We frequently get asked about the uncompressed sizes of dump files.
It should be possible for the dump generator to count the number of
bytes it's pumping out and then have the backup runner take this
information and include it with the download links.
Version: unspecified
Severity: enhancement
URL: http://download.wikimedia.org/
| Status | Subtype | Assigned | Task | ||
|---|---|---|---|---|---|
| Open | Feature | None | T16631 Include uncompressed sizes in dump file RSS info | ||
| Open | Feature | None | T8064 Include uncompressed file sizes in data dump download pages |
I presume you want this only for the files with page content? If we had to we could make this a separate step even. It would be included as a separate file with byte counts, rather than being shovelled into the XML output someplace.
I think it would be most useful for the largest dumps, which I think corresponds pretty well to the ones with actual page content.
It would be nice to have it for the others as well, but definitely not as important.
Moving this to the Dumps Rewrite project; this feature would take too much time to add to the current dumps and still maintain reasonable run speed.
@ArielGlenn: Unassigning task, to avoid cookie-licking. (Feel free to reclaim if you plan to work on this. Thanks.)
Fine by me. (Note that this task is likely infeasible with the current dumps architecture, as mentioned above.)