WD-media-frags-20100624 144 KB
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
<?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" lang="en-us"><head><title>Media Fragments URI 1.0</title><style type="text/css">
        /**/
code           { font-family: monospace; }

div.constraint,
div.issue,
div.note,
div.notice     { margin-left: 2em; }

ol.enumar      { list-style-type: decimal; }
ol.enumla      { list-style-type: lower-alpha; }
ol.enumlr      { list-style-type: lower-roman; }
ol.enumua      { list-style-type: upper-alpha; }
ol.enumur      { list-style-type: upper-roman; }

dt.label       { display: run-in; }

li, p           { margin-top: 0.3em;
                 margin-bottom: 0.3em; }

.diff-chg       { background-color: yellow; }
.diff-del       { background-color: red; text-decoration: line-through;}
.diff-add       { background-color: lime; }

table          { empty-cells: show; }

table caption {
        font-weight: normal;
        font-style: italic;
        text-align: left;
        margin-bottom: .5em;
}

div.issue {
  color: red;
}
.rfc2119 {
  font-variant: small-caps;
}

div.exampleInner pre { margin-left: 1em;
                       margin-top: 0em; margin-bottom: 0em}
div.exampleOuter {border: 4px double gray;
                  margin: 0em; padding: 0em}
div.exampleInner { background-color: #d5dee3;
                   border-top-width: 4px;
                   border-top-style: double;
                   border-top-color: #d3d3d3;
                   border-bottom-width: 4px;
                   border-bottom-style: double;
                   border-bottom-color: #d3d3d3;
                   padding: 4px; margin: 0em }
div.exampleWrapper { margin: 4px }
div.exampleHeader { font-weight: bold;
                    margin: 4px}

div.boxedtext {
   border: solid #bebebe 1px;
   margin: 2em 1em 1em 2em;
 }

span.practicelab {
   margin: 1.5em 0.5em 1em 1em;
   font-weight: bold;
   font-style: italic;
 }

span.practicelab   { background: #dfffff; }

 span.practicelab {
   position: relative;
   padding: 0 0.5em;
   top: -1.5em;
 }
p.practice
{
   margin: 1.5em 0.5em 1em 1em;
 }

@media screen {
 p.practice {
   position: relative;
   top: -2em;
   padding: 0;
   margin: 1.5em 0.5em -1em 1em;
}
}
/**/ </style><link type="text/css" rel="stylesheet" href="http://www.w3.org/StyleSheets/TR/W3C-WD.css" />
</head>
<body><div class="head"><p><a href="http://www.w3.org/"><img width="72" height="48" alt="W3C" src="http://www.w3.org/Icons/w3c_home" /></a></p>
<h1><a id="title" name="title" />Media Fragments URI 1.0</h1>
<h2><a id="w3c-doctype" name="w3c-doctype" />W3C Working Draft 24 June 2010</h2><dl><dt>This version:</dt><dd>
      <a href="http://www.w3.org/TR/2010/WD-media-frags-20100624">http://www.w3.org/TR/2010/WD-media-frags-20100624</a>
    </dd><dt>Latest version:</dt><dd>
      <a href="http://www.w3.org/TR/media-frags">http://www.w3.org/TR/media-frags</a>
    </dd><dt>Previous version:</dt><dd>
      <a href="http://www.w3.org/TR/2010/WD-media-frags-20100413">http://www.w3.org/TR/2010/WD-media-frags-20100413</a>
    </dd><dt>Editors:</dt><dd><a href="http://www.eurecom.fr/~troncy/">
          Raphaël Troncy
        </a>, EURECOM</dd><dd><a href="http://multimedialab.elis.ugent.be/emannens">
          Erik Mannens
        </a>, IBBT Multimedia Lab, University of Ghent</dd><dd><a href="http://blog.gingertech.net/">
          Silvia Pfeiffer
        </a>, W3C Invited Expert</dd><dd><a href="http://multimedialab.elis.ugent.be/dvdeurse">
          Davy Van Deursen
        </a>, IBBT Multimedia Lab, University of Ghent</dd><dt>Contributors:</dt><dd><a href="http://www.deri.ie/about/team/member/Michael_Hausenblas/">
          Michael Hausenblas
        </a>, DERI, National University of Ireland, Galway</dd><dd><a href="mailto:philipj@opera.com">
          Philip Jägenstedt
        </a>, Opera Software</dd><dd><a href="http://www.cwi.nl/~jack/">
          Jack Jansen
        </a>, CWI, Centrum Wiskunde &amp; Informatica, Amsterdam</dd><dd><a href="mailto:ylafon@w3.org">
          Yves Lafon
        </a>, W3C</dd><dd><a href="http://www.kfish.org/">
          Conrad Parker
        </a>, W3C Invited Expert</dd></dl><p class="copyright"><a href="http://www.w3.org/Consortium/Legal/ipr-notice#Copyright">Copyright</a> © 2010 <a href="http://www.w3.org/"><acronym title="World Wide Web Consortium">W3C</acronym></a><sup>®</sup> (<a href="http://www.csail.mit.edu/"><acronym title="Massachusetts Institute of Technology">MIT</acronym></a>, <a href="http://www.ercim.org/"><acronym title="European Research Consortium for Informatics and Mathematics">ERCIM</acronym></a>, <a href="http://www.keio.ac.jp/">Keio</a>), All Rights Reserved. W3C <a href="http://www.w3.org/Consortium/Legal/ipr-notice#Legal_Disclaimer">liability</a>, <a href="http://www.w3.org/Consortium/Legal/ipr-notice#W3C_Trademarks">trademark</a> and <a href="http://www.w3.org/Consortium/Legal/copyright-documents">document use</a> rules apply.</p></div><hr /><div>
<h2><a id="abstract" name="abstract" />Abstract</h2><p>
        This document describes the Media Fragments 1.0 specification. It specifies the syntax for constructing media fragment URIs and explains how to handle them when used over the HTTP protocol. The syntax is based on the specification of particular field-value pairs that can be used in URI fragment and URI query requests to restrict a media resource to a certain fragment.
      </p></div><div>
<h2><a id="status" name="status" />Status of this Document</h2>

  <p><em>This section describes the status of this document at the
  time of its publication. Other documents may supersede this
  document. A list of current W3C publications and the latest revision
  of this technical report can be found in the <a href="http://www.w3.org/TR/">W3C technical reports index</a> at
  http://www.w3.org/TR/.</em></p>
  
  <p>This 24 June 2010 Last Call Working Draft of the Media Fragments URI 1.0 specification is an update of the previous
  <a href="http://www.w3.org/TR/2010/WD-media-frags-20100413/">Media Fragments URI 1.0 of 14 April 2010</a>. It meets the requirements specified 
  in the <a href="http://www.w3.org/TR/media-frags-reqs/">Use cases and requirements for Media Fragments</a> document and incorporates all comments
  received. This document is intended to be published and maintained as a W3C Recommendation after review and refinement.
  It has been produced by the <a href="http://www.w3.org/2008/WebVideo/Fragments/">Media Fragments Working Group</a>, which is part of the
  <a href="http://www.w3.org/2008/WebVideo/">W3C Video on the Web Activity</a>.</p>

  <p>The Media Fragments Working Group believes to have addressed all issues brought forth through previous Working Draft iterations. 
  The Working Group encourages feedback about this document by developers and researchers who have interest in multimedia content addressing
  and retrieval on the web and by developers and researchers who have interest in Semantic Web technologies 
  for content description and annotation. 
  Please send comments about this document to <a href="mailto:public-media-fragment@w3.org">public-media-fragment@w3.org</a>
  mailing list (<a href="http://lists.w3.org/Archives/Public/public-media-fragment/">public
  archive</a>) by <b>27 August 2010</b>.</p>  
  
  <p>
    Publication as a Working Draft does not imply endorsement by the
    W3C Membership. This is a draft document and may be updated,
    replaced or obsoleted by other documents at any time. It is
    inappropriate to cite this document as other than work in
    progress.
  </p><p>
    This document was produced by a group operating under the <a href="http://www.w3.org/Consortium/Patent-Policy-20040205/">
    5 February 2004 W3C Patent Policy</a>. W3C maintains a <a href="http://www.w3.org/2004/01/pp-impl/42785/status">public 
    list of any patent disclosures</a> made in connection with the 
    deliverables of the group; that page also includes instructions 
    for disclosing a patent. An individual who has actual knowledge 
    of a patent which the individual believes contains <a href="http://www.w3.org/Consortium/Patent-Policy-20040205/#def-essential">Essential 
    Claim(s)</a> must disclose the information in accordance with <a href="http://www.w3.org/Consortium/Patent-Policy-20040205/#sec-Disclosure">section 
    6 of the W3C Patent Policy</a>.
  </p></div><div class="toc">
<h2><a id="contents" name="contents" />Table of Contents</h2><p class="toc">1 <a href="#introduction">Introduction</a><br />
2 <a href="#standardisation">Standardisation Issues</a><br />
    2.1 <a href="#standardisation-terminology">Terminology</a><br />
    2.2 <a href="#standardisation-approach">Media Fragments Standardisation</a><br />
        2.2.1 <a href="#standardisation-URI-fragments">URI Fragments</a><br />
        2.2.2 <a href="#standardisation-URI-queries">URI Queries</a><br />
3 <a href="#fragment-query">URI fragment and URI query</a><br />
    3.1 <a href="#URIquery-vs-fragments">When to choose URI fragments? When to choose URI queries?</a><br />
    3.2 <a href="#URIfragment-user-agent">Resolving URI fragments within the user agent</a><br />
    3.3 <a href="#URIfragment-server">Resolving URI fragments with server help</a><br />
    3.4 <a href="#URIfragment-proxies">Resolving URI fragments in a proxy cacheable manner</a><br />
    3.5 <a href="#URIquery-media-fragments">Resolving URI queries</a><br />
    3.6 <a href="#URIquery-URIfragment">Combining URI fragments and URI queries</a><br />
4 <a href="#media-fragment-syntax">Media Fragments Syntax</a><br />
    4.1 <a href="#general-structure">General Structure</a><br />
    4.2 <a href="#url-serialization">URL Serialization</a><br />
    4.3 <a href="#fragment-dimensions">Fragment Dimensions</a><br />
        4.3.1 <a href="#naming-time">Temporal Dimension</a><br />
            4.3.1.1 <a href="#npt-time">Normal Play Time (NPT)</a><br />
            4.3.1.2 <a href="#smpte-time">SMPTE time codes</a><br />
            4.3.1.3 <a href="#clock-time">Wall-clock time code</a><br />
        4.3.2 <a href="#naming-space">Spatial Dimension</a><br />
        4.3.3 <a href="#naming-track">Track Dimension</a><br />
        4.3.4 <a href="#naming-name">Named Dimension</a><br />
        4.3.5 <a href="#common-syntax">Common Syntax</a><br />
5 <a href="#media-fragment-processing">Media Fragments Processing</a><br />
    5.1 <a href="#processing-protocol-frag">Protocol for URI fragment Resolution in HTTP</a><br />
        5.1.1 <a href="#processing-protocol-UA-mapped">UA mapped byte ranges</a><br />
            5.1.1.1 <a href="#processing-protocol-UA-mapped-new">UA requests URI fragment for the first time</a><br />
            5.1.1.2 <a href="#processing-protocol-UA-mapped-unchanged">UA requests URI fragment it already has buffered</a><br />
            5.1.1.3 <a href="#processing-protocol-UA-mapped-changed">UA requests URI fragment of a changed resource</a><br />
        5.1.2 <a href="#processing-protocol-Server-mapped">Server mapped byte ranges</a><br />
            5.1.2.1 <a href="#processing-protocol-server-mapped-default">Server mapped byte ranges with corresponding binary data</a><br />
            5.1.2.2 <a href="#processing-protocol-server-mapped-setup">Server mapped byte ranges with corresponding binary data and codec setup data</a><br />
            5.1.2.3 <a href="#processing-protocol-server-mapped-proxy">Proxy cacheable server mapped byte ranges</a><br />
        5.1.3 <a href="#server-triggered-redirect">Server triggered redirect</a><br />
    5.2 <a href="#processing-protocol-query">Protocol for URI query Resolution in HTTP</a><br />
6 <a href="#media-fragment-semantics">Media Fragments Semantics</a><br />
    6.1 <a href="#error-general">Errors on the General URI level</a><br />
        6.1.1 <a href="#error-general-non-existent">Non-existent dimension:</a><br />
        6.1.2 <a href="#error-general-underspec">Under-specified Dimension</a><br />
    6.2 <a href="#error-temporal">Errors on the temporal dimensions</a><br />
        6.2.1 <a href="#error-temporal-valid">Valid requests</a><br />
        6.2.2 <a href="#error-temporal-empty">Empty</a><br />
        6.2.3 <a href="#error-temporal-non-existent">Non-existent</a><br />
        6.2.4 <a href="#error-temporal-validity">Validity error</a><br />
    6.3 <a href="#error-spatial">Errors on the spatial dimensions</a><br />
    6.4 <a href="#error-track">Errors on the track dimensions</a><br />
    6.5 <a href="#error-named">Errors on the named dimensions</a><br />
7 <a href="#implementor-notes">Notes to Implementors</a><br />
    7.1 <a href="#media-fragment-display">Clients Displaying Media Fragments</a><br />
    7.2 <a href="#media-fragment-clients">All Media Fragment Clients</a><br />
    7.3 <a href="#media-fragment-servers">Media Fragment Servers</a><br />
8 <a href="#conclusions">Conclusions</a><br />
    8.1 <a href="#qualification-resources">Qualification of Media Resources</a><br />
</p>
<h3><a id="appendices" name="appendices" />Appendices</h3><p class="toc">A <a href="#references-normative">References</a><br />
B <a href="#collected-syntax-uri">Collected ABNF Syntax for URI</a> (Non-Normative)<br />
C <a href="#collected-syntax-http">Collected ABNF Syntax for HTTP Headers</a> (Non-Normative)<br />
D <a href="#processing-uri-syntax">Notes on parsing media fragment URIs</a> (Non-Normative)<br />
    D.1 <a href="#processing-name-value-components">Processing name-value components</a><br />
    D.2 <a href="#processing-name-value-lists">Processing name-value lists</a><br />
E <a href="#acknowledgments">Acknowledgements</a> (Non-Normative)<br />
F <a href="#change-log">Change Log</a> (Non-Normative)<br />
</p></div><hr /><div class="body"><div class="div1">
<h2><a id="introduction" name="introduction" />1 Introduction</h2><p>
        Audio and video resources on the World Wide Web are currently treated as "foreign" objects, which can only be embedded using a plugin that is capable of decoding and interacting with the media resource. Specific media servers are generally required to provide for server-side features such as direct access to time offsets into a video without the need to retrieve the entire resource. Support for such media fragment access varies between different media formats and inhibits standard means of dealing with such content on the Web.
      </p><p>
        This specification provides for a media-format independent, standard means of addressing media fragments on the Web using Uniform Resource Identifiers (URI). In the context of this document, media fragments are regarded along three different dimensions: temporal, spatial, and tracks. Further, a fragment can be marked with a name and then addressed through a URI using that name. The specified addressing schemes apply mainly to audio and video resources - the spatial fragment addressing may also be used on images.
      </p><p>
        The aim of this specification is to enhance the Web infrastructure for supporting the addressing and retrieval of subparts of time-based Web resources, as well as the automated processing of such subparts for reuse. Example uses are the sharing of such fragment URIs with friends via email, the automated creation of such fragment URIs in a search engine interface, or the annotation of media fragments with RDF. Such use case examples as well as other side conditions on this specification and a survey of existing media fragment addressing approaches are provided in the requirements <cite><a href="#mf-req">Use cases and requirements for Media Fragments</a></cite> document that accompanies this specification document.
      </p><p>
        The media fragment URIs specified in this document have been implemented and demonstrated to work with media resources over the HTTP protocol.
        This specification is not defining the protocol aspect of RTSP handling of media fragment. We expect the media fragment URI syntax to be
        generic and a possible mapping between this syntax and RTSP messages is currently drafted at <a href="http://www.w3.org/2008/WebVideo/Fragments/wiki/UA_Server_RTSP_Communication">http://www.w3.org/2008/WebVideo/Fragments/wiki/UA_Server_RTSP_Communication</a>.
        This may or may not be part of a future document authored by the Media Fragments Working Group.
        Existing media formats in their current representations and implementations provide varying degrees of support for this specification. 
        It is expected that over time, media formats, media players, Web Browsers, media and Web servers, as well as Web proxies will be extended 
        to adhere to the full specification. This specification will help make video a first-class citizen of the World Wide Web.
      </p></div><div class="div1">
<h2><a id="standardisation" name="standardisation" />2 Standardisation Issues</h2><div class="div2">
<h3><a id="standardisation-terminology" name="standardisation-terminology" />2.1 Terminology</h3><p>
          The keywords <strong>MUST</strong>, <strong>MUST NOT</strong>, <strong>SHOULD</strong> and <strong>SHOULD NOT</strong> are to be interpreted as defined in <cite><a href="#rfc2119">RFC 2119</a></cite>.
        </p><p>
          According to <cite><a href="#rfc3986">RFC 3986</a></cite>, the term "URI" does not include relative
          references. In this document, we consider both URIs and relative
          references. Consequently, we use the term "URI reference" as defined in <cite><a href="#rfc3986">RFC 3986</a></cite> (section 4.1).
          For simplicity reasons, this document, however, only uses the term "media fragment URI" in place of "media fragment URI
          reference".
        </p><p>
          The following terms are used frequently in this document and need to be clearly understood:
          </p>
          <ul><li>URI fragment = anything behind a "#" in a URI</li><li>URI query    = anything behind a "?" and before a "#" in a URI</li><li>media fragment URI = a URI addressing subparts of a media resource - that could be through URI queries or URI fragments</li></ul>
        </div><div class="div2">
<h3><a id="standardisation-approach" name="standardisation-approach" />2.2 Media Fragments Standardisation</h3><p>
          The basis for the standardisation of media fragment URIs is the URI specification, <cite><a href="#rfc3986">RFC 3986</a></cite>. Providing media fragment identification information in URIs refers here to the specification of the structure of a URI fragment or a URI query. This document will explain how URI fragments and URI queries are structured to identify media fragments. It normalises the name-value parameters used in URI fragments and URI queries to address media fragments. These build on existing CGI parameter conventions.
        </p><p>
           In this section, we look at implications of standardising the structure of media fragment URIs.
        </p><div class="div3">
<h4><a id="standardisation-URI-fragments" name="standardisation-URI-fragments" />2.2.1 URI Fragments</h4><p>
            The URI specification <cite><a href="#rfc3986">RFC 3986</a></cite> says about the format of a URI fragment in Section 3.5:
          </p><p>
            <i>"The fragment's format and resolution is [..] dependent on the media type [RFC2046] of a potentially retrieved representation. [..] Fragment identifier semantics are independent of the URI scheme and thus cannot be redefined by scheme specifications."</i>
          </p><p>
            This essentially means that only media type definitions (as registered through the process defined in <cite><a href="#rfc4288">RFC 4288</a></cite>) are able to introduce a standard structure on URI fragments for that mime type. One part of the registration process of a media type can include information about how fragment identifiers in URIs are constructed for use in conjunction with this media type.
          </p><p>
             Note that the registration of URI fragment construction rules as expressed in Section 4.11 of <cite><a href="#rfc4288">RFC 4288</a></cite> is only a SHOULD-requirement. An analysis of all media type registrations showed that there is not a single media type registration in the audio/*, image/*, video/* branches that is currently defining fragments or fragment semantics.
          </p><p>
            The Media Fragment WG has no authority to update registries of all targeted media types. To the best of our knowledge there are only <a href="http://www.w3.org/2008/WebVideo/Fragments/wiki/MediaTypeReview">few media types  that actually have a specified fragment format</a> even if it is not registered with the media type: these include Ogg, MPEG-4, and MPEG-21. Further, only a small number of software packages actually supports these fragment formats. For all others, the semantics of the fragment are  considered to be unknown.
          </p><p>
            As such, the intention of this document is to propose a specification to all media type owners in the audio/*, image/*, and video/* branches for a structured approach to URI fragments and for specification of commonly agreed dimensions to address media fragments (i.e. subparts of a media resource) through URI fragments. We recommend media type owners to harmonize their existing schemes with the ones proposed in this document and update or add the fragment semantics specification to their media type registration.
          </p></div><div class="div3">
<h4><a id="standardisation-URI-queries" name="standardisation-URI-queries" />2.2.2 URI Queries</h4><p>The URI specification <cite><a href="#rfc3986">RFC 3986</a></cite> says about the format of a URI query in Section 3.4:
          </p><p>
            <i>"The query component [..] serves to identify a resource within the scope of the URI's scheme and naming authority (if any). [..] Query components are often used to carry identifying information in the form of "key=value" pairs [..]."</i>
          </p><p>
            URI query specifications are more closely linked to the URI scheme, some of which do not even use a query component. We are mostly concerned with the HTTP <cite><a href="#rfc2616">RFC 2616</a></cite> and the RTP/RTSP <cite><a href="#">rfc2326</a></cite> protocols here, which both support query components. HTTP says nothing about how a URI query has to be interpreted. RTSP explicitly says that fragment and query identifiers do not have a well-defined meaning at this time, with the interpretation left to the RTSP server.
          </p><p>
            The URI specification <cite><a href="#rfc3986">RFC 3986</a></cite> says generally that the data within the URI is often parsed by both the user agent and one or more servers. It refers in particular to HTTP in Section 7.3:
          </p><p>
            <i>"In HTTP, for example, a typical user agent will parse a URI into its five major components, access the authority's server, and send it the data within the authority, path, and query components.  A typical server will take that information, parse the path into segments and the query into key/value pairs, and then invoke implementation-specific handlers to respond to the request."</i>
          </p><p>
            Since the interpretation of query components resides with the functionality of servers, the intention of this document wrt query components is to recommend standard name-value pair formats for use in addressing media fragments through URI queries. We recommend server and server-type software providers to harmonize their existing schemes in use with media resources to support the nomenclature proposed in this specification.
          </p></div></div></div><div class="div1">
<h2><a id="fragment-query" name="fragment-query" />3 URI fragment and URI query</h2><table border="1" summary="Editorial note"><tr><td width="50%" valign="top" align="left"><b>Editorial note</b></td><td width="50%" valign="top" align="right"> </td></tr><tr><td valign="top" align="left" colspan="2">This section is non-normative</td></tr></table><p>
        To address a media fragment, one needs to find ways to convey the fragment information. This specification builds on URIs <cite><a href="#rfc3986">RFC 3986</a></cite>. Every URI is defined as consisting of four parts, as follows:
      </p><div class="exampleOuter"><p>
        &lt;scheme name&gt; : &lt;hierarchical part&gt; [ ? &lt;query&gt; ] [ # &lt;fragment&gt; ]
      </p></div><p>
        There are therefore two possibilities for representing the media fragment addressing in URIs: the <em>URI query part</em> or the <em>URI fragment part</em>.
      </p><div class="div2">
<h3><a id="URIquery-vs-fragments" name="URIquery-vs-fragments" />3.1 When to choose URI fragments? When to choose URI queries?</h3><p>
        For media fragment addressing, both approaches - URI query and URI fragment - are useful.
      </p><p>
        The main difference between a URI query and a URI fragment is that a URI query produces a new resource, while a URI fragment provides a secondary resource that has a relationship to the primary resource. URI fragments are resolved from the primary resource without another retrieval action. This means that a user agent should be capable to resolve a URI fragment on a resource it has already received without having to fetch more data from the server.
      </p><p>
        A further requirement put on a URI fragment is that the media type of the retrieved fragment should be the same as the media type of the primary resource. Among other things, this means that a URI fragment that points to a single video frame out of a longer video results in a one-frame video, not in a still image. To extract a still image, one would need to create a URI query scheme - something not envisaged here, but easy to devise.
      </p><p>
        There are different types of media fragment addressing in this specification. As noted in the <cite><a href="#mf-req">Use cases and requirements for Media Fragments</a></cite> document (section "Fitness Conditions on Media Containers/Resources"): not all container formats and codecs are "fit" for supporting the different types of fragment URIs. "Fitness" relates to the fact that a media fragment can be extracted from the primary resource without syntax element modifications or transcoding of the bitstream.
      </p><p>
        Resources that are "fit" can therefore be addressed with a URI fragment. Resources that are "conditionally fit" can be addressed with a URI fragment with an additional retrieval action that retrieves the modified syntax elements but leaves the codec data untouched. Resources that are "unfit" require transcoding. Such transcoded media fragments cannot be addressed with URI fragments, but only with URI queries.
      </p><p>
        Therefore, when addressing a media fragment with the URI mechanism, the author has to know whether this media fragment can be produced from the (primary) resource itself without any transcoding activities or whether it requires transcoding. In the latter case, the only choice is to use a URI query and to use a server that supports transcoding and delivery of a (primary) derivative resource to satisfy the query.
      </p></div><div class="div2">
<h3><a id="URIfragment-user-agent" name="URIfragment-user-agent" />3.2 Resolving URI fragments within the user agent</h3><p>
        A user agent may itself resolve and control the presentation of media fragment URIs. The simplest case arises where the user agent has
        already downloaded the entire resource and can perform the extraction from its locally cached copy. For some media types, it may also be
        possible to perform the extraction over the network without any special protocol assistance. For temporal fragments this requires a user
        agent to be able to seek on the media resource using existing protocol mechanisms.
      </p><p>
        An example of a URI fragment used to address a media fragment is <code>http://www.example.org/video.ogv#t=60,100</code>. In this case,
        the user agent knows that the primary resource is <code>http://www.example.org/video.ogv</code> and that it is only expected to display
        the portion of the primary resource that relates to the fragment <code>#t=60,100</code>, i.e. seconds 60-100. Thus, the relationship
        between the primary resource and the media fragment is maintained.
      </p><p>
        In traditional URI fragment retrieval, a user agent requests the complete primary resource from the server and then applies the
        fragmentation locally. In the media fragment case, this would result in a retrieval action on the complete media resource, on which the
        user agent would then locally perform its fragment extraction - something generally unviable for such large resources.
      </p><p>
        Therefore, media resources are not always retrieved over HTTP using a single request. They may be retrieved as a sequence of byte range requests
        on the original resource URI, or may be retrieved as a sequence of requests to different URIs each representing a small part of the
        media. The reasons for such mechanisms include bandwidth conservation, where a client chooses to space requests out over time during
        playback in order to maximize bandwidth available for other activities, and bandwidth adaptation, where a client selects among various
        representations with varying bitrate depending on the current bandwidth availability.
      </p><p>
        A user agent that knows how to map media fragments to byte ranges will be able to satisfy a URI fragment request such as the above example
        by itself. This is typically the case for user agents that know how to seek to media fragments over the network.
        For example, a user agent that deals with a media file that includes an index of its seekable structures can resolve the media fragment
        addresses to byte ranges from the index. This is the case e.g. with seekable QuickTime files. Another example is a user agent that
        knows how to seek on a media file through a sequence of byte range requests and eventually receives the correct media fragment. This is
        the case e.g. with Ogg files in Firefox versions above 3.5.
      </p><p>
        Similarly, a user agent that knows how to map media fragments to a sequence of URIs can satisfy a URI fragment request by itself. This
        is typically the case for user agents that perform adaptive streaming. For example, a user agent that deals with a media resource that
        contains a sequence of URIs, each a media file of a few seconds duration, can resolve the media fragment addresses to a subsequence of
        those URIs. This is the case with QuickTime adaptive bitrate streaming or IIS Smooth Streaming.
      </p><p>
        If such a user agent natively supports the media fragment syntax as specified in this document, it is deemed conformant to this
        specification for fragments and for the particular dimension.
      </p></div><div class="div2">
<h3><a id="URIfragment-server" name="URIfragment-server" />3.3 Resolving URI fragments with server help</h3><p>
        For user agents that natively support the media fragment syntax, but have to use their own seeking approach, this specification provides
        an optimisation that can make the byte offset seeking more efficient. It requires a conformant server with which the user agent will follow
        a protocol defined later in this document.
      </p><p>
        In this approach, the user agent asks the server to do the byte range mapping for the media fragment address itself and send back the
        appropriate byte ranges. This can not be done through the URI, but has to be done through adding protocol headers. User agents that
        interact with a conformant server to follow this protocol will receive the appropriate byte ranges directly and will not need to do costly
        seeking over the network.
      </p><p>
        Note that it is important that the server also informs the user agent what actual media fragment range it was able to retrieve. This is
        important since in the compressed domain it is not possible to extract data at an arbitrary resolution, but only at the resolution
        that the data was packaged in. For example, even if a user asked for <code>http://www.example.org/video.ogv#t=60,100</code> and the
        user agent sent a range request of <code>t=60,100</code> to the server, the server may only be able to return the range
        <code>t=58,103</code> as the closest decodable range that encapsulates all the required data.
      </p><p>
        Note that if done right, the native user agent support for media fragments and the improved server support can be integrated without
        problems: the user agent just needs to include the byte range and the media fragment range request in one request. A server that does not
        understand the media fragment range request will only react to the byte ranges, while a server that understands them will ignore the
        byte range request and only reply with the correct byte ranges. The user agent will understand from the response whether it received a
        reply to the byte ranges or the media fragment ranges request and can react accordingly.
      </p></div><div class="div2">
<h3><a id="URIfragment-proxies" name="URIfragment-proxies" />3.4 Resolving URI fragments in a proxy cacheable manner</h3><p>
        The current setup of the World Wide Web relies heavily on the use of caching Web proxies to speed up the delivery of content. In the
        case of URI fragments that are resolved by the server as indicated in the previous section, existing Web proxies have no means of
        caching these requests since they only understand byte ranges.
      </p><p>
        To make use of the existing Web proxy infrastructure of the Web, we need to make sure that the user agent only asks for byte ranges,
        so they can be served from the cache. This is possible if the server - instead of replying with the actual data - replies with the mapped
        byte ranges for the requested media fragment range. Then, the user agent is able to resend his range request this time with bytes only,
        which can possibly already be satisfied from the cache. Details of this will be specified later.
      </p><table border="1" summary="Editorial note: Raphael"><tr><td width="50%" valign="top" align="left"><b>Editorial note: Raphael</b></td><td width="50%" valign="top" align="right"> </td></tr><tr><td valign="top" align="left" colspan="2">Should we not foresee future "smart" media caches that would be able to actually cache range request in other units than bytes?
        </td></tr></table></div><div class="div2">
<h3><a id="URIquery-media-fragments" name="URIquery-media-fragments" />3.5 Resolving URI queries</h3><p>
        The described URI fragment addressing methods only work for byte-identical segments of a media resource, since we assume a simple mapping
        between the media fragment and bytes that each infrastructure element can deal with. Where it is impossible to maintain byte-identity
        and some sort of transcoding of the resource is necessary, the user agent is not able to resolve the fragmentation by itself and a
        server interaction is required. In this case, URI queries have to be used since they result in a server interaction and can deliver a
        transcoded resource.
      </p><p>
        Another use for URI queries is when a user agent actually wants to receive a completely new resource instead of just a byte range from
        an existing (primary) resource. This is, for example, the case for playlists of media fragment resources. Even if a media fragment
        could be resolved through a URI fragment, the URI query may be more desirable since it does not carry with itself the burden
        of the original primary resource - its file headers may be smaller, its duration may be smaller, and it does not automatically allow
        access to the remainder of the original primary resource.
      </p><p>
        When URI queries are used, the retrieval action has to additionally make sure to create a fully valid new resource. For example, for the
        Ogg format, this implies a reconstruction of Ogg headers to accurately describe the new resource (e.g. a non-zero start-time or different
        encoding parameters). Such a resource will be cached in Web proxies as a different resource to the original primary resource.
      </p><p>
        An example URI query that includes a media fragment specification is <code>http://www.example.org/video.ogv?t=60,100</code>. This results
        in a video of duration 40s (assuming the original video was more than 100s long).
      </p><p>
        Note that this resource has no per-se relationship to the original primary resource. As a user agent uses such a URI with e.g. a HTML5
        video element, the browser has no knowledge about the original resource and can only display this video as a 40s long video starting at 0s.
        The context of the original resource is lost.
      </p><p>
        A user agent may want to display the original start time of the resource as the start time of the video in order to be consistent with the
        information in the URI. It is possible to achieve this in one of two ways: either the video file itself has some knowledge that it is an
        extract from a different file and starts at an offset - or the user agent is told through the retrieval action which original primary
        resource the retrieved resource relates to and can find out information about it through another retrieval action. This latter option
        will be regarded later in this document.
      </p><p>
        An example for a media resource that has knowledge about itself of the required kind are Ogg files. Ogg files that have a skeleton
        track and were created correctly from the primary resource will know that their start time is not 0s but 60s in the above example.
        The browser can simply parse this information out of the received bitstream and may display a timeline that starts at 60s and ends at 100s
        in the video controls if it so desires.
      </p><p>
        Another option is that the browser parses the URI and knows about how media resources have a fragment specification that follows a standard.
        Then the browser can interpret the query parameters and extract the correct start and end times and also the original primary resource.
        It can then also display a timeline that starts at 60s and ends at 100s in the video controls. Further it can allow a right-click menu to
        click through to the original resource if required.
      </p><p>
         A use case where the video controls may neither start at 0s nor at 60s is a mashed-up video created through a list of media fragment URIs.
         In such a playlist, the user agent may prefer to display a single continuous timeline across all the media fragments rather than a
         collection of individual timelines for each fragment. Thus, the 60s to 100s fragment may e.g. be mapped to an interval at 3min20 to 4min.
      </p><p>
        No new protocol headers are required to execute a URI query for media fragment retrieval. Some optional protocol headers that improve the
        information exchange will be recommended later in this document.
      </p></div><div class="div2">
<h3><a id="URIquery-URIfragment" name="URIquery-URIfragment" />3.6 Combining URI fragments and URI queries</h3><p>
        A combination of a URI query for a media fragment with a URI fragment yields a URI fragment resolution on top of the newly created resource.
        Since a URI with a query part creates a new resource, we have to do the fragment offset on the new resource. This is simply a conformant
        behaviour to the URI standard <cite><a href="#rfc3986">RFC 3986</a></cite>.
      </p><p>
        For example, <code>http://www.example.org/video.ogv?t=60,100#t=20</code> will lead to the 20s fragment offset being applied to the new
        resource starting at 60 going to 100. Thus, the reply to this is a 40s long resource whose playback will start at an offset of 20s.
      </p><table border="1" summary="Editorial note: Silvia"><tr><td width="50%" valign="top" align="left"><b>Editorial note: Silvia</b></td><td width="50%" valign="top" align="right"> </td></tr><tr><td valign="top" align="left" colspan="2">
          We should at the end of the document set up a table with all the different addressing types and http headers and say what we deem
          is conformant and how to find out whether a server or user agent is conformant or not.
        </td></tr></table></div></div><div class="div1">
<h2><a id="media-fragment-syntax" name="media-fragment-syntax" />4 Media Fragments Syntax</h2><p>
        This section describes the external representation of a media fragment specifier, and how this should be interpreted.
      </p><p>
        Guiding principles for the definition of the media fragments syntax were as follows:</p>
        <ul><li>a. The MF syntax for queries and fragments should be identical</li><li>b. The MF syntax should be unambiguous</li><li>c. The MF syntax should allow any UTF-8 character in track or id names</li><li>d. The MF syntax should adhere to applicable formal standards</li><li>e. The MF syntax should adhere to de-facto usage of queries and fragments</li><li>f. The MF syntax should be as concise as possible, with no unneeded grammatical fluff</li></ul>
      <div class="div2">
<h3><a id="general-structure" name="general-structure" />4.1 General Structure</h3><p>
          The fragment identifier consists of a list of name/value pairs, the
          dimension specifiers, separated by the primary separator
          <code>&amp;</code>. Name and value are separated by an equal
          sign (<code>=</code>).
        </p><p>Here are some examples of URIs with name-value pairs, to
        demonstrate the general structure:</p><div class="exampleInner"><pre>
http://www.example.com/example.ogv#a=b&amp;c=d
http://www.example.com/example.ogv#t=10,20
http://www.example.com/example.ogv#track=audio
http://www.example.com/example.ogv#track=audio&amp;t=10,20</pre></div></div><div class="div2">
<h3><a id="url-serialization" name="url-serialization" />4.2 URL Serialization</h3><p>
          This ABNF syntax defines the structure of media fragment URI components.
        </p><p>Media fragments support fragmenting the media along four dimensions:</p><dl><dt class="label">temporal</dt><dd><p>
                This dimension denotes a specific time range in the
                original media, such as "starting at second 10, continuing
                until second 20";
              </p></dd><dt class="label">spatial</dt><dd><p>
                this dimension denotes a specific range of pixels in
                the original media, such as "a rectangle with size (100,100)
                with its top-left at coordinate (10,10)";
              </p></dd><dt class="label">track</dt><dd><p>
                this dimension denotes one or more tracks in the
                original media, such as "the english audio and the video track";
              </p></dd><dt class="label">named</dt><dd><p>
                this dimension denotes a named section of the original
                media, such as "chapter 2".
              </p></dd></dl><p>
          Note that a general URI fragment or query string specified on a media resource may contain several field-value pairs. They are not restricted to the ones specified here, since applications may want to use additional other parameters to communicate further requests to custom servers.
        </p><table border="1" summary="Editorial note: Philip"><tr><td width="50%" valign="top" align="left"><b>Editorial note: Philip</b></td><td width="50%" valign="top" align="right"> </td></tr><tr><td valign="top" align="left" colspan="2">
            <p>
              We need to decide what do with this and validity. One
              suggested option has been to require non-MF extensions
              to use a vendor prefix.
            </p>
          </td></tr></table><p>
          The temporal, spatial and track dimensions are logically independent and can be combined; the outcome is independent of the order of the dimensions.
        </p><p>
          The track dimension refers to one of a set of parallel media streams (e.g. "the english audio track for a video"), not to a (possibly self-contained) section of the source media (e.g. "Audio track 2 of a CD").
        </p><p>
          The name dimension cannot be combined with the other dimensions, because the semantics depend on the underlying source media format: some media formats support naming of temporal extents, others support naming of groups of tracks, etc. Error semantics are discussed in <a href="#media-fragment-semantics"><b>6 Media Fragments Semantics</b></a>.
        </p><p>
          A conformant server or user agent will need to be able to parse a random URI query or fragment string for a media resource and identify the relevant parts. E.g. the relevant field-value pair out of a media fragment URI like this <code>http://www.example.com/video.ogv#&amp;&amp;=&amp;=tom;jerry=&amp;t=34&amp;t=meow:0#</code> is <code>t=34</code>.
        </p><p>
          A specification of the parsing algorithm to extract these from an actual URI can be found in <a href="#processing-name-value-components"><b>D.1 Processing name-value components</b></a> and <a href="#processing-name-value-lists"><b>D.2 Processing name-value lists</b></a>. Note that the URI works on octet strings, but the parsed name-value pairs are unicode strings, since percent-encoding is resolved. The following  definitions apply to these unicode strings.
        </p></div><div class="div2">
<h3><a id="fragment-dimensions" name="fragment-dimensions" />4.3 Fragment Dimensions</h3><div class="div3">
<h4><a id="naming-time" name="naming-time" />4.3.1 Temporal Dimension</h4><p>
          Temporal clipping is denoted by the name <code>t</code>, and specified as an interval with a begin time and an end time (or an in-point and an out-point, in video editing terms). Either or both may be omitted, with the begin time defaulting to 0 seconds and the end time defaulting to the duration of the source media. The interval is half-open: the begin time is considered part of the interval whereas the end time is considered to be the first time point that is not part of the interval. If a single number only is given, this is the begin time.
          </p><p>Examples:</p><div class="exampleInner"><pre>
t=10,20   # =&gt; results in the time interval [10,20)
t=,20     # =&gt; results in the time interval [0,20)
t=10,     # =&gt; results in the time interval [10,end)
t=10      # =&gt; also results in the time interval [10,end)</pre></div><p>
          Temporal clipping can be specified either as Normal Play Time (npt) <cite><a href="#rtsp">RFC 2326</a></cite>, as SMPTE timecodes, <cite><a href="#smpte">SMPTE</a></cite>, or as real-world clock time (clock) <cite><a href="#rtsp">RFC 2326</a></cite>. Begin and end times are always specified in the same format. The format is specified by name, followed by a colon (<code>:</code>), with <code>npt:</code> being the default.
          </p><div class="exampleInner"><a id="timesegment" name="timesegment" /><pre>
          timeprefix    = %x74                                      ; "t"
          timeparam     = <a href="#npttimedef">npttimedef</a> / <a href="#smptetimedef">smptetimedef</a> / <a href="#clocktimedef">clocktimedef</a>
          </pre></div><p>
          In this version of the media fragments specification there is no extensibility mechanism to add time format specifiers.
          </p><div class="div4">
<h5><a id="npt-time" name="npt-time" />4.3.1.1 Normal Play Time (NPT)</h5><p>
          Normal Play Time can either be specified as seconds, with an optional fractional part to indicate miliseconds, or as colon-separated hours, minutes and seconds (again with an optional fraction). Minutes and seconds must be specified as exactly two digits, hours and fractional seconds can be any number of digits. The hours, minutes and seconds specification for NPT is a convenience only, it does not signal frame accuracy. The specification of the "npt:" identifier is optional since NPT is the default time scheme. This specification builds on the RTSP specification of NPT <cite><a href="#rtsp">RFC 2326</a></cite>.
          </p><div class="exampleInner"><a id="npttimedef" name="npttimedef" /><pre>
npt-sec       = 1*DIGIT [ "." *DIGIT ]                     ; definitions taken
npt-hhmmss    = npt-hh ":" npt-mm ":" npt-ss [ "." *DIGIT] ; from <cite><a href="#rtsp">RFC 2326</a></cite>
npt-hh        =   1*DIGIT     ; any positive number
npt-mm        =   2DIGIT      ; 0-59
npt-ss        =   2DIGIT      ; 0-59

npttimedef    = [ deftimeformat ":"] ( npttime  [ "," npttime ] ) / ( "," npttime )

deftimeformat = %x6E.70.74                                ; "npt"
npttime       = npt-sec / npt-hhmmss
</pre></div><p>Examples:</p><div class="exampleInner"><pre>
t=npt:10,20         # =&gt; results in the time interval [10,20)
t=npt:120,          # =&gt; results in the time interval [120,end)
t=npt:,121.5        # =&gt; results in the time interval [0,121.5)
t=0:02:00,121.5     # =&gt; results in the time interval [120,121.5)
t=npt:120,0:02:01.5 # =&gt; also results in the time interval [120,121.5)</pre></div></div><div class="div4">
<h5><a id="smpte-time" name="smpte-time" />4.3.1.2 SMPTE time codes</h5><p>
          SMPTE time codes are a way to address a specific frame (or field) without running the risk of rounding errors causing a different frame to be selected. The format is always colon-separated hours, minutes, seconds and frames. Frames are optional, defaulting to 00. If the source format has a further subdivison of frames (such as odd/even fields in interlaced video) these can be specified further with a number after a dot (<code>.</code>). The SMPTE format name must always be specified, because the interpretation of the fields depends on the format. The SMPTE formats supported in this version of the specification are:
          </p><ul><li><code>smpte</code>,</li><li><code>smpte-25</code>,</li><li><code>smpte-30</code> and </li><li><code>smpte-30-drop</code>.</li></ul><p>
            <code>smpte</code> is a synonym for <code>smpte-30</code>.
          </p><div class="exampleInner"><a id="smptetimedef" name="smptetimedef" /><pre>
smptetimedef  = smpteformat ":"( frametime [ "," frametime ] ) / ( "," frametime )
smpteformat   = %x73.6D.70.74.65                          ; "smpte"
               / %x73.6D.70.74.65.2D.32.35                ; "smpte-25"
               / %x73.6D.70.74.65.2D.33.30                ; "smpte-30"
               / %x73.6D.70.74.65.2D.33.30.2D.64.72.6F.70 ; "smpte-30-drop"
frametime     = 1*DIGIT ":" 2DIGIT ":" 2DIGIT [ ":" 2DIGIT [ "." 2DIGIT ] ]
</pre></div><p>Examples:</p><div class="exampleInner"><pre>
t=smpte-30:0:02:00,0:02:01:15       # =&gt; results in the time interval [120,121.5)
t=smpte-25:0:02:00:00,0:02:01:12.40 # =&gt; results in the time interval [120,121.5)
                                    #    (80 or 100 subframes per frame seem typical)
</pre></div><p>
          Using SMPTE timecodes may result in frame-accurate begin and end times, but only if the timecode format used in the media fragment specifier is the same as that used in the original media item.
          </p></div><div class="div4">
<h5><a id="clock-time" name="clock-time" />4.3.1.3 Wall-clock time code</h5><p>
          Wall-clock time codes are a way to address real-world clock time that is associated typically with a live video stream. These are the same time codes that are being used by RTSP <cite><a href="#rtsp">RFC 2326</a></cite>, by SMIL <cite><a href="#smil30">SMIL</a></cite>, and by HTML5 <cite><a href="#html5">HTML 5</a></cite>. The scheme uses ISO 8601 UTC timestamps (http://www.iso.org/iso/date_and_time_format). The format separates the date from the time with a "T" character and the string ends with "Z", which includes time zone capabilities. To that effect, the ABNF grammar is referring to <cite><a href="#rfc3339">RFC 3339</a></cite>, which include the relevant part of ISO 8601 in ABNF form. The time scheme identifier is "clock".
          </p><div class="exampleInner"><a id="clocktimedef" name="clocktimedef" /><pre>
datetime      = &lt;date-time, defined in <cite><a href="#rfc3339">RFC 3339</a></cite>&gt;

clocktimedef  = clockformat ":"( clocktime [ "," clocktime ] ) / ( "," clocktime )
clockformat   = %x63.6C.6F.63.6B                          ; "clock"
clocktime     = (datetime / walltime / date)

; WARNING: if your date-time contains '+' (or any other reserved character, per <cite><a href="#rfc3986">RFC 3986</a></cite>),
; it should be percent-encoded when used in a URI.
</pre></div><p>For convenience, the definition is copied here</p><div class="exampleInner"><a id="datetime" name="datetime" /><pre>
; defined in <cite><a href="#rfc3339">RFC 3339</a></cite>
;
date-fullyear   = 4DIGIT
date-month      = 2DIGIT  ; 01-12
date-mday       = 2DIGIT  ; 01-28, 01-29, 01-30, 01-31 based on
                          ; month/year
time-hour       = 2DIGIT  ; 00-23
time-minute     = 2DIGIT  ; 00-59
time-second     = 2DIGIT  ; 00-58, 00-59, 00-60 based on leap second
                          ; rules
time-secfrac    = "." 1*DIGIT
time-numoffset  = ("+" / "-") time-hour ":" time-minute
time-offset     = "Z" / time-numoffset

partial-time    = time-hour ":" time-minute ":" time-second
                  [time-secfrac]
full-date       = date-fullyear "-" date-month "-" date-mday
full-time       = partial-time time-offset

date-time       = full-date "T" full-time
</pre></div><p>Examples:</p><div class="exampleInner"><pre>
t=clock:2009-07-26T11:19:01Z,2009-07-26T11:20:01Z   # =&gt; results in a 1 min interval
                                                    # on 26th Jul 2009 from 11hrs, 19min, 1sec
t=clock:2009-07-26T11:19:01Z                        # =&gt; starts on 26th Jul 2009 from 11hrs, 19min, 1sec
t=clock:,2009-07-26T11:20:01Z                       # =&gt; ends on 26th Jul 2009 from 11hrs, 20min, 1sec</pre></div></div></div><div class="div3">
<h4><a id="naming-space" name="naming-space" />4.3.2 Spatial Dimension</h4><p>
          Spatial clipping selects an area of pixels from visual media streams. For this release of the media fragment specification, only rectangular selections are supported. The rectangle can be specified as pixel coordinates or percentages.
          </p><p>
          Rectangle selection is denoted by the name <code>xywh</code>. The value is an optional format <code>pixel:</code> or <code>percent:</code> (defaulting to pixel) and 4 comma-separated integers. The integers denote x, y, width and height, respectively, with x=0, y=0 being the top left corner of the image. If percent is used, x and width are interpreted as a percentage of the width of the original media, and y and height are interpreted as a percentage of the original height.
          </p><div class="exampleInner"><a id="spacesegment" name="spacesegment" /><pre>
xywhprefix    = %x78.79.77.68                             ; "xywh"
xywhparam     = [ xywhunit ":" ] 1*DIGIT "," 1*DIGIT "," 1*DIGIT "," 1*DIGIT
xywhunit      = %x70.69.78.65.6C                          ; "pixel"
              / %x70.65.72.63.65.6E.74                    ; "percent"
</pre></div><p>Examples:</p><div class="exampleInner"><pre>
xywh=160,120,320,240        # =&gt; results in a 320x240 box at x=160 and y=120
xywh=pixel:160,120,320,240  # =&gt; results in a 320x240 box at x=160 and y=120
xywh=percent:25,25,50,50    # =&gt; results in a 50%x50% box at x=25% and y=25%</pre></div></div><div class="div3">
<h4><a id="naming-track" name="naming-track" />4.3.3 Track Dimension</h4><p>
        Track selection allows the extraction of tracks (audio, video, subtitles, etc) from a media container that supports multiple tracks. Track selection is denoted by the name <code>track</code>. The value is a string. Percent-escaping can be used in the string to specify unsafe characters (including separators such as semi-colon), see the grammar below for details. Multiple track specification is allowed, but requires the specification of multiple track parameters. Interpretation of the string depends on the container format of the original media: some formats allow numbers only, some allow full names.
        </p><div class="exampleInner"><a id="tracksegment" name="tracksegment" /><pre>
trackprefix   = %x74.72.61.63.6B                          ; "track"
trackparams   = trackparam
trackparam    = utf8string
</pre></div><p>Examples:</p><div class="exampleInner"><pre>
track=1                         # =&gt; results in only extracting track 1
track=video&amp;track=subtitle  # =&gt; results in extracting track 'video' and track 'subtitle'
track=Wide%20Angle%20Video      # =&gt; results in only extracting track 'Wide Angle Video'</pre></div><p>
        As the allowed track names are determined by the original source media, this information has to be known before construction of the media fragment. There is no support for generic media type names (audio, video) across container formats: most container formats allow multiple tracks of each media type, which would lead to ambiguities. Note that there are existing discovery mechanisms for retrieving the track names of a media resource, such as the Rich Open multitrack media Exposition format (ROE) <cite><a href="#roe">ROE</a></cite> or the Media Annotations API <cite><a href="#mediaAnnotations">Media Annotations</a></cite>.
      </p><table border="1" summary="Editorial note: Davy"><tr><td width="50%" valign="top" align="left"><b>Editorial note: Davy</b></td><td width="50%" valign="top" align="right"> </td></tr><tr><td valign="top" align="left" colspan="2">
          <p>We can also reference the HTML5 Media Multitrack API here, when it's mentioned in the HTML5 spec.
          </p>
        </td></tr></table></div><div class="div3">
<h4><a id="naming-name" name="naming-name" />4.3.4 Named Dimension</h4><p>
      Name-based selection is denoted by the name <code>id</code>, with the value being a string enclosed in single quotes. Percent-escaping can be used in the string to include unsafe characters such as a single quote, see the grammer below for details. Interpretation of the string depends on the underlying container format: some container formats support named chapters or numbered chapters (leading to temporal clipping), some may support naming of groups of tracks or other objects. As with track selection, determining which names are valid requires knowledge of the original media item.
      </p><div class="exampleInner"><a id="namesegment" name="namesegment" /><pre>
nameprefix    = %x69.64                                   ; "id"
nameparam     = utf8string
</pre></div><p>Examples:</p><div class="exampleInner"><pre>
id=1               # =&gt; results in only extracting the section called '1'
id=chapter-1       # =&gt; results in only extracting the section called 'chapter-1'
id=Airline%20Edit  # =&gt; results in only extracting the section called 'Airline Edit'</pre></div><p>
      Note that, despite the use of the name <code>id</code>, there is no correspondence to XML <code>id</code>: the values are uninterpreted strings, from the point of view of media fragment handling.
      </p></div><div class="div3">
<h4><a id="common-syntax" name="common-syntax" />4.3.5 Common Syntax</h4><p>For convenience, the following definitions are copied here. Only the definitions in the original documents are considered normative</p><div class="exampleInner"><pre>
DIGIT         = &lt;DIGIT, defined in <cite><a href="#rfc4234">RFC 4234</a></cite>&gt;
pchar         = &lt;pchar, defined in <cite><a href="#rfc3986">RFC 3986</a></cite>&gt;
unreserved    = &lt;unreserved, defined in <cite><a href="#rfc3986">RFC 3986</a></cite>&gt;
pct-encoded   = &lt;pct-encoded, defined in <cite><a href="#rfc3986">RFC 3986</a></cite>&gt;
fragment      = &lt;pct-encoded, defined in <cite><a href="#rfc3986">RFC 3986</a></cite>&gt;
utf8string    = *( unreserved / pct-encoded  ":" / "@" )  ; utf-8 character
                                                          ; encoded using <cite><a href="#">rfc3896</a></cite> rules.
</pre></div><div class="exampleInner"><a id="importeddefs" name="importeddefs" /><pre>
ALPHA         =  %x41-5A / %x61-7A   ; A-Z / a-z
DIGIT         =  %x30-39 ; 0-9
HEXDIG        =  DIGIT / "A" / "B" / "C" / "D" / "E" / "F"

unreserved    = ALPHA / DIGIT / "-" / "." / "_" / "~"
pct-encoded   = "%" HEXDIG HEXDIG
sub-delims    = "!" / "$" / "&amp;" / "'" / "(" / ")" / "*" / "+" / "," / ";" / "="
pchar         = unreserved / pct-encoded / sub-delims / ":" / "@"
fragment      = *( pchar / "/" / "?" )
</pre></div></div></div></div><div class="div1">
<h2><a id="media-fragment-processing" name="media-fragment-processing" />5 Media Fragments Processing</h2><p>This section defines the different exchange scenarios for the different situations explained in section <a href="#fragment-query"><b>3 URI fragment and URI query</b></a> over the HTT¨protocol.
    </p><div class="div2">
<h3><a id="processing-protocol-frag" name="processing-protocol-frag" />5.1 Protocol for URI fragment Resolution in HTTP</h3><p>This section defines the protocol steps in HTTP <cite><a href="#rfc2616">RFC 2616</a></cite> to resolve and deliver a media fragment specified as a URI fragment.</p><div class="div3">
<h4><a id="processing-protocol-UA-mapped" name="processing-protocol-UA-mapped" />5.1.1 UA mapped byte ranges</h4><table border="1" summary="Editorial note"><tr><td width="50%" valign="top" align="left"><b>Editorial note</b></td><td width="50%" valign="top" align="right"> </td></tr><tr><td valign="top" align="left" colspan="2">
            <p>This section is ready to implement.</p>
          </td></tr></table><p>As described in section <a href="#URIfragment-user-agent"><b>3.2 Resolving URI fragments within the user agent</b></a>, the most optimal case is a user agent that knows how to map media fragments to byte ranges. This is the case typically where a user agent has already downloaded those parts of a media resource that allow it to do or guess the mapping, e.g. headers or a resource, or an index of a resource.
          </p><p>In this case, the HTTP exchanges are exactly the same as for any other Web resource where byte ranges are requested <cite><a href="#rfc2616">RFC 2616</a></cite>.
          </p><p>How the UA retrieves the byte ranges is dependent on the media type of the media resource.
          We here show examples with only one byte range retrieval per time range, which may
          in practice turn into several such retrieval actions necessary to acquire the correct
          time range.
          </p><p>Here are the three principle cases a media fragment enabled UA and a media Server will encounter:</p><div class="div4">
<h5><a id="processing-protocol-UA-mapped-new" name="processing-protocol-UA-mapped-new" />5.1.1.1 UA requests URI fragment for the first time</h5><p>A user requests a media fragment URI:</p><ul><li><p>User → UA (1):</p><div class="exampleInner"><pre>http://www.example.com/video.ogv#t=10,20</pre></div></li></ul><p>The UA has to check if a local copy of the requested fragment is available in its buffer - not in this case. But it knows how to map the fragment to byte ranges: 19147 - 22890. So, it requests these byte ranges from the server:</p><ul><li><p>UA (1) → Proxy (2) → Origin Server (3):</p><div class="exampleInner"><pre>GET /video.ogv HTTP/1.1
Host: www.example.com
Accept: video/*
Range: bytes=19147-22890</pre></div></li></ul><p>The server extracts the bytes corresponding to the requested range and replies in a 206 HTTP response:</p><ul><li><p>Origin Server (3) → Proxy (4) → UA (5):</p><div class="exampleInner"><pre>HTTP/1.1 206 Partial Content
Accept-Ranges: bytes
Content-Length: 3743
Content-Type: video/ogg
Content-Range: bytes 19147-22880/35614993
Etag: "b7a60-21f7111-46f3219476580"

{binary data}</pre></div></li></ul><p>Assuming the UA has received the byte ranges that it requires to serve t=10,20, which may well be slightly more, it will serve the decoded content to the User from the appropriate time offset. Otherwise it may keep requesting byte ranges to retrieve the required time segments.</p><img src="MF-SD-ClientSide-5.2.1.1.png" alt="Illustration of a UA requesting a URI fragment for the first time" /></div><div class="div4">
<h5><a id="processing-protocol-UA-mapped-unchanged" name="processing-protocol-UA-mapped-unchanged" />5.1.1.2 UA requests URI fragment it already has buffered</h5><p>A user requests a media fragment URI:</p><ul><li><p>User → UA (1):</p><div class="exampleInner"><pre>http://www.example.com/video.ogv#t=10,20</pre></div></li></ul><p>The UA has to check if a local copy of the requested fragment is available in its buffer - it is in this case. But the resource could have changed on the server, so it needs to send a conditional GET. It knows the byte ranges: 19147 - 22890. So, it requests these byte ranges from the server under condition of it having changed:</p><ul><li><p>UA (1) → Proxy (2) → Origin Server (3):</p><div class="exampleInner"><pre>GET /video.ogv HTTP/1.1
Host: www.example.com
Accept: video/*
If-Modified-Since: Sat, 01 Aug 2009 09:34:22 GMT
If-None-Match: "b7a60-21f7111-46f3219476580"
Range: bytes=19147-22890</pre></div></li></ul><p>The server checks if the resource has changed by checking the date - in this case, the resource was not modified. So, the server replies with a 304 HTTP response. (Note that a If-Range header cannot be used, because if the entity has changed, the entire resource would be sent.)</p><ul><li><p>Origin Server (3) → Proxy (4) → UA (5):</p><div class="exampleInner"><pre>HTTP/1.1 304 Not Modified
Accept-Ranges: bytes
Content-Length: 3743
Content-Type: video/ogg
Content-Range: bytes 19147-22880/35614993
Etag: "b7a60-21f7111-46f3219476580"</pre></div></li></ul><p>So, the UA serves the decoded resource to the User our of its existing buffer.</p><img src="MF-SD-ClientSide-5.2.1.2.png" alt="Illustration of a UA requesting a URI fragment it already has buffered" /></div><div class="div4">
<h5><a id="processing-protocol-UA-mapped-changed" name="processing-protocol-UA-mapped-changed" />5.1.1.3 UA requests URI fragment of a changed resource</h5><p>A user requests a media fragment URI and the UA sends the exact same GET request as described in the previous subsection.</p><p>This time, the server checks if the resource has changed by checking the date and it has been modified. Since the byte mapping may not be correct any longer, the server can only tell the UA that the resource has changed and leave all further actions to the UA. So, it sends a 412 HTTP response:</p><ul><li><p>Origin Server (3) → Proxy (4) → UA (5):</p><div class="exampleInner"><pre>HTTP/1.1 412 Precondition Failed
Accept-Ranges: bytes
Content-Length: 3743
Content-Type: video/ogg
Content-Range: bytes 19147-22880/22222222
Etag: "xxxxx-yyyyyyy-zzzzzzzzzzzzz"</pre></div></li></ul><p>So, the UA can only assume the resource has changed and re-retrieve what it needs to get back to being able to retrieve fragments. For most resources this may mean retrieving the header of the file. After this it is possible again to do a byte range retrieval.</p><img src="MF-SD-ClientSide-5.2.1.3.png" alt="Illustration of a UA requesting a URI fragment it has buffered, but that changed" /></div></div><div class="div3">
<h4><a id="processing-protocol-Server-mapped" name="processing-protocol-Server-mapped" />5.1.2 Server mapped byte ranges</h4><p>As described in section <a href="#URIfragment-server"><b>3.3 Resolving URI fragments with server help</b></a>, some User Agents cannot
           undertake the fragment-to-byte mapping themselves, because the mapping is not obvious.
           This typically applies to media formats where the setup of the decoding pipeline does
           not imply knowledge of how to map fragments to byte ranges, e.g. Ogg without OggIndex.
           Thus, the User Agent would be capable of decoding a continuous resource, but would not
          know which bytes to request for a media fragment.
        </p><p>In this case, the User Agent could either guess what byte ranges it has to retrieve
          and the retrieval action would follow the previous case. Or it could hope that the server
          provides a special service, which would allow it to retrieve the byte ranges with a simple
          request of the media fragment ranges. Thus, the HTTP request of the User Agent will include
          a request for the fragment hoping that the server can do the byte range mapping and send
          back the appropriate byte ranges. This is realized by introducing new dimensions for the
          HTTP Range header, next to the byte dimension.
        </p><p>
          The specification for all new Range Request Header dimensions is given through the following
          ABNF as an extension to the HTTP Range Request Header definition (see
          http://www.w3.org/Protocols/rfc2616/rfc2616-sec14.html#sec14.35.2):
        </p><div class="exampleInner"><a id="rangerequestheaderdef" name="rangerequestheaderdef" /><pre>
Range                  = "Range" ":" ranges-specifier
ranges-specifier       = byte-ranges-specifier | fragment-specifier
;
; note that ranges-specifier is extended from <cite><a href="#rfc2616">RFC 2616</a></cite>
; to cover alternate fragment range specifiers
;
fragment-specifier     = "include-setup" | fragment-range *( "," fragment-range )
                                                           [ ";" "include-setup" ]
fragment-range         = time-ranges-specifier | track-ranges-specifier | name-ranges-specifier
;
; note that this doesn't capture the restriction to one fragment dimension occurring
; maximally once only in the fragment-specifier definition.
;
time-ranges-specifier  = npttimeoption / smptetimeoption / clocktimeoption
npttimeoption          = pfxdeftimeformat "=" npt-sec   "-" [ npt-sec ]
smptetimeoption        = pfxsmpteformat   "=" frametime "-" [ frametime ]
clocktimeoption        = pfxclockformat   "=" datetime  "-" [ datetime ]

track-ranges-specifier = trackprefix "=" trackparam *( ";" trackparam )

name-ranges-specifier  = nameprefix  "=" nameparam
</pre></div><p>This specification is meant to be analogous to the one in URIs, but it is a bit stricter.
            The time unit is not optional. For instance, it can be "npt", "smpte", "smpte-25",
            "smpte-30", "smpte-30-drop" or "clock" for temporal. Where "ntp" is used for a temporal
            range, only specification in seconds is possible. Where "clocktime" is used for a temporal
            range, only "datetime" is possible and "walltime" is fully specified in HHMMSS with
            fraction and full timezone. Indeed, all optional elements in the URI specification
            basically become required in the Range header.
          </p><p>There is an optional 'include-setup' flag on the fragment range specifier - this
            flag signals to the server whether delivery of the decoder setup information (i.e.
            typically file header information) is also required as part of the reply to this
            request. This can help avoid an extra roundtrip where a Media Fragment URI is, e.g.
            directly typed into a Web browser.
          </p><p>If there were multiple track parameters provided in the media fragment URI, they are all
            aggregated together here in a single track ranges specifier, where the track names are
            separated by semi-colon. Note that if a track name did include a semi-colon in the media
            fragment URI, it is now percent escaped.
          </p><p>Note that the specification does not foresee a Range dimension for spatial media
            fragments since they are typically resolved and interpreted by the User Agent (i.e.,
            spatial fragment extraction is not performed on server-side) for the following
            reasons:
          </p><ul><li><p>spatial media fragments are typically not expressible in terms of byte ranges.
                Spatial fragment extraction would thus require transcoding operations resulting
                in new resources rather than fragments of the original media resource. As described
                in section <a href="#fragment-query"><b>3 URI fragment and URI query</b></a>, spatial fragment extraction is in this
                case better represented by URI queries.
              </p></li><li><p>When a User Agent receives an extracted spatial media fragment, it is not trivial
                to visualize the context of this fragment (see also section
                <a href="#media-fragment-display"><b>7.1 Clients Displaying Media Fragments</b></a>).
                More specifically, spatial context requires a meaningful background, which will not
                be available at the User Agent when the spatial fragment is extracted by the
                server.
              </p></li></ul><table border="1" summary="Editorial note: Davy"><tr><td width="50%" valign="top" align="left"><b>Editorial note: Davy</b></td><td width="50%" valign="top" align="right"> </td></tr><tr><td valign="top" align="left" colspan="2">
              <p>Special attention should be paid for named fragments and more specifically when a
                named fragment represents a spatial fragment. We should clearly describe 1. what
                named fragments are and 2. how they are resolved.</p>
            </td></tr></table><p>Next to the introduction of new dimensions for the HTTP Range request header, we also
            introduce a new HTTP response header, called Content-Range-Mapping, which provides the
            mapping of the retrieved byte range to the original Range request, which was not in
            bytes. It serves two purposes:</p><ul><li><p>It Indicates the actual mapped range in terms of fragment dimensions. This is
                necessary since the server might not be able to provide a byte range mapping that
                corresponds exactly to the requested range. Therefore, the User Agent needs to be
                aware of this variance.
              </p></li><li><p>It provides context information regarding the parent resource in case the Range
                request contained a temporal dimension. More specifically, the header contains the
                start and end time of the parent resource. This way, the User Agent is able to
                understand and visualize the temporal context of the media fragment.
              </p></li></ul><p>The specification for the Content-Range-Mapping header is based on the specification
            of the Content-Range header
            (see http://www.w3.org/Protocols/rfc2616/rfc2616-sec14.html#sec14.16)
            and is shown below. Note that the Content-Range-Mapping header adds in case of the
            temporal dimension the instance start and end in terms of seconds after a slash
            "/" character in analogy to the Content-Range header. Also, we introduce an extension
            to the Accept-Ranges header
            (see http://www.w3.org/Protocols/rfc2616/rfc2616-sec14.html#sec14.5).
          </p><div class="exampleInner"><a id="contentrangemappingheaderdef" name="contentrangemappingheaderdef" /><pre>
Content-Range-Mapping      = "Content-Range-Mapping" ":" '{'
                             ( content-range-mapping-spec [ ";" def-include-setup ] ) / def-include-setup
                             '}' '=' '{'
                             byte-content-range-mapping-spec '}'

def-include-setup          = %x69.6E.63.6C.75.64.65.2D.73.65.74.75.70  ; "include-setup"

byte-range-mapping-spec    = bytes-unit SP
                             byte-range-resp-spec *( "," byte-range-resp-spec ) "/"
                             ( instance-length / "*" )

content-range-mapping-spec = time-mapping-spec | track-mapping-spec | name-mapping-spec
time-mapping-spec          = timeprefix ":" time-mapping-options
time-mapping-options       = npt-mapping-option / smpte-mapping-option / clock-mapping-option
npt-mapping-option         = deftimeformat SP npt-sec   "-" npt-sec   "/"
                                          [ npt-sec ]   "-" [ npt-sec ]
smpte-mapping-option       = smpteformat   SP frametime "-" frametime "/"
                                          [ frametime ] "-" [ frametime ]
clock-mapping-option       = clockformat   SP datetime  "-" datetime  "/"
                                          [ datetime ]  "-" [ datetime ]

track-mapping-spec         = trackprefix SP trackparam *( ";" trackparam )

name-mapping-spec          = nameprefix  SP nameparam

Accept-Ranges              = "Accept-Ranges" ":" acceptable-ranges
acceptable-ranges          = 1#range-unit *( "," 1#range-unit )| "none"
;
; note this does not represent the restriction that range-units can only appear once at most
;
range-unit                 = bytes-unit | other-range-unit
bytes-unit                 = "bytes"
other-range-unit           = token | timeprefix | trackprefix | nameprefix
</pre></div><p>Three cases can be distinguished when a User Agent needs assistance by a server to
            perform the byte range mapping. In the next subsections, we'll go through the protocol
            exchange action step by step.
          </p><div class="div4">
<h5><a id="processing-protocol-server-mapped-default" name="processing-protocol-server-mapped-default" />5.1.2.1 Server mapped byte ranges with corresponding binary data</h5><ul><li><p>User → UA (1):</p><div class="exampleInner"><pre>http://www.example.com/video.ogv#t=10,20</pre></div></li></ul><p>The UA has to check if a local copy of the requested fragment is available in its
              buffer. If it is, we revert back to the processing described in sections
              <a href="#processing-protocol-UA-mapped-unchanged"><b>5.1.1.2 UA requests URI fragment it already has buffered</b></a>
              and <a href="#processing-protocol-UA-mapped-changed"><b>5.1.1.3 UA requests URI fragment of a changed resource</b></a>, since the UA already
              knows the mapping to byte ranges. If the requested fragment is not available in
              its buffer, the UA sends an HTTP request to the server, including a Range header
              with temporal dimension. The request is shown below:
            </p><ul><li><p>UA (1) → Proxy (2) → Origin Server (3):</p><div class="exampleInner"><pre>GET /video.ogv HTTP/1.1
Host: www.example.com
Accept: video/*
Range: t:npt=10-20</pre></div></li></ul><p>If the server does not understand a Range header, it MUST ignore the header field
              that includes that range-set. This is in sync to the HTTP RFC <cite><a href="#rfc2616">RFC 2616</a></cite>.
              This means that where a server does not support media fragments, the complete resource
              will be delivered. It also means that we can combine both, byte range and fragment range
              headers in one request, since the server will only react to the Range header
              it understands.
            </p><p>Assuming the server can map the given Range to one or more byte ranges, it will
              reply with these in a 206 HTTP response. Where multiple byte ranges are required to
              satisfy the Range request, these are transmitted as a multipart message-body. The
              media type for this purpose is called "multipart/byteranges". This is in sync with
              the HTTP RFC <cite><a href="#rfc2616">RFC 2616</a></cite>.
            </p><p>Here is the reply to the example above, assuming a single byte range is sufficient:</p><ul><li><p>Origin Server (3) → Proxy (4) → UA (5):</p><div class="exampleInner"><pre>HTTP/1.1 206 Partial Content
Accept-Ranges: bytes, t, track, id
Content-Length: 3743
Content-Type: video/ogg
Content-Range: bytes 19147-22880/35614993
Content-Range-Mapping: { t:npt 9.85-21.16/0.0-653.79 } = { bytes 19147-22880/35614993 }
Etag: "b7a60-21f7111-46f3219476580"

{binary data}</pre></div></li></ul><p>Note the presence of the new reply header called Content-Range-Mapping, which provides
              the mapping of the retrieved byte range to the original Content-Range request, which
              was not in bytes. As we return both, byte and temporal ranges, the UA and any
              intermediate caching proxy is enabled to map byte positions with time offsets and fall
              back to byte range request where the fragment is re-requested. Also note that through
              the extended list in the Accept-Ranges it is possible to identify which fragment
              schemes a server supports.
            </p><img src="MF-SD-ServerSide.png" alt="Illustration of a UA requesting a URI time to byte range mapping from the server " /><p>In the case where a media fragment results in a multipart message-body, the
              Content-Range headers will be spread throughout the binary data ranges, but the
              Content-Range-Mapping of the media fragment will only be with the main header.
              Note that requesting track fragments typically result in multipart message-bodies,
              on condition that the parent resource is characterized by interleaved tracks. For
              example:</p><ul><li><p>Origin Server (3) → Proxy (4) → UA (5):</p><div class="exampleInner"><pre>HTTP/1.1 206 Partial Content
Accept-Ranges: bytes, t, track, id
Content-Length: 3743
Content-Type: video/ogg
Content-Range-Mapping: { track audio1;video1 } = { bytes 123-2589, 14560-27891,58909-81230/35614993 }
Content-type: multipart/byteranges; boundary=THIS_STRING_SEPARATES
Etag: "b7a60-21f7111-46f3219476580"

--THIS_STRING_SEPARATES
Content-type: video/ogg
Content-Range: bytes 123-2589/35614993
{binary data}
--THIS_STRING_SEPARATES
Content-type: video/ogg
Content-Range: bytes 14560-27891/35614993
{binary data}
--THIS_STRING_SEPARATES
Content-type: video/ogg
Content-Range: bytes 58909-81230/35614993
{binary data}
--THIS_STRING_SEPARATES--</pre></div></li></ul><p>Note that a caching proxy that does not understand a Range header must not cache
              "206 Partial Content" responses as per HTTP RFC <cite><a href="#rfc2616">RFC 2616</a></cite>. Thus, the
              new Range requests won't be cached by legacy Web proxies.
            </p></div><div class="div4">
<h5><a id="processing-protocol-server-mapped-setup" name="processing-protocol-server-mapped-setup" />5.1.2.2 Server mapped byte ranges with corresponding binary data and codec setup data</h5><p>When the User Agent needs help from the server to setup the initial decoding pipeline
              (i.e., the User Agent has no codec setup information at its disposal), the User Agent
              can request, next to the bytes corresponding to the requested fragment, the bytes
              necessary to setup its decoder. This is possible by adding the 'include-setup' flag to
              the Range header, as illustrated below:
            </p><ul><li><p>UA (1) → Proxy (2) → Origin Server (3):</p><div class="exampleInner"><pre>GET /video.ogv HTTP/1.1
Host: www.example.com
Accept: video/*
Range: t:npt=10-20;include-setup</pre></div></li></ul><p>Analogous to section <a href="#processing-protocol-server-mapped-default"><b>5.1.2.1 Server mapped byte ranges with corresponding binary data</b></a>, the
              server can map the given Range to one or more byte ranges, it will reply with these in
              a 206 HTTP response. Additionally, the server adds the bytes corresponding with the
              requested setup information to the response. Since this setup information usually
              appears in front of a media resource, the response typically results in a multipart
              message-body. The response is shown below:
            </p><ul><li><p>Origin Server (3) → Proxy (4) → UA (5):</p><div class="exampleInner"><pre>HTTP/1.1 206 Partial Content
Accept-Ranges: bytes, t, track, id
Content-Length: 3795
Content-Type: video/ogg
Content-Range-Mapping: { t:npt 11.85-21.16/0.0-653.79;include-setup } = { bytes  0-52,19147-22880/35614993 }
Content-type: multipart/byteranges; boundary=THIS_STRING_SEPARATES
Etag: "b7a60-21f7111-46f3219476580"

--THIS_STRING_SEPARATES
Content-type: video/ogg
Content-Range: bytes 0-52/35614993
{binary data}
--THIS_STRING_SEPARATES
Content-type: video/ogg
Content-Range: bytes 19147-22880/35614993
{binary data}
--THIS_STRING_SEPARATES--</pre></div></li></ul><p>Note that the Content-Range-Mapping header indicates that the codec setup information
              is included in the response. In this example, the response consists of two parts of byte
              ranges: the first part corresponds to the setup information, the second part corresponds
              to the requested fragment.
            </p><img src="MF-SD-ServerSideSetup.png" alt="Illustration of a UA requesting a URI time to byte range mapping from the server, including the codec setup information " /></div><div class="div4">
<h5><a id="processing-protocol-server-mapped-proxy" name="processing-protocol-server-mapped-proxy" />5.1.2.3 Proxy cacheable server mapped byte ranges</h5><p>As described in section <a href="#URIfragment-proxies"><b>3.4 Resolving URI fragments in a proxy cacheable manner</b></a>, the server mapped byte
              ranges approach can be extended to play with existing caching Web proxy infrastructure.
              This is important, since video is a huge bandwidth eater in the current Internet and
              falling back to using existing Web proxy infrastructure is important, particularly
              since progressive download and direct access mechanisms for video rely heavily on this
              functionality. Over time, the proxy infrastructure will learn how to cache media
              fragment URIs directly as described in the previous section and then will not require
              this extra effort.
            </p><p>To enable media-fragment-URI-supporting UAs to make their retrieval cacheable, we
              introduce some extra HTTP headers, which will help tell the server and the proxy what
              to do. There is an Accept-Range-Redirect request header which signals to the server
              that only a redirect to the correct byte ranges is necessary and the result should be
              delivered in the Range-Redirect header.
            </p><p>The ABNF for these additional two HTTP headers is given as follows:</p><div class="exampleInner"><a id="rangeredirectdefs" name="rangeredirectdefs" /><pre>
Accept-Range-Redirect      = "Accept-Range-Redirect" ":" bytes-unit

Range-Redirect             = "Range-Redirect" ":" byte-range-resp-spec *( "," byte-range-resp-spec )
</pre></div><p>Let's play it through on an example. A user requests a media fragment URI:
            </p><ul><li><p>User → UA (1):</p><div class="exampleInner"><pre>http://www.example.com/video.ogv#t=10,20</pre></div></li></ul><p>The UA has to check if a local copy of the requested fragment is available in its
              buffer. In our case here, it is not. If it was, we would revert back to the processing
              described in sections <a href="#processing-protocol-UA-mapped-unchanged"><b>5.1.1.2 UA requests URI fragment it already has buffered</b></a> and
              <a href="#processing-protocol-UA-mapped-changed"><b>5.1.1.3 UA requests URI fragment of a changed resource</b></a>, since the UA already knows the
              mapping to byte ranges. The UA issues a HTTP GET request with the fragment and
              requesting to retrieve just the mapping to byte ranges:
            </p><ul><li><p>UA (1) → Proxy (2) → Origin Server (3):</p><div class="exampleInner"><pre>GET /video.ogv HTTP/1.1
Host: www.example.com
Accept: video/*
Range: t:npt=10-20
Accept-Range-Redirect: bytes</pre></div></li></ul><p>The server converts the given time range to a byte range and sends an empty reply
              that refers the UA to the right byte range for the correct time range.
            </p><ul><li><p>Origin Server (3) → Proxy (4) → UA (5):</p><div class="exampleInner"><pre>HTTP/1.1 307 Temporary Redirect
Location: http://www.example.com/video.ogv
Accept-Ranges: bytes, t, track, id
Content-Length: 0
Content-Type: video/ogg
Content-Range-Mapping: { t:npt 11.85-21.16/0.0-653.79 } = { bytes 19147-22880/* }
Range-Redirect: 19147-22880
Vary: Accept-Range-Redirect</pre></div></li></ul><p>Note that codec setup information can also be requested in combination with the
              Accept-Range-Redirect header, which can be realized by adding the 'include-setup'
              flag to the Range request header.
            </p><p>The UA proceeds to put the actual fragment request through as a normal byte range
              request as in section <a href="#processing-protocol-UA-mapped-new"><b>5.1.1.1 UA requests URI fragment for the first time</b></a>:
            </p><ul><li><p>UA (5) → Proxy (6) → Origin Server (7):</p><div class="exampleInner"><pre>GET /video.ogv HTTP/1.1
Host: www.example.com
Accept: video/*
Range: 19147-22880</pre></div></li></ul><p>The Origin Server puts the data together and sends it to the UA:</p><ul><li><p>Origin Server (7) → Proxy (8) → UA (9):</p><div class="exampleInner"><pre>HTTP/1.1 206 Partial Content
Accept-Ranges: bytes, t, track, id
Content-Length: 3743
Content-Type: video/ogg
Content-Range: bytes 19147-22880/35614993
Etag: "b7a60-21f7111-46f3219476580"

{binary data}</pre></div></li></ul><p>The UA decodes the data and displays it from the requested offset. The caching Web
              proxy in the middle has now cached the byte range, since it adhered to the normal byte
              range request protocol. All existing caching proxies will work with this. New caching
              Web proxies may learn to interpret media fragments natively, so won't require the extra
              packet exchange described in this section.</p><img src="MF-SD-ProxyCacheable.png" alt="Illustration of a UA requesting a URI time to byte range mapping from the server with proxy capability of byte ranges" /></div></div><div class="div3">
<h4><a id="server-triggered-redirect" name="server-triggered-redirect" />5.1.3 Server triggered redirect</h4><p>
            When a server decides not to serve the requested media fragment in terms of byte ranges (i.e., serving the requested media fragment as specified in section <a href="#processing-protocol-Server-mapped"><b>5.1.2 Server mapped byte ranges</b></a>), it can redirect the UA to a representation of this fragment (for instance by transforming the media fragment URI into a media fragment query, as specified in section <a href="#processing-protocol-query"><b>5.2 Protocol for URI query Resolution in HTTP</b></a>). This is particularly useful in cases where too many byte ranges would need to e extracted to satisfy the range request.
          </p><p>A user requests a media fragment URI using a URI fragment:</p><ul><li><p>User → UA (1):</p><div class="exampleInner"><pre>http://www.example.com/video.ogv#track=video1</pre></div></li></ul><p>Subsequently, the UA requests the media fragment from the server using the Range header:</p><ul><li><p>UA (1) → Proxy (2) → Origin Server (3):</p><div class="exampleInner"><pre>GET /video.ogv HTTP/1.1
Host: www.example.com
Accept: video/*
Range: track=video1</pre></div></li></ul><p>
            The server decides not to serve the requested media fragment in terms of byte ranges (for instance, because the track media fragment results in too many byte ranges). The server redirects the UA to an alternate representation. For example, the URI fragment can be transformed into a URI query. Further, a Link header is added stating that the redirected location is a fragment of the originally requested resource.
          </p><ul><li><p>Origin Server (3) → Proxy (4) → UA (5):</p><div class="exampleInner"><pre>HTTP/1.1 307 Temporary Redirect
Location: http://www.example.com/video.ogv?track=video1
Accept-Ranges: bytes, t, track, id
Content-Length: 0
Content-Type: video/ogg
Link: &lt;http://www.example.com/video.ogv#track=video1&gt;; rel="fragment"
Vary: *</pre></div></li></ul><table border="1" summary="Editorial note: Davy"><tr><td width="50%" valign="top" align="left"><b>Editorial note: Davy</b></td><td width="50%" valign="top" align="right"> </td></tr><tr><td valign="top" align="left" colspan="2">
            <p>We need to register the 'fragment' Link Relation <cite><a href="#webLinking">Web Linking</a></cite>.
            </p>
          </td></tr></table><p>Finally, the UA follows the redirect, which in this case corresponds to the process specified in section <a href="#processing-protocol-query"><b>5.2 Protocol for URI query Resolution in HTTP</b></a>.</p><p>The server can also decide to combine a redirect and a media fragment URI:</p><ul><li><p>User → UA (1):</p><div class="exampleInner"><pre>http://www.example.com/video.ogv#track=video1&amp;t=10,20</pre></div></li></ul><p>The UA requests the media fragment to the server using the Range header:</p><ul><li><p>UA (1) → Proxy (2) → Origin Server (3):</p><div class="exampleInner"><pre>GET /video.ogv HTTP/1.1
Host: www.example.com
Accept: video/*
Range: track=video1,t:npt=10-20</pre></div></li></ul><p>
          The server decides not to serve the requested media fragment in terms of byte ranges and redirects the UA to an alternate representation. However, in this case, the server decides to handle the track fragment through a URI query and the temporal fragment through a URI fragment:
        </p><ul><li><p>Origin Server (3) → Proxy (4) → UA (5):</p><div class="exampleInner"><pre>HTTP/1.1 307 Temporary Redirect
Location: http://www.example.com/video.ogv?track=video1#t=10,20
Accept-Ranges: bytes, t, track, id
Content-Length: 0
Content-Type: video/ogg
Link: &lt;http://www.example.com/video.ogv#track=video1&amp;t=10,20&gt;; rel="fragment"
Vary: *</pre></div></li></ul><p>
          Finally, the UA follows the redirect, which in this case corresponds to the process specified in section <a href="#processing-protocol-query"><b>5.2 Protocol for URI query Resolution in HTTP</b></a> for the track fragment, combined with the process specified in section <a href="#processing-protocol-Server-mapped"><b>5.1.2 Server mapped byte ranges</b></a> for the temporal fragment.
        </p></div></div><div class="div2">
<h3><a id="processing-protocol-query" name="processing-protocol-query" />5.2 Protocol for URI query Resolution in HTTP</h3><p>This section describes the protocol steps used in HTTP <cite><a href="#rfc2616">RFC 2616</a></cite> to resolve and deliver a media fragment specified as a URI query.</p><p>A user requests a media fragment URI using a URI query:</p><ul><li><p>User → UA (1):</p><div class="exampleInner"><pre>http://www.example.com/video.ogv?t=10,20</pre></div></li></ul><p>This is a full resource, so it is a simple HTTP retrieval process. The UA has to check if a local copy of the requested resource is available in its buffer. If yes, it does a conditional GET with e.g. an If-Modified-Since and If-None-Match HTTP header.</p><p>Assuming the resource has not been retrieved before, the following is sent to the server:</p><ul><li><p>UA (1) → Proxy (2) → Origin Server (3):</p><div class="exampleInner"><pre>GET /video.ogv?t=10,20 HTTP/1.1
Host: www.example.com
Accept: video/*</pre></div></li></ul><p>If the server doesn't understand these query parameters, it typically ignores them and returns the complete resource. This is not a requirement by the URI or the HTTP standard, but the way it is typically implemented in Web browsers.</p><p>
        A media fragment supporting server has to create a complete media resource for the URI query, which in the case of Ogg requires creation of a new resource by adapting the existing Ogg file headers and combining them with the extracted byte range that relates to the given fragment. Some of the codec data may also need to be re-encoded since, e.g. t=10 does not fall clearly on a decoding boundary, but the retrieved resource must match as closely as possible the URI query. This new resource is sent back as a reply:
      </p><ul><li><p>Origin Server (3) → Proxy (4) → UA (5):</p><div class="exampleInner"><pre>HTTP/1.1 200 OK
Content-Length: 3782
Content-Type: video/ogg
Etag: "b7a60-21f7111-46f3219476580"
Link: &lt;http://www.example.com/video.ogv#t=10,20&gt;; rel="alternate"

{binary data}</pre></div></li></ul><p>Note that a Link header MAY be provided indicating the relationship between the requested URI query and the original media fragment URI. This enables the UA to retrieve further information about the original resource, such as its full length. In this case, the user agent is also enable to choose to display the dimensions of the primary resource or the ones created by the query.</p><p>The UA serves the decoded resource to the user. Caching in Web proxies works as it has always worked - most modern Web servers and UAs implement a caching strategy for URIs that contain a query using one of the three methods for marking freshness: heuristic freshness analysis, the Cache-Control header, or the Expires header. In this case, many copies of different segments of the original resource video.ogv may end up in proxy caches. An intelligent media proxy in future may devise a strategy to buffer such resources in a more efficient manner, where headers and byte ranges are stored differently.</p><p>Further, media fragment URI queries can be extended to enable UAs to use the Range-Redirect HTTP header to also revert back to a byte range request. This is analogous to section <a href="#processing-protocol-server-mapped-proxy"><b>5.1.2.3 Proxy cacheable server mapped byte ranges</b></a>.</p><p>Note that a server that does not support media fragments through either URI fragment or query addressing will return the full resource in either case. It is therefore not possible to first try URI fragment addressing and when that fails to try URI query addressing.</p></div></div><div class="div1">
<h2><a id="media-fragment-semantics" name="media-fragment-semantics" />6 Media Fragments Semantics</h2><p>Errors can occur at several levels:</p><ul><li>the complete URI (e.g. spatial dimension on a audio resource)</li><li>the combination of name-value strings (e.g. name and time dimension combined)</li><li>the name-value string (e.g. invalid percent-encoding)</li><li>the name dimension (e.g. undefined dimension)</li><li>the value (e.g. invalid NPT syntax)</li></ul><p>We will look at errors in the different dimensions and their values in the subsequent sub-sections and we will start with errors on the more general levels.
    </p><div class="div2">
<h3><a id="error-general" name="error-general" />6.1 Errors on the General URI level</h3><div class="div3">
<h4><a id="error-general-non-existent" name="error-general-non-existent" />6.1.1 Non-existent dimension:</h4><p>Attempting to do fragment selection on a dimension that does not exist in the source media, such as temporal clipping on a still image or spatial clipping on an audio file, should be considered a no-op and not throw an error.
        </p></div><div class="div3">
<h4><a id="error-general-underspec" name="error-general-underspec" />6.1.2 Under-specified Dimension</h4><p>The result of doing spatial clipping on a media resource that has multiple video tracks is that the spatial clipping is applied to all tracks.
        </p></div></div><div class="div2">
<h3><a id="error-temporal" name="error-temporal" />6.2 Errors on the temporal dimensions</h3><p>Assuming a single temporal dimension is present, we now analyse what fragment values may be specified here and how they should be handled.
      </p><p>For this, we make the following definitions:</p>
        <ul><li>s: the start point of the media and s &gt;= 0</li><li>e: the end point of the media (i.e. duration = e - s ) and s &lt; e</li><li>a: a positive integer, a &gt;= 0</li><li>b: a positive integer, b &gt;= 0</li></ul>
      <div class="div3">
<h4><a id="error-temporal-valid" name="error-temporal-valid" />6.2.1 Valid requests</h4><p>For t=a,b with a &lt;= b</p><ul><li>t=,                     results in 206: deliver from s to e</li><li>t=a, with s &lt;= a, a &lt; e results in 206: deliver from a to e</li><li>t=a, or t=a with a &lt; s         results in 206: deliver from s to e</li><li>t=,b with s &lt; b, b &lt;= e results in 206: deliver from s to b</li><li>t=,b with e &lt; b         results in 206: deliver from s to e</li><li>t=a,b with s = a, b = e restulst in 206: deliver from s to e</li><li>t=a,b with s &lt;= a, a &lt; b, a &lt; e and b &lt;= e results in 206: deliver from a to b (the normal case)</li><li>t=a,b with s &lt;= a, a &lt; b, a &lt; e and e &lt; b results in 206: deliver from a to e</li><li>t=a,b with a &lt; s, a &lt; b, s &lt; b and b &lt;= e results in 206: deliver from s to b</li><li>t=a,b with a &lt; s, a &lt; b and e &lt; b         results in 206: deliver from s to e</li><li>%74=10,20  resolve percent encoding to t=10,20</li><li>t=%31%30   resolve percent encoding to t=10</li><li>t=10%2C20  resolve percent encoding to t=10,20</li><li>t=%6ept:10 resolve percent encoding to t=npt:10</li><li>t=npt%3a10 resolve percent encoding to t=npt:10</li></ul><table border="1" summary="Editorial note: Raphael"><tr><td width="50%" valign="top" align="left"><b>Editorial note: Raphael</b></td><td width="50%" valign="top" align="right"> </td></tr><tr><td valign="top" align="left" colspan="2">
            <p>The following paragraph is controversial since it could lead to non-interoperable implementations.</p>
          </td></tr></table><table border="1" summary="Editorial note: Silvia"><tr><td width="50%" valign="top" align="left"><b>Editorial note: Silvia</b></td><td width="50%" valign="top" align="right"> </td></tr><tr><td valign="top" align="left" colspan="2">
            <p>
              If the UA needs to retrieve a large part of the resource or even the full resource, it will probably decide to make
              multiple range requests rather than a single one. If the resource is, however, small, it may decide to just retrieve the full
              resource without a range request. The UA should make this choice given context information, e.g. if it knows that it will be a
              lot of data, it will retrieve it in smaller chunks. If it chooses to request the full resource in one go and not make use of a
              Range request, the result will be a 200 rather than a 206.
            </p>
          </td></tr></table></div><div class="div3">
<h4><a id="error-temporal-empty" name="error-temporal-empty" />6.2.2 Empty</h4><p>The resolved time segment is empty.</p><ul><li>t=a,a results in a HTTP Range request with 'include-setup' as specified in section <a href="#processing-protocol-server-mapped-setup"><b>5.1.2.2 Server mapped byte ranges with corresponding binary data and codec setup data</b></a> unless the UA is already set up for this resource in which case it will not undertake an unnecessary retrieval request. Results in 206 for the setup data.</li></ul><p>Effect: retrieve whatever the browser needs to set up playback, but otherwise nothing</p></div><div class="div3">
<h4><a id="error-temporal-non-existent" name="error-temporal-non-existent" />6.2.3 Non-existent</h4><p>The value resolves to a non-existent fragment.
        </p><p> If the UA is already set up for decoding the resource and it can identify that the fragment is non-existent (i.e. knows about start and end times), it will avoid undertaking an unnecessary retrieval action. Otherwise it will undertake the RANGE retrieval request with the 'include-setup' as specified in section <a href="#processing-protocol-server-mapped-setup"><b>5.1.2.2 Server mapped byte ranges with corresponding binary data and codec setup data</b></a> and will receive a 206 with just the setup data. If the UA is set up for decoding, but cannot identify that the fragment is non-existent and does the retrieval action without the 'include-setup', it will result in a 416.
        </p><ul><li>t=a,b with s &lt; a, a &lt; b, a &gt;= e and b &gt; e retrieves nothing since interval beyond end of the resource</li><li>t=a,b with a &lt; s, a &lt; b, b &lt;= s and b &lt; e  retrieves nothing since interval ahead of start of the resource</li><li>t=a, with a &gt;= e  retrieves nothing since interval beyond end of the resource</li><li>t=,a with a &lt;= s  retrieves nothing since interval ahead of start of the resource</li></ul><p>Effect: retrieve whatever the browser needs to set up playback, but otherwise nothing</p></div><div class="div3">
<h4><a id="error-temporal-validity" name="error-temporal-validity" />6.2.4 Validity error</h4><p>The value cannot be parsed for the dimension.</p><p>If the UA is already set up for decoding the resource, it will identify that the fragment is invalid and avoid undertaking an unnecessary retrieval action. Otherwise it will undertake the RANGE retrieval request with the 'include-setup' as specified in section <a href="#processing-protocol-server-mapped-setup"><b>5.1.2.2 Server mapped byte ranges with corresponding binary data and codec setup data</b></a> and will receive a 206 with just the setup data.
        </p><p>Examples:</p><ul><li>t=a,b with a &gt; b  retrieves nothing since inverted interval</li><li>t=asdf</li><li>t=5,ekj</li><li>t=agk,9</li><li>t='0'</li><li>t=10-20</li><li>t=10:20</li><li>t=10,20,40</li><li>t%3D10 where %3D is equivalent to =; percent encoding does not resolve</li></ul><p>Effect: retrieve whatever the browser needs to set up playback, but otherwise nothing</p></div></div><div class="div2">
<h3><a id="error-spatial" name="error-spatial" />6.3 Errors on the spatial dimensions</h3><p>Assuming a single spatial dimension is present, we now analyse what content can appear here and how it should be handled.
      </p><table border="1" summary="Editorial note: Silvia"><tr><td width="50%" valign="top" align="left"><b>Editorial note: Silvia</b></td><td width="50%" valign="top" align="right"> </td></tr><tr><td valign="top" align="left" colspan="2">
        <p>This list still has to be provided.</p>
        </td></tr></table></div><div class="div2">
<h3><a id="error-track" name="error-track" />6.4 Errors on the track dimensions</h3><p>Assuming a single track dimension is present, we now analyse what content can appear here and how it should be handled.
      </p><table border="1" summary="Editorial note: Silvia"><tr><td width="50%" valign="top" align="left"><b>Editorial note: Silvia</b></td><td width="50%" valign="top" align="right"> </td></tr><tr><td valign="top" align="left" colspan="2">
        <p>This list still has to be provided.</p>
        </td></tr></table></div><div class="div2">
<h3><a id="error-named" name="error-named" />6.5 Errors on the named dimensions</h3><p>Assuming a single named dimension is present, we now analyse what content can appear here and how it should be handled.
      </p><table border="1" summary="Editorial note: Silvia"><tr><td width="50%" valign="top" align="left"><b>Editorial note: Silvia</b></td><td width="50%" valign="top" align="right"> </td></tr><tr><td valign="top" align="left" colspan="2">
        <p>This list still has to be provided.</p>
        </td></tr></table></div></div><div class="div1">
<h2><a id="implementor-notes" name="implementor-notes" />7 Notes to Implementors</h2><p>
      This section contains notes to implementors. Some of the information here is already
      stated formally elsewhere in the document, and the reference here is mainly a heads-up.
      Other items are really outside the scope of this specification, but the notes here
      reflect what the authors think would be good practice.
    </p><div class="div2">
<h3><a id="media-fragment-display" name="media-fragment-display" />7.1 Clients Displaying Media Fragments</h3><p>
        When dealing with media fragments, there is a question whether to display the media fragment in
        context or without context. In general, it is recommended to display a URI fragment in context since
        it is part of a larger resource. On the other hand, a URI query results in a new resource, so it is
        recommended to display it as a complete resource without context. The next paragraphs discuss for each
        axis the context of a media fragment and provides suggestions regarding the visualization of the
        URI fragment within its context.
      </p><p>
        For a temporal URI fragment, it is recommended to start playback at a time offset that equals
        to the start of the fragment and pause at the end of the fragment. When the "play" button is
        hit again, the resource will continue loading and play back beyond the end of the fragment.
        When seeking to specific offsets, the resource will load and play back from those seek points.
        It is also recommended to introduce a "reload" button to replay just the URI fragment. In this
        way, a URI fragment basically stands for "focusing attention". Additionally, temporal URI
        fragments could be highlighted on the transport bar.
      </p><p>
        For a spatial URI fragment, we foresee two distinct use cases: highlighting the spatial
        region in-context and cropping to the region. In the first case, the spatial region could
        be indicated by means of a bounding box or the background (i.e., all the pixels that are
        not contained within the region) could be blurred or darkened. In the second case, the region
        alone would be presented as a cropped area. How a document author specifies which use case
        is intended is outside the scope of this specification, we suggest implementors of the
        specification provide a means for this, for example through attributes or stylesheet elements.
      </p><p>
        Finally, for track URI fragments, it is recommended to play only the tracks identified by the
        track URI fragment. If no tracks are specified, the default tracks should be played.
        Different tracks could be selected using drop-down boxes or buttons; the selected tracks are
        then highlighted during playback. The way the UA retrieves information regarding the available
        tracks of a particular resource is out of scope for this specification.
      </p></div><div class="div2">
<h3><a id="media-fragment-clients" name="media-fragment-clients" />7.2 All Media Fragment Clients</h3><p>
        <em>Resolution Order:</em> Where multiple dimensions are combined in one URI fragment request, implementations are expected to first do track and temporal selection on the container level, and then do spatial clipping on the codec level. Named selection is done for whatever the name stands for: a track, a temporal section, or a spatial region.
      </p><p>
        <em>Media Fragment Grammar:</em> Note that the grammar for Media Fragment URI only specifies
        the grammar for features standardised by this specification. If a string does not parse
        correctly it does not necessarily mean the URI is wrong, it only means it is not a Media
        Fragment according to this specification. It may be correct for some extended form,
        or for a completely different fragment specification method. For this reason, error
        recovery on syntax errors in media fragment specifiers is unwise.
      </p><p>
        <em>External Clipping:</em> There is no obligatory resolution method for a situation where a media fragment URI is being used in the context of another clipping method. Formally, it is up to the context embedding the media fragment URI to decide whether the outside clipping method overrides the media fragment URI or cascades, i.e. is defined on the resulting resource. In the absence of strong reasons to do otherwise we suggest cascading. An example is a SMIL element as follows: <code>&lt;smil:video clipBegin="5" clipEnd="15" src="http://www.example.com/example.mp4#t=100,200"/&gt;</code>. This should start playback of the original media resource at second 105, and stop at 115.
      </p><p>
        <em>Content-Range-Mapping:</em> The Content-Range-Mapping header returned sometimes
        refers to a completely different range than the one that was specified as the
        <em>Range:</em> in the request. This can happen if a byte-based range is requested from
        a cache server that is not Media Fragment aware, and that server had previously cached
        the data as a result of a time range request. Technically, the information in the
        Content-Range-Mapping header is still correct, but it is completely unrelated to the
        request issued.
      </p></div><div class="div2">
<h3><a id="media-fragment-servers" name="media-fragment-servers" />7.3 Media Fragment Servers</h3><p>
        <em>Media type:</em> The media type of a resource retrieved through a URI fragment request is the same as that of the primary resource. Thus, retrieval of e.g. a single frame from a video will result in a one-frame-long video. Or, retrieval of all the audio tracks from a video resource will result in a video and not a audio resource. When using a URI query approach, media type changes are possible. E.g. a spatial fragment from a video at a certain time offset could be retrieved as a jpeg using a specific HTTP "Accept" header in the request.
      </p><p>
        <em>Synchronisation:</em> Synchronisation between different tracks of a media resource needs to be maintained when retrieving media fragments of that resource. This is true for both, URI fragment and URI query retrieval. With URI queries, when transcoding is required, a non-perceivable change in the synchronisation is acceptable.
      </p><p>
        <em>Embedded Timecodes:</em> When a media resource contains embedded time codes, these need
        to be maintained for media fragment retrieval, in particular when the URI fragment method
        is used. When URI queries are used and transcoding takes place, the embedded time codes
        should remain when they are useful and required.
      </p><p>
        <em>SMPTE Timecodes:</em> Standardisation of SMPTE timecodes in this document is
        primarily intended to allow frame-accurate references to sections of video files, they
        can be seen as a form of content-based addressing.
      </p><p>
        <em>Reasonable Clipping:</em> Temporal clipping
        needs to be as close as reasonably possible to what the media fragment
        specified, and not omit any requested data.
        "Reasonably close" means the nearest compression entity to
        the requested fragment that completely contains the requested fragment.
        This means, e.g. for temporal fragments if a request is made for
        <code>http://www.example.org/video.ogv#t=60,100</code>, but the closest
        decodable range is <code>t=58,102</code> because this is where a packet
        boundary lies for audio and video, then it will be this range that is
        returned. The UA is then capable of
        displaying only the requested subpart, and should also just do that.
        For some container formats this is a non-issue, because the container
        format allows specification of logical begin and end.
      </p><p>
        <em>Reasonable byte ranges:</em> If a single temporal range request would result
        in a disproportionally large number of byte ranges it may be better if the server
        returns a redirect to the query form of the media fragment. This situation could happen,
        for example, if the underlying media file is organized in a strange way.
      </p></div></div><div class="div1">
<h2><a id="conclusions" name="conclusions" />8 Conclusions</h2><div class="div2">
<h3><a id="qualification-resources" name="qualification-resources" />8.1 Qualification of Media Resources</h3><p>
        HTTP byte ranges can only be used to request media fragments if these media fragments can be expressed in terms of byte ranges. This restriction implies that media resources should fulfil the following conditions:
      </p><ul><li><p>The media fragments can be extracted in the compressed domain;</p></li><li><p>No syntax element modifications in the bitstream are needed to perform the extraction.</p></li></ul><p>
        Not all media formats will be compliant with these two conditions. Hence, we distinguish the following categories:
      </p><ol class="enumar"><li><p>The media resource meets the two conditions (i.e., fragments can be extracted in the compressed domain and no syntax element modifications are necessary). In this case, caching media fragments of such media resources is possible using HTTP byte ranges, because their media fragments are addressable in terms of byte ranges.
          </p></li><li><p>
            Media fragments can be extracted in the compressed domain, but syntax element modifications are required. These media fragments are cacheable using HTTP byte ranges on condition that the syntax element modifications are needed in media-headers applying to the whole media resource/fragment. In this case, those media-headers could be sent to the client in the first response of the server, which is a response to a request on a specific resource different from the byte-range content.
          </p></li><li><p>
            Media fragments cannot be extracted in the compressed domain. In this case, transcoding operations are necessary to extract media fragments. Since these media fragments are not expressible in terms of byte ranges, it is not possible to cache these media fragments using HTTP byte ranges. Note that media formats which enable extracting fragments in the compressed domain, but are not compliant with category 2 (i.e., syntax element modifications are not only applicable to the whole media resource), also belong to this category.
          </p></li></ol></div></div></div><div class="back"><div class="div1">
<h2><a id="references-normative" name="references-normative" />A References</h2><dl><dt class="label"><a name="rfc2119" />[RFC 2119] </dt><dd>S. Bradner. <a href="http://www.ietf.org/rfc/rfc2119.txt"><cite>Key Words for use in RFCs to Indicate Requirement Levels</cite></a>. IETF RFC 2119, March 1997. Available at  <a href="http://www.ietf.org/rfc/rfc2119.txt">http://www.ietf.org/rfc/rfc2119.txt</a>.
        </dd><dt class="label"><a name="rtsp" />[RFC 2326] </dt><dd>
          <cite>Real Time Streaming Protocol (RTSP)</cite>. IETF RFC 2326, April 1998. Available at <a href="http://www.ietf.org/rfc/rfc2326.txt">http://www.ietf.org/rfc/rfc2326.txt</a>.
        </dd><dt class="label"><a name="rfc2616" />[RFC 2616] </dt><dd>
          <cite>Hypertext Transfer Protocol -- HTTP/1.1</cite>. IETF RFC 2616, June 1999. Available at <a href="http://www.ietf.org/rfc/rfc2616.txt">http://www.ietf.org/rfc/rfc2616.txt</a>.
        </dd><dt class="label"><a name="rfc3339" />[RFC 3339] </dt><dd>G. Klyne and C. Newman.
          <cite>Date and Time on the Internet: Timestamps</cite>. IETF RFC 3339, July 2002. Available at <a href="http://www.ietf.org/rfc/rfc3339.txt">http://www.ietf.org/rfc/rfc3339.txt</a>.
        </dd><dt class="label"><a name="ogg" />[RFC 3533] </dt><dd>
          <cite>The Ogg Encapsulation Format Version 0</cite>. IETF RFC 3533, May 2003. Available at <a href="http://www.ietf.org/rfc/rfc3533.txt">http://www.ietf.org/rfc/rfc3533.txt</a>.
        </dd><dt class="label"><a name="rfc3986" />[RFC 3986] </dt><dd>T. Berners-Lee and R. Fielding and L. Masinter. <a href="http://www.ietf.org/rfc/rfc3986.txt"><cite> Uniform Resource Identifier (URI): Generic Syntax</cite></a>. IETF RFC 3986, January 2005. Available at  <a href="http://www.ietf.org/rfc/rfc3986.txt">http://www.ietf.org/rfc/rfc3986.txt</a>.
        </dd><dt class="label"><a name="rfc4234" />[RFC 4234] </dt><dd>D. Crocker, Ed. <a href="http://www.ietf.org/rfc/rfc4234.txt"><cite>Augmented BNF for Syntax Specifications: ABNF</cite></a>. IETF RFC 4234, October 2005. Available at <a href="http://www.ietf.org/rfc/rfc4234.txt">http://www.ietf.org/rfc/rfc4234.txt</a>.
        </dd><dt class="label"><a name="rfc4288" />[RFC 4288] </dt><dd>N. Freed and J. Klensin <a href="http://www.ietf.org/rfc/rfc4288.txt"><cite>Media Type Specifications and Registration Procedures</cite></a>. IETF RFC 4288, December 2005. Available at <a href="http://www.ietf.org/rfc/rfc4288.txt">http://www.ietf.org/rfc/rfc4288.txt</a>.
        </dd><dt class="label"><a name="rfc5147" />[RFC 5147] </dt><dd>
          E. Wilde and M. Duerst.<a href="http://tools.ietf.org/html/rfc5147"><cite>URI Fragment Identifiers for the text/plain Media Type</cite></a>. IETF RFC 5147, April 2008. Available at  <a href="http://tools.ietf.org/html/rfc5147">http://tools.ietf.org/html/rfc5147</a>.
        </dd><dt class="label"><a name="html40" />[HTML 4.0] </dt><dd>D. Ragett and A. Le Hors and I. Jacobs.<a href="http://www.w3.org/TR/REC-html40/intro/intro.html#fragment-uri"><cite>HTML Fragment identifiers</cite></a>. W3C Rec, December 1999. Available at  <a href="http://www.w3.org/TR/REC-html40/intro/intro.html#fragment-uri">http://www.w3.org/TR/REC-html40/intro/intro.html#fragment-uri</a>.
        </dd><dt class="label"><a name="html5" />[HTML 5] </dt><dd>Ian Hickson, Google (ed).<a href="http://www.w3.org/TR/2009/WD-html5-20090825/"><cite>HTML5</cite></a>. W3C Working Draft, 25th August 2009. Available at <a href="http://www.w3.org/TR/2009/WD-html5-20090825/">http://www.w3.org/TR/2009/WD-html5-20090825/</a>.
        </dd><dt class="label"><a name="svg" />[SVG] </dt><dd>J. Ferraiolo.<a href="http://www.w3.org/TR/2001/REC-SVG-20010904/linking#FragmentIdentifiersSVG"><cite>SVG Fragment identifiers</cite></a>. W3C Rec, September 2001. Available at  <a href="http://www.w3.org/TR/2001/REC-SVG-20010904/linking#FragmentIdentifiersSVG">http://www.w3.org/TR/2001/REC-SVG-20010904/linking#FragmentIdentifiersSVG</a>.
        </dd><dt class="label"><a name="smil30" />[SMIL] </dt><dd>Sjoerd Mullender, CWI (ed).<a href="http://www.w3.org/TR/2008/REC-SMIL3-20081201/"><cite>Synchronized Multimedia Integration Language (SMIL 3.0)</cite></a>. W3C Recommendation 01 December 2008. Available at  <a href="http://www.w3.org/TR/2008/REC-SMIL3-20081201/">http://www.w3.org/TR/2008/REC-SMIL3-20081201/</a>.
        </dd><dt class="label"><a name="xpointer" />[xpointer] </dt><dd>P. Grosso and E. Maler and J. Marsh and N. Walsh.<a href="http://www.w3.org/TR/xptr-framework/"><cite>XPointer Framework</cite></a>. W3C Rec, March 2003. Available at  <a href="http://www.w3.org/TR/xptr-framework/">http://www.w3.org/TR/xptr-framework/</a>.
        </dd><dt class="label"><a name="mpeg-7" />[MPEG-7] </dt><dd>
          <cite>Information Technology - Multimedia Content Description Interface (MPEG-7)</cite>. Standard No. ISO/IEC 15938:2001, International Organization for Standardization(ISO), 2001.
        </dd><dt class="label"><a name="temporalURI" />[temporal URI] </dt><dd>
          S. Pfeiffer and C. Parker and A. Pang.<a href="http://annodex.net/TR/draft-pfeiffer-temporal-fragments-03.html"><cite>Specifying time intervals in URI queries and fragments of time-based Web resources</cite></a>. Internet Draft, March 2005. Available at  <a href="http://annodex.net/TR/draft-pfeiffer-temporal-fragments-03.html">http://annodex.net/TR/draft-pfeiffer-temporal-fragments-03.html</a>.
        </dd><dt class="label"><a name="cmml" />[CMML] </dt><dd>
          <cite>Continuous Media Markup Language (CMML), Version 2.1</cite>. IETF Internet-Draft 4th March 2006 <a href="http://www.annodex.net/TR/draft-pfeiffer-cmml-03.txt">http://www.annodex.net/TR/draft-pfeiffer-cmml-03.txt</a>.
        </dd><dt class="label"><a name="roe" />[ROE] </dt><dd>
          <cite>Rich Open multitrack media Exposition (ROE)</cite>. Xiph Wiki. Retrieved 13 April 2009 at <a href="http://wiki.xiph.org/index.php/ROE">http://wiki.xiph.org/index.php/ROE</a>.
        </dd><dt class="label"><a name="skeleton" />[Skeleton] </dt><dd>
          <cite>Ogg Skeleton</cite>. Xiph Wiki. Retrieved 13 April 2009 at <a href="http://wiki.xiph.org/OggSkeleton">http://wiki.xiph.org/OggSkeleton</a>.
        </dd><dt class="label"><a name="mpeg21" />[MPEG-21] </dt><dd>
          <cite>Information Technology - Multimedia Framework (MPEG-21)</cite>. Standard No. ISO/IEC 21000:2002, International Organization for Standardization(ISO), 2002. Available at  <a href="http://www.chiariglione.org/mpeg/working_documents/mpeg-21/fid/fid-is.zip">http://www.chiariglione.org/mpeg/working_documents/mpeg-21/fid/fid-is.zip</a>.
        </dd><dt class="label"><a name="smpte" />[SMPTE] </dt><dd>
          <cite>SMPTE RP 136 Time and Control Codes for 24, 25 or 30 Frame-Per-Second Motion-Picture Systems</cite>
        </dd><dt class="label"><a name="abnf" />[ABNF] </dt><dd>
          <cite>Augmented BNF for Syntax Specifications: ABNF</cite>, Internet STD 68 (as of April 2009: <a href="http://tools.ietf.org/html/rfc5234">RFC 5234</a>).
        </dd><dt class="label"><a name="isoBaseMediaFF" />[ISO Base Media File Format] </dt><dd>
          <cite>Information technology - Coding of audio-visual objects - Part 12: ISO base media file format</cite>. Retrieved 13 April 2009 at <a href="http://standards.iso.org/ittf/PubliclyAvailableStandards/c051533_ISO_IEC_14496-12_2008.zip">http://standards.iso.org/ittf/PubliclyAvailableStandards/c051533_ISO_IEC_14496-12_2008.zip</a>
        </dd><dt class="label"><a name="mf-req" />[Use cases and requirements for Media Fragments] </dt><dd>
          <cite>Use cases and requirements for Media Fragments</cite>. W3C Working Draft 30 April 2009: <a href="http://www.w3.org/2008/WebVideo/Fragments/WD-media-fragments-reqs/">http://www.w3.org/2008/WebVideo/Fragments/WD-media-fragments-reqs/</a>
        </dd><dt class="label"><a name="ecma-262" />[ECMA-262 5th edition] </dt><dd>
          <cite>ECMA-262 5th edition</cite>: <a href="http://www.ecma-international.org/publications/standards/Ecma-262.htm">http://www.ecma-international.org/publications/standards/Ecma-262.htm</a>
        </dd><dt class="label"><a name="mediaAnnotations" />[Media Annotations] </dt><dd>
          <cite>API for Media Resource 1.0</cite>: <a href="http://www.w3.org/TR/mediaont-api-1.0/">http://www.w3.org/TR/mediaont-api-1.0/</a>
        </dd><dt class="label"><a name="webLinking" />[Web Linking] </dt><dd>
          <cite>Web Linking</cite>: <a href="http://tools.ietf.org/html/draft-nottingham-http-link-header-10">http://tools.ietf.org/html/draft-nottingham-http-link-header-10</a>
        </dd></dl></div><div class="div1">
<h2><a id="collected-syntax-uri" name="collected-syntax-uri" />B Collected ABNF Syntax for URI (Non-Normative)</h2><div class="exampleInner"><pre>
; defined in <cite><a href="#rfc4234">RFC 4234</a></cite>
ALPHA         =  %x41-5A / %x61-7A   ; A-Z / a-z
DIGIT         =  %x30-39 ; 0-9
HEXDIG        =  DIGIT / "A" / "B" / "C" / "D" / "E" / "F"

; defined in <cite><a href="#rfc3986">RFC 3986</a></cite>
unreserved    = ALPHA / DIGIT / "-" / "." / "_" / "~"
pct-encoded   = "%" HEXDIG HEXDIG
sub-delims    = "!" / "$" / "&amp;" / "'" / "(" / ")" / "*" / "+" / "," / ";" / "="
pchar         = unreserved / pct-encoded / sub-delims / ":" / "@"

; defined in <cite><a href="#rtsp">RFC 2326</a></cite>
npt-sec       = 1*DIGIT [ "." *DIGIT ]                     ; definitions taken
npt-hhmmss    = npt-hh ":" npt-mm ":" npt-ss [ "." *DIGIT] ; from <cite><a href="#rtsp">RFC 2326</a></cite>
npt-hh        =   1*DIGIT     ; any positive number
npt-mm        =   2DIGIT      ; 0-59
npt-ss        =   2DIGIT      ; 0-59

; defined in <cite><a href="#rfc3339">RFC 3339</a></cite>
date-fullyear   = 4DIGIT
date-month      = 2DIGIT  ; 01-12
date-mday       = 2DIGIT  ; 01-28, 01-29, 01-30, 01-31 based on
                          ; month/year
time-hour       = 2DIGIT  ; 00-23
time-minute     = 2DIGIT  ; 00-59
time-second     = 2DIGIT  ; 00-58, 00-59, 00-60 based on leap second
                          ; rules
time-secfrac    = "." 1*DIGIT
time-numoffset  = ("+" / "-") time-hour ":" time-minute
time-offset     = "Z" / time-numoffset

partial-time    = time-hour ":" time-minute ":" time-second
                  [time-secfrac]
full-date       = date-fullyear "-" date-month "-" date-mday
full-time       = partial-time time-offset

date-time       = full-date "T" full-time

; Mediafragment definitions
segment       = mediasegment / *( pchar / "/" / "?" )     ; augmented fragment
                                                          ; definition taken from
                                                          ; <cite><a href="#rfc3986">RFC 3986</a></cite>

utf8string    = *( unreserved / pct-encoded / ":" / "@" ) ; utf-8 character
                                                          ; encoded using <cite><a href="#">rfc3896</a></cite> rules.

;
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; Common Prefixes ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;

deftimeformat    = %x6E.70.74                                      ; "npt"
pfxdeftimeformat = %x74.3A.6E.70.74                                ; "t:npt"
smpteformat      = %x73.6D.70.74.65                                ; "smpte"
                  / %x73.6D.70.74.65.2D.32.35                      ; "smpte-25"
                  / %x73.6D.70.74.65.2D.33.30                      ; "smpte-30"
                  / %x73.6D.70.74.65.2D.33.30.2D.64.72.6F.70       ; "smpte-30-drop"
pfxsmpteformat   = %x74.3A.73.6D.70.74.65                          ; "t:smpte"
                  / %x74.3A.73.6D.70.74.65.2D.32.35                ; "t:smpte-25"
                  / %x74.3A.73.6D.70.74.65.2D.33.30                ; "t:smpte-30"
                  / %x74.3A.73.6D.70.74.65.2D.33.30.2D.64.72.6F.70 ; "t:smpte-30-drop"
clockformat      = %x63.6C.6F.63.6B                                ; "clock"
pfxclockformat   = %x74.3A.63.6C.6F.63.6B                          ; "clock"

;
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; Media Segment ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;
mediasegment     = namesegment / axissegment
axissegment      = ( <a href="#timesegment">timesegment</a> / <a href="#spacesegment">spacesegment</a> / <a href="#tracksegment">tracksegment</a> )
               *( "&amp;" ( <a href="#timesegment">timesegment</a> / <a href="#spacesegment">spacesegment</a> / <a href="#tracksegment">tracksegment</a> )
;
; note that this does not capture the restriction of only one timesegment or spacesegment
; in the axisfragment definition, unless we list explicitely all the cases,
;
timesegment      = timeprefix "=" timeparam
timeprefix       = %x74                                      ; "t"
timeparam        = <a href="#npttimedef">npttimedef</a> / <a href="#smptetimedef">smptetimedef</a> / <a href="#clocktimedef">clocktimedef</a>
npttimedef       = [ deftimeformat ":"] ( npttime  [ "," npttime ] ) / ( "," npttime )

npttime       = npt-sec / npt-hhmmss

smptetimedef  = smpteformat ":"( frametime [ "," frametime ] ) / ( "," frametime )
frametime     = 1*DIGIT ":" 2DIGIT ":" 2DIGIT [ ":" 2DIGIT [ "." 2DIGIT ] ]

clocktimedef  = clockformat ":"( clocktime [ "," clocktime ] ) / ( "," clocktime )
clocktime     = (datetime / walltime / date)
datetime      = date-time                                 ; inclusion of <cite><a href="#rfc3339">RFC 3339</a></cite>

spacesegment  = xywhprefix   "=" xywhparam
xywhprefix    = %x78.79.77.68                             ; "xywh"
xywhparam     = [ xywhunit ":" ] 1*DIGIT "," 1*DIGIT "," 1*DIGIT "," 1*DIGIT
xywhunit      = %x70.69.78.65.6C                          ; "pixel"
              / %x70.65.72.63.65.6E.74                    ; "percent"

tracksegment  = trackprefix "=" trackparam
trackprefix   = %x74.72.61.63.6B                          ; "track"
trackparam    = utf8string

namesegment   = nameprefix "=" nameparam
nameprefix    = %x69.64                                   ; "id"
nameparam     = utf8string
      </pre></div></div><div class="div1">
<h2><a id="collected-syntax-http" name="collected-syntax-http" />C Collected ABNF Syntax for HTTP Headers (Non-Normative)</h2><div class="exampleInner"><pre>
; defined in <cite><a href="#rfc2616">RFC 2616</a></cite>
CHAR                   = [any US-ASCII character (octets 0 - 127)]
token                  = 1*[any CHAR except CTLs or separators]`
first-byte-pos         = 1*DIGIT
last-byte-pos          = 1*DIGIT
bytes-unit             = "bytes"
range-unit             = bytes-unit | other-range-unit
byte-range-resp-spec   = (first-byte-pos "-" last-byte-pos)
Range                  = "Range" ":" ranges-specifier
Accept-Ranges          = "Accept-Ranges" ":" acceptable-ranges

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; HTTP Request Headers ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;;
ranges-specifier       = byte-ranges-specifier | fragment-specifier
;
; note that ranges-specifier is extended from <cite><a href="#rfc2616">RFC 2616</a></cite>
; to cover alternate fragment range specifiers
;
fragment-specifier     = "include-setup" | fragment-range *( "," fragment-range )
                                                           [ ";" "include-setup" ]
fragment-range         = time-ranges-specifier | track-ranges-specifier | name-ranges-specifier
;
; note that this doesn't capture the restriction to one fragment dimension occurring
; maximally once only in the fragment-specifier definition.
;
time-ranges-specifier  = timeprefix ":" time-ranges-options
time-ranges-options    = npttimeoption / smptetimeoption / clocktimeoption
npttimeoption          = deftimeformat "=" npt-sec   "-" [ npt-sec ]
smptetimeoption        = smpteformat   "=" frametime "-" [ frametime ]
clocktimeoption        = clockformat   "=" datetime  "-" [ datetime ]

track-ranges-specifier = trackprefix "=" trackparam *( ";" trackparam )

name-ranges-specifier  = nameprefix  "=" nameparam

;;
Accept-Range-Redirect  = "Accept-Range-Redirect" ":" bytes-unit
;
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; HTTP Response Headers ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;
Content-Range-Mapping      = "Content-Range-Mapping" ":" '{'
                             ( content-range-mapping-spec [ ";" def-include-setup ] ) / def-include-setup
                             '}' '=' '{'
                             byte-content-range-mapping-spec '}'

def-include-setup          = %x69.6E.63.6C.75.64.65.2D.73.65.74.75.70  ; "include-setup"

byte-range-mapping-spec    = bytes-unit SP
                             byte-range-resp-spec *( "," byte-range-resp-spec ) "/"
                             ( instance-length / "*" )

content-range-mapping-spec = time-mapping-spec | track-mapping-spec | name-mapping-spec
time-mapping-spec          = timeprefix ":" time-mapping-options
time-mapping-options       = npt-mapping-option / smpte-mapping-option / clock-mapping-option
npt-mapping-option         = deftimeformat SP npt-sec   "-" npt-sec   "/"
                                          [ npt-sec ]   "-" [ npt-sec ]
smpte-mapping-option       = smpteformat   SP frametime "-" frametime "/"
                                          [ frametime ] "-" [ frametime ]
clock-mapping-option       = clockformat   SP datetime  "-" datetime  "/"
                                          [ datetime ]  "-" [ datetime ]

track-mapping-spec         = trackprefix SP trackparam *( ";" trackparam )

name-mapping-spec          = nameprefix  SP nameparam

;;
acceptable-ranges          = 1#range-unit *( "," 1#range-unit )| "none"
;
; note this does not represent the restriction that range-units can only appear once at most;
; this has also been adapted from <cite><a href="#rfc2616">RFC 2616</a></cite>
; to allow multiple range units.
;
other-range-unit           = token | timeprefix | trackprefix | nameprefix

;;
Range-Redirect             = "Range-Redirect" ":" byte-range-resp-spec *( "," byte-range-resp-spec )
      </pre></div></div><div class="div1">
<h2><a id="processing-uri-syntax" name="processing-uri-syntax" />D Notes on parsing media fragment URIs (Non-Normative)</h2><p>
        This appendix contains pseudo-code translation of the ABNF syntax, along with notes on
        some of the caveats to be aware of. Implementors are free to use any equivalent technique(s).
      </p><table border="1" summary="Editorial note: Raphael"><tr><td width="50%" valign="top" align="left"><b>Editorial note: Raphael</b></td><td width="50%" valign="top" align="right"> </td></tr><tr><td valign="top" align="left" colspan="2">To generate a simple figure that shows this processing:
        URI parsing (percent decoding) =&gt;  name=value pairs =&gt; (rfc2047encoding) HTTP</td></tr></table><div class="div2">
<h3><a id="processing-name-value-components" name="processing-name-value-components" />D.1 Processing name-value components</h3><p>
          This is the pseudo-code translation of the ABNF syntax. Implementors are free to use any equivalent technique(s).
        </p><p>
          This section defines how to parse an octet string into an
          ordered list of name-value pairs of unicode strings.
          The octet string consists of name-value pairs separated by
          the "&amp;" character.
        </p><p id="parse-a-name-value-component">
          To <b>parse a name-value component</b>, perform the
          following steps:
        </p><ol class="enumar"><li>
            Let <var>input</var> be the octet string to be parsed.
          </li><li>
            Let <var>pairs</var> be a list of 2-tuples, initially empty.
          </li><li>
            For each octet string <var>name-value</var>
            in <var>input</var> delimited by and not including
            "&amp;", the beginning of <var>input</var> and the end
            of <var>input</var>, in the order they appear
            in <var>input</var>, perform the following steps:
            <ol class="enumla"><li>
                Let <var>pct-name</var> be longest substring from the
                beginning of <var>name-value</var> that does not
                include "=".
              </li><li>
                Let <var>pct-value</var> be the substring from after
                the first "=" in <var>name-value</var> to the end
                of <var>name-value</var>, or the empty string
                if <var>name-value</var> does not include "=".
              </li><li>
                Let <var>name</var> and <var>value</var> be the
                result of
                <a title="" href="#decode-a-percent-encoded-string">
                  decoding
                  percent-encoding
                </a> of <var>pct-name</var>
                and <var>pct-value</var> respectively. If either
                decoding fails, skip the following step.
              </li><li>
                Append the 2-tuple
                (<var>name</var>, <var>value</var>)
                to <var>pairs</var>.
              </li></ol></li><li>
            Return <var>pairs</var>.
          </li></ol><p id="decode-a-percent-encoded-string">
          To <b>decode a percent-encoded string</b>, perform
          the following steps:
        </p><ol class="enumar"><li>
            Let <var>input</var> be the string to be decoded.
          </li><li>
            Let <var>reserved</var> be the empty string.
          </li><li>
            Let <var>output</var> be the result of calling
            Decode(<var>input</var>, <var>reserved</var>).
            If <b>URIError</b> was thrown, abort these steps;
            decoding fails. <cite><a href="#ecma-262">ECMA-262 5th edition</a></cite></li><li>
            Return <var>output</var>.
          </li></ol><p>
          Note that when <a title="" href="#parse-a-name-value-component">
            parsing a name-value component
          </a>,the output is well defined for any input.
          Examples:
        </p><table border="1"><tr><th>input</th><th>output</th><th>comments</th></tr><tr><td>"t=1"</td><td>[("t", "1")]</td><td>simple case</td></tr><tr><td>"t=1&amp;t=2"</td><td>[("t", "1"), ("t", "2")]</td><td>repeated name</td></tr><tr><td>"a=b=c"</td><td>[("a", "b=c")]</td><td>"=" in value</td></tr><tr><td>"a&amp;b=c"</td><td>[("a", ""), ("b", "c")]</td><td>missing value</td></tr><tr><td>"%74=%6ept%3A%310"</td><td>[("t", "npt:10")]</td><td>unnecssary percent-encoding</td></tr><tr><td>"id=J%E4genstedt&amp;t=1"</td><td>[("t", "1")]</td><td>invalid percent-encoding (not UTF-8)</td></tr></table><p>
          While the algorithms defined in this section are designed
          to be largely compatible with the parsing of the URI query
          component in many HTTP server environments, there are
          incompatible differences that implementors should be aware
          of:
        </p><ul><li>
            "&amp;" is the only primary separator for name-value
            pairs, but some server-side languages also treat ";" as
            a separator.
          </li><li>
            name-value pairs with invalid percent-encoding should be
            ignored, but some server-side languages silently mask
            such errors.
          </li><li>
            The "+" character should not be treated specially, but
            some server-side languages replace it with a space (" ")
            character.
          </li><li>
            Multiple occurrences of the same name must be preserved,
            but some server-side languages only preserve the last
            occurrence.
          </li></ul></div><div class="div2">
<h3><a id="processing-name-value-lists" name="processing-name-value-lists" />D.2 Processing name-value lists</h3><p>
          This section defines how to convert an ordered list of
          name-value pairs of unicode strings into a set of media
          fragment dimensions.
        </p><p id="convert-name-value-pairs">
          To <b>
            convert a name-value list into a set of media
            fragment dimensions
          </b>, perform the following steps:
        </p><ol class="enumar"><li>
            Let <var>input</var> be the ordered list of name-value
            pairs to be converted.
          </li><li>
            Let <var>dimensions</var> be a set of dimensions,
            initially empty.
          </li><li>
            For each pair (<var>name</var>, <var>value</var>)
            in <var>input</var>, perform the following steps:
            <ol class="enumla"><li>
                If <var>name</var> is a valid production of
                the <a href="#">timeprefix</a> syntax
                and <var>value</var> is a valid production of
                the <a href="#">timeparam</a> syntax, perform
                the following steps, let the temporal dimension
                of <var>dimensions</var> be the time range represented
                by <var>value</var> (it is an error if a value was previously set).
              </li><li>
                If <var>name</var> is a valid production of
                the <a href="#">xywhprefix</a> syntax
                and <var>value</var> is a valid production of
                the <a href="#">xywhparam</a> syntax, perform
                the following steps, let the spatial dimension
                of <var>dimensions</var> be the area represented
                by <var>value</var> (it is an error if a value was previously set).
              </li><li>
                If <var>name</var> is a valid production of
                the <a href="#">trackprefix</a> syntax
                and <var>value</var> is a valid production of
                the <a href="#">trackparam</a> syntax,
                perform the following steps, let a track dimension
                of <var>dimensions</var> be the track represented
                by <var>value</var>.
              </li><li>
                If <var>name</var> is a valid production of
                the <a href="#">nameprefix</a> syntax
                and <var>value</var> is a valid production of
                the <a href="#">nameparam</a> syntax, perform
                the following steps, let the named dimension
                of <var>dimensions</var> be the name represented
                by <var>value</var> (it is an error if a value was previously set).
              </li><li>
                Otherwise, this is a invalid name-value pair per this
                specification. Validators must emit an error.
              </li></ol></li><li>
            Return <var>dimensions</var>.
          </li></ol></div></div><div class="div1">
<h2><a id="acknowledgments" name="acknowledgments" />E Acknowledgements (Non-Normative)</h2><p>
    This document is the work of the <a href="http://www.w3.org/2008/WebVideo/Fragments/">W3C Media Fragments Working Group</a>. Members of the Working Group are
    (at the time of writing, and in alphabetical order):
    Eric Carlson (Apple, Inc.), 
Michael Hausenblas (DERI Galway at the National University of Ireland, Galway, Ireland),
Philip Jägenstedt (Opera Software),
Jack Jansen (CWI),
Yves Lafon (W3C),
Wonsuk Lee (Electronics and Telecommunications Research Institute),
Erik Mannens (IBBT),
Thierry Michel (W3C/ERCIM),
Guillaume (Jean-Louis) Olivrin (Meraka Institute),
Soohong Daniel Park (Samsung Electronics Co., Ltd.),
Conrad Parker (W3C Invited Experts),
Silvia Pfeiffer (W3C Invited Experts),
David Singer (Apple, Inc.),
Raphaël Troncy (EURECOM),
Davy Van Deursen (IBBT),

  </p><p>
    The people who have contributed to <a href="http://lists.w3.org/Archives/Public/public-media-fragment/">
      discussions on public-media-fragment@w3.org
    </a> are also gratefully acknowledged. In particular:
    Olivier Aubert, Werner Bailer, Pierre-Antoine Champin, Cyril Concolato, Franck Denoual, Martin J. Dürst,
    Jean Pierre Evain, Ken Harrenstien, Kilroy Hughes, Ryo Kawaguchi, Wim Van Lancker,
    Véronique Malaisé, Henrik Nordstrom, Yannick Prié, Yves Raimond, Julian Reschke, Geoffrey Sneddon, 
    Felix Sasaki, Jakub Sendor, Philip Taylor, Christian Timmerer, Jorrit Vermeiren, Jeroen Wijering and Munjo Yu.
  </p></div><div class="div1">
<h2><a id="change-log" name="change-log" />F Change Log (Non-Normative)</h2><p>@@This paragraph will be replaced by the change log. DO NOT TOUCH@@</p></div></div></body></html>