response

1 files changed, 23 insertions, 0 deletions

diff --git a/doc/bugs/importfeed_does_not_work_with_socks_proxy/comment_1_9797d8de55cf91288ecfc64a7e96645d._comment b/doc/bugs/importfeed_does_not_work_with_socks_proxy/comment_1_9797d8de55cf91288ecfc64a7e96645d._comment
new file mode 100644
index 000000000..3db7a070d
--- /dev/null
+++ b/doc/bugs/importfeed_does_not_work_with_socks_proxy/comment_1_9797d8de55cf91288ecfc64a7e96645d._comment
@@ -0,0 +1,23 @@
+[[!comment format=mdwn
+ username="joey"
+ subject="""comment 1"""
+ date="2018-02-22T16:32:33Z"
+ content="""
+It's true that importfeed calls Url.download to download the feed file, 
+without bothering to support annex.web-download-command. That could easily
+be changed.
+
+However, it also uses Url.getUrlInfo, which does not and cannot use the
+annex.web-download-command interface, and which is too complicated an
+interface to make into a hook.
+
+Indeed, annex.web-download-command was never intended to cover all
+the ways git-annex uses http, but only uses of http to download
+large file contents. And importfeed does use it for such downloads,
+but not for its other http needs.
+
+To configure use of a proxy, you would probably be best served by using
+the `http_proxy` and `https_proxy` environment variables, which are
+supported by wget, curl, and by the haskell http library that git-annex
+also uses.
+"""]]