File: comment_1_a20bbe49f2f23d23131221ec79c8b973._comment

package info (click to toggle)
git-annex 10.20230126-3
  • links: PTS, VCS
  • area: main
  • in suites: bookworm
  • size: 69,344 kB
  • sloc: haskell: 74,654; javascript: 9,103; sh: 1,304; makefile: 203; perl: 136; ansic: 44
file content (28 lines) | stat: -rw-r--r-- 1,153 bytes parent folder | download | duplicates (8)
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
[[!comment format=mdwn
 username="joey"
 subject="""comment 1"""
 date="2014-12-02T18:39:34Z"
 content="""
There's no particular size limit for metadata. But things may not scale
well past some size.

Having a big metadata value could make git-annex be a little slow when
dealing with metadata, since it has to parse the metadata log file.
So if you stored say, 1 mb in metadata, that's one mb of disk IO and data
processing every time git-annex looks at the metadata for that file.

All the metadata for a file will also be buffered in memory when git-annex
is looking at that file's metadata. So `git-annex view` would use 1 mb or
so more memory in the above example. (But git-annex only looks at one
file's metadata at a time, so its memory use won't grow if you have a lot
of files with metadata.)

Also of course, the metadata is stored in git, and so it will make your git
repository bigger.

I would not expect any reasonable description of a photo to be large enough
for its size to be a problem.

I'd even feel ok with putting the full text of a license in as metadata
(GPL is 30kb, and git would deduplicate redundant license metadata).
"""]]