Actually I will post a new version as this version doesn't apply anymore (the context which overlaps an unrelated function has changed) Best wishes Mark On Sun, 30 Jun 2013, Mark Walters <markwalters1009@gmail.com> wrote: > Would anyone be able to review (or just test) this pair of patches: they > are smaller and simpler than the diffstat suggests: the second patch is > just 3 extra lines of code (with some whitespace change and commments). > > This pair of patches does remove the worst piece of code in pick: a > sleep loop waiting for the correct message to arrive. > > I should have said that it applies on top of > id:1369550458-30562-1-git-send-email-markwalters1009@gmail.com > > (which is also very simple) > > Many thanks > > Mark > > > > > On Sun, 26 May 2013, Mark Walters <markwalters1009@gmail.com> wrote: >> Pick used to have one horrible hack: if the user asked it to open the >> first matching message it had to check whether that had arrived (as >> the search is asynchronous) and if not wait and try again. Now the >> opening of the first matching message is called via the pick process >> filter this hack can be removed. >> >> This did reveal the followibg small bug. Pick shows the subject line >> in the output but if it is the same as the previous line (ignoring re: >> etc) it shows ... If a single message is refreshed (eg for a tag >> update) this was got wrong. The change above triggered this and made >> the test fail as the unread tag was removed from the first matching >> message when it was displayed. >> >> Patch 2/2 fixes this by storing the previous subject with the search result. >> >> Best wishes >> >> Mark >> >> >> Mark Walters (2): >> contrib: pick: remove hack notmuch-pick-show-match-message-with-wait >> contrib: pick: fix refresh result >> >> contrib/notmuch-pick/notmuch-pick.el | 60 ++++++++++++++++----------------- >> 1 files changed, 29 insertions(+), 31 deletions(-) >> >> -- >> 1.7.9.1