Re: [Evolution] Move mail to where?



On Mon, 2020-01-20 at 18:19 +0100, Milan Crha via evolution-list wrote:
[My comments interspersed in Milan's -GNR]
On Mon, 2020-01-20 at 17:20 +0100, George N. Reeke wrote:
There is one other way that would let me go back to "type-to-
search"
and that would be if the folders are always searched in some known
order, like alphabetically.  Then I could get what I want by
renaming
my folders, for example "mailA2020", "mailB2019" etc.  Would that
always open "mailA2020" first for the search and return the first
box found in that folder that matched the typed string, if any?

      Hi,
the type-to-search is fully handled by gtk+. If I'm not mistaken,
then
it searches from the currently selected row downwards.

---No it doesn't, it searches from the top of the tree--seems
to be alphabetical, but I haven't seen it enough times to be sure.

I briefly looked into the code, just in case I'd forget of some
option
or anything like that, and I saw that there is no option for the
"expand-all", it's always expanding it. There is an option to avoid
expansion of archive folders.

Neither of it is what you want, right? You basically want an option
to
always open the copy/move dialog with certain folder preselected.
With
that, you'll navigate whenever you want, but this navigation will not
change the folder to have the copy/move dialog opened with the next
time, right? Like you'd select the top archive folder and navigate
always from it.

---I suggested this behavior as a solution that would be ideal for me,
but there has been no support for it on this list, so I will drop it.
The other alternative I suggested was to leave it like it was in the
old evolution I was using (2.32) where type-to-search seemed to start
in whatever top level was used last--not perfect, but OK.

I know you basically want to "revert" (by added option) the previous
change, but I'm wondering whether it's really the best thing to do.
Thinking that you use it as archiving, I'd rather change how
archiving
itself works. Currently, when you right-click the folder in the
folder
tree and pick Properties, there's an Archive tab. You can set there
how
the Message->Archive (Ctrl+Alt+A) should work. There are three
options:
1) move to default archive folder (per-account option); 2) move to
specified folder; 3) delete messages. Maybe this is a pure nonsense,
but would the 4th option "always pick folder where to archive", which
would disable the "auto-cleanup option" and which would always open
the
folder selector in the archive folder set for the account, do the
trick?
To me, archiving is what I do when I move old tree branches offline
for permanent storage.  In my system, these are conveniently top-level
folders named according to year.  What I do every day is file or save
emails in current folders, except every once in a great while I want
to save one in an older folder (that has not yet been archived).  So
your new suggestion, while interesting, seems unduly complex for me
(more clicks on everyday mail) so I would not push for something like
it.


I look on it as a similarity when saving files in a web browser, I
prefer to always pick where to save the file, but I also prefer to
have
preselected the directory I used the last time (which you do not seem
to like).
No, I am fine with that (even though not ideal because it messes up
after one of those abnormal saves).  My original complaint was simply
that keyboard searches don't do that anymore [but mouse searches do].
So I am now retraining my brain to use the mouse instead of the
keyboard.  Maybe I could appeal to those of you who are programmers
to ask why is it that that mouse searches now behave differently from
keyboard searches, and how hard would it be to make a change that
would make keyboard searches useful again and would have no effect at
all on those who prefer to use the mouse?
   Anyway, I think this topic has gone as far as it can, I will no
longer post on it except to answer any questions directed at me.
George Reeke

This is more complicated than tweaking the copy/move dialog, code-
wise
speaking too, but I think it's cleaner solution. Maybe not.
      Bye,
      Milan





[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]