errors such as an HTTP 408 “request timeout”
/// are retried automatically with an exponential backoff mechanism, and do not count towards
/// this upper limit. The default is to fail after 30 non-transient network errors over the
/// whole download.
pub fn max_error_count(mut self, count: u32) -> DashDownloader {
self.max_error_count = count;
self
}
/// Specify a number of seconds to sleep between network requests (default 0).
pub fn sleep_between_requests(mut self, seconds: u8) -> DashDownloader {
self.sleep_between_requests = seconds;
self
}
/// Specify whether to attempt to download from a “live” stream, or dynamic DASH manifest.
/// Default is false.
///
/// Downloading from a genuinely live stream won’t work well, because this library doesn’t
/// implement the clock-related throttling needed to only download media segments when they
/// become available. However, some media sources publish pseudo-live streams where all media
/// segments are in fact available, which we will be able to download. You might also have some
/// success in combination with the `sleep_between_requests()` method.
///
/// You may also need to force a duration for the live stream using method
/// `force_duration()`, because live streams often don’t specify a duration.
pub fn allow_live_streams(mut self, value: bool) -> DashDownloader {
self.allow_live_streams = value;
self
}
/// Specify the number of seconds to capture from the media stream, overriding the duration
/// specified in the DASH manifest. This is mostly useful for live streams, for which the
/// duration is often not specified. It can also be used to capture only the first part of a
/// normal (static/on-demand) media stream.
pub fn force_duration(mut self, seconds: f64) -> DashDownloader {
self.force_duration = Some(seconds);
self
}
/// A maximal limit on the network bandwidth consumed to download media segments, expressed in
/// octets (bytes) per second. No limit on bandwidth if set to zero (the default value).
/// Limiting bandwidth below 50kB/s is not recommended, as the downloader may fail to respect
/// this limit.
pub fn with_rate_limit(mut self, bps: u64) -> DashDownloader {
if bps < 10 * 1024 {
warn!("Limiting bandwidth below 10kB/s is unlikely to be stable");
}
if self.verbosity > 1 {
info!("Limiting bandwidth to {} kB/s", bps/1024);
}
self.rate_limit = bps;
// Our rate_limit is in bytes/second, but the governor::RateLimiter can only handle an u32 rate.
// We express our cells in the RateLimiter in kB/s instead of bytes/second, to allow for numbing
// future bandwidth capacities. We need to be careful to allow a quota burst size which
// corresponds to the size (in kB) of the largest media segments we are going to be retrieving,
// because that's the number of bucket cells that will be consumed for each downloaded segment.
let mut kps = 1 + bps / 1024;
if kps > u32::MAX.into() {
warn!("Throttling bandwidth limit");
kps = u32::MAX.into();
}
if let Some(bw_limit) = NonZeroU32::new(kps as u32) {
if let Some(burst) = NonZeroU32::new(10 * 1024) {
let bw_quota = Quota::per_second(bw_limit)
.allow_burst(burst);
self.bw_limiter = Some(RateLimiter::direct(bw_quota));
}
}
self
}
/// Set the verbosity level of the download process.
///
/// # Arguments
///
/// * Level - an integer specifying the verbosity level.
/// - 0: no information is printed
/// - 1: basic information on the number of Periods and bandwidth of selected representations
/// - 2: information above + segment addressing mode
/// - 3 or larger: information above + size of each downloaded segment
pub fn verbosity(mut self, level: u8) -> DashDownloader {
self.verbosity = level;
self
}
/// Specify whether to record metainformation concerning the media content (origin URL, title,
/// source and copyright metainformation) as extended attributes in the output file, assuming
/// this information is present in the DASH manifest.
pub fn record_metainformation(mut self, record: bool) -> DashDownloader {
self.record_metainformation = record;
self
}
/// When muxing audio and video streams to a container of type `container`, try muxing
/// applications following the order given by `ordering`.
///
/// This function may be called multiple times to specify the ordering for different container
/// types. If called more than once for the same container type, the ordering specified in the
/// last call is retained.
///
/// # Arguments
///
/// * `container`: the container type (e.g. "mp4", "mkv", "avi")
/// * `ordering`: the comma-separated order of preference for trying muxing applications (e.g.
/// "ffmpeg,vlc,mp4box")
///
/// # Example
///
/// ```rust
/// let out = DashDownloader::new(url)
/// .with_muxer_preference("mkv", "ffmpeg")
/// .download_to("wonderful.mkv")
/// .await?;
/// ```
pub fn with_muxer_preference(mut self, container: &str, ordering: &str) -> DashDownloader {
self.muxer_preference.insert(container.to_string(), ordering.to_string());
self
}
/// When concatenating streams from a multi-period manifest to a container of type `container`,
/// try concat helper applications following the order given by `ordering`.
///
/// This function may be called multiple times to specify the ordering for different container
/// types. If called more than once for the same container type, the ordering specified in the
/// last call is retained.
///
/// # Arguments
///
/// * `container`: the container type (e.g. "mp4", "mkv", "avi")
/// * `ordering`: the comma-separated order of preference for trying concat helper applications.
/// Valid possibilities are "ffmpeg" (the ffmpeg concat filter, slow), "ffmpegdemuxer" (the
/// ffmpeg concat demuxer, fast but less robust), "mkvmerge" (fast but not robust), and "mp4box".
///
/// # Example
///
/// ```rust
/// let out = DashDownloader::new(url)
/// .with_concat_preference("mkv", "ffmpeg,mkvmerge")
/// .download_to("wonderful.mkv")
/// .await?;
/// ```
pub fn with_concat_preference(mut self, container: &str, ordering: &str) -> DashDownloader {
self.concat_preference.insert(container.to_string(), ordering.to_string());
self
}
/// Specify the commandline application to be used to decrypt media which has been enriched with
/// ContentProtection (DRM).
///
/// # Arguments
///
/// * `decryption_tool`: either "mp4decrypt" or "shaka"
pub fn with_decryptor_preference(mut self, decryption_tool: &str) -> DashDownloader {
self.decryptor_preference = decryption_tool.to_string();
self
}
/// Specify the location of the `ffmpeg` application, if not located in PATH.
///
/// # Arguments
///
/// * `ffmpeg_path`: the path to the ffmpeg application. If it does not specify an absolute
/// path, the `PATH` environment variable will be searched in a platform-specific way
/// (implemented in `std::process::Command`).
///
/// # Example
///
/// ```rust
/// #[cfg(target_os = "unix")]
/// let ddl = ddl.with_ffmpeg("/opt/ffmpeg-next/bin/ffmpeg");
/// ```
pub fn with_ffmpeg(mut self, ffmpeg_path: &str) -> DashDownloader {
self.ffmpeg_location = ffmpeg_path.to_string();
self
}
/// Specify the location of the VLC application, if not located in PATH.
///
/// # Arguments
///
/// * `vlc_path`: the path to the VLC application. If it does not specify an absolute
/// path, the `PATH` environment variable will be searched in a platform-specific way
/// (implemented in `std::process::Command`).
///
/// # Example
///
/// ```rust
/// #[cfg(target_os = "windows")]
/// let ddl = ddl.with_vlc("C:/Program Files/VideoLAN/VLC/vlc.exe");
/// ```
pub fn with_vlc(mut self, vlc_path: &str) -> DashDownloader {
self.vlc_location = vlc_path.to_string();
self
}
/// Specify the location of the mkvmerge application, if not located in PATH.
///
/// # Arguments
///
/// * `path`: the path to the mkvmerge application. If it does not specify an absolute
/// path, the `PATH` environment variable will be searched in a platform-specific way
/// (implemented in `std::process::Command`).
pub fn with_mkvmerge(mut self, path: &str) -> DashDownloader {
self.mkvmerge_location = path.to_string();
self
}
/// Specify the location of the MP4Box application, if not located in PATH.
///
/// # Arguments
///
/// * `path`: the path to the MP4Box application. If it does not specify an absolute
/// path, the `PATH` environment variable will be searched in a platform-specific way
/// (implemented in `std::process::Command`).
pub fn with_mp4box(mut self, path: &str) -> DashDownloader {
self.mp4box_location = path.to_string();
self
}
/// Specify the location of the Bento4 mp4decrypt application, if not located in PATH.
///
/// # Arguments
///
/// * `path`: the path to the mp4decrypt application. If it does not specify an absolute
/// path, the `PATH` environment variable will be searched in a platform-specific way
/// (implemented in `std::process::Command`).
pub fn with_mp4decrypt(mut self, path: &str) -> DashDownloader {
self.mp4decrypt_location = path.to_string();
self
}
/// Specify the location of the shaka-packager application, if not located in PATH.
///
/// # Arguments
///
/// * `path`: the path to the shaka-packager application. If it does not specify an absolute
/// path, the `PATH` environment variable will be searched in a platform-specific way
/// (implemented in `std::process::Command`).
pub fn with_shaka_packager(mut self, path: &str) -> DashDownloader {
self.shaka_packager_location = path.to_string();
self
}
/// Download DASH streaming media content to the file named by `out`. If the output file `out`
/// already exists, its content will be overwritten.
///
/// Note that the media container format used when muxing audio and video streams depends on the
/// filename extension of the path `out`. If the filename extension is `.mp4`, an MPEG-4
/// container will be used; if it is `.mkv` a Matroska container will be used, for `.webm` a
/// WebM container (specific type of Matroska) will be used, and otherwise the heuristics
/// implemented by the selected muxer (by default ffmpeg) will apply (e.g. an `.avi` extension
/// will generate an AVI container).
pub async fn download_to<P: Into<PathBuf>>(mut self, out: P) -> Result<PathBuf, DashMpdError> {
self.output_path = Some(out.into());
if self.http_client.is_none() {
let client = reqwest::Client::builder()
.timeout(Duration::new(30, 0))
.cookie_store(true)
.build()
.map_err(|_| DashMpdError::Network(String::from("building HTTP client")))?;
self.http_client = Some(client);
}
fetch_mpd(&mut self).await
}
/// Download DASH streaming media content to a file in the current working directory and return
/// the corresponding `PathBuf`.
///
/// The name of the output file is derived from the manifest URL. The output file will be
/// overwritten if it already exists. The downloaded media will be placed in an MPEG-4
/// container. To select another media container, see the `download_to` function.
pub async fn download(mut self) -> Result<PathBuf, DashMpdError> {
let cwd = env::current_dir()
.map_err(|e| DashMpdError::Io(e, String::from("obtaining current directory")))?;
let filename = generate_filename_from_url(&self.mpd_url);
let outpath = cwd.join(filename);
self.output_path = Some(outpath);
if self.http_client.is_none() {
let client = reqwest::Client::builder()
.timeout(Duration::new(30, 0))
.cookie_store(true)
.build()
.map_err(|_| DashMpdError::Network(String::from("building HTTP client")))?;
self.http_client = Some(client);
}
fetch_mpd(&mut self).await
}
}
// Parse a range specifier, such as Initialization@range or SegmentBase@indexRange attributes, of
// the form "45-67"
fn parse_range(range: &str) -> Result<(u64, u64), DashMpdError> {
let v: Vec<&str> = range.split_terminator('-').collect();
if v.len() != 2 {
return Err(DashMpdError::Parsing(format!("invalid range specifier: {range}")));
}
#[allow(clippy::indexing_slicing)]
let start: u64 = v[0].parse()
.map_err(|_| DashMpdError::Parsing(String::from("invalid start for range specifier")))?;
#[allow(clippy::indexing_slicing)]
let end: u64 = v[1].parse()
.map_err(|_| DashMpdError::Parsing(String::from("invalid end for range specifier")))?;
Ok((start, end))
}
#[derive(Debug)]
struct MediaFragment {
period: u8,
url: Url,
start_byte: Option<u64>,
end_byte: Option<u64>,
is_init: bool,
timeout: Option<Duration>,
}
#[derive(Debug)]
struct MediaFragmentBuilder {
period: u8,
url: Url,
start_byte: Option<u64>,
end_byte: Option<u64>,
is_init: bool,
timeout: Option<Duration>,
}
impl MediaFragmentBuilder {
pub fn new(period: u8, url: Url) -> MediaFragmentBuilder {
MediaFragmentBuilder {
period, url, start_byte: None, end_byte: None, is_init: false, timeout: None
}
}
pub fn with_range(mut self, start_byte: Option<u64>, end_byte: Option<u64>) -> MediaFragmentBuilder {
self.start_byte = start_byte;
self.end_byte = end_byte;
self
}
pub fn with_timeout(mut self, timeout: Duration) -> MediaFragmentBuilder {
self.timeout = Some(timeout);
self
}
pub fn set_init(mut self) -> MediaFragmentBuilder {
self.is_init = true;
self
}
pub fn build(self) -> MediaFragment {
MediaFragment {
period: self.period,
url: self.url,
start_byte: self.start_byte,
end_byte: self.end_byte,
is_init: self.is_init,
timeout: self.timeout
}
}
}
// This struct is used to share information concerning the media fragments identified while parsing
// a Period as being wanted for download, alongside any diagnostics information that we collected
// while parsing the Period (in particular, any ContentProtection details).
#[derive(Debug, Default)]
struct PeriodOutputs {
fragments: Vec<MediaFragment>,
diagnostics: Vec<String>,
subtitle_formats: Vec<SubtitleType>,
}
#[derive(Debug, Default)]
struct PeriodDownloads {
audio_fragments: Vec<MediaFragment>,
video_fragments: Vec<MediaFragment>,
subtitle_fragments: Vec<MediaFragment>,
subtitle_formats: Vec<SubtitleType>,
period_counter: u8,
id: Option<String>,
}
fn period_fragment_count(pd: &PeriodDownloads) -> usize {
pd.audio_fragments.len() +
pd.video_fragments.len() +
pd.subtitle_fragments.len()
}
async fn throttle_download_rate(downloader: &DashDownloader, size: u32) -> Result<(), DashMpdError> {
if downloader.rate_limit > 0 {
if let Some(cells) = NonZeroU32::new(size) {
if let Some(limiter) = downloader.bw_limiter.as_ref() {
#[allow(clippy::redundant_pattern_matching)]
if let Err(_) = limiter.until_n_ready(cells).await {
return Err(DashMpdError::Other(
"Bandwidth limit is too low".to_string()));
}
}
}
}
Ok(())
}
fn generate_filename_from_url(url: &str) -> PathBuf {
use sanitise_file_name::{sanitise_with_options, Options};
let mut path = url;
if let Some(p) = path.strip_prefix("http://") {
path = p;
} else if let Some(p) = path.strip_prefix("https://") {
path = p;
} else if let Some(p) = path.strip_prefix("file://") {
path = p;
}
if let Some(p) = path.strip_prefix("www.") {
path = p;
}
if let Some(p) = path.strip_prefix("ftp.") {
path = p;
}
if let Some(p) = path.strip_suffix(".mpd") {
path = p;
}
let mut sanitize_opts = Options::DEFAULT;
sanitize_opts.length_limit = 150;
// We could also enable sanitize_opts.url_safe here.
// We currently default to an MP4 container (could default to Matroska which is more flexible,
// and less patent-encumbered, but perhaps less commonly supported).
PathBuf::from(sanitise_with_options(path, &sanitize_opts) + ".mp4")
}
// A manifest containing a single Period will be saved to the output name requested by calling
// download_to("outputname.mp4") or to a name determined by generate_filename_from_url() above from
// the MPD URL.
//
// A manifest containing multiple Periods will be saved (in the general case where each period has a
// different resolution) to files whose name is built from the outputname, including the period name
// as a stem suffix (e.g. "outputname-p3.mp4" for the third period). The content of the first Period
// will be saved to a file with the requested outputname ("outputname.mp4" in this example).
//
// In the special case where each period has the same resolution (meaning that it is possible to
// concatenate the Periods into a single media container, re-encoding if the codecs used in each
// period differ), the content will be saved to a single file named as for a single Period.
//
// Illustration for a three-Period manifest with differing resolutions:
//
// download_to("foo.mkv") => foo.mkv (Period 1), foo-p2.mkv (Period 2), foo-p3.mkv (Period 3)
fn output_path_for_period(base: &Path, period: u8) -> PathBuf {
assert!(period > 0);
if period == 1 {
base.to_path_buf()
} else {
if let Some(stem) = base.file_stem() {
if let Some(ext) = base.extension() {
let fname = format!("{}-p{period}.{}", stem.to_string_lossy(), ext.to_string_lossy());
return base.with_file_name(fname);
}
}
let p = format!("dashmpd-p{period}");
tmp_file_path(&p, base.extension().unwrap_or(OsStr::new("mp4")))
.unwrap_or_else(|_| p.into())
}
}
fn is_absolute_url(s: &str) -> bool {
s.starts_with("http://") ||
s.starts_with("https://") ||
s.starts_with("file://") ||
s.starts_with("ftp://")
}
fn merge_baseurls(current: &Url, new: &str) -> Result<Url, DashMpdError> {
if is_absolute_url(new) {
Url::parse(new)
.map_err(|e| parse_error("parsing BaseURL", e))
} else {
// We are careful to merge the query portion of the current URL (which is either the
// original manifest URL, or the URL that it redirected to, or the value of a BaseURL
// element in the manifest) with the new URL. But if the new URL already has a query string,
// it takes precedence.
//
// Examples
//
// merge_baseurls(https://example.com/manifest.mpd?auth=secret, /video42.mp4) =>
// https://example.com/video42.mp4?auth=secret
//
// merge_baseurls(https://example.com/manifest.mpd?auth=old, /video42.mp4?auth=new) =>
// https://example.com/video42.mp4?auth=new
let mut merged = current.join(new)
.map_err(|e| parse_error("joining base with BaseURL", e))?;
if merged.query().is_none() {
merged.set_query(current.query());
}
Ok(merged)
}
}
// Return true if the response includes a content-type header corresponding to audio. We need to
// allow "video/" MIME types because some servers return "video/mp4" content-type for audio segments
// in an MP4 container, and we accept application/octet-stream headers because some servers are
// poorly configured.
fn content_type_audio_p(response: &reqwest::Response) -> bool {
match response.headers().get("content-type") {
Some(ct) => {
let ctb = ct.as_bytes();
ctb.starts_with(b"audio/") ||
ctb.starts_with(b"video/") ||
ctb.starts_with(b"application/octet-stream")
},
None => false,
}
}
// Return true if the response includes a content-type header corresponding to video.
fn content_type_video_p(response: &reqwest::Response) -> bool {
match response.headers().get("content-type") {
Some(ct) => {
let ctb = ct.as_bytes();
ctb.starts_with(b"video/") ||
ctb.starts_with(b"application/octet-stream")
},
None => false,
}
}
// Return a measure of the distance between this AdaptationSet's lang attribute and the language
// code specified by language_preference. If the AdaptationSet node has no lang attribute, return an
// arbitrary large distance.
fn adaptation_lang_distance(a: &AdaptationSet, language_preference: &str) -> u8 {
if let Some(lang) = &a.lang {
if lang.eq(language_preference) {
return 0;
}
if lang[0..2].eq(&language_preference[0..2]) {
return 5;
}
100
} else {
100
}
}
// We can have a <Role value="foobles"> element directly within the AdaptationSet element, or within
// a ContentComponent element in the AdaptationSet.
fn adaptation_roles(a: &AdaptationSet) -> Vec<String> {
let mut roles = Vec::new();
for r in &a.Role {
if let Some(rv) = &r.value {
roles.push(String::from(rv));
}
}
for cc in &a.ContentComponent {
for r in &cc.Role {
if let Some(rv) = &r.value {
roles.push(String::from(rv));
}
}
}
roles
}
// Best possible "score" is zero.
fn adaptation_role_distance(a: &AdaptationSet, role_preference: &[String]) -> u8 {
adaptation_roles(a).iter()
.map(|r| role_preference.binary_search(r).unwrap_or(u8::MAX.into()))
.map(|u| u8::try_from(u).unwrap_or(u8::MAX))
.min()
.unwrap_or(u8::MAX)
}
// We select the AdaptationSets that correspond to our language preference, and if there are several
// with our language preference, that with the role according to role_preference, and if no
// role_preference, return all adaptations.
//
// Start by getting a Vec of adaptation_lang_distance
// Take the min and collect all Adaptations where dist = min_distance
// then apply role_preference
fn select_preferred_adaptations<'a>(
adaptations: Vec<&'a AdaptationSet>,
downloader: &DashDownloader) -> Vec<&'a AdaptationSet>
{
let mut preferred: Vec<&'a AdaptationSet>;
if let Some(ref lang) = downloader.language_preference {
preferred = Vec::new();
let distance: Vec<u8> = adaptations.iter()
.map(|a| adaptation_lang_distance(a, lang))
.collect();
let min_distance = distance.iter().min().unwrap_or(&0);
for (i, a) in adaptations.iter().enumerate() {
if let Some(di) = distance.get(i) {
if di == min_distance {
preferred.push(a);
}
}
}
} else {
preferred = adaptations;
}
// Apply the role_preference. For example, a role_preference of ["main", "alternate",
// "supplementary", "commentary"] means we should prefer an AdaptationSet with role=main, and
// return only that AdaptationSet. If there are no role annotations on the AdaptationSets, or
// the specified roles don't match anything in our role_preference ordering, then all
// AdaptationSets will receive the maximum distance and they will all be returned.
let role_distance: Vec<u8> = preferred.iter()
.map(|a| adaptation_role_distance(a, &downloader.role_preference))
.collect();
let role_distance_min = role_distance.iter().min().unwrap_or(&0);
let mut best = Vec::new();
for (i, a) in preferred.into_iter().enumerate() {
if let Some(rdi) = role_distance.get(i) {
if rdi == role_distance_min {
best.push(a);
}
}
}
best
}
// A manifest often contains multiple video Representations with different bandwidths and video
// resolutions. We select the Representation to download by ranking following the user's specified
// quality preference. We first rank following the @qualityRanking attribute if it is present, and
// otherwise by the bandwidth specified. Note that quality ranking may be different from bandwidth
// ranking when different codecs are used.
fn select_preferred_representation<'a>(
representations: Vec<&'a Representation>,
downloader: &DashDownloader) -> Option<&'a Representation>
{
if representations.iter().all(|x| x.qualityRanking.is_some()) {
// rank according to the @qualityRanking attribute (lower values represent
// higher quality content)
match downloader.quality_preference {
QualityPreference::Lowest =>
representations.iter()
.max_by_key(|r| r.qualityRanking.unwrap_or(u8::MAX))
.copied(),
QualityPreference::Highest =>
representations.iter().min_by_key(|r| r.qualityRanking.unwrap_or(0))
.copied(),
QualityPreference::Intermediate => {
let count = representations.len();
match count {
0 => None,
1 => Some(representations[0]),
_ => {
let mut ranking: Vec<u8> = representations.iter()
.map(|r| r.qualityRanking.unwrap_or(u8::MAX))
.collect();
ranking.sort_unstable();
if let Some(want_ranking) = ranking.get(count / 2) {
representations.iter()
.find(|r| r.qualityRanking.unwrap_or(u8::MAX) == *want_ranking)
.copied()
} else {
representations.first().copied()
}
},
}
},
}
} else {
// rank according to the bandwidth attribute (lower values imply lower quality)
match downloader.quality_preference {
QualityPreference::Lowest => representations.iter()
.min_by_key(|r| r.bandwidth.unwrap_or(1_000_000_000))
.copied(),
QualityPreference::Highest => representations.iter()
.max_by_key(|r| r.bandwidth.unwrap_or(0))
.copied(),
QualityPreference::Intermediate => {
let count = representations.len();
match count {
0 => None,
1 => Some(representations[0]),
_ => {
let mut ranking: Vec<u64> = representations.iter()
.map(|r| r.bandwidth.unwrap_or(100_000_000))
.collect();
ranking.sort_unstable();
if let Some(want_ranking) = ranking.get(count / 2) {
representations.iter()
.find(|r| r.bandwidth.unwrap_or(100_000_000) == *want_ranking)
.copied()
} else {
representations.first().copied()
}
},
}
},
}
}
}
// The AdaptationSet a is the parent of the Representation r.
fn print_available_subtitles_representation(r: &Representation, a: &AdaptationSet) {
let unspecified = "<unspecified>".to_string();
let empty = "".to_string();
let lang = r.lang.as_ref().unwrap_or(a.lang.as_ref().unwrap_or(&unspecified));
let codecs = r.codecs.as_ref().unwrap_or(a.codecs.as_ref().unwrap_or(&empty));
let typ = subtitle_type(&a);
let stype = if !codecs.is_empty() {
format!("{typ:?}/{codecs}")
} else {
format!("{typ:?}")
};
let role = a.Role.first()
.map_or_else(|| String::from(""),
|r| r.value.as_ref().map_or_else(|| String::from(""), |v| format!(" role={v}")));
let label = a.Label.first()
.map_or_else(|| String::from(""), |l| format!(" label={}", l.clone().content));
info!(" subs {stype:>18} | {lang:>10} |{role}{label}");
}
fn print_available_subtitles_adaptation(a: &AdaptationSet) {
a.representations.iter()
.for_each(|r| print_available_subtitles_representation(r, a));
}
// The AdaptationSet a is the parent of the Representation r.
fn print_available_streams_representation(r: &Representation, a: &AdaptationSet, typ: &str) {
// for now, we ignore the Vec representation.SubRepresentation which could contain width, height, bw etc.
let unspecified = "<unspecified>".to_string();
let w = r.width.unwrap_or(a.width.unwrap_or(0));
let h = r.height.unwrap_or(a.height.unwrap_or(0));
let codec = r.codecs.as_ref().unwrap_or(a.codecs.as_ref().unwrap_or(&unspecified));
let bw = r.bandwidth.unwrap_or(a.maxBandwidth.unwrap_or(0));
let fmt = if typ.eq("audio") {
let unknown = String::from("?");
format!("lang={}", r.lang.as_ref().unwrap_or(a.lang.as_ref().unwrap_or(&unknown)))
} else if w == 0 || h == 0 {
// Some MPDs do not specify width and height, such as
// https://dash.akamaized.net/fokus/adinsertion-samples/scte/dash.mpd
String::from("")
} else {
format!("{w}x{h}")
};
let role = a.Role.first()
.map_or_else(|| String::from(""),
|r| r.value.as_ref().map_or_else(|| String::from(""), |v| format!(" role={v}")));
let label = a.Label.first()
.map_or_else(|| String::from(""), |l| format!(" label={}", l.clone().content));
info!(" {typ} {codec:17} | {:5} Kbps | {fmt:>9}{role}{label}", bw / 1024);
}
fn print_available_streams_adaptation(a: &AdaptationSet, typ: &str) {
a.representations.iter()
.for_each(|r| print_available_streams_representation(r, a, typ));
}
fn print_available_streams_period(p: &Period) {
p.adaptations.iter()
.filter(is_audio_adaptation)
.for_each(|a| print_available_streams_adaptation(a, "audio"));
p.adaptations.iter()
.filter(is_video_adaptation)
.for_each(|a| print_available_streams_adaptation(a, "video"));
p.adaptations.iter()
.filter(is_subtitle_adaptation)
.for_each(print_available_subtitles_adaptation);
}
#[tracing::instrument(level="trace", skip_all)]
fn print_available_streams(mpd: &MPD) {
let mut counter = 0;
for p in &mpd.periods {
let mut period_duration_secs: f64 = 0.0;
if let Some(d) = mpd.mediaPresentationDuration {
period_duration_secs = d.as_secs_f64();
}
if let Some(d) = &p.duration {
period_duration_secs = d.as_secs_f64();
}
counter += 1;
if let Some(id) = p.id.as_ref() {
info!("Streams in period {id} (#{counter}), duration {period_duration_secs:.3}s:");
} else {
info!("Streams in period #{counter}, duration {period_duration_secs:.3}s:");
}
print_available_streams_period(p);
}
}
async fn extract_init_pssh(downloader: &DashDownloader, init_url: Url) -> Option<Vec<u8>> {
use bstr::ByteSlice;
use hex_literal::hex;
if let Some(client) = downloader.http_client.as_ref() {
let mut req = client.get(init_url);
if let Some(referer) = &downloader.referer {
req = req.header("Referer", referer);
}
if let Some(username) = &downloader.auth_username {
if let Some(password) = &downloader.auth_password {
req = req.basic_auth(username, Some(password));
}
}
if let Some(token) = &downloader.auth_bearer_token {
req = req.bearer_auth(token);
}
if let Ok(mut resp) = req.send().await {
// We only download the first bytes of the init segment, because it may be very large in the
// case of indexRange adressing, and we don't want to fill up RAM.
let mut chunk_counter = 0;
let mut segment_first_bytes = Vec::<u8>::new();
while let Ok(Some(chunk)) = resp.chunk().await {
let size = min((chunk.len()/1024+1) as u32, u32::MAX);
#[allow(clippy::redundant_pattern_matching)]
if let Err(_) = throttle_download_rate(downloader, size).await {
return None;
}
segment_first_bytes.append(&mut chunk.to_vec());
chunk_counter += 1;
if chunk_counter > 20 {
break;
}
}
let needle = b"pssh";
for offset in segment_first_bytes.find_iter(needle) {
#[allow(clippy::needless_range_loop)]
for i in offset-4..offset+2 {
if let Some(b) = segment_first_bytes.get(i) {
if *b != 0 {
continue;
}
}
}
#[allow(clippy::needless_range_loop)]
for i in offset+4..offset+8 {
if let Some(b) = segment_first_bytes.get(i) {
if *b != 0 {
continue;
}
}
}
if offset+24 > segment_first_bytes.len() {
continue;
}
// const PLAYREADY_SYSID: [u8; 16] = hex!("9a04f07998404286ab92e65be0885f95");
const WIDEVINE_SYSID: [u8; 16] = hex!("edef8ba979d64acea3c827dcd51d21ed");
if let Some(sysid) = segment_first_bytes.get((offset+8)..(offset+24)) {
if !sysid.eq(&WIDEVINE_SYSID) {
continue;
}
}
if let Some(length) = segment_first_bytes.get(offset-1) {
let start = offset - 4;
let end = start + *length as usize;
if let Some(pssh) = &segment_first_bytes.get(start..end) {
return Some(pssh.to_vec());
}
}
}
}
None
} else {
None
}
}
// From https://dashif.org/docs/DASH-IF-IOP-v4.3.pdf:
// "For the avoidance of doubt, only %0[width]d is permitted and no other identifiers. The reason
// is that such a string replacement can be easily implemented without requiring a specific library."
//
// Instead of pulling in C printf() or a reimplementation such as the printf_compat crate, we reimplement
// this functionality directly.
//
// Example template: "$RepresentationID$/$Number%06d$.m4s"
lazy_static! {
static ref URL_TEMPLATE_IDS: Vec<(&'static str, String, Regex)> = {
vec!["RepresentationID", "Number", "Time", "Bandwidth"].into_iter()
.map(|k| (k, format!("${k}$"), Regex::new(&format!("\\${k}%0([\\d])d\\$")).unwrap()))
.collect()
};
}
fn resolve_url_template(template: &str, params: &HashMap<&str, String>) -> String {
let mut result = template.to_string();
for (k, ident, rx) in URL_TEMPLATE_IDS.iter() {
// first check for simple cases such as $Number$
if result.contains(ident) {
if let Some(value) = params.get(k as &str) {
result = result.replace(ident, value);
}
}
// now check for complex cases such as $Number%06d$
if let Some(cap) = rx.captures(&result) {
if let Some(value) = params.get(k as &str) {
if let Ok(width) = cap[1].parse::<usize>() {
if let Some(m) = rx.find(&result) {
let count = format!("{value:0>width$}");
result = result[..m.start()].to_owned() + &count + &result[m.end()..];
}
}
}
}
}
result
}
fn reqwest_error_transient_p(e: &reqwest::Error) -> bool {
if e.is_timeout() {
return true;
}
if let Some(s) = e.status() {
if s == reqwest::StatusCode::REQUEST_TIMEOUT ||
s == reqwest::StatusCode::TOO_MANY_REQUESTS ||
s == reqwest::StatusCode::SERVICE_UNAVAILABLE ||
s == reqwest::StatusCode::GATEWAY_TIMEOUT {
return true;
}
}
false
}
fn categorize_reqwest_error(e: reqwest::Error) -> backoff::Error<reqwest::Error> {
if reqwest_error_transient_p(&e) {
backoff::Error::retry_after(e, Duration::new(5, 0))
} else {
backoff::Error::permanent(e)
}
}
fn notify_transient<E: std::fmt::Debug>(err: E, dur: Duration) {
warn!("Transient error after {dur:?}: {err:?}");
}
fn network_error(why: &str, e: reqwest::Error) -> DashMpdError {
if e.is_timeout() {
DashMpdError::NetworkTimeout(format!("{why}: {e:?}"))
} else if e.is_connect() {
DashMpdError::NetworkConnect(format!("{why}: {e:?}"))
} else {
DashMpdError::Network(format!("{why}: {e:?}"))
}
}
fn parse_error(why: &str, e: impl std::error::Error) -> DashMpdError {
DashMpdError::Parsing(format!("{why}: {e:#?}"))
}
// This would be easier with middleware such as https://lib.rs/crates/tower-reqwest or
// https://lib.rs/crates/reqwest-retry or https://docs.rs/again/latest/again/
// or https://github.com/naomijub/tokio-retry
async fn reqwest_bytes_with_retries(
client: &reqwest::Client,
req: reqwest::Request,
retry_count: u32) -> Result<Bytes, reqwest::Error>
{
let mut last_error = None;
for _ in 0..retry_count {
if let Some(rqw) = req.try_clone() {
match client.execute(rqw).await {
Ok(response) => {
match response.error_for_status() {
Ok(resp) => {
match resp.bytes().await {
Ok(bytes) => return Ok(bytes),
Err(e) => {
info!("Retrying after HTTP error {e:?}");
last_error = Some(e);
},
}
},
Err(e) => {
info!("Retrying after HTTP error {e:?}");
last_error = Some(e);
},
}
},
Err(e) => {
info!("Retrying after HTTP error {e:?}");
last_error = Some(e);
},
}
}
}
Err(last_error.unwrap())
}
// As per https://www.freedesktop.org/wiki/CommonExtendedAttributes/, set extended filesystem
// attributes indicating metadata such as the origin URL, title, source and copyright, if
// specified in the MPD manifest. This functionality is only active on platforms where the xattr
// crate supports extended attributes (currently Android, Linux, MacOS, FreeBSD, and NetBSD); on
// unsupported Unix platforms it's a no-op. On other non-Unix platforms the crate doesn't build.
//
// TODO: on Windows, could use NTFS Alternate Data Streams
// https://en.wikipedia.org/wiki/NTFS#Alternate_data_stream_(ADS)
//
// We could also include a certain amount of metainformation (title, copyright) in the video
// container metadata, though this would have to be implemented separately by each muxing helper and
// each concat helper application in the ffmpeg module.
#[allow(unused_variables)]
fn maybe_record_metainformation(path: &Path, downloader: &DashDownloader, mpd: &MPD) {
#[cfg(target_family = "unix")]
if downloader.record_metainformation && (downloader.fetch_audio || downloader.fetch_video) {
if let Ok(origin_url) = Url::parse(&downloader.mpd_url) {
// Don't record the origin URL if it contains sensitive information such as passwords
#[allow(clippy::collapsible_if)]
if origin_url.username().is_empty() && origin_url.password().is_none() {
#[cfg(target_family = "unix")]
if xattr::set(path, "user.xdg.origin.url", downloader.mpd_url.as_bytes()).is_err() {
info!("Failed to set user.xdg.origin.url xattr on output file");
}
}
if let Some(pi) = &mpd.ProgramInformation {
if let Some(t) = &pi.Title {
if let Some(tc) = &t.content {
if xattr::set(path, "user.dublincore.title", tc.as_bytes()).is_err() {
info!("Failed to set user.dublincore.title xattr on output file");
}
}
}
if let Some(source) = &pi.Source {
if let Some(sc) = &source.content {
if xattr::set(path, "user.dublincore.source", sc.as_bytes()).is_err() {
info!("Failed to set user.dublincore.source xattr on output file");
}
}
}
if let Some(copyright) = &pi.Copyright {
if let Some(cc) = ©right.content {
if xattr::set(path, "user.dublincore.rights", cc.as_bytes()).is_err() {
info!("Failed to set user.dublincore.rights xattr on output file");
}
}
}
}
}
}
}
// From the DASH-IF-IOP-v4.0 specification, "If the value of the @xlink:href attribute is
// urn:mpeg:dash:resolve-to-zero:2013, HTTP GET request is not issued, and the in-MPD element shall
// be removed from the MPD."
fn fetchable_xlink_href(href: &str) -> bool {
(!href.is_empty()) && href.ne("urn:mpeg:dash:resolve-to-zero:2013")
}
fn element_resolves_to_zero(element: &xmltree::Element) -> bool {
element.attributes.get("href")
.is_some_and(|hr| hr.eq("urn:mpeg:dash:resolve-to-zero:2013"))
}
#[derive(Debug)]
struct PendingInsertion {
target: xmltree::XMLNode,
insertions: Vec<xmltree::XMLNode>,
}
fn do_pending_insertions_recurse(
element: &mut xmltree::Element,
pending: &Vec<PendingInsertion>)
{
for pi in pending {
if let Some(idx) = element.children.iter().position(|c| *c == pi.target) {
if pi.insertions.len() == 1 {
element.children[idx] = pi.insertions[0].clone();
} else {
element.children[idx] = pi.insertions[0].clone();
for (i, ins) in pi.insertions[1..].iter().enumerate() {
element.children.insert(idx+i, ins.clone());
}
}
}
}
for child in element.children.iter_mut() {
if let Some(ce) = child.as_mut_element() {
do_pending_insertions_recurse(ce, pending);
}
}
}
// Walk the XML tree recursively to resolve any XLink references in any nodes.
//
// Maintenance note: the xot crate might be a good alternative to the xmltree crate.
#[async_recursion]
async fn resolve_xlink_references_recurse(
downloader: &DashDownloader,
element: &mut xmltree::Element) -> Result<Vec<PendingInsertion>, DashMpdError>
{
let mut pending_insertions = Vec::new();
if let Some(href) = element.attributes.remove("href") {
if fetchable_xlink_href(&href) {
let xlink_url = if is_absolute_url(&href) {
Url::parse(&href)
.map_err(|e| parse_error(&format!("parsing XLink on {}", element.name), e))?
} else {
// Note that we are joining against the original/redirected URL for the MPD, and
// not against the currently scoped BaseURL
let mut merged = downloader.redirected_url.join(&href)
.map_err(|e| parse_error(&format!("parsing XLink on {}", element.name), e))?;
merged.set_query(downloader.redirected_url.query());
merged
};
let client = downloader.http_client.as_ref().unwrap();
trace!("Fetching XLinked element {}", xlink_url.clone());
let mut req = client.get(xlink_url.clone())
.header("Accept", "application/dash+xml,video/vnd.mpeg.dash.mpd")
.header("Accept-Language", "en-US,en")
.header("Sec-Fetch-Mode", "navigate");
if let Some(referer) = &downloader.referer {
req = req.header("Referer", referer);
} else {
req = req.header("Referer", downloader.redirected_url.to_string());
}
if let Some(username) = &downloader.auth_username {
if let Some(password) = &downloader.auth_password {
req = req.basic_auth(username, Some(password));
}
}
if let Some(token) = &downloader.auth_bearer_token {
req = req.bearer_auth(token);
}
let xml = req.send().await
.map_err(|e| network_error(&format!("fetching XLink for {}", element.name), e))?
.error_for_status()
.map_err(|e| network_error(&format!("fetching XLink for {}", element.name), e))?
.text().await
.map_err(|e| network_error(&format!("resolving XLink on {}", element.name), e))?;
if downloader.verbosity > 2 {
info!(" Resolved onLoad XLink {xlink_url} on {} -> {} octets",
element.name, xml.len());
}
// The difficulty here is that the XML fragment received may contain multiple elements,
// for example a Period with xlink resolves to two Period elements. For a single
// resolved element we can simply replace the original element by its resolved
// counterpart. When the xlink resolves to multiple elements, we can't insert them back
// into the parent node directly, but need to return them to the caller for later insertion.
let nodes = xmltree::Element::parse_all(xml.as_bytes())
.map_err(|e| parse_error("xmltree parsing", e))?;
pending_insertions.push(PendingInsertion {
target: xmltree::XMLNode::Element(element.clone()),
insertions: nodes,
});
}
}
// Delete any child Elements that have XLink resolve-to-zero semantics.
element.children.retain(
|n| n.as_element().is_none() ||
n.as_element().is_some_and(|e| !element_resolves_to_zero(e)));
for child in element.children.iter_mut() {
if let Some(ce) = child.as_mut_element() {
let pending = resolve_xlink_references_recurse(downloader, ce).await?;
for p in pending {
pending_insertions.push(p);
}
}
}
Ok(pending_insertions)
}
#[tracing::instrument(level="trace", skip_all)]
pub async fn parse_resolving_xlinks(
downloader: &DashDownloader,
xml: &[u8]) -> Result<MPD, DashMpdError>
{
let mut doc = xmltree::Element::parse(xml)
.map_err(|e| parse_error("xmltree parsing", e))?;
if !doc.name.eq("MPD") {
return Err(DashMpdError::Parsing(format!("root element is {}, expecting <MPD>", doc.name)));
}
// The remote XLink fragments may contain further XLink references. However, we only repeat the
// resolution 5 times to avoid potential infloop DoS attacks.
for _ in 1..5 {
let pending = resolve_xlink_references_recurse(downloader, &mut doc).await?;
do_pending_insertions_recurse(&mut doc, &pending);
}
let mut buf = Vec::new();
doc.write(&mut buf)
.map_err(|e| parse_error("serializing rewritten manifest", e))?;
// Run user-specified XSLT stylesheets on the manifest, using xsltproc (a component of libxslt)
// as a commandline filter application. Existing XSLT implementations in Rust are incomplete
// (but improving; hopefully we will one day be able to use the xrust crate).
for ss in &downloader.xslt_stylesheets {
if downloader.verbosity > 0 {
info!(" Applying XSLT stylesheet {} with xsltproc", ss.display());
}
let tmpmpd = tmp_file_path("dashxslt", OsStr::new("xslt"))?;
fs::write(&tmpmpd, &buf)
.map_err(|e| DashMpdError::Io(e, String::from("writing MPD")))?;
let xsltproc = Command::new("xsltproc")
.args([ss, &tmpmpd])
.output()
.map_err(|e| DashMpdError::Io(e, String::from("spawning xsltproc")))?;
if !xsltproc.status.success() {
let msg = format!("xsltproc returned {}", xsltproc.status);
let out = partial_process_output(&xsltproc.stderr).to_string();
return Err(DashMpdError::Io(std::io::Error::new(std::io::ErrorKind::Other, msg), out));
}
if let Err(e) = fs::remove_file(&tmpmpd) {
warn!("Error removing temporary MPD after XSLT processing: {e:?}");
}
buf.clone_from(&xsltproc.stdout);
}
let rewritten = std::str::from_utf8(&buf)
.map_err(|e| parse_error("parsing UTF-8", e))?;
// Here using the quick-xml serde support to deserialize into Rust structs.
let mpd = parse(rewritten)?;
if downloader.conformity_checks {
for emsg in check_conformity(&mpd) {
warn!("DASH conformity error in manifest: {emsg}");
}
}
Ok(mpd)
}
async fn do_segmentbase_indexrange(
downloader: &DashDownloader,
period_counter: u8,
base_url: Url,
sb: &SegmentBase,
dict: &HashMap<&str, String>
) -> Result<Vec<MediaFragment>, DashMpdError>
{
// Something like the following
//
// <SegmentBase indexRange="839-3534" timescale="12288">
// <Initialization range="0-838"/>
// </SegmentBase>
//
// The SegmentBase@indexRange attribute points to a byte range in the media file
// that contains index information (an sidx box for MPEG files, or a Cues entry for
// a DASH-WebM stream). There are two possible strategies to implement when downloading this content:
//
// - Simply download the full content specified by the BaseURL element for this
// segment (ignoring the indexRange attribute).
//
// - Download the sidx box using a Range request, parse the segment references it
// contains, and download each one using a different Range request, and
// concatenate the full contents.
//
// The first option is what a browser-based player does. It avoids making a huge
// segment download that will fill up our RAM if chunked download is not offered by
// the server. It works with web servers that prevent direct access to the full
// MP4/WebM file by blocking requests without a limited byte range. Its more
// correct, because in theory the content at BaseURL might contain lots of
// irrelevant information which is not pointed to by any of the sidx byte ranges.
// However, it is a little more fragile because some MP4 elements that are necessary
// to create a valid MP4 file (e.g. trex, trun, tfhd boxes) might not be included in
// the sidx-referenced byte ranges.
//
// In practice, it seems that the indexRange information is mostly provided by DASH
// encoders to allow clients to rewind and fast-forward a stream, and both
// strategies work. We default to using the indexRange information, but include the
// option parse_index_range to allow fallback to the simpler "download-it-all"
// strategy.
let mut fragments = Vec::new();
let mut start_byte: Option<u64> = None;
let mut end_byte: Option<u64> = None;
let mut indexable_segments = false;
if downloader.use_index_range {
if let Some(ir) = &sb.indexRange {
// Fetch the octet slice corresponding to the (sidx) index.
let (s, e) = parse_range(ir)?;
trace!("Fetching sidx for {}", base_url.clone());
let mut req = downloader.http_client.as_ref()
.unwrap()
.get(base_url.clone())
.header(RANGE, format!("bytes={s}-{e}"))
.header("Referer", downloader.redirected_url.to_string())
.header("Sec-Fetch-Mode", "navigate");
if let Some(username) = &downloader.auth_username {
if let Some(password) = &downloader.auth_password {
req = req.basic_auth(username, Some(password));
}
}
if let Some(token) = &downloader.auth_bearer_token {
req = req.bearer_auth(token);
}
let mut resp = req.send().await
.map_err(|e| network_error("fetching index data", e))?
.error_for_status()
.map_err(|e| network_error("fetching index data", e))?;
let headers = std::mem::take(resp.headers_mut());
if let Some(content_type) = headers.get(CONTENT_TYPE) {
let idx = resp.bytes().await
.map_err(|e| network_error("fetching index data", e))?;
if idx.len() as u64 != e - s + 1 {
warn!(" HTTP server does not support Range requests; can't use indexRange addressing");
} else {
#[allow(clippy::collapsible_else_if)]
if content_type.eq("video/mp4") ||
content_type.eq("audio/mp4") {
// Handle as ISOBMFF. First prepare to save the index data itself
// and any leading bytes (from byte positions 0 to s) to the output
// container, because it may contain other types of MP4 boxes than
// only sidx boxes (eg. trex, trun tfhd boxes), which are necessary
// to play the media content. Then prepare to save each referenced
// segment chunk to the output container.
let mf = MediaFragmentBuilder::new(period_counter, base_url.clone())
.with_range(Some(0), Some(e))
.build();
fragments.push(mf);
let mut max_chunk_pos = 0;
if let Ok(segment_chunks) = crate::sidx::from_isobmff_sidx(&idx, e+1) {
trace!("Have {} segment chunks in sidx data", segment_chunks.len());
for chunk in segment_chunks {
let mf = MediaFragmentBuilder::new(period_counter, base_url.clone())
.with_range(Some(chunk.start), Some(chunk.end))
.build();
fragments.push(mf);
if chunk.end > max_chunk_pos {
max_chunk_pos = chunk.end;
}
}
indexable_segments = true;
}
}
// In theory we should also be able to handle Cue data in a WebM media
// stream similarly to chunks specified by an sidx box in an ISOBMFF/MP4
// container. However, simply appending the content pointed to by the
// different Cue elements in the WebM file leads to an invalid media
// file. We need to implement more complicated logic to reconstruct a
// valid WebM file from chunks of content.
}
}
}
}
if indexable_segments {
if let Some(init) = &sb.initialization
{
if let Some(range) = &init.range {
let (s, e) = parse_range(range)?;
start_byte = Some(s);
end_byte = Some(e);
}
if let Some(su) = &init.sourceURL {
let path = resolve_url_template(su, dict);
let u = merge_baseurls(&base_url, &path)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.with_range(start_byte, end_byte)
.set_init()
.build();
fragments.push(mf);
} else {
// Use the current BaseURL
let mf = MediaFragmentBuilder::new(period_counter, base_url.clone())
.with_range(start_byte, end_byte)
.set_init()
.build();
fragments.push(mf);
}
}
} else {
// If anything prevented us from handling this SegmentBase@indexRange element using
// HTTP Range requests, just download the whole segment as a single chunk. This is
// likely to be a large HTTP request (for instance, the full video content as a
// single MP4 file), so we increase our network request timeout.
trace!("Falling back to retrieving full SegmentBase for {}", base_url.clone());
let mf = MediaFragmentBuilder::new(period_counter, base_url.clone())
.with_timeout(Duration::new(10_000, 0))
.build();
fragments.push(mf);
}
Ok(fragments)
}
#[tracing::instrument(level="trace", skip_all)]
async fn do_period_audio(
downloader: &DashDownloader,
mpd: &MPD,
period: &Period,
period_counter: u8,
base_url: Url
) -> Result<PeriodOutputs, DashMpdError>
{
let mut fragments = Vec::new();
let mut diagnostics = Vec::new();
let mut opt_init: Option<String> = None;
let mut opt_media: Option<String> = None;
let mut opt_duration: Option<f64> = None;
let mut timescale = 1;
let mut start_number = 1;
// The period_duration is specified either by the <Period> duration attribute, or by the
// mediaPresentationDuration of the top-level MPD node.
let mut period_duration_secs: f64 = 0.0;
if let Some(d) = mpd.mediaPresentationDuration {
period_duration_secs = d.as_secs_f64();
}
if let Some(d) = period.duration {
period_duration_secs = d.as_secs_f64();
}
if let Some(s) = downloader.force_duration {
period_duration_secs = s;
}
// SegmentTemplate as a direct child of a Period element. This can specify some common attribute
// values (media, timescale, duration, startNumber) for child SegmentTemplate nodes in an
// enclosed AdaptationSet or Representation node.
if let Some(st) = &period.SegmentTemplate {
if let Some(i) = &st.initialization {
opt_init = Some(i.to_string());
}
if let Some(m) = &st.media {
opt_media = Some(m.to_string());
}
if let Some(d) = st.duration {
opt_duration = Some(d);
}
if let Some(ts) = st.timescale {
timescale = ts;
}
if let Some(s) = st.startNumber {
start_number = s;
}
}
// Handle the AdaptationSet with audio content. Note that some streams don't separate out
// audio and video streams, so this might be None.
let audio_adaptations: Vec<&AdaptationSet> = period.adaptations.iter()
.filter(is_audio_adaptation)
.collect();
let representations: Vec<&Representation> = select_preferred_adaptations(audio_adaptations, downloader)
.iter()
.flat_map(|a| a.representations.iter())
.collect();
if let Some(audio_repr) = select_preferred_representation(representations, downloader) {
// Find the AdaptationSet that is the parent of the selected Representation. This may be
// needed for certain Representation attributes whose value can be located higher in the XML
// tree.
let audio_adaptation = period.adaptations.iter()
.find(|a| a.representations.iter().any(|r| r.eq(audio_repr)))
.unwrap();
// The AdaptationSet may have a BaseURL (e.g. the test BBC streams). We use a local variable
// to make sure we don't "corrupt" the base_url for the video segments.
let mut base_url = base_url.clone();
if let Some(bu) = &audio_adaptation.BaseURL.first() {
base_url = merge_baseurls(&base_url, &bu.base)?;
}
if let Some(bu) = audio_repr.BaseURL.first() {
base_url = merge_baseurls(&base_url, &bu.base)?;
}
if downloader.verbosity > 0 {
let bw = if let Some(bw) = audio_repr.bandwidth {
format!("bw={} Kbps ", bw / 1024)
} else {
String::from("")
};
let unknown = String::from("?");
let lang = audio_repr.lang.as_ref()
.unwrap_or(audio_adaptation.lang.as_ref()
.unwrap_or(&unknown));
let codec = audio_repr.codecs.as_ref()
.unwrap_or(audio_adaptation.codecs.as_ref()
.unwrap_or(&unknown));
diagnostics.push(format!(" Audio stream selected: {bw}lang={lang} codec={codec}"));
// Check for ContentProtection on the selected Representation/Adaptation
for cp in audio_repr.ContentProtection.iter()
.chain(audio_adaptation.ContentProtection.iter())
{
diagnostics.push(format!(" ContentProtection: {}", content_protection_type(cp)));
if let Some(kid) = &cp.default_KID {
diagnostics.push(format!(" KID: {}", kid.replace('-', "")));
}
for pssh_element in cp.cenc_pssh.iter() {
if let Some(pssh_b64) = &pssh_element.content {
diagnostics.push(format!(" PSSH (from manifest): {pssh_b64}"));
if let Ok(pssh) = pssh_box::from_base64(pssh_b64) {
diagnostics.push(format!(" {pssh}"));
}
}
}
}
}
// SegmentTemplate as a direct child of an Adaptation node. This can specify some common
// attribute values (media, timescale, duration, startNumber) for child SegmentTemplate
// nodes in an enclosed Representation node. Don't download media segments here, only
// download for SegmentTemplate nodes that are children of a Representation node.
if let Some(st) = &audio_adaptation.SegmentTemplate {
if let Some(i) = &st.initialization {
opt_init = Some(i.to_string());
}
if let Some(m) = &st.media {
opt_media = Some(m.to_string());
}
if let Some(d) = st.duration {
opt_duration = Some(d);
}
if let Some(ts) = st.timescale {
timescale = ts;
}
if let Some(s) = st.startNumber {
start_number = s;
}
}
let mut dict = HashMap::new();
if let Some(rid) = &audio_repr.id {
dict.insert("RepresentationID", rid.to_string());
}
if let Some(b) = &audio_repr.bandwidth {
dict.insert("Bandwidth", b.to_string());
}
// Now the 6 possible addressing modes: (1) SegmentList,
// (2) SegmentTemplate+SegmentTimeline, (3) SegmentTemplate@duration,
// (4) SegmentTemplate@index, (5) SegmentBase@indexRange, (6) plain BaseURL
// Though SegmentBase and SegmentList addressing modes are supposed to be
// mutually exclusive, some manifests in the wild use both. So we try to work
// around the brokenness.
// Example: http://ftp.itec.aau.at/datasets/mmsys12/ElephantsDream/MPDs/ElephantsDreamNonSeg_6s_isoffmain_DIS_23009_1_v_2_1c2_2011_08_30.mpd
if let Some(sl) = &audio_adaptation.SegmentList {
// (1) AdaptationSet>SegmentList addressing mode (can be used in conjunction
// with Representation>SegmentList addressing mode)
if downloader.verbosity > 1 {
info!(" {}", "Using AdaptationSet>SegmentList addressing mode for audio representation".italic());
}
let mut start_byte: Option<u64> = None;
let mut end_byte: Option<u64> = None;
if let Some(init) = &sl.Initialization {
if let Some(range) = &init.range {
let (s, e) = parse_range(range)?;
start_byte = Some(s);
end_byte = Some(e);
}
if let Some(su) = &init.sourceURL {
let path = resolve_url_template(su, &dict);
let init_url = merge_baseurls(&base_url, &path)?;
let mf = MediaFragmentBuilder::new(period_counter, init_url)
.with_range(start_byte, end_byte)
.set_init()
.build();
fragments.push(mf);
} else {
let mf = MediaFragmentBuilder::new(period_counter, base_url.clone())
.with_range(start_byte, end_byte)
.set_init()
.build();
fragments.push(mf);
}
}
for su in sl.segment_urls.iter() {
start_byte = None;
end_byte = None;
// we are ignoring SegmentURL@indexRange
if let Some(range) = &su.mediaRange {
let (s, e) = parse_range(range)?;
start_byte = Some(s);
end_byte = Some(e);
}
if let Some(m) = &su.media {
let u = merge_baseurls(&base_url, m)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.with_range(start_byte, end_byte)
.build();
fragments.push(mf);
} else if let Some(bu) = audio_adaptation.BaseURL.first() {
let u = merge_baseurls(&base_url, &bu.base)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.with_range(start_byte, end_byte)
.build();
fragments.push(mf);
}
}
}
if let Some(sl) = &audio_repr.SegmentList {
// (1) Representation>SegmentList addressing mode
if downloader.verbosity > 1 {
info!(" {}", "Using Representation>SegmentList addressing mode for audio representation".italic());
}
let mut start_byte: Option<u64> = None;
let mut end_byte: Option<u64> = None;
if let Some(init) = &sl.Initialization {
if let Some(range) = &init.range {
let (s, e) = parse_range(range)?;
start_byte = Some(s);
end_byte = Some(e);
}
if let Some(su) = &init.sourceURL {
let path = resolve_url_template(su, &dict);
let init_url = merge_baseurls(&base_url, &path)?;
let mf = MediaFragmentBuilder::new(period_counter, init_url)
.with_range(start_byte, end_byte)
.set_init()
.build();
fragments.push(mf);
} else {
let mf = MediaFragmentBuilder::new(period_counter, base_url.clone())
.with_range(start_byte, end_byte)
.set_init()
.build();
fragments.push(mf);
}
}
for su in sl.segment_urls.iter() {
start_byte = None;
end_byte = None;
// we are ignoring SegmentURL@indexRange
if let Some(range) = &su.mediaRange {
let (s, e) = parse_range(range)?;
start_byte = Some(s);
end_byte = Some(e);
}
if let Some(m) = &su.media {
let u = merge_baseurls(&base_url, m)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.with_range(start_byte, end_byte)
.build();
fragments.push(mf);
} else if let Some(bu) = audio_repr.BaseURL.first() {
let u = merge_baseurls(&base_url, &bu.base)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.with_range(start_byte, end_byte)
.build();
fragments.push(mf);
}
}
} else if audio_repr.SegmentTemplate.is_some() ||
audio_adaptation.SegmentTemplate.is_some()
{
// Here we are either looking at a Representation.SegmentTemplate, or a
// higher-level AdaptationSet.SegmentTemplate
let st;
if let Some(it) = &audio_repr.SegmentTemplate {
st = it;
} else if let Some(it) = &audio_adaptation.SegmentTemplate {
st = it;
} else {
panic!("unreachable");
}
if let Some(i) = &st.initialization {
opt_init = Some(i.to_string());
}
if let Some(m) = &st.media {
opt_media = Some(m.to_string());
}
if let Some(ts) = st.timescale {
timescale = ts;
}
if let Some(sn) = st.startNumber {
start_number = sn;
}
if let Some(stl) = &audio_repr.SegmentTemplate.as_ref().and_then(|st| st.SegmentTimeline.clone())
.or(audio_adaptation.SegmentTemplate.as_ref().and_then(|st| st.SegmentTimeline.clone()))
{
// (2) SegmentTemplate with SegmentTimeline addressing mode (also called
// "explicit addressing" in certain DASH-IF documents)
if downloader.verbosity > 1 {
info!(" {}", "Using SegmentTemplate+SegmentTimeline addressing mode for audio representation".italic());
}
if let Some(init) = opt_init {
let path = resolve_url_template(&init, &dict);
let u = merge_baseurls(&base_url, &path)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.set_init()
.build();
fragments.push(mf);
}
if let Some(media) = opt_media {
let audio_path = resolve_url_template(&media, &dict);
let mut segment_time = 0;
let mut segment_duration;
let mut number = start_number;
for s in &stl.segments {
if let Some(t) = s.t {
segment_time = t;
}
segment_duration = s.d;
// the URLTemplate may be based on $Time$, or on $Number$
let dict = HashMap::from([("Time", segment_time.to_string()),
("Number", number.to_string())]);
let path = resolve_url_template(&audio_path, &dict);
let u = merge_baseurls(&base_url, &path)?;
fragments.push(MediaFragmentBuilder::new(period_counter, u).build());
number += 1;
if let Some(r) = s.r {
let mut count = 0i64;
// FIXME perhaps we also need to account for startTime?
let end_time = period_duration_secs * timescale as f64;
loop {
count += 1;
// Exit from the loop after @r iterations (if @r is
// positive). A negative value of the @r attribute indicates
// that the duration indicated in @d attribute repeats until
// the start of the next S element, the end of the Period or
// until the next MPD update.
if r >= 0 {
if count > r {
break;
}
if downloader.force_duration.is_some() && segment_time as f64 > end_time {
break;
}
} else if segment_time as f64 > end_time {
break;
}
segment_time += segment_duration;
let dict = HashMap::from([("Time", segment_time.to_string()),
("Number", number.to_string())]);
let path = resolve_url_template(&audio_path, &dict);
let u = merge_baseurls(&base_url, &path)?;
fragments.push(MediaFragmentBuilder::new(period_counter, u).build());
number += 1;
}
}
segment_time += segment_duration;
}
} else {
return Err(DashMpdError::UnhandledMediaStream(
"SegmentTimeline without a media attribute".to_string()));
}
} else { // no SegmentTimeline element
// (3) SegmentTemplate@duration addressing mode or (4) SegmentTemplate@index
// addressing mode (also called "simple addressing" in certain DASH-IF
// documents)
if downloader.verbosity > 1 {
info!(" {}", "Using SegmentTemplate addressing mode for audio representation".italic());
}
let mut total_number = 0i64;
if let Some(init) = opt_init {
let path = resolve_url_template(&init, &dict);
let u = merge_baseurls(&base_url, &path)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.set_init()
.build();
fragments.push(mf);
}
if let Some(media) = opt_media {
let audio_path = resolve_url_template(&media, &dict);
let timescale = st.timescale.unwrap_or(timescale);
let mut segment_duration: f64 = -1.0;
if let Some(d) = opt_duration {
// it was set on the Period.SegmentTemplate node
segment_duration = d;
}
if let Some(std) = st.duration {
segment_duration = std / timescale as f64;
}
if segment_duration < 0.0 {
return Err(DashMpdError::UnhandledMediaStream(
"Audio representation is missing SegmentTemplate@duration attribute".to_string()));
}
total_number += (period_duration_secs / segment_duration).round() as i64;
let mut number = start_number;
for _ in 1..=total_number {
let dict = HashMap::from([("Number", number.to_string())]);
let path = resolve_url_template(&audio_path, &dict);
let u = merge_baseurls(&base_url, &path)?;
fragments.push(MediaFragmentBuilder::new(period_counter, u).build());
number += 1;
}
}
}
} else if let Some(sb) = &audio_repr.SegmentBase {
// (5) SegmentBase@indexRange addressing mode
if downloader.verbosity > 1 {
info!(" {}", "Using SegmentBase@indexRange addressing mode for audio representation".italic());
}
let mf = do_segmentbase_indexrange(downloader, period_counter, base_url, sb, &dict).await?;
fragments.extend(mf);
} else if fragments.is_empty() {
if let Some(bu) = audio_repr.BaseURL.first() {
// (6) plain BaseURL addressing mode
if downloader.verbosity > 1 {
info!(" {}", "Using BaseURL addressing mode for audio representation".italic());
}
let u = merge_baseurls(&base_url, &bu.base)?;
fragments.push(MediaFragmentBuilder::new(period_counter, u).build());
}
}
if fragments.is_empty() {
return Err(DashMpdError::UnhandledMediaStream(
"no usable addressing mode identified for audio representation".to_string()));
}
}
Ok(PeriodOutputs { fragments, diagnostics, subtitle_formats: Vec::new() })
}
#[tracing::instrument(level="trace", skip_all)]
async fn do_period_video(
downloader: &DashDownloader,
mpd: &MPD,
period: &Period,
period_counter: u8,
base_url: Url
) -> Result<PeriodOutputs, DashMpdError>
{
let mut fragments = Vec::new();
let mut diagnostics = Vec::new();
let mut period_duration_secs: f64 = 0.0;
let mut opt_init: Option<String> = None;
let mut opt_media: Option<String> = None;
let mut opt_duration: Option<f64> = None;
let mut timescale = 1;
let mut start_number = 1;
if let Some(d) = mpd.mediaPresentationDuration {
period_duration_secs = d.as_secs_f64();
}
if let Some(d) = period.duration {
period_duration_secs = d.as_secs_f64();
}
if let Some(s) = downloader.force_duration {
period_duration_secs = s;
}
// SegmentTemplate as a direct child of a Period element. This can specify some common attribute
// values (media, timescale, duration, startNumber) for child SegmentTemplate nodes in an
// enclosed AdaptationSet or Representation node.
if let Some(st) = &period.SegmentTemplate {
if let Some(i) = &st.initialization {
opt_init = Some(i.to_string());
}
if let Some(m) = &st.media {
opt_media = Some(m.to_string());
}
if let Some(d) = st.duration {
opt_duration = Some(d);
}
if let Some(ts) = st.timescale {
timescale = ts;
}
if let Some(s) = st.startNumber {
start_number = s;
}
}
// A manifest may contain multiple AdaptationSets with video content (in particular, when
// different codecs are offered). Each AdaptationSet often contains multiple video
// Representations with different bandwidths and video resolutions. We select the Representation
// to download by ranking the available streams according to the preferred width specified by
// the user, or by the preferred height specified by the user, or by the user's specified
// quality preference.
let video_adaptations: Vec<&AdaptationSet> = period.adaptations.iter()
.filter(is_video_adaptation)
.collect();
let representations: Vec<&Representation> = select_preferred_adaptations(video_adaptations, downloader)
.iter()
.flat_map(|a| a.representations.iter())
.collect();
let maybe_video_repr = if let Some(want) = downloader.video_width_preference {
representations.iter()
.min_by_key(|x| if let Some(w) = x.width { want.abs_diff(w) } else { u64::MAX })
.copied()
} else if let Some(want) = downloader.video_height_preference {
representations.iter()
.min_by_key(|x| if let Some(h) = x.height { want.abs_diff(h) } else { u64::MAX })
.copied()
} else {
select_preferred_representation(representations, downloader)
};
if let Some(video_repr) = maybe_video_repr {
// Find the AdaptationSet that is the parent of the selected Representation. This may be
// needed for certain Representation attributes whose value can be located higher in the XML
// tree.
let video_adaptation = period.adaptations.iter()
.find(|a| a.representations.iter().any(|r| r.eq(video_repr)))
.unwrap();
// The AdaptationSet may have a BaseURL. We use a local variable to make sure we
// don't "corrupt" the base_url for the subtitle segments.
let mut base_url = base_url.clone();
if let Some(bu) = &video_adaptation.BaseURL.first() {
base_url = merge_baseurls(&base_url, &bu.base)?;
}
if let Some(bu) = &video_repr.BaseURL.first() {
base_url = merge_baseurls(&base_url, &bu.base)?;
}
if downloader.verbosity > 0 {
let bw = if let Some(bw) = video_repr.bandwidth.or(video_adaptation.maxBandwidth) {
format!("bw={} Kbps ", bw / 1024)
} else {
String::from("")
};
let unknown = String::from("?");
let w = video_repr.width.unwrap_or(video_adaptation.width.unwrap_or(0));
let h = video_repr.height.unwrap_or(video_adaptation.height.unwrap_or(0));
let fmt = if w == 0 || h == 0 {
String::from("")
} else {
format!("resolution={w}x{h} ")
};
let codec = video_repr.codecs.as_ref()
.unwrap_or(video_adaptation.codecs.as_ref().unwrap_or(&unknown));
diagnostics.push(format!(" Video stream selected: {bw}{fmt}codec={codec}"));
// Check for ContentProtection on the selected Representation/Adaptation
for cp in video_repr.ContentProtection.iter()
.chain(video_adaptation.ContentProtection.iter())
{
diagnostics.push(format!(" ContentProtection: {}", content_protection_type(cp)));
if let Some(kid) = &cp.default_KID {
diagnostics.push(format!(" KID: {}", kid.replace('-', "")));
}
for pssh_element in cp.cenc_pssh.iter() {
if let Some(pssh_b64) = &pssh_element.content {
diagnostics.push(format!(" PSSH (from manifest): {pssh_b64}"));
if let Ok(pssh) = pssh_box::from_base64(pssh_b64) {
diagnostics.push(format!(" {pssh}"));
}
}
}
}
}
let mut dict = HashMap::new();
if let Some(rid) = &video_repr.id {
dict.insert("RepresentationID", rid.to_string());
}
if let Some(b) = &video_repr.bandwidth {
dict.insert("Bandwidth", b.to_string());
}
// SegmentTemplate as a direct child of an Adaptation node. This can specify some common
// attribute values (media, timescale, duration, startNumber) for child SegmentTemplate
// nodes in an enclosed Representation node. Don't download media segments here, only
// download for SegmentTemplate nodes that are children of a Representation node.
if let Some(st) = &video_adaptation.SegmentTemplate {
if let Some(i) = &st.initialization {
opt_init = Some(i.to_string());
}
if let Some(m) = &st.media {
opt_media = Some(m.to_string());
}
if let Some(d) = st.duration {
opt_duration = Some(d);
}
if let Some(ts) = st.timescale {
timescale = ts;
}
if let Some(s) = st.startNumber {
start_number = s;
}
}
// Now the 6 possible addressing modes: (1) SegmentList,
// (2) SegmentTemplate+SegmentTimeline, (3) SegmentTemplate@duration,
// (4) SegmentTemplate@index, (5) SegmentBase@indexRange, (6) plain BaseURL
if let Some(sl) = &video_adaptation.SegmentList {
// (1) AdaptationSet>SegmentList addressing mode
if downloader.verbosity > 1 {
info!(" {}", "Using AdaptationSet>SegmentList addressing mode for video representation".italic());
}
let mut start_byte: Option<u64> = None;
let mut end_byte: Option<u64> = None;
if let Some(init) = &sl.Initialization {
if let Some(range) = &init.range {
let (s, e) = parse_range(range)?;
start_byte = Some(s);
end_byte = Some(e);
}
if let Some(su) = &init.sourceURL {
let path = resolve_url_template(su, &dict);
let u = merge_baseurls(&base_url, &path)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.with_range(start_byte, end_byte)
.set_init()
.build();
fragments.push(mf);
}
} else {
let mf = MediaFragmentBuilder::new(period_counter, base_url.clone())
.with_range(start_byte, end_byte)
.set_init()
.build();
fragments.push(mf);
}
for su in sl.segment_urls.iter() {
start_byte = None;
end_byte = None;
// we are ignoring @indexRange
if let Some(range) = &su.mediaRange {
let (s, e) = parse_range(range)?;
start_byte = Some(s);
end_byte = Some(e);
}
if let Some(m) = &su.media {
let u = merge_baseurls(&base_url, m)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.with_range(start_byte, end_byte)
.build();
fragments.push(mf);
} else if let Some(bu) = video_adaptation.BaseURL.first() {
let u = merge_baseurls(&base_url, &bu.base)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.with_range(start_byte, end_byte)
.build();
fragments.push(mf);
}
}
}
if let Some(sl) = &video_repr.SegmentList {
// (1) Representation>SegmentList addressing mode
if downloader.verbosity > 1 {
info!(" {}", "Using Representation>SegmentList addressing mode for video representation".italic());
}
let mut start_byte: Option<u64> = None;
let mut end_byte: Option<u64> = None;
if let Some(init) = &sl.Initialization {
if let Some(range) = &init.range {
let (s, e) = parse_range(range)?;
start_byte = Some(s);
end_byte = Some(e);
}
if let Some(su) = &init.sourceURL {
let path = resolve_url_template(su, &dict);
let u = merge_baseurls(&base_url, &path)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.with_range(start_byte, end_byte)
.set_init()
.build();
fragments.push(mf);
} else {
let mf = MediaFragmentBuilder::new(period_counter, base_url.clone())
.with_range(start_byte, end_byte)
.set_init()
.build();
fragments.push(mf);
}
}
for su in sl.segment_urls.iter() {
start_byte = None;
end_byte = None;
// we are ignoring @indexRange
if let Some(range) = &su.mediaRange {
let (s, e) = parse_range(range)?;
start_byte = Some(s);
end_byte = Some(e);
}
if let Some(m) = &su.media {
let u = merge_baseurls(&base_url, m)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.with_range(start_byte, end_byte)
.build();
fragments.push(mf);
} else if let Some(bu) = video_repr.BaseURL.first() {
let u = merge_baseurls(&base_url, &bu.base)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.with_range(start_byte, end_byte)
.build();
fragments.push(mf);
}
}
} else if video_repr.SegmentTemplate.is_some() ||
video_adaptation.SegmentTemplate.is_some() {
// Here we are either looking at a Representation.SegmentTemplate, or a
// higher-level AdaptationSet.SegmentTemplate
let st;
if let Some(it) = &video_repr.SegmentTemplate {
st = it;
} else if let Some(it) = &video_adaptation.SegmentTemplate {
st = it;
} else {
panic!("impossible");
}
if let Some(i) = &st.initialization {
opt_init = Some(i.to_string());
}
if let Some(m) = &st.media {
opt_media = Some(m.to_string());
}
if let Some(ts) = st.timescale {
timescale = ts;
}
if let Some(sn) = st.startNumber {
start_number = sn;
}
if let Some(stl) = &video_repr.SegmentTemplate.as_ref().and_then(|st| st.SegmentTimeline.clone())
.or(video_adaptation.SegmentTemplate.as_ref().and_then(|st| st.SegmentTimeline.clone()))
{
// (2) SegmentTemplate with SegmentTimeline addressing mode
if downloader.verbosity > 1 {
info!(" {}", "Using SegmentTemplate+SegmentTimeline addressing mode for video representation".italic());
}
if let Some(init) = opt_init {
let path = resolve_url_template(&init, &dict);
let u = merge_baseurls(&base_url, &path)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.set_init()
.build();
fragments.push(mf);
}
if let Some(media) = opt_media {
let video_path = resolve_url_template(&media, &dict);
let mut segment_time = 0;
let mut segment_duration;
let mut number = start_number;
// FIXME for a live manifest, need to look at the time elapsed since now and
// the mpd.availabilityStartTime to determine the correct value for
// startNumber, based on duration and timescale.
for s in &stl.segments {
if let Some(t) = s.t {
segment_time = t;
}
segment_duration = s.d;
// the URLTemplate may be based on $Time$, or on $Number$
let dict = HashMap::from([("Time", segment_time.to_string()),
("Number", number.to_string())]);
let path = resolve_url_template(&video_path, &dict);
let u = merge_baseurls(&base_url, &path)?;
let mf = MediaFragmentBuilder::new(period_counter, u).build();
fragments.push(mf);
number += 1;
if let Some(r) = s.r {
let mut count = 0i64;
// FIXME perhaps we also need to account for startTime?
let end_time = period_duration_secs * timescale as f64;
loop {
count += 1;
// Exit from the loop after @r iterations (if @r is
// positive). A negative value of the @r attribute indicates
// that the duration indicated in @d attribute repeats until
// the start of the next S element, the end of the Period or
// until the next MPD update.
if r >= 0 {
if count > r {
break;
}
if downloader.force_duration.is_some() && segment_time as f64 > end_time {
break;
}
} else if segment_time as f64 > end_time {
break;
}
segment_time += segment_duration;
let dict = HashMap::from([("Time", segment_time.to_string()),
("Number", number.to_string())]);
let path = resolve_url_template(&video_path, &dict);
let u = merge_baseurls(&base_url, &path)?;
let mf = MediaFragmentBuilder::new(period_counter, u).build();
fragments.push(mf);
number += 1;
}
}
segment_time += segment_duration;
}
} else {
return Err(DashMpdError::UnhandledMediaStream(
"SegmentTimeline without a media attribute".to_string()));
}
} else { // no SegmentTimeline element
// (3) SegmentTemplate@duration addressing mode or (4) SegmentTemplate@index addressing mode
if downloader.verbosity > 1 {
info!(" {}", "Using SegmentTemplate addressing mode for video representation".italic());
}
let mut total_number = 0i64;
if let Some(init) = opt_init {
let path = resolve_url_template(&init, &dict);
let u = merge_baseurls(&base_url, &path)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.set_init()
.build();
fragments.push(mf);
}
if let Some(media) = opt_media {
let video_path = resolve_url_template(&media, &dict);
let timescale = st.timescale.unwrap_or(timescale);
let mut segment_duration: f64 = -1.0;
if let Some(d) = opt_duration {
// it was set on the Period.SegmentTemplate node
segment_duration = d;
}
if let Some(std) = st.duration {
segment_duration = std / timescale as f64;
}
if segment_duration < 0.0 {
return Err(DashMpdError::UnhandledMediaStream(
"Video representation is missing SegmentTemplate@duration attribute".to_string()));
}
total_number += (period_duration_secs / segment_duration).round() as i64;
let mut number = start_number;
for _ in 1..=total_number {
let dict = HashMap::from([("Number", number.to_string())]);
let path = resolve_url_template(&video_path, &dict);
let u = merge_baseurls(&base_url, &path)?;
let mf = MediaFragmentBuilder::new(period_counter, u).build();
fragments.push(mf);
number += 1;
}
}
}
} else if let Some(sb) = &video_repr.SegmentBase {
// (5) SegmentBase@indexRange addressing mode
if downloader.verbosity > 1 {
info!(" {}", "Using SegmentBase@indexRange addressing mode for video representation".italic());
}
let mf = do_segmentbase_indexrange(downloader, period_counter, base_url, sb, &dict).await?;
fragments.extend(mf);
} else if fragments.is_empty() {
if let Some(bu) = video_repr.BaseURL.first() {
// (6) BaseURL addressing mode
if downloader.verbosity > 1 {
info!(" {}", "Using BaseURL addressing mode for video representation".italic());
}
let u = merge_baseurls(&base_url, &bu.base)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.with_timeout(Duration::new(10000, 0))
.build();
fragments.push(mf);
}
}
if fragments.is_empty() {
return Err(DashMpdError::UnhandledMediaStream(
"no usable addressing mode identified for video representation".to_string()));
}
}
// FIXME we aren't correctly handling manifests without a Representation node
// eg https://raw.githubusercontent.com/zencoder/go-dash/master/mpd/fixtures/newperiod.mpd
Ok(PeriodOutputs { fragments, diagnostics, subtitle_formats: Vec::new() })
}
#[tracing::instrument(level="trace", skip_all)]
async fn do_period_subtitles(
downloader: &DashDownloader,
mpd: &MPD,
period: &Period,
period_counter: u8,
base_url: Url
) -> Result<PeriodOutputs, DashMpdError>
{
let client = downloader.http_client.as_ref().unwrap();
let output_path = &downloader.output_path.as_ref().unwrap().clone();
let period_output_path = output_path_for_period(output_path, period_counter);
let mut fragments = Vec::new();
let mut subtitle_formats = Vec::new();
let mut period_duration_secs: f64 = 0.0;
if let Some(d) = mpd.mediaPresentationDuration {
period_duration_secs = d.as_secs_f64();
}
if let Some(d) = period.duration {
period_duration_secs = d.as_secs_f64();
}
let maybe_subtitle_adaptation = if let Some(ref lang) = downloader.language_preference {
period.adaptations.iter().filter(is_subtitle_adaptation)
.min_by_key(|a| adaptation_lang_distance(a, lang))
} else {
// returns the first subtitle adaptation found
period.adaptations.iter().find(is_subtitle_adaptation)
};
if downloader.fetch_subtitles {
if let Some(subtitle_adaptation) = maybe_subtitle_adaptation {
let subtitle_format = subtitle_type(&subtitle_adaptation);
subtitle_formats.push(subtitle_format);
if downloader.verbosity > 1 && downloader.fetch_subtitles {
info!(" Retrieving subtitles in format {subtitle_format:?}");
}
// The AdaptationSet may have a BaseURL. We use a local variable to make sure we
// don't "corrupt" the base_url for the subtitle segments.
let mut base_url = base_url.clone();
if let Some(bu) = &subtitle_adaptation.BaseURL.first() {
base_url = merge_baseurls(&base_url, &bu.base)?;
}
// We don't do any ranking on subtitle Representations, because there is probably only a
// single one for our selected Adaptation.
if let Some(rep) = subtitle_adaptation.representations.first() {
if !rep.BaseURL.is_empty() {
for st_bu in rep.BaseURL.iter() {
let st_url = merge_baseurls(&base_url, &st_bu.base)?;
let mut req = client.get(st_url.clone());
if let Some(referer) = &downloader.referer {
req = req.header("Referer", referer);
} else {
req = req.header("Referer", base_url.to_string());
}
let rqw = req.build()
.map_err(|e| network_error("building request", e))?;
let subs = reqwest_bytes_with_retries(client, rqw, 5).await
.map_err(|e| network_error("fetching subtitles", e))?;
let mut subs_path = period_output_path.clone();
let subtitle_format = subtitle_type(&subtitle_adaptation);
match subtitle_format {
SubtitleType::Vtt => subs_path.set_extension("vtt"),
SubtitleType::Srt => subs_path.set_extension("srt"),
SubtitleType::Ttml => subs_path.set_extension("ttml"),
SubtitleType::Sami => subs_path.set_extension("sami"),
SubtitleType::Wvtt => subs_path.set_extension("wvtt"),
SubtitleType::Stpp => subs_path.set_extension("stpp"),
_ => subs_path.set_extension("sub"),
};
subtitle_formats.push(subtitle_format);
let mut subs_file = File::create(subs_path.clone())
.map_err(|e| DashMpdError::Io(e, String::from("creating subtitle file")))?;
if downloader.verbosity > 2 {
info!(" Subtitle {st_url} -> {} octets", subs.len());
}
match subs_file.write_all(&subs) {
Ok(()) => {
if downloader.verbosity > 0 {
info!(" Downloaded subtitles ({subtitle_format:?}) to {}",
subs_path.display());
}
},
Err(e) => {
error!("Unable to write subtitle file: {e:?}");
return Err(DashMpdError::Io(e, String::from("writing subtitle data")));
},
}
if subtitle_formats.contains(&SubtitleType::Wvtt) ||
subtitle_formats.contains(&SubtitleType::Ttxt)
{
if downloader.verbosity > 0 {
info!(" Converting subtitles to SRT format with MP4Box ");
}
let out = subs_path.with_extension("srt");
// We try to convert this to SRT format, which is more widely supported,
// using MP4Box. However, it's not a fatal error if MP4Box is not
// installed or the conversion fails.
//
// Could also try to convert to WebVTT with
// MP4Box -raw "0:output=output.vtt" input.mp4
let out_str = out.to_string_lossy();
let subs_str = subs_path.to_string_lossy();
let args = vec![
"-srt", "1",
"-out", &out_str,
&subs_str];
if downloader.verbosity > 0 {
info!(" Running MPBox {}", args.join(" "));
}
if let Ok(mp4box) = Command::new(downloader.mp4box_location.clone())
.args(args)
.output()
{
let msg = partial_process_output(&mp4box.stdout);
if msg.len() > 0 {
info!("MP4Box stdout: {msg}");
}
let msg = partial_process_output(&mp4box.stderr);
if msg.len() > 0 {
info!("MP4Box stderr: {msg}");
}
if mp4box.status.success() {
info!(" Converted subtitles to SRT");
} else {
warn!("Error running MP4Box to convert subtitles");
}
}
}
}
} else if rep.SegmentTemplate.is_some() || subtitle_adaptation.SegmentTemplate.is_some() {
let mut opt_init: Option<String> = None;
let mut opt_media: Option<String> = None;
let mut opt_duration: Option<f64> = None;
let mut timescale = 1;
let mut start_number = 1;
// SegmentTemplate as a direct child of an Adaptation node. This can specify some common
// attribute values (media, timescale, duration, startNumber) for child SegmentTemplate
// nodes in an enclosed Representation node. Don't download media segments here, only
// download for SegmentTemplate nodes that are children of a Representation node.
if let Some(st) = &rep.SegmentTemplate {
if let Some(i) = &st.initialization {
opt_init = Some(i.to_string());
}
if let Some(m) = &st.media {
opt_media = Some(m.to_string());
}
if let Some(d) = st.duration {
opt_duration = Some(d);
}
if let Some(ts) = st.timescale {
timescale = ts;
}
if let Some(s) = st.startNumber {
start_number = s;
}
}
let rid = match &rep.id {
Some(id) => id,
None => return Err(
DashMpdError::UnhandledMediaStream(
"Missing @id on Representation node".to_string())),
};
let mut dict = HashMap::from([("RepresentationID", rid.to_string())]);
if let Some(b) = &rep.bandwidth {
dict.insert("Bandwidth", b.to_string());
}
// Now the 6 possible addressing modes: (1) SegmentList,
// (2) SegmentTemplate+SegmentTimeline, (3) SegmentTemplate@duration,
// (4) SegmentTemplate@index, (5) SegmentBase@indexRange, (6) plain BaseURL
if let Some(sl) = &rep.SegmentList {
// (1) AdaptationSet>SegmentList addressing mode (can be used in conjunction
// with Representation>SegmentList addressing mode)
if downloader.verbosity > 1 {
info!(" {}", "Using AdaptationSet>SegmentList addressing mode for subtitle representation".italic());
}
let mut start_byte: Option<u64> = None;
let mut end_byte: Option<u64> = None;
if let Some(init) = &sl.Initialization {
if let Some(range) = &init.range {
let (s, e) = parse_range(range)?;
start_byte = Some(s);
end_byte = Some(e);
}
if let Some(su) = &init.sourceURL {
let path = resolve_url_template(su, &dict);
let u = merge_baseurls(&base_url, &path)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.with_range(start_byte, end_byte)
.set_init()
.build();
fragments.push(mf);
} else {
let mf = MediaFragmentBuilder::new(period_counter, base_url.clone())
.with_range(start_byte, end_byte)
.set_init()
.build();
fragments.push(mf);
}
}
for su in sl.segment_urls.iter() {
start_byte = None;
end_byte = None;
// we are ignoring SegmentURL@indexRange
if let Some(range) = &su.mediaRange {
let (s, e) = parse_range(range)?;
start_byte = Some(s);
end_byte = Some(e);
}
if let Some(m) = &su.media {
let u = merge_baseurls(&base_url, m)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.with_range(start_byte, end_byte)
.build();
fragments.push(mf);
} else if let Some(bu) = subtitle_adaptation.BaseURL.first() {
let u = merge_baseurls(&base_url, &bu.base)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.with_range(start_byte, end_byte)
.build();
fragments.push(mf);
}
}
}
if let Some(sl) = &rep.SegmentList {
// (1) Representation>SegmentList addressing mode
if downloader.verbosity > 1 {
info!(" {}", "Using Representation>SegmentList addressing mode for subtitle representation".italic());
}
let mut start_byte: Option<u64> = None;
let mut end_byte: Option<u64> = None;
if let Some(init) = &sl.Initialization {
if let Some(range) = &init.range {
let (s, e) = parse_range(range)?;
start_byte = Some(s);
end_byte = Some(e);
}
if let Some(su) = &init.sourceURL {
let path = resolve_url_template(su, &dict);
let u = merge_baseurls(&base_url, &path)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.with_range(start_byte, end_byte)
.set_init()
.build();
fragments.push(mf);
} else {
let mf = MediaFragmentBuilder::new(period_counter, base_url.clone())
.with_range(start_byte, end_byte)
.set_init()
.build();
fragments.push(mf);
}
}
for su in sl.segment_urls.iter() {
start_byte = None;
end_byte = None;
// we are ignoring SegmentURL@indexRange
if let Some(range) = &su.mediaRange {
let (s, e) = parse_range(range)?;
start_byte = Some(s);
end_byte = Some(e);
}
if let Some(m) = &su.media {
let u = merge_baseurls(&base_url, m)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.with_range(start_byte, end_byte)
.build();
fragments.push(mf);
} else if let Some(bu) = &rep.BaseURL.first() {
let u = merge_baseurls(&base_url, &bu.base)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.with_range(start_byte, end_byte)
.build();
fragments.push(mf);
};
}
} else if rep.SegmentTemplate.is_some() ||
subtitle_adaptation.SegmentTemplate.is_some()
{
// Here we are either looking at a Representation.SegmentTemplate, or a
// higher-level AdaptationSet.SegmentTemplate
let st;
if let Some(it) = &rep.SegmentTemplate {
st = it;
} else if let Some(it) = &subtitle_adaptation.SegmentTemplate {
st = it;
} else {
panic!("unreachable");
}
if let Some(i) = &st.initialization {
opt_init = Some(i.to_string());
}
if let Some(m) = &st.media {
opt_media = Some(m.to_string());
}
if let Some(ts) = st.timescale {
timescale = ts;
}
if let Some(sn) = st.startNumber {
start_number = sn;
}
if let Some(stl) = &rep.SegmentTemplate.as_ref().and_then(|st| st.SegmentTimeline.clone())
.or(subtitle_adaptation.SegmentTemplate.as_ref().and_then(|st| st.SegmentTimeline.clone()))
{
// (2) SegmentTemplate with SegmentTimeline addressing mode (also called
// "explicit addressing" in certain DASH-IF documents)
if downloader.verbosity > 1 {
info!(" {}", "Using SegmentTemplate+SegmentTimeline addressing mode for subtitle representation".italic());
}
if let Some(init) = opt_init {
let path = resolve_url_template(&init, &dict);
let u = merge_baseurls(&base_url, &path)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.set_init()
.build();
fragments.push(mf);
}
if let Some(media) = opt_media {
let sub_path = resolve_url_template(&media, &dict);
let mut segment_time = 0;
let mut segment_duration;
let mut number = start_number;
for s in &stl.segments {
if let Some(t) = s.t {
segment_time = t;
}
segment_duration = s.d;
// the URLTemplate may be based on $Time$, or on $Number$
let dict = HashMap::from([("Time", segment_time.to_string()),
("Number", number.to_string())]);
let path = resolve_url_template(&sub_path, &dict);
let u = merge_baseurls(&base_url, &path)?;
let mf = MediaFragmentBuilder::new(period_counter, u).build();
fragments.push(mf);
number += 1;
if let Some(r) = s.r {
let mut count = 0i64;
// FIXME perhaps we also need to account for startTime?
let end_time = period_duration_secs * timescale as f64;
loop {
count += 1;
// Exit from the loop after @r iterations (if @r is
// positive). A negative value of the @r attribute indicates
// that the duration indicated in @d attribute repeats until
// the start of the next S element, the end of the Period or
// until the next MPD update.
if r >= 0 {
if count > r {
break;
}
if downloader.force_duration.is_some() &&
segment_time as f64 > end_time
{
break;
}
} else if segment_time as f64 > end_time {
break;
}
segment_time += segment_duration;
let dict = HashMap::from([("Time", segment_time.to_string()),
("Number", number.to_string())]);
let path = resolve_url_template(&sub_path, &dict);
let u = merge_baseurls(&base_url, &path)?;
let mf = MediaFragmentBuilder::new(period_counter, u).build();
fragments.push(mf);
number += 1;
}
}
segment_time += segment_duration;
}
} else {
return Err(DashMpdError::UnhandledMediaStream(
"SegmentTimeline without a media attribute".to_string()));
}
} else { // no SegmentTimeline element
// (3) SegmentTemplate@duration addressing mode or (4) SegmentTemplate@index
// addressing mode (also called "simple addressing" in certain DASH-IF
// documents)
if downloader.verbosity > 0 {
info!(" {}", "Using SegmentTemplate addressing mode for stpp subtitles".italic());
}
if let Some(i) = &st.initialization {
opt_init = Some(i.to_string());
}
if let Some(m) = &st.media {
opt_media = Some(m.to_string());
}
if let Some(d) = st.duration {
opt_duration = Some(d);
}
if let Some(ts) = st.timescale {
timescale = ts;
}
if let Some(s) = st.startNumber {
start_number = s;
}
let rid = match &rep.id {
Some(id) => id,
None => return Err(
DashMpdError::UnhandledMediaStream(
"Missing @id on Representation node".to_string())),
};
let mut dict = HashMap::from([("RepresentationID", rid.to_string())]);
if let Some(b) = &rep.bandwidth {
dict.insert("Bandwidth", b.to_string());
}
let mut total_number = 0i64;
if let Some(init) = opt_init {
let path = resolve_url_template(&init, &dict);
let u = merge_baseurls(&base_url, &path)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.set_init()
.build();
fragments.push(mf);
}
if let Some(media) = opt_media {
let sub_path = resolve_url_template(&media, &dict);
let mut segment_duration: f64 = -1.0;
if let Some(d) = opt_duration {
// it was set on the Period.SegmentTemplate node
segment_duration = d;
}
if let Some(std) = st.duration {
segment_duration = std / timescale as f64;
}
if segment_duration < 0.0 {
return Err(DashMpdError::UnhandledMediaStream(
"Subtitle representation is missing SegmentTemplate@duration".to_string()));
}
total_number += (period_duration_secs / segment_duration).ceil() as i64;
let mut number = start_number;
for _ in 1..=total_number {
let dict = HashMap::from([("Number", number.to_string())]);
let path = resolve_url_template(&sub_path, &dict);
let u = merge_baseurls(&base_url, &path)?;
let mf = MediaFragmentBuilder::new(period_counter, u).build();
fragments.push(mf);
number += 1;
}
}
}
} else if let Some(sb) = &rep.SegmentBase {
// SegmentBase@indexRange addressing mode
println!("Using SegmentBase@indexRange for subs");
if downloader.verbosity > 1 {
info!(" {}", "Using SegmentBase@indexRange addressing mode for subtitle representation".italic());
}
let mut start_byte: Option<u64> = None;
let mut end_byte: Option<u64> = None;
if let Some(init) = &sb.initialization {
if let Some(range) = &init.range {
let (s, e) = parse_range(range)?;
start_byte = Some(s);
end_byte = Some(e);
}
if let Some(su) = &init.sourceURL {
let path = resolve_url_template(su, &dict);
let u = merge_baseurls(&base_url, &path)?;
let mf = MediaFragmentBuilder::new(period_counter, u)
.with_range(start_byte, end_byte)
.set_init()
.build();
fragments.push(mf);
}
}
let mf = MediaFragmentBuilder::new(period_counter, base_url.clone())
.set_init()
.build();
fragments.push(mf);
// TODO also implement SegmentBase addressing mode for subtitles
// (sample MPD: https://usp-cmaf-test.s3.eu-central-1.amazonaws.com/tears-of-steel-ttml.mpd)
}
}
}
}
}
Ok(PeriodOutputs { fragments, diagnostics: Vec::new(), subtitle_formats })
}
// This is a complement to the DashDownloader struct, intended to contain the mutable state
// associated with a download. We have chosen an API where the DashDownloader is not mutable.
struct DownloadState {
period_counter: u8,
segment_count: usize,
segment_counter: usize,
download_errors: u32
}
// Fetch a media fragment at URL frag.url, using the reqwest client in downloader.http_client.
// Network bandwidth is throttled according to downloader.rate_limit. Transient network failures are
// retried.
//
// Note: We return a File instead of a Bytes buffer, because some streams using SegmentBase indexing
// have huge segments that can fill up RAM.
#[tracing::instrument(level="trace", skip_all)]
async fn fetch_fragment(
downloader: &DashDownloader,
frag: &MediaFragment,
fragment_type: &str,
progress_percent: u32) -> Result<std::fs::File, DashMpdError>
{
let send_request = || async {
trace!("send_request {}", frag.url.clone());
// Don't use only "audio/*" or "video/*" in Accept header because some web servers (eg.
// media.axprod.net) are misconfigured and reject requests for valid audio content (eg .m4s)
let mut req = downloader.http_client.as_ref().unwrap()
.get(frag.url.clone())
.header("Accept", format!("{}/*;q=0.9,*/*;q=0.5", fragment_type))
.header("Sec-Fetch-Mode", "navigate");
if let Some(sb) = &frag.start_byte {
if let Some(eb) = &frag.end_byte {
req = req.header(RANGE, format!("bytes={sb}-{eb}"));
}
}
if let Some(ts) = &frag.timeout {
req = req.timeout(*ts);
}
if let Some(referer) = &downloader.referer {
req = req.header("Referer", referer);
} else {
req = req.header("Referer", downloader.redirected_url.to_string());
}
if let Some(username) = &downloader.auth_username {
if let Some(password) = &downloader.auth_password {
req = req.basic_auth(username, Some(password));
}
}
if let Some(token) = &downloader.auth_bearer_token {
req = req.bearer_auth(token);
}
req.send().await
.map_err(categorize_reqwest_error)?
.error_for_status()
.map_err(categorize_reqwest_error)
};
// FIXME we should be tracking this in our DashDownloader, rather than here. If each fragment
// takes very little time to download, we will not be reporting recent bandwidth in a
// satisfactory manner.
let mut bw_estimator_started = Instant::now();
let mut bw_estimator_bytes = 0;
match retry_notify(ExponentialBackoff::default(), send_request, notify_transient).await {
Ok(response) => {
match response.error_for_status() {
Ok(mut resp) => {
let mut tmp_out = tempfile::tempfile()
.map_err(|e| DashMpdError::Io(e, String::from("creating tmpfile for fragment")))?;
let content_type_checker = if fragment_type.eq("audio") {
content_type_audio_p
} else if fragment_type.eq("video") {
content_type_video_p
} else {
panic!("fragment_type not audio or video");
};
if !downloader.content_type_checks || content_type_checker(&resp) {
let mut fragment_out: Option<File> = None;
if let Some(ref fragment_path) = downloader.fragment_path {
if let Some(path) = frag.url.path_segments()
.unwrap_or_else(|| "".split(' '))
.last()
{
let vf_file = fragment_path.clone().join(fragment_type).join(path);
if let Ok(f) = File::create(vf_file) {
fragment_out = Some(f)
}
}
}
let mut segment_size = 0;
// Download in chunked format instead of using reqwest's .bytes() API, in
// order to avoid saturating RAM with a large media segment. This is
// important for DASH manifests that use indexRange addressing, which we
// don't download using byte range requests as a normal DASH client would
// do, but rather download using a single network request.
while let Some(chunk) = resp.chunk().await
.map_err(|e| network_error(&format!("fetching DASH {fragment_type} segment"), e))?
{
segment_size += chunk.len();
bw_estimator_bytes += chunk.len();
let size = min((chunk.len()/1024+1) as u32, u32::MAX);
throttle_download_rate(downloader, size).await?;
if let Err(e) = tmp_out.write_all(&chunk) {
return Err(DashMpdError::Io(e, format!("writing DASH {fragment_type} data")));
}
if let Some(ref mut fout) = fragment_out {
fout.write_all(&chunk)
.map_err(|e| DashMpdError::Io(e, format!("writing {fragment_type} fragment")))?;
}
let elapsed = bw_estimator_started.elapsed().as_secs_f64();
if (elapsed > 1.5) || (bw_estimator_bytes > 100_000) {
let bw = bw_estimator_bytes as f64 / (1e6 * elapsed);
let msg = if bw > 0.5 {
format!("Fetching {fragment_type} segments ({bw:.1} MB/s)")
} else {
let kbs = (bw * 1000.0).round() as u64;
format!("Fetching {fragment_type} segments ({kbs:3} kB/s)")
};
for observer in &downloader.progress_observers {
observer.update(progress_percent, &msg);
}
bw_estimator_started = Instant::now();
bw_estimator_bytes = 0;
}
}
if downloader.verbosity > 2 {
if let Some(sb) = &frag.start_byte {
if let Some(eb) = &frag.end_byte {
info!(" {fragment_type} segment {} range {sb}-{eb} -> {} octets",
frag.url, segment_size);
}
} else {
info!(" {fragment_type} segment {} -> {segment_size} octets", &frag.url);
}
}
} else {
warn!("{} {} with non-{fragment_type} content-type", "Ignoring segment".red(), frag.url);
};
tmp_out.sync_all()
.map_err(|e| DashMpdError::Io(e, format!("syncing {fragment_type} fragment")))?;
Ok(tmp_out)
},
Err(e) => Err(network_error("HTTP error", e)),
}
},
Err(e) => Err(network_error(&format!("{e:?}"), e)),
}
}
// Retrieve the audio segments for period `period_counter` and concatenate them to a file at tmppath.
#[tracing::instrument(level="trace", skip_all)]
async fn fetch_period_audio(
downloader: &DashDownloader,
tmppath: PathBuf,
audio_fragments: &[MediaFragment],
ds: &mut DownloadState) -> Result<bool, DashMpdError>
{
let start_download = Instant::now();
let mut have_audio = false;
{
// We need a local scope for our temporary File, so that the file is closed when we later
// optionally call the decryption application (which requires exclusive access to its input
// file on Windows).
let tmpfile_audio = File::create(tmppath.clone())
.map_err(|e| DashMpdError::Io(e, String::from("creating audio tmpfile")))?;
let mut tmpfile_audio = BufWriter::new(tmpfile_audio);
// Optionally create the directory to which we will save the audio fragments.
if let Some(ref fragment_path) = downloader.fragment_path {
let audio_fragment_dir = fragment_path.join("audio");
if !audio_fragment_dir.exists() {
fs::create_dir_all(audio_fragment_dir)
.map_err(|e| DashMpdError::Io(e, String::from("creating audio fragment dir")))?;
}
}
// FIXME: in DASH, the init segment contains headers that are necessary to generate a valid MP4
// file, so we should always abort if the first segment cannot be fetched. However, we could
// tolerate loss of subsequent segments.
for frag in audio_fragments.iter().filter(|f| f.period == ds.period_counter) {
ds.segment_counter += 1;
let progress_percent = (100.0 * ds.segment_counter as f32 / (2.0 + ds.segment_count as f32)).ceil() as u32;
let url = &frag.url;
// A manifest may use a data URL (RFC 2397) to embed media content such as the
// initialization segment directly in the manifest (recommended by YouTube for live
// streaming, but uncommon in practice).
if url.scheme() == "data" {
let us = &url.to_string();
let du = DataUrl::process(us)
.map_err(|_| DashMpdError::Parsing(String::from("parsing data URL")))?;
if du.mime_type().type_ != "audio" {
return Err(DashMpdError::UnhandledMediaStream(
String::from("expecting audio content in data URL")));
}
let (body, _fragment) = du.decode_to_vec()
.map_err(|_| DashMpdError::Parsing(String::from("decoding data URL")))?;
if downloader.verbosity > 2 {
info!(" Audio segment data URL -> {} octets", body.len());
}
if let Err(e) = tmpfile_audio.write_all(&body) {
error!("Unable to write DASH audio data: {e:?}");
return Err(DashMpdError::Io(e, String::from("writing DASH audio data")));
}
have_audio = true;
} else {
// We could download these segments in parallel, but that might upset some servers.
'done: for _ in 0..downloader.fragment_retry_count {
match fetch_fragment(downloader, frag, "audio", progress_percent).await {
Ok(mut frag_file) => {
frag_file.rewind()
.map_err(|e| DashMpdError::Io(e, String::from("rewinding fragment tempfile")))?;
let mut buf = Vec::new();
frag_file.read_to_end(&mut buf)
.map_err(|e| DashMpdError::Io(e, String::from("reading fragment tempfile")))?;
if let Err(e) = tmpfile_audio.write_all(&buf) {
error!("Unable to write DASH audio data: {e:?}");
return Err(DashMpdError::Io(e, String::from("writing DASH audio data")));
}
have_audio = true;
break 'done;
},
Err(e) => {
if downloader.verbosity > 0 {
error!("Error fetching audio segment {url}: {e:?}");
}
ds.download_errors += 1;
if ds.download_errors > downloader.max_error_count {
error!("max_error_count network errors encountered");
return Err(DashMpdError::Network(
String::from("more than max_error_count network errors")));
}
},
}
info!(" Retrying audio segment {url}");
if downloader.sleep_between_requests > 0 {
tokio::time::sleep(Duration::new(downloader.sleep_between_requests.into(), 0)).await;
}
}
}
}
tmpfile_audio.flush().map_err(|e| {
error!("Couldn't flush DASH audio file: {e}");
DashMpdError::Io(e, String::from("flushing DASH audio file"))
})?;
} // end local scope for the FileHandle
if !downloader.decryption_keys.is_empty() {
if downloader.verbosity > 0 {
let metadata = fs::metadata(tmppath.clone())
.map_err(|e| DashMpdError::Io(e, String::from("reading encrypted audio metadata")))?;
info!(" Attempting to decrypt audio stream ({} kB) with {}",
metadata.len() / 1024,
downloader.decryptor_preference);
}
let out_ext = downloader.output_path.as_ref().unwrap()
.extension()
.unwrap_or(OsStr::new("mp4"));
let decrypted = tmp_file_path("dashmpd-decrypted-audio", out_ext)?;
if downloader.decryptor_preference.eq("mp4decrypt") {
let mut args = Vec::new();
for (k, v) in downloader.decryption_keys.iter() {
args.push("--key".to_string());
args.push(format!("{k}:{v}"));
}
args.push(String::from(tmppath.to_string_lossy()));
args.push(String::from(decrypted.to_string_lossy()));
if downloader.verbosity > 1 {
info!(" Running mp4decrypt {}", args.join(" "));
}
let out = Command::new(downloader.mp4decrypt_location.clone())
.args(args)
.output()
.map_err(|e| DashMpdError::Io(e, String::from("spawning mp4decrypt")))?;
let mut no_output = true;
if let Ok(metadata) = fs::metadata(decrypted.clone()) {
if downloader.verbosity > 0 {
info!(" Decrypted audio stream of size {} kB.", metadata.len() / 1024);
}
no_output = false;
}
if !out.status.success() || no_output {
warn!(" mp4decrypt subprocess failed");
let msg = partial_process_output(&out.stdout);
if msg.len() > 0 {
warn!(" mp4decrypt stdout: {msg}");
}
let msg = partial_process_output(&out.stderr);
if msg.len() > 0 {
warn!(" mp4decrypt stderr: {msg}");
}
}
if no_output {
error!("{}", "Failed to decrypt audio stream with mp4decrypt".red());
warn!(" Undecrypted audio left in {}", tmppath.display());
return Err(DashMpdError::Decrypting(String::from("audio stream")));
}
} else if downloader.decryptor_preference.eq("shaka") {
let mut args = Vec::new();
let mut keys = Vec::new();
if downloader.verbosity < 1 {
args.push("--quiet".to_string());
}
args.push(format!("in={},stream=audio,output={}", tmppath.display(), decrypted.display()));
let mut drm_label = 0;
#[allow(clippy::explicit_counter_loop)]
for (k, v) in downloader.decryption_keys.iter() {
keys.push(format!("label=lbl{drm_label}:key_id={k}:key={v}"));
drm_label += 1;
}
args.push("--enable_raw_key_decryption".to_string());
args.push("--keys".to_string());
args.push(keys.join(","));
if downloader.verbosity > 1 {
info!(" Running shaka-packager {}", args.join(" "));
}
let out = Command::new(downloader.shaka_packager_location.clone())
.args(args)
.output()
.map_err(|e| DashMpdError::Io(e, String::from("spawning shaka-packager")))?;
let mut no_output = false;
if let Ok(metadata) = fs::metadata(decrypted.clone()) {
if downloader.verbosity > 0 {
info!(" Decrypted audio stream of size {} kB.", metadata.len() / 1024);
}
if metadata.len() == 0 {
no_output = true;
}
} else {
no_output = true;
}
if !out.status.success() || no_output {
warn!(" shaka-packager subprocess failed");
let msg = partial_process_output(&out.stdout);
if msg.len() > 0 {
warn!(" shaka-packager stdout: {msg}");
}
let msg = partial_process_output(&out.stderr);
if msg.len() > 0 {
warn!(" shaka-packager stderr: {msg}");
}
}
if no_output {
error!(" {}", "Failed to decrypt audio stream with shaka-packager".red());
warn!(" Undecrypted audio stream left in {}", tmppath.display());
return Err(DashMpdError::Decrypting(String::from("audio stream")));
}
} else {
return Err(DashMpdError::Decrypting(String::from("unknown decryption application")));
}
fs::rename(decrypted, tmppath.clone())
.map_err(|e| DashMpdError::Io(e, String::from("renaming decrypted audio")))?;
}
if let Ok(metadata) = fs::metadata(tmppath.clone()) {
if downloader.verbosity > 1 {
let mbytes = metadata.len() as f64 / (1024.0 * 1024.0);
let elapsed = start_download.elapsed();
info!(" Wrote {mbytes:.1}MB to DASH audio file ({:.1} MB/s)",
mbytes / elapsed.as_secs_f64());
}
}
Ok(have_audio)
}
// Retrieve the video segments for period `period_counter` and concatenate them to a file at tmppath.
#[tracing::instrument(level="trace", skip_all)]
async fn fetch_period_video(
downloader: &DashDownloader,
tmppath: PathBuf,
video_fragments: &[MediaFragment],
ds: &mut DownloadState) -> Result<bool, DashMpdError>
{
let start_download = Instant::now();
let mut have_video = false;
{
// We need a local scope for our tmpfile_video File, so that the file is closed when
// we later call mp4decrypt (which requires exclusive access to its input file on Windows).
let tmpfile_video = File::create(tmppath.clone())
.map_err(|e| DashMpdError::Io(e, String::from("creating video tmpfile")))?;
let mut tmpfile_video = BufWriter::new(tmpfile_video);
// Optionally create the directory to which we will save the video fragments.
if let Some(ref fragment_path) = downloader.fragment_path {
let video_fragment_dir = fragment_path.join("video");
if !video_fragment_dir.exists() {
fs::create_dir_all(video_fragment_dir)
.map_err(|e| DashMpdError::Io(e, String::from("creating video fragment dir")))?;
}
}
for frag in video_fragments.iter().filter(|f| f.period == ds.period_counter) {
ds.segment_counter += 1;
let progress_percent = (100.0 * ds.segment_counter as f32 / ds.segment_count as f32).ceil() as u32;
if frag.url.scheme() == "data" {
let us = &frag.url.to_string();
let du = DataUrl::process(us)
.map_err(|_| DashMpdError::Parsing(String::from("parsing data URL")))?;
if du.mime_type().type_ != "video" {
return Err(DashMpdError::UnhandledMediaStream(
String::from("expecting video content in data URL")));
}
let (body, _fragment) = du.decode_to_vec()
.map_err(|_| DashMpdError::Parsing(String::from("decoding data URL")))?;
if downloader.verbosity > 2 {
info!(" Video segment data URL -> {} octets", body.len());
}
if let Err(e) = tmpfile_video.write_all(&body) {
error!("Unable to write DASH video data: {e:?}");
return Err(DashMpdError::Io(e, String::from("writing DASH video data")));
}
have_video = true;
} else {
'done: for _ in 0..downloader.fragment_retry_count {
match fetch_fragment(downloader, frag, "video", progress_percent).await {
Ok(mut frag_file) => {
frag_file.rewind()
.map_err(|e| DashMpdError::Io(e, String::from("rewinding fragment tempfile")))?;
let mut buf = Vec::new();
frag_file.read_to_end(&mut buf)
.map_err(|e| DashMpdError::Io(e, String::from("reading fragment tempfile")))?;
if let Err(e) = tmpfile_video.write_all(&buf) {
error!("Unable to write DASH video data: {e:?}");
return Err(DashMpdError::Io(e, String::from("writing DASH video data")));
}
have_video = true;
break 'done;
},
Err(e) => {
if downloader.verbosity > 0 {
error!(" Error fetching video segment {}: {e:?}", frag.url);
}
ds.download_errors += 1;
if ds.download_errors > downloader.max_error_count {
return Err(DashMpdError::Network(
String::from("more than max_error_count network errors")));
}
},
}
info!(" Retrying video segment {}", frag.url);
if downloader.sleep_between_requests > 0 {
tokio::time::sleep(Duration::new(downloader.sleep_between_requests.into(), 0)).await;
}
}
}
}
tmpfile_video.flush().map_err(|e| {
error!(" Couldn't flush video file: {e}");
DashMpdError::Io(e, String::from("flushing video file"))
})?;
} // end local scope for tmpfile_video File
if !downloader.decryption_keys.is_empty() {
if downloader.verbosity > 0 {
let metadata = fs::metadata(tmppath.clone())
.map_err(|e| DashMpdError::Io(e, String::from("reading encrypted video metadata")))?;
info!(" Attempting to decrypt video stream ({} kB) with {}",
metadata.len() / 1024,
downloader.decryptor_preference);
}
let out_ext = downloader.output_path.as_ref().unwrap()
.extension()
.unwrap_or(OsStr::new("mp4"));
let decrypted = tmp_file_path("dashmpd-decrypted-video", out_ext)?;
if downloader.decryptor_preference.eq("mp4decrypt") {
let mut args = Vec::new();
for (k, v) in downloader.decryption_keys.iter() {
args.push("--key".to_string());
args.push(format!("{k}:{v}"));
}
args.push(tmppath.to_string_lossy().to_string());
args.push(decrypted.to_string_lossy().to_string());
if downloader.verbosity > 1 {
info!(" Running mp4decrypt {}", args.join(" "));
}
let out = Command::new(downloader.mp4decrypt_location.clone())
.args(args)
.output()
.map_err(|e| DashMpdError::Io(e, String::from("spawning mp4decrypt")))?;
let mut no_output = false;
if let Ok(metadata) = fs::metadata(decrypted.clone()) {
if downloader.verbosity > 0 {
info!(" Decrypted video stream of size {} kB.", metadata.len() / 1024);
}
if metadata.len() == 0 {
no_output = true;
}
} else {
no_output = true;
}
if !out.status.success() || no_output {
error!(" mp4decrypt subprocess failed");
let msg = partial_process_output(&out.stdout);
if msg.len() > 0 {
warn!(" mp4decrypt stdout: {msg}");
}
let msg = partial_process_output(&out.stderr);
if msg.len() > 0 {
warn!(" mp4decrypt stderr: {msg}");
}
}
if no_output {
error!(" {}", "Failed to decrypt video stream with mp4decrypt".red());
warn!(" Undecrypted video stream left in {}", tmppath.display());
return Err(DashMpdError::Decrypting(String::from("video stream")));
}
} else if downloader.decryptor_preference.eq("shaka") {
let mut args = Vec::new();
let mut keys = Vec::new();
if downloader.verbosity < 1 {
args.push("--quiet".to_string());
}
args.push(format!("in={},stream=video,output={}", tmppath.display(), decrypted.display()));
let mut drm_label = 0;
#[allow(clippy::explicit_counter_loop)]
for (k, v) in downloader.decryption_keys.iter() {
keys.push(format!("label=lbl{drm_label}:key_id={k}:key={v}"));
drm_label += 1;
}
args.push("--enable_raw_key_decryption".to_string());
args.push("--keys".to_string());
args.push(keys.join(","));
if downloader.verbosity > 1 {
info!(" Running shaka-packager {}", args.join(" "));
}
let out = Command::new(downloader.shaka_packager_location.clone())
.args(args)
.output()
.map_err(|e| DashMpdError::Io(e, String::from("spawning shaka-packager")))?;
let mut no_output = true;
if let Ok(metadata) = fs::metadata(decrypted.clone()) {
if downloader.verbosity > 0 {
info!(" Decrypted video stream of size {} kB.", metadata.len() / 1024);
}
no_output = false;
}
if !out.status.success() || no_output {
warn!(" shaka-packager subprocess failed");
let msg = partial_process_output(&out.stdout);
if msg.len() > 0 {
warn!(" shaka-packager stdout: {msg}");
}
let msg = partial_process_output(&out.stderr);
if msg.len() > 0 {
warn!(" shaka-packager stderr: {msg}");
}
}
if no_output {
error!(" {}", "Failed to decrypt video stream with shaka-packager".red());
warn!(" Undecrypted video left in {}", tmppath.display());
return Err(DashMpdError::Decrypting(String::from("video stream")));
}
} else {
return Err(DashMpdError::Decrypting(String::from("unknown decryption application")));
}
fs::rename(decrypted, tmppath.clone())
.map_err(|e| DashMpdError::Io(e, String::from("renaming decrypted video")))?;
}
if let Ok(metadata) = fs::metadata(tmppath.clone()) {
if downloader.verbosity > 1 {
let mbytes = metadata.len() as f64 / (1024.0 * 1024.0);
let elapsed = start_download.elapsed();
info!(" Wrote {mbytes:.1}MB to DASH video file ({:.1} MB/s)",
mbytes / elapsed.as_secs_f64());
}
}
Ok(have_video)
}
// Retrieve the video segments for period `ds.period_counter` and concatenate them to a file at `tmppath`.
#[tracing::instrument(level="trace", skip_all)]
async fn fetch_period_subtitles(
downloader: &DashDownloader,
tmppath: PathBuf,
subtitle_fragments: &[MediaFragment],
subtitle_formats: &[SubtitleType],
ds: &mut DownloadState) -> Result<bool, DashMpdError>
{
let client = downloader.http_client.clone().unwrap();
let start_download = Instant::now();
let mut have_subtitles = false;
{
let tmpfile_subs = File::create(tmppath.clone())
.map_err(|e| DashMpdError::Io(e, String::from("creating subs tmpfile")))?;
let mut tmpfile_subs = BufWriter::new(tmpfile_subs);
for frag in subtitle_fragments {
// Update any ProgressObservers
ds.segment_counter += 1;
let progress_percent = (100.0 * ds.segment_counter as f32 / ds.segment_count as f32).ceil() as u32;
for observer in &downloader.progress_observers {
observer.update(progress_percent, "Fetching subtitle segments");
}
if frag.url.scheme() == "data" {
let us = &frag.url.to_string();
let du = DataUrl::process(us)
.map_err(|_| DashMpdError::Parsing(String::from("parsing data URL")))?;
if du.mime_type().type_ != "video" {
return Err(DashMpdError::UnhandledMediaStream(
String::from("expecting video content in data URL")));
}
let (body, _fragment) = du.decode_to_vec()
.map_err(|_| DashMpdError::Parsing(String::from("decoding data URL")))?;
if downloader.verbosity > 2 {
info!(" Subtitle segment data URL -> {} octets", body.len());
}
if let Err(e) = tmpfile_subs.write_all(&body) {
error!("Unable to write DASH subtitle data: {e:?}");
return Err(DashMpdError::Io(e, String::from("writing DASH subtitle data")));
}
have_subtitles = true;
} else {
let fetch = || async {
let mut req = client.get(frag.url.clone())
.header("Sec-Fetch-Mode", "navigate");
if let Some(sb) = &frag.start_byte {
if let Some(eb) = &frag.end_byte {
req = req.header(RANGE, format!("bytes={sb}-{eb}"));
}
}
if let Some(referer) = &downloader.referer {
req = req.header("Referer", referer);
} else {
req = req.header("Referer", downloader.redirected_url.to_string());
}
if let Some(username) = &downloader.auth_username {
if let Some(password) = &downloader.auth_password {
req = req.basic_auth(username, Some(password));
}
}
if let Some(token) = &downloader.auth_bearer_token {
req = req.bearer_auth(token);
}
req.send().await
.map_err(categorize_reqwest_error)?
.error_for_status()
.map_err(categorize_reqwest_error)
};
let mut failure = None;
match retry_notify(ExponentialBackoff::default(), fetch, notify_transient).await {
Ok(response) => {
if response.status().is_success() {
let dash_bytes = response.bytes().await
.map_err(|e| network_error("fetching DASH subtitle segment", e))?;
if downloader.verbosity > 2 {
if let Some(sb) = &frag.start_byte {
if let Some(eb) = &frag.end_byte {
info!(" Subtitle segment {} range {sb}-{eb} -> {} octets",
&frag.url, dash_bytes.len());
}
} else {
info!(" Subtitle segment {} -> {} octets", &frag.url, dash_bytes.len());
}
}
let size = min((dash_bytes.len()/1024 + 1) as u32, u32::MAX);
throttle_download_rate(downloader, size).await?;
if let Err(e) = tmpfile_subs.write_all(&dash_bytes) {
return Err(DashMpdError::Io(e, String::from("writing DASH subtitle data")));
}
have_subtitles = true;
} else {
failure = Some(format!("HTTP error {}", response.status().as_str()));
}
},
Err(e) => failure = Some(format!("{e}")),
}
if let Some(f) = failure {
if downloader.verbosity > 0 {
error!("{} fetching subtitle segment {}", f.red(), &frag.url);
}
ds.download_errors += 1;
if ds.download_errors > downloader.max_error_count {
return Err(DashMpdError::Network(
String::from("more than max_error_count network errors")));
}
}
}
if downloader.sleep_between_requests > 0 {
tokio::time::sleep(Duration::new(downloader.sleep_between_requests.into(), 0)).await;
}
}
tmpfile_subs.flush().map_err(|e| {
error!("Couldn't flush subs file: {e}");
DashMpdError::Io(e, String::from("flushing subtitle file"))
})?;
} // end local scope for tmpfile_subs File
if have_subtitles {
if let Ok(metadata) = fs::metadata(tmppath.clone()) {
if downloader.verbosity > 1 {
let mbytes = metadata.len() as f64 / (1024.0 * 1024.0);
let elapsed = start_download.elapsed();
info!(" Wrote {mbytes:.1}MB to DASH subtitle file ({:.1} MB/s)",
mbytes / elapsed.as_secs_f64());
}
}
// TODO: for subtitle_formats sub and srt we could also try to embed them in the output
// file, for example using MP4Box or mkvmerge
if subtitle_formats.contains(&SubtitleType::Wvtt) ||
subtitle_formats.contains(&SubtitleType::Ttxt)
{
// We can extract these from the MP4 container in .srt format, using MP4Box.
if downloader.verbosity > 0 {
if let Some(fmt) = subtitle_formats.first() {
info!(" Downloaded media contains subtitles in {fmt:?} format");
}
info!(" {}", "Running MP4Box to extract subtitles".italic());
}
let out = downloader.output_path.as_ref().unwrap()
.with_extension("srt");
let out_str = out.to_string_lossy();
let tmp_str = tmppath.to_string_lossy();
let args = vec![
"-srt", "1",
"-out", &out_str,
&tmp_str];
if downloader.verbosity > 0 {
info!(" Running MP4Box {}", args.join(" "));
}
if let Ok(mp4box) = Command::new(downloader.mp4box_location.clone())
.args(args)
.output()
{
let msg = partial_process_output(&mp4box.stdout);
if msg.len() > 0 {
info!(" MP4Box stdout: {msg}");
}
let msg = partial_process_output(&mp4box.stderr);
if msg.len() > 0 {
info!(" MP4Box stderr: {msg}");
}
if mp4box.status.success() {
info!(" Extracted subtitles as SRT");
} else {
warn!(" Error running MP4Box to extract subtitles");
}
} else {
warn!(" Failed to spawn MP4Box to extract subtitles");
}
}
if subtitle_formats.contains(&SubtitleType::Stpp) {
if downloader.verbosity > 0 {
info!(" Converting STPP subtitles to TTML format with ffmpeg");
}
let out = downloader.output_path.as_ref().unwrap()
.with_extension("ttml");
let tmppath_arg = &tmppath.to_string_lossy();
let out_arg = &out.to_string_lossy();
let ffmpeg_args = vec![
"-hide_banner",
"-nostats",
"-loglevel", "error",
"-y", // overwrite output file if it exists
"-nostdin",
"-i", tmppath_arg,
"-f", "data",
"-map", "0",
"-c", "copy",
out_arg];
if downloader.verbosity > 0 {
info!(" Running ffmpeg {}", ffmpeg_args.join(" "));
}
if let Ok(ffmpeg) = Command::new(downloader.ffmpeg_location.clone())
.args(ffmpeg_args)
.output()
{
let msg = partial_process_output(&ffmpeg.stdout);
if msg.len() > 0 {
info!(" ffmpeg stdout: {msg}");
}
let msg = partial_process_output(&ffmpeg.stderr);
if msg.len() > 0 {
info!(" ffmpeg stderr: {msg}");
}
if ffmpeg.status.success() {
info!(" Converted STPP subtitles to TTML format");
} else {
warn!(" Error running ffmpeg to convert subtitles");
}
}
// TODO: it would be useful to also convert the subtitles to SRT/WebVTT format, as they tend
// to be better supported. However, ffmpeg does not seem able to convert from SPTT to
// these formats. We could perhaps use the Python ttconv package, or below with MP4Box.
}
}
Ok(have_subtitles)
}
// Fetch XML content of manifest from an HTTP/HTTPS URL
async fn fetch_mpd_http(downloader: &mut DashDownloader) -> Result<Bytes, DashMpdError> {
let client = &downloader.http_client.clone().unwrap();
let send_request = || async {
let mut req = client.get(&downloader.mpd_url)
.header("Accept", "application/dash+xml,video/vnd.mpeg.dash.mpd")
.header("Accept-Language", "en-US,en")
.header("Upgrade-Insecure-Requests", "1")
.header("Sec-Fetch-Mode", "navigate");
if let Some(referer) = &downloader.referer {
req = req.header("Referer", referer);
}
if let Some(username) = &downloader.auth_username {
if let Some(password) = &downloader.auth_password {
req = req.basic_auth(username, Some(password));
}
}
if let Some(token) = &downloader.auth_bearer_token {
req = req.bearer_auth(token);
}
req.send().await
.map_err(categorize_reqwest_error)?
.error_for_status()
.map_err(categorize_reqwest_error)
};
for observer in &downloader.progress_observers {
observer.update(1, "Fetching DASH manifest");
}
if downloader.verbosity > 0 {
if !downloader.fetch_audio && !downloader.fetch_video && !downloader.fetch_subtitles {
info!("Only simulating media downloads");
}
info!("Fetching the DASH manifest");
}
let response = retry_notify(ExponentialBackoff::default(), send_request, notify_transient)
.await
.map_err(|e| network_error("requesting DASH manifest", e))?;
if !response.status().is_success() {
let msg = format!("fetching DASH manifest (HTTP {})", response.status().as_str());
return Err(DashMpdError::Network(msg));
}
downloader.redirected_url = response.url().clone();
response.bytes().await
.map_err(|e| network_error("fetching DASH manifest", e))
}
// Fetch XML content of manifest from a file:// URL. The reqwest library is not able to download
// from this URL type.
async fn fetch_mpd_file(downloader: &mut DashDownloader) -> Result<Bytes, DashMpdError> {
if ! &downloader.mpd_url.starts_with("file://") {
return Err(DashMpdError::Other(String::from("expecting file:// URL scheme")));
}
let url = Url::parse(&downloader.mpd_url)
.map_err(|_| DashMpdError::Other(String::from("parsing MPD URL")))?;
let path = url.to_file_path()
.map_err(|_| DashMpdError::Other(String::from("extracting path from file:// URL")))?;
let octets = fs::read(path)
.map_err(|_| DashMpdError::Other(String::from("reading from file:// URL")))?;
Ok(Bytes::from(octets))
}
#[tracing::instrument(level="trace", skip_all)]
async fn fetch_mpd(downloader: &mut DashDownloader) -> Result<PathBuf, DashMpdError> {
let xml = if downloader.mpd_url.starts_with("file://") {
fetch_mpd_file(downloader).await?
} else {
fetch_mpd_http(downloader).await?
};
let mut mpd: MPD = parse_resolving_xlinks(downloader, &xml).await
.map_err(|e| parse_error("parsing DASH XML", e))?;
// From the DASH specification: "If at least one MPD.Location element is present, the value of
// any MPD.Location element is used as the MPD request". We make a new request to the URI and reparse.
let client = &downloader.http_client.clone().unwrap();
if let Some(new_location) = &mpd.locations.first() {
let new_url = &new_location.url;
if downloader.verbosity > 0 {
info!("Redirecting to new manifest <Location> {new_url}");
}
let send_request = || async {
let mut req = client.get(new_url)
.header("Accept", "application/dash+xml,video/vnd.mpeg.dash.mpd")
.header("Accept-Language", "en-US,en")
.header("Sec-Fetch-Mode", "navigate");
if let Some(referer) = &downloader.referer {
req = req.header("Referer", referer);
} else {
req = req.header("Referer", downloader.redirected_url.to_string());
}
if let Some(username) = &downloader.auth_username {
if let Some(password) = &downloader.auth_password {
req = req.basic_auth(username, Some(password));
}
}
if let Some(token) = &downloader.auth_bearer_token {
req = req.bearer_auth(token);
}
req.send().await
.map_err(categorize_reqwest_error)?
.error_for_status()
.map_err(categorize_reqwest_error)
};
let response = retry_notify(ExponentialBackoff::default(), send_request, notify_transient)
.await
.map_err(|e| network_error("requesting relocated DASH manifest", e))?;
if !response.status().is_success() {
let msg = format!("fetching DASH manifest (HTTP {})", response.status().as_str());
return Err(DashMpdError::Network(msg));
}
downloader.redirected_url = response.url().clone();
let xml = response.bytes().await
.map_err(|e| network_error("fetching relocated DASH manifest", e))?;
mpd = parse_resolving_xlinks(downloader, &xml).await
.map_err(|e| parse_error("parsing relocated DASH XML", e))?;
}
if let Some(mpdtype) = mpd.mpdtype.as_ref() {
if mpdtype.eq("dynamic") {
// TODO: look at algorithm used in function segment_numbers at
// https://github.com/streamlink/streamlink/blob/master/src/streamlink/stream/dash_manifest.py
if downloader.allow_live_streams {
if downloader.verbosity > 0 {
warn!("Attempting to download from live stream (this may not work).");
}
} else {
return Err(DashMpdError::UnhandledMediaStream("Don't know how to download dynamic MPD".to_string()));
}
}
}
let mut toplevel_base_url = downloader.redirected_url.clone();
// There may be several BaseURL tags in the MPD, but we don't currently implement failover
if let Some(bu) = &mpd.base_url.first() {
toplevel_base_url = merge_baseurls(&downloader.redirected_url, &bu.base)?;
}
if downloader.verbosity > 0 {
let pcount = mpd.periods.len();
info!("DASH manifest has {pcount} period{}", if pcount > 1 { "s" } else { "" });
print_available_streams(&mpd);
}
// Analyse the content of each Period in the manifest. We need to ensure that we associate media
// segments with the correct period, because segments in each Period may use different codecs,
// so they can't be concatenated together directly without reencoding. The main purpose for this
// iteration of Periods (which is then followed by an iteration over Periods where we retrieve
// the media segments and concatenate them) is to obtain a count of the total number of media
// fragments that we are going to retrieve, so that the ProgressBar shows information relevant
// to the total download (we don't want a per-Period ProgressBar).
let mut pds: Vec<PeriodDownloads> = Vec::new();
let mut period_counter = 0;
for mpd_period in &mpd.periods {
let period = mpd_period.clone();
period_counter += 1;
if let Some(min) = downloader.minimum_period_duration {
if let Some(duration) = period.duration {
if duration < min {
if let Some(id) = period.id.as_ref() {
info!("Skipping period {id} (#{period_counter}): duration is less than requested minimum");
} else {
info!("Skipping period #{period_counter}: duration is less than requested minimum");
}
continue;
}
}
}
let mut pd = PeriodDownloads { period_counter, ..Default::default() };
if let Some(id) = period.id.as_ref() {
pd.id = Some(id.clone());
}
if downloader.verbosity > 0 {
if let Some(id) = period.id.as_ref() {
info!("Preparing download for period {id} (#{period_counter})");
} else {
info!("Preparing download for period #{period_counter}");
}
}
let mut base_url = toplevel_base_url.clone();
// A BaseURL could be specified for each Period
if let Some(bu) = period.BaseURL.first() {
base_url = merge_baseurls(&base_url, &bu.base)?;
}
let mut audio_outputs = PeriodOutputs::default();
if downloader.fetch_audio {
audio_outputs = do_period_audio(downloader, &mpd, &period, period_counter, base_url.clone()).await?;
for f in audio_outputs.fragments {
pd.audio_fragments.push(f);
}
}
let mut video_outputs = PeriodOutputs::default();
if downloader.fetch_video {
video_outputs = do_period_video(downloader, &mpd, &period, period_counter, base_url.clone()).await?;
for f in video_outputs.fragments {
pd.video_fragments.push(f);
}
}
match do_period_subtitles(downloader, &mpd, &period, period_counter, base_url.clone()).await {
Ok(subtitle_outputs) => {
for f in subtitle_outputs.fragments {
pd.subtitle_fragments.push(f);
}
for f in subtitle_outputs.subtitle_formats {
pd.subtitle_formats.push(f);
}
},
Err(e) => warn!(" Ignoring error triggered while processing subtitles: {e}"),
}
// Print some diagnostics information on the selected streams
if downloader.verbosity > 0 {
use base64::prelude::{Engine as _, BASE64_STANDARD};
audio_outputs.diagnostics.iter().for_each(|msg| info!("{}", msg));
for f in pd.audio_fragments.iter().filter(|f| f.is_init) {
if let Some(pssh_bytes) = extract_init_pssh(downloader, f.url.clone()).await {
info!(" PSSH (from init segment): {}", BASE64_STANDARD.encode(&pssh_bytes));
if let Ok(pssh) = pssh_box::from_bytes(&pssh_bytes) {
info!(" {}", pssh.to_string());
}
}
}
video_outputs.diagnostics.iter().for_each(|msg| info!("{}", msg));
for f in pd.video_fragments.iter().filter(|f| f.is_init) {
if let Some(pssh_bytes) = extract_init_pssh(downloader, f.url.clone()).await {
info!(" PSSH (from init segment): {}", BASE64_STANDARD.encode(&pssh_bytes));
if let Ok(pssh) = pssh_box::from_bytes(&pssh_bytes) {
info!(" {}", pssh.to_string());
}
}
}
}
pds.push(pd);
} // loop over Periods
// To collect the muxed audio and video segments for each Period in the MPD, before their
// final concatenation-with-reencoding.
let output_path = &downloader.output_path.as_ref().unwrap().clone();
let mut period_output_paths: Vec<PathBuf> = Vec::new();
let mut ds = DownloadState {
period_counter: 0,
// The additional +2 is for our initial .mpd fetch action and final muxing action
segment_count: pds.iter().map(period_fragment_count).sum(),
segment_counter: 0,
download_errors: 0
};
for pd in pds {
let mut have_audio = false;
let mut have_video = false;
let mut have_subtitles = false;
ds.period_counter = pd.period_counter;
let period_output_path = output_path_for_period(output_path, pd.period_counter);
#[allow(clippy::collapsible_if)]
if downloader.verbosity > 0 {
if downloader.fetch_audio || downloader.fetch_video || downloader.fetch_subtitles {
let idnum = if let Some(id) = pd.id {
format!("id={} (#{})", id, pd.period_counter)
} else {
format!("#{}", pd.period_counter)
};
info!("Period {idnum}: fetching {} audio, {} video and {} subtitle segments",
pd.audio_fragments.len(),
pd.video_fragments.len(),
pd.subtitle_fragments.len());
}
}
let output_ext = downloader.output_path.as_ref().unwrap()
.extension()
.unwrap_or(OsStr::new("mp4"));
let tmppath_audio = if let Some(ref path) = downloader.keep_audio {
path.clone()
} else {
tmp_file_path("dashmpd-audio", output_ext)?
};
let tmppath_video = if let Some(ref path) = downloader.keep_video {
path.clone()
} else {
tmp_file_path("dashmpd-video", output_ext)?
};
let tmppath_subs = tmp_file_path("dashmpd-subs", OsStr::new("sub"))?;
if downloader.fetch_audio && !pd.audio_fragments.is_empty() {
have_audio = fetch_period_audio(downloader,
tmppath_audio.clone(), &pd.audio_fragments,
&mut ds).await?;
}
if downloader.fetch_video && !pd.video_fragments.is_empty() {
have_video = fetch_period_video(downloader,
tmppath_video.clone(), &pd.video_fragments,
&mut ds).await?;
}
// Here we handle subtitles that are distributed in fragmented MP4 segments, rather than as a
// single .srt or .vtt file file. This is the case for WVTT (WebVTT) and STPP (which should be
// formatted as EBU-TT for DASH media) formats.
if downloader.fetch_subtitles && !pd.subtitle_fragments.is_empty() {
have_subtitles = fetch_period_subtitles(downloader,
tmppath_subs.clone(),
&pd.subtitle_fragments,
&pd.subtitle_formats,
&mut ds).await?;
}
// The output file for this Period is either a mux of the audio and video streams, if both
// are present, or just the audio stream, or just the video stream.
if have_audio && have_video {
for observer in &downloader.progress_observers {
observer.update(99, "Muxing audio and video");
}
if downloader.verbosity > 1 {
info!(" {}", "Muxing audio and video streams".italic());
}
mux_audio_video(downloader, &period_output_path, &tmppath_audio, &tmppath_video)?;
if pd.subtitle_formats.contains(&SubtitleType::Stpp) {
let container = match &period_output_path.extension() {
Some(ext) => ext.to_str().unwrap_or("mp4"),
None => "mp4",
};
if container.eq("mp4") {
if downloader.verbosity > 1 {
if let Some(fmt) = &pd.subtitle_formats.first() {
info!(" Downloaded media contains subtitles in {fmt:?} format");
}
info!(" Running MP4Box to merge subtitles with output MP4 container");
}
// We can try to add the subtitles to the MP4 container, using MP4Box. Only
// works with MP4 containers.
let tmp_str = tmppath_subs.to_string_lossy();
let period_output_str = period_output_path.to_string_lossy();
let args = vec!["-add", &tmp_str, &period_output_str];
if downloader.verbosity > 0 {
info!(" Running MP4Box {}", args.join(" "));
}
if let Ok(mp4box) = Command::new(downloader.mp4box_location.clone())
.args(args)
.output()
{
let msg = partial_process_output(&mp4box.stdout);
if msg.len() > 0 {
info!(" MP4Box stdout: {msg}");
}
let msg = partial_process_output(&mp4box.stderr);
if msg.len() > 0 {
info!(" MP4Box stderr: {msg}");
}
if mp4box.status.success() {
info!(" Merged subtitles with MP4 container");
} else {
warn!(" Error running MP4Box to merge subtitles");
}
} else {
warn!(" Failed to spawn MP4Box to merge subtitles");
}
} else if container.eq("mkv") || container.eq("webm") {
// Try using mkvmerge to add a subtitle track. mkvmerge does not seem to be able
// to merge STPP subtitles, but can merge SRT if we have managed to convert
// them.
//
// We mkvmerge to a temporary output file, and if the command succeeds we copy
// that to the original output path. Note that mkvmerge on Windows is compiled
// using MinGW and isn't able to handle native pathnames (for instance files
// created with tempfile::Builder), so we use temporary_outpath() which will create a
// temporary file in the current directory on Windows.
//
// mkvmerge -o output.mkv input.mkv subs.srt
let srt = period_output_path.with_extension("srt");
if srt.exists() {
if downloader.verbosity > 0 {
info!(" Running mkvmerge to merge subtitles with output Matroska container");
}
let tmppath = temporary_outpath(".mkv")?;
let pop_arg = &period_output_path.to_string_lossy();
let srt_arg = &srt.to_string_lossy();
let mkvmerge_args = vec!["-o", &tmppath, pop_arg, srt_arg];
if downloader.verbosity > 0 {
info!(" Running mkvmerge {}", mkvmerge_args.join(" "));
}
if let Ok(mkvmerge) = Command::new(downloader.mkvmerge_location.clone())
.args(mkvmerge_args)
.output()
{
let msg = partial_process_output(&mkvmerge.stdout);
if msg.len() > 0 {
info!(" mkvmerge stdout: {msg}");
}
let msg = partial_process_output(&mkvmerge.stderr);
if msg.len() > 0 {
info!(" mkvmerge stderr: {msg}");
}
if mkvmerge.status.success() {
info!(" Merged subtitles with Matroska container");
// Copy the output file from mkvmerge to the period_output_path
// local scope so that tmppath is not busy on Windows and can be deleted
{
let tmpfile = File::open(tmppath.clone())
.map_err(|e| DashMpdError::Io(
e, String::from("opening mkvmerge output")))?;
let mut merged = BufReader::new(tmpfile);
// This will truncate the period_output_path
let outfile = File::create(period_output_path.clone())
.map_err(|e| DashMpdError::Io(
e, String::from("creating output file")))?;
let mut sink = BufWriter::new(outfile);
io::copy(&mut merged, &mut sink)
.map_err(|e| DashMpdError::Io(
e, String::from("copying mkvmerge output to output file")))?;
}
if let Err(e) = fs::remove_file(tmppath) {
warn!(" Error deleting temporary mkvmerge output: {e}");
}
} else {
warn!(" Error running mkvmerge to merge subtitles");
}
}
}
}
}
} else if have_audio {
copy_audio_to_container(downloader, &period_output_path, &tmppath_audio)?;
} else if have_video {
copy_video_to_container(downloader, &period_output_path, &tmppath_video)?;
} else if downloader.fetch_video && downloader.fetch_audio {
return Err(DashMpdError::UnhandledMediaStream("no audio or video streams found".to_string()));
} else if downloader.fetch_video {
return Err(DashMpdError::UnhandledMediaStream("no video streams found".to_string()));
} else if downloader.fetch_audio {
return Err(DashMpdError::UnhandledMediaStream("no audio streams found".to_string()));
}
#[allow(clippy::collapsible_if)]
if downloader.keep_audio.is_none() && downloader.fetch_audio {
if tmppath_audio.exists() && fs::remove_file(tmppath_audio).is_err() {
info!(" Failed to delete temporary file for audio stream");
}
}
#[allow(clippy::collapsible_if)]
if downloader.keep_video.is_none() && downloader.fetch_video {
if tmppath_video.exists() && fs::remove_file(tmppath_video).is_err() {
info!(" Failed to delete temporary file for video stream");
}
}
#[allow(clippy::collapsible_if)]
if downloader.fetch_subtitles && tmppath_subs.exists() && fs::remove_file(tmppath_subs).is_err() {
info!(" Failed to delete temporary file for subtitles");
}
if downloader.verbosity > 1 && (downloader.fetch_audio || downloader.fetch_video || have_subtitles) {
if let Ok(metadata) = fs::metadata(period_output_path.clone()) {
info!(" Wrote {:.1}MB to media file", metadata.len() as f64 / (1024.0 * 1024.0));
}
}
if have_audio || have_video {
period_output_paths.push(period_output_path);
}
} // Period iterator
#[allow(clippy::comparison_chain)]
if period_output_paths.len() == 1 {
// We already arranged to write directly to the requested output_path.
maybe_record_metainformation(output_path, downloader, &mpd);
} else if period_output_paths.len() > 1 {
// If the streams for the different periods are all of the same resolution, we can
// concatenate them (with reencoding) into a single media file. Otherwise, we can't
// concatenate without rescaling and loss of quality, so we leave them in separate files.
// This feature isn't implemented using libav instead of ffmpeg as a subprocess.
#[allow(unused_mut)]
let mut concatenated = false;
#[cfg(not(feature = "libav"))]
if downloader.concatenate_periods && video_containers_concatable(downloader, &period_output_paths) {
info!("Preparing to concatenate multiple Periods into one output file");
concat_output_files(downloader, &period_output_paths)?;
for p in &period_output_paths[1..] {
if fs::remove_file(p).is_err() {
warn!(" Failed to delete temporary file {}", p.display());
}
}
concatenated = true;
if let Some(pop) = period_output_paths.first() {
maybe_record_metainformation(pop, downloader, &mpd);
}
}
if !concatenated {
info!("Media content has been saved in a separate file for each period:");
// FIXME this is not the original period number if we have dropped periods
period_counter = 0;
for p in period_output_paths {
period_counter += 1;
info!(" Period #{period_counter}: {}", p.display());
maybe_record_metainformation(&p, downloader, &mpd);
}
}
}
let have_content_protection = mpd.periods.iter().any(
|p| p.adaptations.iter().any(
|a| (!a.ContentProtection.is_empty()) ||
a.representations.iter().any(
|r| !r.ContentProtection.is_empty())));
if have_content_protection && downloader.decryption_keys.is_empty() {
warn!("Manifest seems to use ContentProtection (DRM), but you didn't provide decryption keys.");
}
for observer in &downloader.progress_observers {
observer.update(100, "Done");
}
Ok(PathBuf::from(output_path))
}
#[cfg(test)]
mod tests {
#[test]
fn test_resolve_url_template() {
use std::collections::HashMap;
use super::resolve_url_template;
assert_eq!(resolve_url_template("AA$Time$BB", &HashMap::from([("Time", "ZZZ".to_string())])),
"AAZZZBB");
assert_eq!(resolve_url_template("AA$Number%06d$BB", &HashMap::from([("Number", "42".to_string())])),
"AA000042BB");
let dict = HashMap::from([("RepresentationID", "640x480".to_string()),
("Number", "42".to_string()),
("Time", "ZZZ".to_string())]);
assert_eq!(resolve_url_template("AA/$RepresentationID$/segment-$Number%05d$.mp4", &dict),
"AA/640x480/segment-00042.mp4");
}
}