Organisation internationale de normalisation


m35090 Clarification on fragment identifiers for ISOBMFF



Yüklə 9,04 Mb.
səhifə24/277
tarix02.01.2022
ölçüsü9,04 Mb.
#24054
1   ...   20   21   22   23   24   25   26   27   ...   277

m35090 Clarification on fragment identifiers for ISOBMFF


The uniqueness requirement at a given level (file, movie, track) seems implied (e.g. the replacement rule for movie fragments) but embarrassingly nowhere stated. Into the defect report [Cyril/Jean]. Item_IDs shall be unique at a given level; this is in alignment with the ‘replacement rule’ for items with the same ID in movie fragments. We explicitly document what happens when there is a reference to an item by item_id and the item is updated in movie fragments (‘use the latest version you have received at the same level’). Do we allow meta boxes at file level in movie fragments? This should also be clarified (no?). The word ‘file’ level is ambiguous (fragment file, or the file that contains the movie box? We think the latter.)

The question of meco’s is rather hard. If the mere box says that two meta boxes are semantically equivalent, could they share item_IDs? The whole interaction of mere/meco with the evolved design of meta boxes needs a careful examination. (We note that meco/mere have no reference software or conformance streams). Experts are urged to look at the issues around meco/mere, movie fragments, URL fragment identifiers, and the item_id uniqueness requirement.

Then for the clarification on URL fragment forms we agree. Into the defect report or TuC? Some existing cases are ambiguous which suggests COR, but some of this looks like new features perhaps? For now, into the defect report. [Jean/Cyril]


      1. Yüklə 9,04 Mb.

        Dostları ilə paylaş:
1   ...   20   21   22   23   24   25   26   27   ...   277




Verilənlər bazası müəlliflik hüququ ilə müdafiə olunur ©muhaz.org 2024
rəhbərliyinə müraciət

gir | qeydiyyatdan keç
    Ana səhifə


yükləyin