mirrors/git-annex
2
0
Fork 0
Code Issues Pull requests Projects Releases Packages 1 Wiki Activity Actions

clean up url removal presence update

Browse source 
* rmurl: Fix a case where removing the last url left git-annex thinking
  content was still present in the web special remote.
* SETURLPRESENT, SETURIPRESENT, SETURLMISSING, and SETURIMISSING
  used to update the presence information of the external special remote
  that called them; this was not documented behavior and is no longer done.

Done by making setUrlPresent and setUrlMissing only update presence info
for the web, and only when the url is a web url. See the comment for
reasoning about why that's the right thing to do.

In AddUrl, had to make it update location tracking, to handle the
non-web-url case.

This commit was sponsored by Ewen McNeill on Patreon.
This commit is contained in:
Joey Hess 2018-10-04 17:33:25 -04:00
parent ca66e7b66a
commit 451171b7c1
No known key found for this signature in database
GPG key ID: DB12DB0FF05F8F38
14 changed files with 77 additions and 26 deletions
Download patch file Download diff file Expand all files Collapse all files

2
doc/bugs/suggests_to_enable_web_remote_even_when_there_is_no_web_urls_for_the_file.mdwn
View file

@ -39,4 +39,4 @@ ok
[[!meta name=yoh]]
> closing as yoh says "nothing to fix" [[done]] --[[Joey]]
> I think I've now understood and fixed this, so [[done]] --[[Joey]]

35
doc/bugs/suggests_to_enable_web_remote_even_when_there_is_no_web_urls_for_the_file/comment_4_6dff7befbaacbff573c5f72688966af5._comment Normal file
View file

@ -0,0 +1,35 @@
[[!comment format=mdwn
username="joey"
subject="""comment 4"""
date="2018-10-04T20:20:43Z"
content="""
I think that making rmurl filter out urls with an OtherDownloader
when removing an url claimed by the web is sufficient.
Then SETURIPRESENT won't confuse rmurl from the web.
But what about SETURLPRESENT? Well, that url ought to be a url that can be
downloaded by the web special remote. And if the web special remote is marked
as present, it will try to download from that url along with whatever other urls
are recorded. So, it makes sense for rmurl to treat that url as an indication
that the web still has content, when removing some other url.
But then if an url is added to the web, and then an external special remote
uses SETURLPRESENT with another url, and then rmurl removes the first url,
and then SETURLMISSING removes the other url, the content will still be marked
as present in the web. So it seem that SETURLMISSING should mark the
content as not present in the web if that was the last url. (So should other
special remotes that remove regular urls, like bittorrent.)
Currently the content is marked as not present in the remote making the change,
which is the wrong thing! (And SETURLPRESENT/SETURIPRESENT mark content
as present in the external remote, which also seems unncessary.)
The remaining case is an rmurl on an url with an OtherDownloader,
which is being removed from some other special remote than web, when there are
some more urls with OtherDownloader that were set by other special remotes.
I think that it doesn't actually make sense for rmurl of such an url to
mark the content as not being present in the special remote. The user should
drop it if they want to remove it from the special remote. It's ok for it to
remove the url but not update the presence information.
"""]]