sup

community/pipermail-archives/sup-talk/2008-03.txt (126949B) - raw

 From wmorgan-sup@masanjin.net  Sat Mar  1 15:37:35 2008
 From: wmorgan-sup@masanjin.net (William Morgan)
 Date: Sat, 01 Mar 2008 12:37:35 -0800
 Subject: [sup-talk] [PATCH] shell commands are now run in a child process
 In-Reply-To: <1645600603.20080229224209@rice.edu>
 References: <1203961671-sup-8158@south> <1645600603.20080229224209@rice.edu>
 Message-ID: <1204401464-sup-9155@south>
 
 Reformatted excerpts from Christopher Warrington's message of 2008-02-29:
 > I've been getting more of these when playing with offlineimap. I
 > submitted a bug upstream:
 > http://ferret.davebalmain.com/trac/ticket/343
 
 And it looks like Dave is back maintaining Ferret, so this might
 actually get addressed. That's great.
 
 -- 
 William <wmorgan-sup at masanjin.net>
 
 From wmorgan-sup@masanjin.net  Sun Mar  2 13:08:58 2008
 From: wmorgan-sup@masanjin.net (William Morgan)
 Date: Sun, 02 Mar 2008 10:08:58 -0800
 Subject: [sup-talk] [PATCH] First draft of attachment processing for
 	more gmail style searches
 In-Reply-To: <1204232994-sup-628@tomsk>
 References: <1203972458-sup-5906@tomsk> <1204220051-sup-129@south>
 	<1204232994-sup-628@tomsk>
 Message-ID: <1204479552-sup-4100@south>
 
 Reformatted excerpts from Marcus Williams's message of 2008-02-28:
 > The only thing I'm a little wary of is the join() I do of the
 > attachment filenames for the index (like labels). This means that
 > ferret doesnt actually know the difference between two files called
 > file1 and file2 and a single file called "file1 file2". Not sure it
 > matters that much for this usage though.
 
 The answer here is to escape the spaces and to use a Ferret custom
 analyzer for this field in the index, one that will split only on
 non-escaped spaces.
 
 Something like this (needs testing):
 
   irb(main):055:0> a = Ferret::Analysis::RegExpAnalyzer.new /([^\s\\]|(\\\s))+/, false
   => #<Ferret::Analysis::RegExpAnalyzer:0xb79740fc>
   irb(main):056:0> t = a.token_stream :potato, "one\\ two three\\ four"=> #<Ferret::Analysis::TokenStream:0xb79705d8>
   irb(main):057:0> t.next
   => token["one\ two":0:8:1]
   irb(main):058:0> t.next
   => token["three\ four":9:20:1]
 
 Then assign that analyzer to the :attachments field in index.rb circa
 line 37, just like I do for :subject and :body.
 
 You'll have to make sure to do the escaping properly both on user input
 at query time, and at storage time to the index.
 
 > Also I dont repopulate the attachments attribute on the message object
 > and I couldnt figure out quite how you do it for labels (through the
 > initialise?). 
 
 Not quite sure what you mean here, but the answer might be: index.rb
 line 371 is where we build a Message object from an index entry, and
 you'll need to pass in an :attachments attribute (and handle it within
 Message#initialize).
 
 -- 
 William <wmorgan-sup at masanjin.net>
 
 From chrisw@rice.edu  Mon Mar  3 02:07:55 2008
 From: chrisw@rice.edu (Christopher Warrington)
 Date: Mon, 03 Mar 2008 01:07:55 -0600
 Subject: [sup-talk] Maildir source with one message
 Message-ID: <1204528063-sup-7345@chris-tablet>
 
 In playing with offline IMAP and Maildir sources (much faster than a native IMAP source and it can be used offline!), I noticed an odd bug.
 
 If there is only one message in a Maildir source, neither sup not
 sup-sync can see this message. Adding another message makes both
 visible messages visible.
 
 --- here begins speculation ---
 
 I *THINK* that this has to do with the definitions Source.start_offset
 and Source.end_offset. As I understand it, they should define a range
 of message ids [start_offset, end_offset).
 
 Now, when start_offset = end_offset--as is the case when there is only
 one message, bad things happen. As I understand the math:
 (a,a) = {}
 [a,a) = {} *our case
 (a,a] = {}
 [a,a] = {a.
 
 Changing maildir.rb:126 from:
     @ids.last
 to
     @ids.last + 1
 appears to have fixed the problem without ill-effect.
 
 Simple inspection of imap.rp leads be to believe that it will have the
 same problem.
 
 A mbox source will have the same problem iff the mbox is empty.
 
 --- here ends speculation ---
 
 -- 
 Christopher Warrington <chrisw at rice.edu>
 
 From chrisw@rice.edu  Mon Mar  3 02:42:53 2008
 From: chrisw@rice.edu (Christopher Warrington)
 Date: Mon, 03 Mar 2008 01:42:53 -0600
 Subject: [sup-talk] new in next: faster saving and bigger indexes
 In-Reply-To: <1204219351-sup-3340@south>
 References: <1203871982-sup-7942@south> <1204138450-sup-1391@chris-tablet>
 	<1204219351-sup-3340@south>
 Message-ID: <1204530080-sup-987@chris-tablet>
 
 Excerpts from William Morgan's message of Thu Feb 28 11:23:29 -0600 2008:
 >> Now, we have offline(-ish) mode for IMAP. Horray! I can do e-mail
 >> on the plane.
 > You could always have used offlineimap. :)
 
 Wow! That's an amazing program! I'm using it now to read and write
 e-mail using sup at 30,000 feet (and much quicker than IMAP, too!).
 
 (Message sent at a paltry 200 feet...)
 
 -- 
 Christopher Warrington <chrisw at rice.edu>
 
 From chrisw@rice.edu  Mon Mar  3 03:00:33 2008
 From: chrisw@rice.edu (Christopher Warrington)
 Date: Mon,  3 Mar 2008 02:00:33 -0600
 Subject: [sup-talk] [PATCH] fixed off-by-one error in imap.rb and maildir.rb
 In-Reply-To: <1204528063-sup-7345@chris-tablet>
 References: <1204528063-sup-7345@chris-tablet>
 Message-ID: <1204531233-2092-1-git-send-email-chrisw@rice.edu>
 
 The end_offset reported by imap and maildir sources was incorrect if there was
 only one message in the source. Since end_offset is exclusive, we must add one to
 the last known id to get include all valid message ids in the range.
 ---
  lib/sup/imap.rb    |    2 +-
  lib/sup/maildir.rb |    2 +-
  2 files changed, 2 insertions(+), 2 deletions(-)
 
 diff --git a/lib/sup/imap.rb b/lib/sup/imap.rb
 index e785597..1d36976 100644
 --- a/lib/sup/imap.rb
 +++ b/lib/sup/imap.rb
 @@ -176,7 +176,7 @@ class IMAP < Source
  
    def end_offset
      unsynchronized_scan_mailbox
 -    @ids.last
 +    @ids.last + 1
    end
    synchronized :end_offset
  
 diff --git a/lib/sup/maildir.rb b/lib/sup/maildir.rb
 index 5c9600d..584e657 100644
 --- a/lib/sup/maildir.rb
 +++ b/lib/sup/maildir.rb
 @@ -123,7 +123,7 @@ class Maildir < Source
  
    def end_offset
      scan_mailbox :rescan => true
 -    @ids.last
 +    @ids.last + 1
    end
  
    def pct_done; 100.0 * (@ids.index(cur_offset) || 0).to_f / (@ids.length - 1).to_f; end
 -- 
 1.5.3.7
 
 
 From chrisw@rice.edu  Mon Mar  3 03:01:20 2008
 From: chrisw@rice.edu (Christopher Warrington)
 Date: Mon,  3 Mar 2008 02:01:20 -0600
 Subject: [sup-talk] [PATCH] updated text and keybinding in
 	label-search-results-mode
 Message-ID: <1204531280-4048-1-git-send-email-chrisw@rice.edu>
 
 The text in label-search-results-mode now matches that of search-results-mode.
 ---
  lib/sup/modes/label-search-results-mode.rb |    4 ++--
  1 files changed, 2 insertions(+), 2 deletions(-)
 
 diff --git a/lib/sup/modes/label-search-results-mode.rb b/lib/sup/modes/label-search-results-mode.rb
 index af6831c..bca51d4 100644
 --- a/lib/sup/modes/label-search-results-mode.rb
 +++ b/lib/sup/modes/label-search-results-mode.rb
 @@ -10,12 +10,12 @@ class LabelSearchResultsMode < ThreadIndexMode
    end
  
    register_keymap do |k|
 -    k.add :refine_search, "Refine search", '.'
 +    k.add :refine_search, "Refine search", '|'
    end
  
    def refine_search
      label_query = @labels.size > 1 ? "(#{@labels.join('||')})" : @labels.first
 -    query = BufferManager.ask :search, "query: ", "+label:#{label_query} "
 +    query = BufferManager.ask :search, "refine query: ", "+label:#{label_query} "
      return unless query && query !~ /^\s*$/
      SearchResultsMode.spawn_from_query query
    end
 -- 
 1.5.3.7
 
 
 From chrisw@rice.edu  Mon Mar  3 04:33:15 2008
 From: chrisw@rice.edu (Christopher Warrington)
 Date: Mon, 03 Mar 2008 03:33:15 -0600
 Subject: [sup-talk] Newline Issues
 Message-ID: <1204536487-sup-9413@chris-tablet>
 
 Now, I'm running into newline issues.
 
 When I use offlineimap under Cgywin, the messages are delivered with
 CRLF. Sup cannot parse these: it thinks that the body is empty. I know
 that sup is seeing the entire message. If I press H, I see all of the
 message.
 
 Shouldn't sup handle CRLF delimited messages? As I understand RFC822,
 all headers should be terminated with CRLF.
 
 Failing that, is anyone well versed enough with offlineimap or python
 to tell me how to get it to deliver with LF instead of CRLF?
 
 -- 
 Christopher Warrington <chrisw at rice.edu>
 
 From marcus-sup@bar-coded.net  Wed Mar  5 05:01:14 2008
 From: marcus-sup@bar-coded.net (Marcus Williams)
 Date: Wed, 05 Mar 2008 10:01:14 +0000
 Subject: [sup-talk] [PATCH] First draft of attachment processing for
 	more gmail style searches
 In-Reply-To: <1204479552-sup-4100@south>
 References: <1203972458-sup-5906@tomsk> <1204220051-sup-129@south>
 	<1204232994-sup-628@tomsk> <1204479552-sup-4100@south>
 Message-ID: <1204711082-sup-3541@tomsk>
 
 On 2.3.2008, William Morgan wrote:
 > The answer here is to escape the spaces and to use a Ferret custom
 > analyzer for this field in the index, one that will split only on
 > non-escaped spaces.
 [snip]
 
 Ah, right. Should be easy enough (sup-sync here we come)
 
 > Not quite sure what you mean here, but the answer might be: index.rb
 > line 371 is where we build a Message object from an index entry, and
 > you'll need to pass in an :attachments attribute (and handle it within
 > Message#initialize).
 
 Ok thats what I figured in the end.
 
 ANother question - how do I get sent/drafts to get the attachment
 labels? They dont seem to get set up when I attach a file to a
 message. Should I just be adding/deleting them in the methods that
 deal with adding/deleting attachments in reply mode?
 
 Thanks
 
 Marcus
 
 From marcus-sup@bar-coded.net  Wed Mar  5 05:11:13 2008
 From: marcus-sup@bar-coded.net (Marcus Williams)
 Date: Wed, 05 Mar 2008 10:11:13 +0000
 Subject: [sup-talk] [PATCH] Fix for icon title in gnome terminal (possibly
 	others)
 Message-ID: <1204711708-sup-3638@tomsk>
 
 The escape code used for terminal titles works mostly but it doesnt set
 the icon title as well [1], which is useful when minimised. This patch makes
 the escape code set both window and icon title so the task bar updates
 correctly in the gnome terminal (and probably others).
 
 [1] http://rtfm.etla.org/xterm/ctlseq.html (see Operating System Controls)
 ---
  lib/sup/buffer.rb |    2 +-
  1 files changed, 1 insertions(+), 1 deletions(-)
 
 diff --git a/lib/sup/buffer.rb b/lib/sup/buffer.rb
 index d40a626..978e104 100644
 --- a/lib/sup/buffer.rb
 +++ b/lib/sup/buffer.rb
 @@ -263,7 +263,7 @@ EOS
          get_status_and_title @focus_buf # must be called outside of the ncurses lock
        end
  
 -    print "\033]2;#{title}\07" if title && @in_x
 +    print "\033]0;#{title}\07" if title && @in_x
  
      Ncurses.mutex.lock unless opts[:sync] == false
  
 -- 
 1.5.4.1
 
 
 From rodkoch@gmail.com  Thu Mar  6 18:21:18 2008
 From: rodkoch@gmail.com (rodneyk)
 Date: Thu, 6 Mar 2008 15:21:18 -0800 (PST)
 Subject: [sup-talk]  How to search for and mass tag thousands of emails?
 Message-ID: <15885391.post@talk.nabble.com>
 
 
 Hi,
 I just started using sup, after years of muttng I must say I like it a lot!
 I have a giant inbox mbox with thousands of emails which I want to
 starting using in sup. I can easily search and tag to categorize them all,
 however, I can only tag the current screen full of search results. Is there
 a way to tag all the search results, and not just the results on screen?
 
 Being able to import existing mail and quickly organize is probably
 something
 most new users will need to do. I also checked out sup from the git repo,
 and can run
 devel/console, so if this mass tagging can only be done manually, I can do
 it, as I'm
 pretty proficient with ruby.
 
 --Rodney
 -- 
 View this message in context: http://www.nabble.com/How-to-search-for-and-mass-tag-thousands-of-emails--tp15885391p15885391.html
 Sent from the SUP Talk mailing list archive at Nabble.com.
 
 
 From manish@gslab.com  Fri Mar  7 02:15:09 2008
 From: manish@gslab.com (Manish Sapariya)
 Date: Fri, 07 Mar 2008 12:45:09 +0530
 Subject: [sup-talk] How to search for and mass tag thousands of emails?
 In-Reply-To: <15885391.post@talk.nabble.com>
 References: <15885391.post@talk.nabble.com>
 Message-ID: <1204874020-sup-2295@alioth.gs-lab.com>
 
 I did it by writing before-add-message.rb hook and resyncing all
 messages. The before-add-message hook allows you to add delete
 labels before it indexes the message.
 
 Hope this helps.
 Regards,
 Manish
 
 Excerpts from rodneyk's message of Fri Mar 07 04:51:18 +0530 2008:
 > 
 > Hi,
 > I just started using sup, after years of muttng I must say I like it a lot!
 > I have a giant inbox mbox with thousands of emails which I want to
 > starting using in sup. I can easily search and tag to categorize them all,
 > however, I can only tag the current screen full of search results. Is there
 > a way to tag all the search results, and not just the results on screen?
 > 
 > Being able to import existing mail and quickly organize is probably
 > something
 > most new users will need to do. I also checked out sup from the git repo,
 > and can run
 > devel/console, so if this mass tagging can only be done manually, I can do
 > it, as I'm
 > pretty proficient with ruby.
 > 
 > --Rodney
 
 -- 
 
 From wmorgan-sup@masanjin.net  Fri Mar  7 10:52:40 2008
 From: wmorgan-sup@masanjin.net (William Morgan)
 Date: Fri, 07 Mar 2008 07:52:40 -0800
 Subject: [sup-talk] How to search for and mass tag thousands of emails?
 In-Reply-To: <15885391.post@talk.nabble.com>
 References: <15885391.post@talk.nabble.com>
 Message-ID: <1204904692-sup-7926@south>
 
 Reformatted excerpts from rodneyk's message of 2008-03-06:
 > I have a giant inbox mbox with thousands of emails which I want to
 > starting using in sup. I can easily search and tag to categorize them
 > all, however, I can only tag the current screen full of search
 > results. Is there a way to tag all the search results, and not just
 > the results on screen?
 
 You could do it through the GUI, but that probably won't scale well.
 A better approach is to use the before-add-hook as Manish said, or to
 use the development console to modify the index directly.
 
 Feed devel/console.sh something like (untested):
 
   Index.index.search_each("potato") do |id, score|
     m = Index.build_message(id)
     puts "[#{id}] [#{m.from}] [#{m.subj}]"
     m.add_label :potato
     m.save! Index
   end
 
 But there's not really much of a point to labeling the results of a
 (simple) search. Labels are good for things that are hard to search for;
 for everything else, just trust in the search!
 
 > Being able to import existing mail and quickly organize is probably
 > something most new users will need to do.
 
 Sup supports auto-labeling folders out of the box, which I think is the
 most common transition case.
 
 > I also checked out sup from the git repo, and can run devel/console,
 > so if this mass tagging can only be done manually, I can do it, as I'm
 > pretty proficient with ruby.
 
 The big advantaage of the git repo (the next branch, at least) right now
 is that adding labels to a message is much faster.
 
 -- 
 William <wmorgan-sup at masanjin.net>
 
 From wmorgan-sup@masanjin.net  Sat Mar  8 17:02:06 2008
 From: wmorgan-sup@masanjin.net (William Morgan)
 Date: Sat, 08 Mar 2008 14:02:06 -0800
 Subject: [sup-talk] [PATCH] First draft of attachment processing for
 	more gmail style searches
 In-Reply-To: <1204711082-sup-3541@tomsk>
 References: <1203972458-sup-5906@tomsk> <1204220051-sup-129@south>
 	<1204232994-sup-628@tomsk> <1204479552-sup-4100@south>
 	<1204711082-sup-3541@tomsk>
 Message-ID: <1205013484-sup-7653@south>
 
 Reformatted excerpts from Marcus Williams's message of 2008-03-05:
 > ANother question - how do I get sent/drafts to get the attachment
 > labels? They dont seem to get set up when I attach a file to a
 > message. Should I just be adding/deleting them in the methods that
 > deal with adding/deleting attachments in reply mode?
 
 Once the message is sent, SentManager.write_sent_message will call
 PollManager.add_messages_from, which in turn calls the index.rb code you
 tweaked to build the message object that's actually used outside of
 edit-message-mode. 
 
 I've just merged the topic branch that actually make SentManager work
 this way down to master (gotta love it when the same changes fix two
 different problems!), so if you rebase now, you should be good to go.
 
 Sorry for the general delay in replying. I've been coding up a little
 distributed issue tracker that I think will help me manage Sup a little
 better... or at least ensure I don't forget people's suggestions.
 
 -- 
 William <wmorgan-sup at masanjin.net>
 
 From jh@chabran.fr  Sun Mar  9 09:28:21 2008
 From: jh@chabran.fr (Jean-Hadrien CHABRAN)
 Date: Sun, 9 Mar 2008 14:28:21 +0100
 Subject: [sup-talk] startup crash
 Message-ID: <345330030803090628u2ac33bfcn3b8ab30da6b16be3@mail.gmail.com>
 
 Hello,
 
 I just discovered sup, and I wanted to give it a try, but sup-config /
 sup crash on startup, without any useful informations ( for me )  to
 discover why.
 
 Some informations :
 
 uname -a : Linux Gradius 2.6.24-gentoo-r2 #3 SMP Wed Feb 20 17:39:36
 CET 2008 i686 Intel(R) Pentium(R) 4 CPU 2.60GHz GenuineIntel GNU/Linux
 
 ruby : ruby 1.8.6 (2007-12-03 patchlevel 113) [i686-linux]
 
 using sup v0.4, installed with "gem i sup"
 
 backtrace :
 
 jh at gradius $ sup-config
 /usr/lib/ruby/gems/1.8/gems/sup-0.4/lib/sup/account.rb:7:in
 `initialize': no name for account (ArgumentError)
         from /usr/lib/ruby/gems/1.8/gems/sup-0.4/lib/sup/account.rb:44:in `new'
         from /usr/lib/ruby/gems/1.8/gems/sup-0.4/lib/sup/account.rb:44:in
 `add_account'
         from /usr/lib/ruby/gems/1.8/gems/sup-0.4/lib/sup/account.rb:26:in
 `initialize'
         from /usr/lib/ruby/gems/1.8/gems/sup-0.4/lib/sup.rb:98:in `new'
         from /usr/lib/ruby/gems/1.8/gems/sup-0.4/lib/sup.rb:98:in `start'
         from /usr/lib/ruby/gems/1.8/gems/sup-0.4/bin/sup-config:153
         from /usr/bin/sup-config:16:in `load'
         from /usr/bin/sup-config:16
 
 tested on my three gentoo boxes, got same results :(
 
 -- 
 Jean-Hadrien Chabran
 
 From jh@chabran.fr  Sun Mar  9 10:45:34 2008
 From: jh@chabran.fr (Jean-Hadrien CHABRAN)
 Date: Sun, 9 Mar 2008 15:45:34 +0100
 Subject: [sup-talk] startup crash
 In-Reply-To: <345330030803090628u2ac33bfcn3b8ab30da6b16be3@mail.gmail.com>
 References: <345330030803090628u2ac33bfcn3b8ab30da6b16be3@mail.gmail.com>
 Message-ID: <345330030803090745p1f0c6a14if40f95da893c07c0@mail.gmail.com>
 
 I just solved the problem on my FreeBSD 6.3, by installing
 ruby18-iconv-1.8.6.111,1.
 
 On my gentoo boxes, iconv seems to be correctly installed
 /usr/lib64/ruby/1.8/x86_64-linux/iconv.so, but this is now a
 distribution issue.
 
 But I'm still curious, why this dependency problem is showing itself
 in a such odd way ?
 
 On 3/9/08, Jean-Hadrien CHABRAN <jh at chabran.fr> wrote:
 > Hello,
 >
 >  I just discovered sup, and I wanted to give it a try, but sup-config /
 >  sup crash on startup, without any useful informations ( for me )  to
 >  discover why.
 ---- 8< --- 8< -------------------------------------------------------------------
 >
 
 
 -- 
 Jean-Hadrien Chabran
 Maitre-Oeuvre SEP EFREI
 Reponsable Tutorat-L
 
 From wmorgan-sup@masanjin.net  Sun Mar  9 13:44:07 2008
 From: wmorgan-sup@masanjin.net (William Morgan)
 Date: Sun, 09 Mar 2008 10:44:07 -0700
 Subject: [sup-talk] startup crash
 In-Reply-To: <345330030803090745p1f0c6a14if40f95da893c07c0@mail.gmail.com>
 References: <345330030803090628u2ac33bfcn3b8ab30da6b16be3@mail.gmail.com>
 	<345330030803090745p1f0c6a14if40f95da893c07c0@mail.gmail.com>
 Message-ID: <1205082815-sup-7754@south>
 
 Reformatted excerpts from Jean-Hadrien CHABRAN's message of 2008-03-09:
 > But I'm still curious, why this dependency problem is showing itself
 > in a such odd way ?
 
 Not sure. Sup generates the default name value using:
 
   require 'etc'
   Etc.getpwnam(ENV["USER"]).gecos.split(/,/).first
 
 What does that evaluate to on your system?
 
 Now that actually could evaluate to nil for a variety of reasons, so
 I've fixed the logic to avoid the crash if so. But I'm not sure why it
 would be nil due to iconv not being installed.
 
 -- 
 William <wmorgan-sup at masanjin.net>
 
 From nicolas.pouillard@gmail.com  Mon Mar 10 04:11:17 2008
 From: nicolas.pouillard@gmail.com (Nicolas Pouillard)
 Date: Mon, 10 Mar 2008 09:11:17 +0100
 Subject: [sup-talk] [PATCH] Sort contacts when saving them.
 Message-ID: <12051366771380-git-send-email-nicolas.pouillard@gmail.com>
 
 This makes sup behave more nicely with versionning.
 ---
  lib/sup/contact.rb |    2 +-
  1 files changed, 1 insertions(+), 1 deletions(-)
 
 diff --git a/lib/sup/contact.rb b/lib/sup/contact.rb
 index 8050c06..b0c272e 100644
 --- a/lib/sup/contact.rb
 +++ b/lib/sup/contact.rb
 @@ -51,7 +51,7 @@ class ContactManager
  
    def save
      File.open(@fn, "w") do |f|
 -      @p2a.each do |p, a|
 +      @p2a.sort_by { |(p, a)| [p.full_address, a] }.each do |(p, a)|
          f.puts "#{a || ''}: #{p.full_address}"
        end
      end
 -- 
 1.5.3.1.109.gacd69
 
 
 From DMack@Tickets.com  Mon Mar 10 10:01:14 2008
 From: DMack@Tickets.com (Mack, Daemian)
 Date: Mon, 10 Mar 2008 10:01:14 -0400
 Subject: [sup-talk] startup crash
 In-Reply-To: <20080310134823.GA24401@lenin>
 Message-ID: <61B99676296C5044A1A088C0D0EA1BA801B3C342@syr2k3mail>
 
 > > But I'm still curious, why this dependency problem is showing itself
 
 > > in a such odd way ?
 > 
 > Not sure. Sup generates the default name value using:
 > 
 >   require 'etc'
 >   Etc.getpwnam(ENV["USER"]).gecos.split(/,/).first
 > 
 > What does that evaluate to on your system?
 > 
 > Now that actually could evaluate to nil for a variety of 
 > reasons, so I've fixed the logic to avoid the crash if so. 
 > But I'm not sure why it would be nil due to iconv not being installed.
 
 I was getting the 'no name for account' message recently too, when
 setting up a test sup instance to play with the indexing, and wondered
 if it might be related to gecos info. 
 
 I've tried the test snippet above on three boxes (Debian Etch, Debian
 Sarge, Ubuntu 7.10), and it returns the current username when run under
 an existing user account. If I add a test user, and run the snippet as
 that user, it returns nil. This is true for all three boxes. I
 eventually got around it by hardcoding a dummy account name in
 account.rb, as I haven't had time to Do The Right Thing and figure out
 why ENV["USER"] isn't getting set for new accounts.
 
 Thanks for sup, incidentally! I'm experimenting with using it to manage
 a huge influx of business emails that Outlook makes somewhat annoying.
 
 
 
 DJFM
 --
 Daemian Mack
 Configuration Management
 Tickets.com
 344 W Genesee St
 Syracuse NY, 13202
 315.479.6663 x476
 AIM: daemianmack
  
 
 From wmorgan-sup@masanjin.net  Mon Mar 10 13:12:45 2008
 From: wmorgan-sup@masanjin.net (William Morgan)
 Date: Mon, 10 Mar 2008 10:12:45 -0700
 Subject: [sup-talk] startup crash
 In-Reply-To: <61B99676296C5044A1A088C0D0EA1BA801B3C342@syr2k3mail>
 References: <61B99676296C5044A1A088C0D0EA1BA801B3C342@syr2k3mail>
 Message-ID: <1205168755-sup-7027@south>
 
 Excerpts from Mack, Daemian's message of Mon Mar 10 07:01:14 -0700 2008:
 > This is true for all three boxes. I eventually got around it by
 > hardcoding a dummy account name in account.rb, as I haven't had time
 > to Do The Right Thing and figure out why ENV["USER"] isn't getting set
 > for new accounts.
 
 Is it really ENV["USER"], or is it the gecos entry being empty?
 
 At any rate, I think this should also be fixed in the newest git.
 
 -- 
 William <wmorgan-sup at masanjin.net>
 
 From alexandra@walford.id.au  Tue Mar 11 00:45:43 2008
 From: alexandra@walford.id.au (Alexandra Walford)
 Date: Tue, 11 Mar 2008 14:45:43 +1000
 Subject: [sup-talk] [Crash] "thread_by_subject: true" leads to "wrong id
 	called on nil"
 Message-ID: <1205210652-sup-1364@amethyst>
 
 Hi all,
 
 I am getting frequent sup crashes with "wrong id called on nil" (using the
 latest git master), when thread_by_subject is true.
 
 The crash generally happens when the thread index is being built as a result of
 a search, or when opening the inbox (which is really just another sort of
 search).
 
 I have managed to narrow down one particular crash to two messages sent to the
 sup-talk list.  If I set up a new maildir containing only those two messages
 and a new sup config, accepting all sup-config defaults and changing
 thread_by_subject to true, then I can reproduce the crash reliably.  The second
 message References: the first message, although the subjects are dissimilar.
 
 I have attached the two messages, a transcript of my test case session
 (including backtrace), and sup-exception-log.txt.  I am not yet familiar enough
 with sup or ruby to offer a decent patch; my current fix just hides any thread
 which will cause a crash, on line 208 of lib/sup/modes/thread-index-mode.rb in
 git master, which is a hopeless workaround really.
 
 Thanks,
 
 Alexandra
 
 -- 
 alexandra at walford.id.au
 -------------- next part --------------
 An embedded and charset-unspecified text was scrubbed...
 Name: message1.txt
 Url: http://rubyforge.org/pipermail/sup-talk/attachments/20080311/3d1a1cb1/attachment.txt 
 -------------- next part --------------
 An embedded and charset-unspecified text was scrubbed...
 Name: message2.txt
 Url: http://rubyforge.org/pipermail/sup-talk/attachments/20080311/3d1a1cb1/attachment-0001.txt 
 -------------- next part --------------
 An embedded and charset-unspecified text was scrubbed...
 Name: sup-exception-log.txt
 Url: http://rubyforge.org/pipermail/sup-talk/attachments/20080311/3d1a1cb1/attachment-0002.txt 
 -------------- next part --------------
 An embedded and charset-unspecified text was scrubbed...
 Name: sup-test-case.txt
 Url: http://rubyforge.org/pipermail/sup-talk/attachments/20080311/3d1a1cb1/attachment-0003.txt 
 
 From chrisw@rice.edu  Tue Mar 11 03:06:08 2008
 From: chrisw@rice.edu (Christopher Warrington)
 Date: Tue, 11 Mar 2008 02:06:08 -0500
 Subject: [sup-talk] [Crash] "thread_by_subject: true" leads to "wrong id
 	called on nil"
 In-Reply-To: <1205210652-sup-1364@amethyst>
 References: <1205210652-sup-1364@amethyst>
 Message-ID: <521619824.20080311020608@rice.edu>
 
 
 Alexandra Walford @ 2008-3-10 11:45:43 PM
 "[sup-talk] [Crash] "thread_by_subject: true" leads to "wrong id  called on nil"" <mid:1205210652-sup-1364 at amethyst>
 
 > I have managed to narrow down one particular crash to two messages
 > sent to the sup-talk list. If I set up a new maildir containing only
 > those two messages and a new sup config, accepting all sup-config
 > defaults and changing thread_by_subject to true, then I can
 > reproduce the crash reliably. The second message References: the
 > first message, although the subjects are dissimilar.
 
 Sorry! :-)
 
 The first message was created with sup. The second with git's
 send-email command. I've not looked at them closely, but the messages
 may be malformed.
 
 -- 
 Christopher Warrington <chrisw at rice.edu>
 
 "The next best thing to being clever is being able to quote someone
 who is."
 -Mary Poole
 -------------- next part --------------
 A non-text attachment was scrubbed...
 Name: not available
 Type: application/pgp-signature
 Size: 183 bytes
 Desc: not available
 Url : http://rubyforge.org/pipermail/sup-talk/attachments/20080311/3ee207f6/attachment.bin 
 
 From sup-talk@gurski.org  Tue Mar 11 20:51:46 2008
 From: sup-talk@gurski.org (Michael Gurski)
 Date: Tue, 11 Mar 2008 20:51:46 -0400
 Subject: [sup-talk] [Crash] "thread_by_subject: true" leads to "wrong id
 	called on nil"
 In-Reply-To: <521619824.20080311020608@rice.edu>
 References: <1205210652-sup-1364@amethyst> <521619824.20080311020608@rice.edu>
 Message-ID: <1205282947-sup-6038@ln>
 
 Excerpts from Christopher Warrington's message of Tue Mar 11 03:06:08 -0400 2008:
 > 
 > Alexandra Walford @ 2008-3-10 11:45:43 PM
 > "[sup-talk] [Crash] "thread_by_subject: true" leads to "wrong id  called on
 > nil"" <mid:1205210652-sup-1364 at amethyst>
 > 
 > > I have managed to narrow down one particular crash to two messages
 > > sent to the sup-talk list. If I set up a new maildir containing only
 > > those two messages and a new sup config, accepting all sup-config
 > > defaults and changing thread_by_subject to true, then I can
 > > reproduce the crash reliably. The second message References: the
 > > first message, although the subjects are dissimilar.
 > 
 > Sorry! :-)
 > 
 > The first message was created with sup. The second with git's
 > send-email command. I've not looked at them closely, but the messages
 > may be malformed.
 > 
 
 Interestingly, the crashes I was seeing with the same error seem to
 have disappeared now that I've disabled thread_by_subject as well.
 Once I get past the huge amount of backlog in mail that I have (who'd
 have thought getting married would eat into my copious free time?),
 I'll rerun sup and try to get the debugging output from the patch that
 I applied to create a sendme.txt when this happens.
 
 Mike
 
 
 -- 
 Michael A. Gurski (opt. [first].)[last]@pobox.com  http://www.pobox.com/~[last]
 1024R/39B5BADD PGP: 34 93 A9 94 B1 59 48 B7  17 57 1E 4E 62 56 45 70
 1024D/1166213E GPG: 628F 37A4 62AF 1475 45DB  AD81 ADC9 E606 1166 213E
 4096R/C0B4F04B GPG: 5B3E 75D7 43CF CF34 4042  7788 1DCE B5EE C0B4 F04B
 Views expressed by the host do not reflect the staff, management or sponsors.
 
 
 From marcus-sup@bar-coded.net  Thu Mar 13 06:38:57 2008
 From: marcus-sup@bar-coded.net (Marcus Williams)
 Date: Thu, 13 Mar 2008 10:38:57 +0000
 Subject: [sup-talk] Switching from IMAP to offlineimap and maintaining status
 Message-ID: <1205404558-sup-1495@tomsk>
 
 Hi -
 
 Having switched to offlineimap at work using sup I want to make the
 switch for my out-of-work accounts. Finally had enough of IMAP speed.
 Is there any way to maintain my sup status (starred/spam/etc) when I
 make the switch? I know I can dump the status of the db, but if I
 switch sources is this dump going to be of any use?
 
 Thanks
 
 Marcus
 
 From johnbent@lanl.gov  Thu Mar 13 12:06:30 2008
 From: johnbent@lanl.gov (John Bent)
 Date: Thu, 13 Mar 2008 10:06:30 -0600
 Subject: [sup-talk] forwarded messages
 Message-ID: <1205424317-sup-1180@tangerine.lanl.gov>
 
 Whenever I forward a message, it appears as a new thread.  I always then tag
 it and the original and merge them.  Is there a way to make this happen
 automagically?  Should that rather be the default behavior?
 
 Thanks,
 
 John
 
 From mml@pobox.com  Fri Mar 14 11:01:41 2008
 From: mml@pobox.com (Matt Liggett)
 Date: Fri, 14 Mar 2008 11:01:41 -0400
 Subject: [sup-talk] viewing threads oldest-first
 Message-ID: <1205506459-sup-9596@chain.local>
 
 Forgive me; I'm just getting started with sup and I'm trying to get
 the hang of it, but I have not looked at the source yet, beyond a
 cursory glance.
 
 In every mailer I've used before my gmail days, I always viewed my
 email oldest first, especially in any inboxen.  Is there a config
 option (I don't see a keyboard shortcut) to do this?
 
 As I write this, I realize that oldest-first might mean different
 things in a threaded view.  Is that the oldest first message in the
 thread, or the oldest last message?  Do I generally want inbox threads
 sorted by unread messages (that is, ordering by oldest unread message
 in each thread)?  I'm not sure I know the answer to either question,
 but I'm sure I'd like to be able to try all the choices.
 
 I will try to start digging in to the code in the next few days,
 because I am interested in contributing and making this better, not
 least of all for myself.  In the meantime, can anyone give me a
 pointer as to where in the code I might start digging regarding such a
 feature?
 
 FWIW, I was a long time mutt user but switched to gmail for search and
 tagging.  I missed a lot of things, and I generally felt less able to
 deal with large numbers of email conversations on a daily basis, but I
 lived with it because "just archive and search" (and to a lesser
 extent, tagging) was killer.
 
 I tried mutt+mairix for awhile, but it was kludgy and slow.  sup looks
 like it might be the best of both worlds, enabling me to deal with a
 large volume of mail and still giving me the tagging and search that I
 love from gmail.  Thanks, William!
 
 -- 
 Matt Liggett <mml at pobox.com>                          http://mml.name/
 
 From wmorgan-sup@masanjin.net  Fri Mar 14 14:43:21 2008
 From: wmorgan-sup@masanjin.net (William Morgan)
 Date: Fri, 14 Mar 2008 11:43:21 -0700
 Subject: [sup-talk] Switching from IMAP to offlineimap and maintaining
 	status
 In-Reply-To: <1205404558-sup-1495@tomsk>
 References: <1205404558-sup-1495@tomsk>
 Message-ID: <1205520043-sup-2321@south>
 
 Reformatted excerpts from Marcus Williams's message of 2008-03-13:
 > Is there any way to maintain my sup status (starred/spam/etc) when I
 > make the switch? I know I can dump the status of the db, but if I
 > switch sources is this dump going to be of any use?
 
 The dump should do that for you, since it's keyed directly off of
 message id.
 
 There might be a way to do this without needing to rescan the messages,
 but it'll only be useful if the messages have been reindexed since the
 speedy-index stuff has been in. LMK if that's the case.
 
 -- 
 William <wmorgan-sup at masanjin.net>
 
 From wmorgan-sup@masanjin.net  Fri Mar 14 14:51:31 2008
 From: wmorgan-sup@masanjin.net (William Morgan)
 Date: Fri, 14 Mar 2008 11:51:31 -0700
 Subject: [sup-talk] [Crash] "thread_by_subject: true" leads to "wrong id
 	called on nil"
 In-Reply-To: <1205210652-sup-1364@amethyst>
 References: <1205210652-sup-1364@amethyst>
 Message-ID: <1205520291-sup-7442@south>
 
 Reformatted excerpts from Alexandra Walford's message of 2008-03-10:
 > I am getting frequent sup crashes with "wrong id called on nil" (using
 > the latest git master), when thread_by_subject is true.
 
 The thread_by_subject stuff is broken and I have a patch to fix it, I
 just haven't merged it in yet. I'll try and do this soon. In the mean
 time, as you've discovered, turning off thread_by_subj is the
 workaround.
 
 -- 
 William <wmorgan-sup at masanjin.net>
 
 From wmorgan-sup@masanjin.net  Fri Mar 14 14:30:58 2008
 From: wmorgan-sup@masanjin.net (William Morgan)
 Date: Fri, 14 Mar 2008 11:30:58 -0700
 Subject: [sup-talk] viewing threads oldest-first
 In-Reply-To: <1205506459-sup-9596@chain.local>
 References: <1205506459-sup-9596@chain.local>
 Message-ID: <1205517246-sup-9385@south>
 
 Reformatted excerpts from Matt Liggett's message of 2008-03-14:
 > In every mailer I've used before my gmail days, I always viewed my
 > email oldest first, especially in any inboxen.  Is there a config
 > option (I don't see a keyboard shortcut) to do this?
 
 There's no option for this. I don't think it makes quite as much sense
 in a world where mailboxes can be arbitrarily large (a Sup mailbox is a
 search result across your entire set of email) and where loading is
 fairly slow.
 
 But there's no technical reason why this couldn't happen, and other
 people have asked for it before, so I'd be willing to add it.
 
 > As I write this, I realize that oldest-first might mean different
 > things in a threaded view.  Is that the oldest first message in the
 > thread, or the oldest last message?
 
 Currently a thread's timestamp is the timestamp of the most recent
 message in it, and I would be loathe to change that.
 
 > Do I generally want inbox threads sorted by unread messages (that is,
 > ordering by oldest unread message in each thread)? 
 
 That would be interesting, but unfortunately is not feasible in Sup. The
 index currently knows nothing about threads, just individual messages,
 so we can only sort by properties of messages, not by properties of
 threads.
 
 The way we build a threadset now is we search for messages which meet
 some search criterea, then we build a thread for each message.
 
 > In the meantime, can anyone give me a pointer as to where in the code
 > I might start digging regarding such a feature?
 
 Normally I would want a hook, but since there are really only two
 options that are technically feasible now, I would add a config variable
 called "sort_oldest_first" or something like that.  It will follow a
 very similar codepath to thread_by_subj (so grep for that): it should be
 initialized in sup.rb for new configurations, passed into a ThreadSet
 constructor from ThreadIndexMode#initialize_threads, and passed into
 Index#each_id_by_date, where it will change the query we send to Ferret.
 Finally, ThreadIndexMode#update (which handles the UI component of
 sorting) has got to respect this variable.
 
 > I tried mutt+mairix for awhile, but it was kludgy and slow.  sup looks
 > like it might be the best of both worlds, enabling me to deal with a
 > large volume of mail and still giving me the tagging and search that I
 > love from gmail.
 
 Yep, you're my target audience. :)
 
 -- 
 William <wmorgan-sup at masanjin.net>
 
 From wmorgan-sup@masanjin.net  Fri Mar 14 14:40:32 2008
 From: wmorgan-sup@masanjin.net (William Morgan)
 Date: Fri, 14 Mar 2008 11:40:32 -0700
 Subject: [sup-talk] forwarded messages
 In-Reply-To: <1205424317-sup-1180@tangerine.lanl.gov>
 References: <1205424317-sup-1180@tangerine.lanl.gov>
 Message-ID: <1205519620-sup-1897@south>
 
 Reformatted excerpts from John Bent's message of 2008-03-13:
 > Whenever I forward a message, it appears as a new thread.  I always
 > then tag it and the original and merge them.  Is there a way to make
 > this happen automagically?  Should that rather be the default
 > behavior?
 
 I've thought about having this be the default behavior because I often
 do that as well. There are two ways it could go: actually add a
 References: header (or a in-reply-to?), which would be pretty
 nonstandard, I think, or just do it in the index itself.
 
 The dirty secret of the thread joining (which I've just realized now) is
 that the joins won't be preserved if you reindex those messages. I need
 to add another blob of information somewhere to preserve that stuff.  If
 I added the header to the outgoing message, the threading would be
 preserved, of course...
 
 -- 
 William <wmorgan-sup at masanjin.net>
 
 From alexandra@walford.id.au  Fri Mar 14 18:28:39 2008
 From: alexandra@walford.id.au (Alexandra Walford)
 Date: Sat, 15 Mar 2008 08:28:39 +1000
 Subject: [sup-talk] [Crash] "thread_by_subject: true" leads to "wrong id
 	called on nil"
 In-Reply-To: <1205520291-sup-7442@south>
 References: <1205210652-sup-1364@amethyst> <1205520291-sup-7442@south>
 Message-ID: <1205533710-sup-2188@amethyst>
 
 Excerpts from William Morgan's message of Sat Mar 15 04:51:31 +1000 2008:
 > The thread_by_subject stuff is broken and I have a patch to fix it, I
 > just haven't merged it in yet. I'll try and do this soon. In the mean
 > time, as you've discovered, turning off thread_by_subj is the
 > workaround.
 
 Sounds good!  Thanks :)
 
 -- 
 alexandra at walford.id.au
 
 From wmorgan-sup@masanjin.net  Mon Mar 17 12:37:41 2008
 From: wmorgan-sup@masanjin.net (William Morgan)
 Date: Mon, 17 Mar 2008 09:37:41 -0700
 Subject: [sup-talk] Crash on startup, "wrong id called on nil"
 In-Reply-To: <1203172019-sup-1133@south>
 References: <20080207172725.GA9157@gurski.org> <1203172019-sup-1133@south>
 Message-ID: <1205771845-sup-9353@south>
 
 Reformatted excerpts from William Morgan's message of 2008-02-16:
 > Reformatted excerpts from Michael Gurski's message of 2008-02-07:
 > > It's been a while since I launched sup, and updating today from
 > > master, I get a crash on startup:
 > > 
 > > --- RuntimeError from thread: load threads for thread-index-mode wrong
 > > id called on nil
 > 
 > Hm, this is a bit tricky. It looks like on of the thread invariants I
 > maintain is actually a variant.
 
 I think this should be fixed at this point.
 
 -- 
 William <wmorgan-sup at masanjin.net>
 
 From wmorgan-sup@masanjin.net  Mon Mar 17 12:41:55 2008
 From: wmorgan-sup@masanjin.net (William Morgan)
 Date: Mon, 17 Mar 2008 09:41:55 -0700
 Subject: [sup-talk] [Crash] "thread_by_subject: true" leads to "wrong id
 	called on nil"
 In-Reply-To: <1205520291-sup-7442@south>
 References: <1205210652-sup-1364@amethyst> <1205520291-sup-7442@south>
 Message-ID: <1205771997-sup-226@south>
 
 Reformatted excerpts from William Morgan's message of 2008-03-14:
 > The thread_by_subject stuff is broken and I have a patch to fix it, I
 > just haven't merged it in yet. I'll try and do this soon.
 
 Ok, I've merged the changes into next. Enable thread-by-subject and see
 what happens.
 
 If you're still seeing messages with "Re: " subjects in separate
 threads, try reindexing those messages.
 
 -- 
 William <wmorgan-sup at masanjin.net>
 
 From kendall@clarkparsia.com  Mon Mar 17 12:50:38 2008
 From: kendall@clarkparsia.com (Kendall Grant Clark)
 Date: Mon, 17 Mar 2008 12:50:38 -0400
 Subject: [sup-talk] Kill thread doesn't seem to work
 Message-ID: <1205772559-sup-6593@k-desktop.int.clarkparsia.com>
 
 Wm., et. al.,
 
 I'm using v0.4 on Fedora and when I &-mark a thread to kill it, I'm
 expecting that new replies to that thread will *not* come back to my
 inbox view. Except, that they do come back, and I'm puzzled as to
 whether this is a bug, an artifact of my setup, etc. I looked through
 sup-talk archives, but didn't see anything about this issue.
 
 I have a single source (imaps://...) being fed to sup; I sup-sync
 every morning from a cron job, and, otherwise, everything "just
 works".
 
 Any pointers or requests for further info would be great. And, FWIW,
 other than this issue, I love-love-love sup! Thx!!
 
 -- 
 Cheers,
 Kendall
 
 From alexandra@walford.id.au  Tue Mar 18 08:35:24 2008
 From: alexandra@walford.id.au (Alexandra Walford)
 Date: Tue, 18 Mar 2008 22:35:24 +1000
 Subject: [sup-talk] [Crash] "thread_by_subject: true" leads to "wrong id
 	called on nil"
 In-Reply-To: <1205771997-sup-226@south>
 References: <1205210652-sup-1364@amethyst> <1205520291-sup-7442@south>
 	<1205771997-sup-226@south>
 Message-ID: <1205843716-sup-9243@amethyst>
 
 Excerpts from William Morgan's message of Tue Mar 18 02:41:55 +1000 2008:
 > Ok, I've merged the changes into next. Enable thread-by-subject and see
 > what happens.
 
 Nice, I'll give it a go :)
 
 -- 
 alexandra at walford.id.au
 
 From wmorgan-sup@masanjin.net  Tue Mar 18 22:16:02 2008
 From: wmorgan-sup@masanjin.net (William Morgan)
 Date: Tue, 18 Mar 2008 19:16:02 -0700
 Subject: [sup-talk] Kill thread doesn't seem to work
 In-Reply-To: <1205772559-sup-6593@k-desktop.int.clarkparsia.com>
 References: <1205772559-sup-6593@k-desktop.int.clarkparsia.com>
 Message-ID: <1205892920-sup-9762@south>
 
 Reformatted excerpts from Kendall Grant Clark's message of 2008-03-17:
 > I'm using v0.4 on Fedora and when I &-mark a thread to kill it, I'm
 > expecting that new replies to that thread will *not* come back to my
 > inbox view. Except, that they do come back, and I'm puzzled as to
 > whether this is a bug, an artifact of my setup, etc. I looked through
 > sup-talk archives, but didn't see anything about this issue.
 
 This should work. Can you try against git next? There have been a couple
 commits that probably affected this.
 
 -- 
 William <wmorgan-sup at masanjin.net>
 
 From kendall@clarkparsia.com  Wed Mar 19 09:59:14 2008
 From: kendall@clarkparsia.com (Kendall Grant Clark)
 Date: Wed, 19 Mar 2008 09:59:14 -0400
 Subject: [sup-talk] Kill thread doesn't seem to work
 In-Reply-To: <1205892920-sup-9762@south>
 References: <1205772559-sup-6593@k-desktop.int.clarkparsia.com>
 	<1205892920-sup-9762@south>
 Message-ID: <1205935094-sup-6162@k-desktop.int.clarkparsia.com>
 
 Excerpts from William Morgan's message of Tue Mar 18 22:16:02 -0400 2008:
 > Reformatted excerpts from Kendall Grant Clark's message of 2008-03-17:
 > > I'm using v0.4 on Fedora and when I &-mark a thread to kill it, I'm
 > > expecting that new replies to that thread will *not* come back to my
 > > inbox view. Except, that they do come back, and I'm puzzled as to
 > > whether this is a bug, an artifact of my setup, etc. I looked through
 > > sup-talk archives, but didn't see anything about this issue.
 > 
 > This should work. Can you try against git next? There have been a couple
 > commits that probably affected this.
 
 I'm a Python programmer, only marginally a Ruby src code reader... Is
 there some kind of sandbox I can use to test the git next version,
 w/out borking my existing (and working, except for this one issue) sup
 install?
 
 -- 
 Cheers,
 Kendall
 
 From johnbent@lanl.gov  Wed Mar 19 10:51:29 2008
 From: johnbent@lanl.gov (John Bent)
 Date: Wed, 19 Mar 2008 08:51:29 -0600
 Subject: [sup-talk] Kill thread doesn't seem to work
 In-Reply-To: <1205935094-sup-6162@k-desktop.int.clarkparsia.com>
 References: <1205772559-sup-6593@k-desktop.int.clarkparsia.com>
 	<1205892920-sup-9762@south>
 	<1205935094-sup-6162@k-desktop.int.clarkparsia.com>
 Message-ID: <1205937908-sup-4264@tangerine.lanl.gov>
 
 Excerpts from Kendall Grant Clark's message of Wed Mar 19 07:59:14 -0600 2008:
 > I'm a Python programmer, only marginally a Ruby src code reader... Is
 > there some kind of sandbox I can use to test the git next version,
 > w/out borking my existing (and working, except for this one issue) sup
 > install?
 > 
 So this is from: http://sup.rubyforge.org/wiki/wiki.pl?Contributing
 I set this up previously, so I'm not positive about the initial setup
 syntax but the rest of it is definitely correct:
 
 # initial setup
 git clone git://repo.or.cz/sup.git 
 git branch --track next origin/next
 git checkout next
 
 # getting current, building, running local copy
 git pull
 rake gem
 ruby -I lib -w bin/sup
 
 This will then use a local next version w/out affecting your existing sup
 install.  If you then decide you want to fully install the next version:
 sudo gem install pkg/sup-999.gem
 
 John
 
 From johnbent@lanl.gov  Wed Mar 19 11:03:39 2008
 From: johnbent@lanl.gov (John Bent)
 Date: Wed, 19 Mar 2008 09:03:39 -0600
 Subject: [sup-talk] update contact, compose new mail => crash
 Message-ID: <1205938994-sup-2758@tangerine.lanl.gov>
 
 If I am reading an email, hit 'i' to capture a contact, hit 'x' to
 get back to my inbox, hit 'c' to compose a new email, sup (next) crashes.
 Exception log contents:
 
 --- NoMethodError from thread: main
 undefined method `<=>' for nil:NilClass
 /opt/local/lib/ruby/gems/1.8/gems/sup-999/lib/sup/buffer.rb:494:in `sort'
 /opt/local/lib/ruby/gems/1.8/gems/sup-999/lib/sup/buffer.rb:494:in `ask_for_contacts'
 /opt/local/lib/ruby/gems/1.8/gems/sup-999/lib/sup/util.rb:497:in `send'
 /opt/local/lib/ruby/gems/1.8/gems/sup-999/lib/sup/util.rb:497:in `method_missing'
 /opt/local/lib/ruby/gems/1.8/gems/sup-999/lib/sup/modes/compose-mode.rb:23:in `spawn_nicely'
 /opt/local/lib/ruby/gems/1.8/gems/sup-999/bin/sup:274
 /opt/local/bin/sup:16:in `load'
 /opt/local/bin/sup:16
 
 From kendall@clarkparsia.com  Wed Mar 19 11:17:07 2008
 From: kendall@clarkparsia.com (Kendall Grant Clark)
 Date: Wed, 19 Mar 2008 11:17:07 -0400
 Subject: [sup-talk] Kill thread doesn't seem to work
 In-Reply-To: <1205937908-sup-4264@tangerine.lanl.gov>
 References: <1205772559-sup-6593@k-desktop.int.clarkparsia.com>
 	<1205892920-sup-9762@south>
 	<1205935094-sup-6162@k-desktop.int.clarkparsia.com>
 	<1205937908-sup-4264@tangerine.lanl.gov>
 Message-ID: <1205939782-sup-1932@k-desktop.int.clarkparsia.com>
 
 Excerpts from John Bent's message of Wed Mar 19 10:51:29 -0400 2008:
 
 > So this is from: http://sup.rubyforge.org/wiki/wiki.pl?Contributing
 
 Doh. Sorry, I should have rechecked the wiki for this.
 
 > I set this up previously, so I'm not positive about the initial setup
 > syntax but the rest of it is definitely correct:
 
 It worked for me; thx! Testing now. :>
 
 -- 
 Cheers,
 Kendall
 
 From kendall@clarkparsia.com  Wed Mar 19 11:20:37 2008
 From: kendall@clarkparsia.com (Kendall Grant Clark)
 Date: Wed, 19 Mar 2008 11:20:37 -0400
 Subject: [sup-talk] Kill thread doesn't seem to work
 In-Reply-To: <1205892920-sup-9762@south>
 References: <1205772559-sup-6593@k-desktop.int.clarkparsia.com>
 	<1205892920-sup-9762@south>
 Message-ID: <1205939956-sup-9857@k-desktop.int.clarkparsia.com>
 
 Excerpts from William Morgan's message of Tue Mar 18 22:16:02 -0400 2008:
 > This should work. Can you try against git next? There have been a couple
 > commits that probably affected this.
 
 Doesn't work for me w/ git next either. I asked a coworker to send me
 a message, I replied. Then I killed the thread. He replied to my
 reply, and it popped right up in my INBOX with the +killed tag on it.
 
 -- 
 Cheers,
 Kendall
 
 From patroclo7@gmail.com  Wed Mar 19 11:24:10 2008
 From: patroclo7@gmail.com (Giorgio Lando)
 Date: Wed, 19 Mar 2008 16:24:10 +0100
 Subject: [sup-talk] Kill thread doesn't seem to work
 In-Reply-To: <1205939956-sup-9857@k-desktop.int.clarkparsia.com>
 References: <1205772559-sup-6593@k-desktop.int.clarkparsia.com>
 	<1205892920-sup-9762@south>
 	<1205939956-sup-9857@k-desktop.int.clarkparsia.com>
 Message-ID: <1205940116-sup-6578@clarabella.clarabella>
 
 Excerpts from Kendall Grant Clark's message of Wed Mar 19 16:20:37 +0100 2008:
 > Doesn't work for me w/ git next either. I asked a coworker to send me
 > a message, I replied. Then I killed the thread. He replied to my
 > reply, and it popped right up in my INBOX with the +killed tag on it.
 
 I confirm, at some point killed threads stopped
 working for me (they are back in the INBOX when a new message is
 received). Both git next and git master have this problem for me.
 
 -- 
 Giorgio Lando <patroclo7 at gmail dot com>
 
 From wmorgan-sup@masanjin.net  Wed Mar 19 12:20:50 2008
 From: wmorgan-sup@masanjin.net (William Morgan)
 Date: Wed, 19 Mar 2008 09:20:50 -0700
 Subject: [sup-talk] Kill thread doesn't seem to work
 In-Reply-To: <1205935094-sup-6162@k-desktop.int.clarkparsia.com>
 References: <1205772559-sup-6593@k-desktop.int.clarkparsia.com>
 	<1205892920-sup-9762@south>
 	<1205935094-sup-6162@k-desktop.int.clarkparsia.com>
 Message-ID: <1205943578-sup-2011@south>
 
 Reformatted excerpts from Kendall Grant Clark's message of 2008-03-19:
 > I'm a Python programmer, only marginally a Ruby src code reader... Is
 > there some kind of sandbox I can use to test the git next version,
 > w/out borking my existing (and working, except for this one issue) sup
 > install?
 
 Once you have git next in a directory somewhere (top part of
 http://sup.rubyforge.org/wiki/wiki.pl?Contributing), the HACKING file
 has instructions on how to run it locally. Please let me know if
 anything is unclear in either...
 
 -- 
 William <wmorgan-sup at masanjin.net>
 
 From kendall@clarkparsia.com  Wed Mar 19 12:45:25 2008
 From: kendall@clarkparsia.com (Kendall Grant Clark)
 Date: Wed, 19 Mar 2008 12:45:25 -0400
 Subject: [sup-talk] Kill thread doesn't seem to work
 In-Reply-To: <1205943578-sup-2011@south>
 References: <1205772559-sup-6593@k-desktop.int.clarkparsia.com>
 	<1205892920-sup-9762@south>
 	<1205935094-sup-6162@k-desktop.int.clarkparsia.com>
 	<1205943578-sup-2011@south>
 Message-ID: <1205945088-sup-5986@k-desktop.int.clarkparsia.com>
 
 Excerpts from William Morgan's message of Wed Mar 19 12:20:50 -0400 2008:
 
 > Once you have git next in a directory somewhere (top part of
 > http://sup.rubyforge.org/wiki/wiki.pl?Contributing), the HACKING file
 > has instructions on how to run it locally. Please let me know if
 > anything is unclear in either...
 
 Yes, I ran git next version and it still doesn't handle +killed tagged
 threads correctly; that is, they show up in searches, which I expect,
 but new messages in the thread also cause the thread to reappear in
 INBOX. (FWIW, this test thread is tagged: +/INBOX +inbox +killed
 +sent.)
 
 
 -- 
 Cheers,
 Kendall
 
 From wmorgan-sup@masanjin.net  Wed Mar 19 15:23:02 2008
 From: wmorgan-sup@masanjin.net (William Morgan)
 Date: Wed, 19 Mar 2008 12:23:02 -0700
 Subject: [sup-talk] Kill thread doesn't seem to work
 In-Reply-To: <1205940116-sup-6578@clarabella.clarabella>
 References: <1205772559-sup-6593@k-desktop.int.clarkparsia.com>
 	<1205892920-sup-9762@south>
 	<1205939956-sup-9857@k-desktop.int.clarkparsia.com>
 	<1205940116-sup-6578@clarabella.clarabella>
 Message-ID: <1205954571-sup-5093@south>
 
 Reformatted excerpts from Giorgio Lando's message of 2008-03-19:
 > I confirm, at some point killed threads stopped working for me (they
 > are back in the INBOX when a new message is received). Both git next
 > and git master have this problem for me.
 
 Ok, I'll take a look.
 
 -- 
 William <wmorgan-sup at masanjin.net>
 
 From kendall@clarkparsia.com  Thu Mar 20 10:58:59 2008
 From: kendall@clarkparsia.com (Kendall Grant Clark)
 Date: Thu, 20 Mar 2008 10:58:59 -0400
 Subject: [sup-talk] File saving wart...
 Message-ID: <1206024540-sup-8048@k-desktop.int.clarkparsia.com>
 
 I lazily haven't looked in sup-talk archives to see if this has been
 reported, and I can't find a bug/ticket tracking thing to check
 there...
 
 When saving an attachment (or message, for that matter) the built-in
 filesystem navigator thingie has this annoying wart where it will
 expand "~/" when you tab, but if you just give "~/" in the path to
 save to, it doesn't do an expansion and complains that there's no such
 directory or file.
 
 This is a minor wart, not really a bug, so not a big deal, but it is a
 daily annoyance, since I always remember that "~/" works but I never
 seem to remember that it only works for tab expansion, but not for
 direct saving.
 
 -- 
 Cheers,
 Kendall
 
 From johnbent@lanl.gov  Thu Mar 20 15:05:03 2008
 From: johnbent@lanl.gov (John Bent)
 Date: Thu, 20 Mar 2008 13:05:03 -0600
 Subject: [sup-talk] manually merging threads not persistent in next
 Message-ID: <1206039801-sup-1759@tangerine.lanl.gov>
 
 I'm running next and I've got ":thread_by_subject: true" in my config.
 I've noticed that manually merging threads works in my current session but
 is not persistent across a quit and restart (i.e. they reappear as separate
 threads).  Has anyone else noticed this?
 
 Thanks,
 
 John
 
 From rodkoch@gmail.com  Thu Mar 20 14:57:55 2008
 From: rodkoch@gmail.com (rodneyk)
 Date: Thu, 20 Mar 2008 11:57:55 -0700 (PDT)
 Subject: [sup-talk] Support for non-black background in transparent
  terminals?
 Message-ID: <16186298.post@talk.nabble.com>
 
 
 Hi,
 I was wondering if it's possible to re-configure the black background to be
 another color,
 or even work with a transparent terminal? I checked the Colormap in bin/sup
 but there's nothing
 related to the background.
 
 Also, regarding the Colormap, if I wanted to change the inbox color for the
 'date' and 'from' fields,
 which item handles that? I'm not really sure what most of the items in the
 Colormap do.
 
 Thanks,
 --Rodney
 -- 
 View this message in context: http://www.nabble.com/Support-for-non-black-background-in-transparent-terminals--tp16186298p16186298.html
 Sent from the SUP Talk mailing list archive at Nabble.com.
 
 
 From johnbent@lanl.gov  Thu Mar 20 15:24:08 2008
 From: johnbent@lanl.gov (John Bent)
 Date: Thu, 20 Mar 2008 13:24:08 -0600
 Subject: [sup-talk] default domain name
 Message-ID: <1206040874-sup-2650@tangerine.lanl.gov>
 
 My apologies if this is not a sup question.  But I feel like this used to
 work in pine ... My putmail config has server = server1.lanl.gov which 
 works for sending mail.  However, if I just send to an address without
 a domain attached (e.g. to johnbent instead of to johnbent at lanl.gov), then 
 I get a bounce from server1.lanl.gov saying "there is no 
 johnbent at server1.lanl.gov."  Is there some way to config sup (or the rest of
 my email infrastructure) to teach it a particular default email address?
 
 Thanks,
 
 John
 
 From johnbent@lanl.gov  Thu Mar 20 18:37:42 2008
 From: johnbent@lanl.gov (John Bent)
 Date: Thu, 20 Mar 2008 16:37:42 -0600
 Subject: [sup-talk] can't remove draft label
 In-Reply-To: <1202013739-sup-550@south>
 References: <1201716067-sup-3586@tangerine.lanl.gov>
 	<1201754138-sup-7276@chris-tablet>
 	<1201759596-sup-4274@tangerine.lanl.gov> <1202013739-sup-550@south>
 Message-ID: <1206052546-sup-3667@tangerine.lanl.gov>
 
 Excerpts from William Morgan's message of Sat Feb 02 21:44:41 -0700 2008:
 > I'm not sure why that label would be there, but you can perform some
 > index surgery to remove it.
 > 
 > $ sh devel/console.sh 
 > irb(main):001:0>
 > r=Index.ferret.search("message_id:1201759596-sup-4274 at tangerine.lanl.gov")
 > => #<struct Ferret::Search::TopDocs total_hits=1, hits=[#<struct
 > Ferret::Search::Hit doc=92571, score=10.82630443573>],
 > max_score=10.82630443573, searcher=#<Ferret::Search::Searcher:0xb740dff0>>
 > irb(main):002:0> id=r.hits.first.doc
 > => 92571
 > irb(main):003:0> m = Index.build_message id
 > [...]
 > irb(main):004:0> m.remove_label :draft
 > => true
 > irb(main):005:0> m.save Index
 > => false
 > irb(main):006:0> exit
 > 
 Two months later and I finally did this... :)  It worked great.  The only
 slight problem was that somehow all 25 messages in that thread were falsely
 labeled so I had to repeat the above 25 times.  Although I did eventually
 realize that the console accepted semicolons so I combined them into a 
 single line and then just did arrow up 25 times so it was pretty fast.
 
 Thanks!
 
 John
 
 From marcus-sup@bar-coded.net  Sun Mar 23 17:13:58 2008
 From: marcus-sup@bar-coded.net (Marcus Williams)
 Date: Sun, 23 Mar 2008 21:13:58 +0000
 Subject: [sup-talk] [PATCH] First draft of attachment processing for
 	more gmail style searches
 In-Reply-To: <1205013484-sup-7653@south>
 References: <1203972458-sup-5906@tomsk> <1204220051-sup-129@south>
 	<1204232994-sup-628@tomsk> <1204479552-sup-4100@south>
 	<1204711082-sup-3541@tomsk> <1205013484-sup-7653@south>
 Message-ID: <1206306720-sup-2563@tomsk>
 
 On 8.3.2008, William Morgan wrote:
 > Sorry for the general delay in replying. I've been coding up a little
 > distributed issue tracker that I think will help me manage Sup a little
 > better... or at least ensure I don't forget people's suggestions.
 
 Mmmmm - I'm on the lookout for a decent issue tracker :) Any more info?
 
 Marcus
 
 From 5srmspw02@sneakemail.com  Wed Mar 26 01:57:05 2008
 From: 5srmspw02@sneakemail.com (Guarded Identity)
 Date: Wed, 26 Mar 2008 00:57:05 -0500
 Subject: [sup-talk] New User Questions
 In-Reply-To: <1199944172-sup-660@south>
 References: <29393-78803@sneakemail.com> <1199944172-sup-660@south>
 Message-ID: <13352-37236@sneakemail.com>
 
 Hi,
 
 I've been using Sup for a few months now, read up a little more on Ruby, and
 thought I might be in more of a position to respond to this thread.
 
 Excerpts from William Morgan's message of Thu Jan 10 00:34:03 -0600 2008:
 > Excerpts from Guarded Identity's message of Tue Jan 08 19:39:18 -0800 2008:
 
 > > Will migration of maildir mail from new/ to cur/ necessitate an execution
 > > of sup-sync?
 >
 > Hm. I don't think so. As long as the mtime and size of the file are
 > preserved (which I believe they are during a move) you should be ok.
 
 I was thinking about the algorithm for calculating the message's unique
 "source_info" for Maildir mail (mtime appended to filesize).  I know altering
 the backend might really be something you don't want to support, so maybe I'm
 voicing a moot concern.  It's just that sometimes I get annoyed at people
 sending me mail with large attachments (especially at work where I just can't
 control this kind of thing too well).  In the past, I'd use mutt to purge mail
 of their large attachments.
 
 However, I guess if I do this en mass, I loose my labeling of these messages
 when I do a sup-sync because the message's source_info changes, right?, but
 have you considered using a message's "Message-Id" header as part of the
 source_info?  I'm thinking you might have.  For the most part, this should be
 pretty unique, right?  I know the Message-Id is sometimes missing, but we can
 default to the old source_id if things are really wacky (which should be
 rarely, right?).  The only reason I could think you wouldn't use the Message-Id
 is because maybe it's less efficient to open up a message than look at it's
 file stats.  Or maybe there's another reason?
 
 > > I had done this before, actually, but I noticed that Sup indexing was
 > > slowed down quite a bit (rss2email and mailing lists generate a lot of new
 > > mail).
 >
 > If that's true, then it must be due to IMAP transmission costs. I'm
 > certainly willing to believe that pulling a message from IMAP is
 > significantly slower than a disk read.
 
 I was thinking about this too.  Is there anything we can do to pull down
 headers first to fill the index (backgrounding the indexing)?  Maybe that might
 make the delay with IMAP less of a nuisance?
 
 However, I won't harp on this too much.  Because I wasn't comfortable
 synchronizing (say with Unison) sup Ferret indexes across hosts, I'm ssh'ing
 into the machine with the index and using "maildir" sources.  But this
 unfortunately causes me a problem with viewing graphical attachments because X
 forwarding is really slow.  So for now I'm saving attachments and scp'ing them
 over.  I'm not sure there's much anyone can do about any of this.
 
 > > I would really like to have time-based auto-expiry (excluding starred or
 > > special-labeled items) for some of my mail (primarily mailing lists and rss
 > > feeds). With mboxes, I guess I could do some "deleted"-labeling with a Ruby
 > > script followed by a call to sup-sync-back.
 >
 > That sounds like the right approach.
 >
 > > 6. What are my scripting options for mail expiry with Maildirs?  The search
 > > to get the message objects is pretty straight forward.  How much further
 > > work is it to delete the message from the index and to get a filename of
 > > the message to delete from the maildir?
 >
 > Trivial, although the filename/IMAP id require minor API changes to expose
 > those functions. But with a working sup-sync-back that applies to all source
 > types, you won't have to do anything other than inject :deleted labels
 > everywhere and simply call sup-sync-back.
 
 Okay,
 
 So I've been using a helper script that does some things to help me manage my
 Maildir mail (all other source types are ignored).  My primary idea was to have
 a script to expire mail, but allowing me to save threads labeled "starred" or
 "save".  Eventually I added in a function to manipulate labels.  Unlike
 sup-tweak-labels, this script gathers it's pool of messages from the index
 using a Ferret query (Chronic allowed) rather than using source URIs.
 
 I got around some of the encapsulation inherent in the API by just brute force
 breaking encapsulation (which Ruby makes kind of ridiculously easy it seems).
 
 I've attached the script, and would like to see any feedback if people have
 some.  In particular, I'm eager to know of any concerns people have with the
 approach or defects people catch.  I'm still pretty new to Ruby, so if I did
 anything flat out stupid with the language, please let me know.
 
 By the way, I was thinking that there seems to be a good deal of healthy
 participation in this list.  Does anybody see any value to hanging out in #sup
 on Freenode?  Sometimes it's just nice to have an IRC forum.
 
 -Sukant
 -------------- next part --------------
 A non-text attachment was scrubbed...
 Name: sup-tweak-maildir
 Type: application/octet-stream
 Size: 6652 bytes
 Desc: not available
 Url : http://rubyforge.org/pipermail/sup-talk/attachments/20080326/cef1c568/attachment.obj 
 
 From wmorgan-sup@masanjin.net  Wed Mar 26 12:04:10 2008
 From: wmorgan-sup@masanjin.net (William Morgan)
 Date: Wed, 26 Mar 2008 09:04:10 -0700
 Subject: [sup-talk] [PATCH] fixed off-by-one error in imap.rb and
 	maildir.rb
 In-Reply-To: <1204531233-2092-1-git-send-email-chrisw@rice.edu>
 References: <1204528063-sup-7345@chris-tablet>
 	<1204531233-2092-1-git-send-email-chrisw@rice.edu>
 Message-ID: <1206547432-sup-4074@south>
 
 Reformatted excerpts from Christopher Warrington's message of 2008-03-03:
 > The end_offset reported by imap and maildir sources was incorrect if
 > there was only one message in the source. Since end_offset is
 > exclusive, we must add one to the last known id to get include all
 > valid message ids in the range.
 
 Precisely so. Applied. Thanks!
 
 -- 
 William <wmorgan-sup at masanjin.net>
 
 From wmorgan-sup@masanjin.net  Wed Mar 26 12:07:20 2008
 From: wmorgan-sup@masanjin.net (William Morgan)
 Date: Wed, 26 Mar 2008 09:07:20 -0700
 Subject: [sup-talk] [PATCH] updated text and keybinding in
 	label-search-results-mode
 In-Reply-To: <1204531280-4048-1-git-send-email-chrisw@rice.edu>
 References: <1204531280-4048-1-git-send-email-chrisw@rice.edu>
 Message-ID: <1206547632-sup-4441@south>
 
 Reformatted excerpts from Christopher Warrington's message of 2008-03-03:
 > The text in label-search-results-mode now matches that of search-results-mode.
 
 Applied, thanks!
 
 -- 
 William <wmorgan-sup at masanjin.net>
 
 From wmorgan-sup@masanjin.net  Wed Mar 26 12:11:49 2008
 From: wmorgan-sup@masanjin.net (William Morgan)
 Date: Wed, 26 Mar 2008 09:11:49 -0700
 Subject: [sup-talk] [PATCH] Fix for icon title in gnome terminal
 	(possibly others)
 In-Reply-To: <1204711708-sup-3638@tomsk>
 References: <1204711708-sup-3638@tomsk>
 Message-ID: <1206547886-sup-1106@south>
 
 Reformatted excerpts from Marcus Williams's message of 2008-03-05:
 > The escape code used for terminal titles works mostly but it doesnt
 > set the icon title as well [1], which is useful when minimised. This
 > patch makes the escape code set both window and icon title so the task
 > bar updates correctly in the gnome terminal (and probably others).
 
 Reworked slightly (to avoid duplicating the conditional) and applied.
 Thanks!
 
 -- 
 William <wmorgan-sup at masanjin.net>
 
 From wmorgan-sup@masanjin.net  Wed Mar 26 12:13:53 2008
 From: wmorgan-sup@masanjin.net (William Morgan)
 Date: Wed, 26 Mar 2008 09:13:53 -0700
 Subject: [sup-talk] [PATCH] Sort contacts when saving them.
 In-Reply-To: <12051366771380-git-send-email-nicolas.pouillard@gmail.com>
 References: <12051366771380-git-send-email-nicolas.pouillard@gmail.com>
 Message-ID: <1206548025-sup-37@south>
 
 Reformatted excerpts from nicolas.pouillard's message of 2008-03-10:
 > This makes sup behave more nicely with versionning.
 
 Applied. Thanks!
 
 -- 
 William <wmorgan-sup at masanjin.net>
 
 From wmorgan-sup@masanjin.net  Wed Mar 26 12:40:24 2008
 From: wmorgan-sup@masanjin.net (William Morgan)
 Date: Wed, 26 Mar 2008 09:40:24 -0700
 Subject: [sup-talk] Newline Issues
 In-Reply-To: <1204536487-sup-9413@chris-tablet>
 References: <1204536487-sup-9413@chris-tablet>
 Message-ID: <1206549226-sup-8056@south>
 
 Reformatted excerpts from Christopher Warrington's message of 2008-03-03:
 > When I use offlineimap under Cgywin, the messages are delivered with
 > CRLF. Sup cannot parse these: it thinks that the body is empty. I know
 > that sup is seeing the entire message. If I press H, I see all of the
 > message.
 
 I've finally had a chance to look into this. I've just patched Sup to
 handle carriage returns in mbox files, but things are still broken
 because of RubyMail.
 
 It looks like there's an outstanding RubyMail patch here:
 
 http://rubyforge.org/tracker/index.php?func=detail&aid=2821&group_id=446&atid=1756
 
 If you apply that and get the latest git, I think things should work.
 
 -- 
 William <wmorgan-sup at masanjin.net>
 
 From cjs@weisshuhn.de  Thu Mar 27 00:49:02 2008
 From: cjs@weisshuhn.de (Carsten =?iso-8859-1?Q?Sch=F6lzki?=)
 Date: Thu, 27 Mar 2008 05:49:02 +0100
 Subject: [sup-talk] ISO 8859-15 at the header
 Message-ID: <20080327044901.GA12106@weisshuhn.de>
 
 Hi,
 
 I started trying sup today and it seems to be wonderful for me. Till
 now I'm a mutt user for over 10 years - but know I found a reason to
 change...
 
 But there is a little problem with the header-encoding. Seems, that
 sup doesn't de- or encode ISO 8859-15 correct.
 
 There are no problems with the body or with ISO 8859-1 in the
 header.
 
 Did I messed up some settings or is this bug?
 
 I'm using sup v0.4 on MacOS X 10.5.2. - encoding in mutt works fine.
 
 
 Thanks, Carsten
 
 From marc.hartstein@alum.vassar.edu  Thu Mar 27 20:22:09 2008
 From: marc.hartstein@alum.vassar.edu (Marc Hartstein)
 Date: Thu, 27 Mar 2008 20:22:09 -0400
 Subject: [sup-talk] New User Questions
 Message-ID: <1206662265-sup-7439@cabinet>
 
 Sorry if any of this has been discussed elsewhere.  I've read the docs
 on the website, skimmed the wiki, and skimmed the mailing list archives
 for the past few months, but I could easily have missed something.
 
 First, I ran into the same problem discussed in the thread "startup
 crash" started on 2008-03-09.  (I'm also running Gentoo and installed
 the latest from portage)  I poked around the source a bit, and it
 seems that when AccountManager.initialize is called out of
 Redwood::start (sup line 113), the code to set up the default
 configuration (lib/sup.rb, starts at line 166) has *not* been called, so
 the $config object is empty.
 
 Just in case the information helps; I'm not a Ruby person (yet), and I
 didn't poke around enough to be clear on how it's supposed to be
 working.
 
 2. sup is pretty cool.  I'm particularly intrigued by this tag (sorry,
 label) concept.
 
 3. Is there an IRC channel?  It would be nice to be able to lurk and
 drop questions if anybody happens to be around.
 
 4. I use maildir.  Is there any way sup can mark messages as read (and,
 perhaps more importantly, no longer new) once I've read them and hit
 '$'?  I can deal with mutt being confused (I know sup doesn't play well
 with others anyway), but it's really confusing my mail monitor, and
 that's bugging me.
 
 5. Can sup be set to automatically poll local sources for new messages
 every n seconds?  Mutt does this, and it's nice not having to tell it
 that I know there are new messages and it should go find them.
 
 6. Mutt can pass text/html attachments through an external program and
 display the result in its internal pager.  Is there a way to get sup to
 do the equivalent?
 
 7. How does sup choose the email address to use as "From" when replying
 to an email?  It's definitely choosing the wrong thing in my setup.  My
 primary address forwards to gmail, sup is configured with the primary as
 my :email:, but whenever I hit reply it sets "From" to be my gmail
 address.  I'd like it to be whatever alternate of mine is in To/CC if
 any, and my primary address otherwise.
 
 8. Is there any way to specify custom keybindings in a configuration
 file, or would it require hacking the source?
 
 8.1 I'd really like a one-touch reply-all like I had in mutt.
 
 9. Is there a way to default to making all messages PGP-signed?
 
 10. Is there (planned) a compose-hook?  I'd like to be able to switch to
 sign-and-encrypt when I have a public key for every recipient.
 
 Thanks for the cool project.  I'll probably have more "I miss this
 feature from mutt" thoughts in the future, at least if I don't go
 running back.
 -------------- next part --------------
 A non-text attachment was scrubbed...
 Name: signature.asc
 Type: application/pgp-signature
 Size: 189 bytes
 Desc: not available
 Url : http://rubyforge.org/pipermail/sup-talk/attachments/20080327/ca129127/attachment.bin 
 
 From marcus-sup@bar-coded.net  Fri Mar 28 06:33:45 2008
 From: marcus-sup@bar-coded.net (Marcus Williams)
 Date: Fri, 28 Mar 2008 10:33:45 +0000
 Subject: [sup-talk] New User Questions
 In-Reply-To: <1206662265-sup-7439@cabinet>
 References: <1206662265-sup-7439@cabinet>
 Message-ID: <1206699001-sup-9770@tomsk>
 
 On 28.3.2008, Marc Hartstein wrote:
 > 3. Is there an IRC channel?  It would be nice to be able to lurk and
 > drop questions if anybody happens to be around.
 
 Someone has suggested #sup on freenode but no takers so far. I'd
 probably lurk on a channel if we can agree on one :)
 
 > 4. I use maildir.  Is there any way sup can mark messages as read (and,
 > perhaps more importantly, no longer new) once I've read them and hit
 > '$'?  I can deal with mutt being confused (I know sup doesn't play well
 > with others anyway), but it's really confusing my mail monitor, and
 > that's bugging me.
 
 Technically not (although you could probably write a hook to do this).
 However, you could change the way your monitor works if you're using
 sup locally. See the before/after poll hooks [1]
 
 > 5. Can sup be set to automatically poll local sources for new messages
 > every n seconds?  Mutt does this, and it's nice not having to tell it
 > that I know there are new messages and it should go find them.
 
 It should do already. I dont know offhand what the delay is though.
 
 > 6. Mutt can pass text/html attachments through an external program and
 > display the result in its internal pager.  Is there a way to get sup to
 > do the equivalent?
 
 Theres more than one way to do this. Via a hook for auto-demangling
 (mime-decode.rb) [1] and via the pipe attachment/message keypress '|'
 in thread view mode. Basically if you want to view an attachment put
 your cursor line over the attachment line and hit '|'. sup will pipe
 the attachment to what ever command line you type in here. html2text
 works well (as does antiword for word docs). If your cursor is in the
 message somewhere other than an attachment line it will pipe the raw
 message to the command.
 
 > 7. How does sup choose the email address to use as "From" when replying
 > to an email?  It's definitely choosing the wrong thing in my setup.  My
 > primary address forwards to gmail, sup is configured with the primary as
 > my :email:, but whenever I hit reply it sets "From" to be my gmail
 > address.  I'd like it to be whatever alternate of mine is in To/CC if
 > any, and my primary address otherwise.
 
 The :alternates: setting is probably what you want here. I think you
 should just have to add your gmail address as an alternate in your
 default account. At least this is how I think it should work!
 Basically the alternate emails list tell sup you receive email at this
 account as these addresses as well, but that you reply using the
 default :email: setting.
 
 You may also be interested in the :regexen: list - this allows you to
 setup mail extensions (like marcus-sup that I use) and get sup to
 reply using the extension.
 
 > 8. Is there any way to specify custom keybindings in a configuration
 > file, or would it require hacking the source?
 
 Hack the source afaik
 
 > 8.1 I'd really like a one-touch reply-all like I had in mutt.
 
 I'm about to submit a patch for this.
 
 > 9. Is there a way to default to making all messages PGP-signed?
 
 I dont think so at the moment (again, it wouldnt be hard to add)
 
 > 10. Is there (planned) a compose-hook?  I'd like to be able to
 > switch to sign-and-encrypt when I have a public key for every
 > recipient.
 
 Theres certainly a before-edit hook, which you may be able to do
 things like this in. I havnt played with the pgp stuff yet so someone
 else probably knows more here.
 
 HTH 
 
 Marcus
 
 [1] http://sup.rubyforge.org/wiki/wiki.pl?Hooks
 
 From vasudeva@linkswarm.com  Fri Mar 28 07:24:47 2008
 From: vasudeva@linkswarm.com (vasudeva)
 Date: Fri, 28 Mar 2008 07:24:47 -0400
 Subject: [sup-talk] New User Questions
 In-Reply-To: <1206699001-sup-9770@tomsk>
 References: <1206662265-sup-7439@cabinet> <1206699001-sup-9770@tomsk>
 Message-ID: <1206703335-sup-6607@lenin>
 
 Excerpts from Marcus Williams's message of Fri Mar 28 06:33:45 -0400 2008:
 > On 28.3.2008, Marc Hartstein wrote:
 > > 3. Is there an IRC channel?  It would be nice to be able to lurk and
 > > drop questions if anybody happens to be around.
 > 
 > Someone has suggested #sup on freenode but no takers so far. I'd
 > probably lurk on a channel if we can agree on one :)
 
 Actually, there are a couple of us lurking there now. Come on in.
 
 
 
 
 -- 
 linkswarm.com :: Collaborative Insolence
 vasudeva.linkswarm.com/gallery :: For The Faint of Heart
 
 
 From 5srmspw02@sneakemail.com  Fri Mar 28 11:54:11 2008
 From: 5srmspw02@sneakemail.com (Guarded Identity)
 Date: Fri, 28 Mar 2008 10:54:11 -0500
 Subject: [sup-talk] Wiki updates, more people on #sup
 Message-ID: <18635-64358@sneakemail.com>
 
 Hey everyone,
 
 So now we've got five people lurking in #sup on Freenode (only two of us are
 quasi-active, though).  I was going to post some of my recent configurations
 for Sup to the mailing list, but then vasudeva noted the Wiki was
 under-utilized, so I made two new pages there [1,2].  One thing to note about
 these configurations are that I'm relying on other tools to solve problems
 rather than asking William to "improve" Sup.  However, regarding [2], I guess I
 wish that contact tab-complete could be customizable (say to hook into ldbdq).
 I'm completely fine with the solution in [1] as-is, though.
 
 [1] http://sup.rubyforge.org/wiki/wiki.pl?WebBrowserIntegration
 [2] http://sup.rubyforge.org/wiki/wiki.pl?LbdbIntegration
 
 Regards,
 Sukant
 
 From marc.hartstein@alum.vassar.edu  Fri Mar 28 13:23:29 2008
 From: marc.hartstein@alum.vassar.edu (Marc Hartstein)
 Date: Fri, 28 Mar 2008 13:23:29 -0400
 Subject: [sup-talk] New User Questions
 In-Reply-To: <1206699001-sup-9770@tomsk>
 References: <1206662265-sup-7439@cabinet> <1206699001-sup-9770@tomsk>
 Message-ID: <1206724477-sup-3121@cabinet>
 
 Excerpts from Marcus Williams's message of Fri Mar 28 06:33:45 -0400 2008:
 > On 28.3.2008, Marc Hartstein wrote:
 > 
 > > 5. Can sup be set to automatically poll local sources for new messages
 > > every n seconds?  Mutt does this, and it's nice not having to tell it
 > > that I know there are new messages and it should go find them.
 > 
 > It should do already. I dont know offhand what the delay is though.
 
 So it does, I think it's just less frequent than I'm accustomed to.
 This should probably go on the list of things which should eventually be
 configurable.
 
 > > 6. Mutt can pass text/html attachments through an external program and
 > > display the result in its internal pager.  Is there a way to get sup to
 > > do the equivalent?
 > 
 > Theres more than one way to do this. Via a hook for auto-demangling
 > (mime-decode.rb)
 
 Thanks, that's exactly what I was looking for.  And should be nicely
 extensible if I ever want.
 
 > and via the pipe attachment/message keypress '|' in thread view mode.
 > Basically if you want to view an attachment put your cursor line over
 > the attachment line and hit '|'. sup will pipe the attachment to what
 > ever command line you type in here.
 
 Probably too much effort unless it can consult mailcap or something and
 fill in an appropriate default.  I wonder how difficult that would be to
 add....
 
 > > 7. How does sup choose the email address to use as "From" when replying
 > > to an email?  It's definitely choosing the wrong thing in my setup.  My
 > > primary address forwards to gmail, sup is configured with the primary as
 > > my :email:, but whenever I hit reply it sets "From" to be my gmail
 > > address.  I'd like it to be whatever alternate of mine is in To/CC if
 > > any, and my primary address otherwise.
 > 
 > The :alternates: setting is probably what you want here. I think you
 > should just have to add your gmail address as an alternate in your
 > default account. At least this is how I think it should work!
 > Basically the alternate emails list tell sup you receive email at this
 > account as these addresses as well, but that you reply using the
 > default :email: setting.
 
 Nope.  The gmail address is an alternate, the primary is the :email:, it
 tries to reply with the gmail address.  But only sometimes.  I still
 haven't figured out what's causing it.
 
 > You may also be interested in the :regexen: list - this allows you to
 > setup mail extensions (like marcus-sup that I use) and get sup to
 > reply using the extension.
 
 I'll look into this, thanks.
 
 > > 8. Is there any way to specify custom keybindings in a configuration
 > > file, or would it require hacking the source?
 > 
 > Hack the source afaik
 
 Feh, hope that's on the planned features list.
 
 > > 9. Is there a way to default to making all messages PGP-signed?
 > 
 > I dont think so at the moment (again, it wouldnt be hard to add)
 
 Should probably be a config setting.
 
 > Theres certainly a before-edit hook, which you may be able to do
 > things like this in. I havnt played with the pgp stuff yet so someone
 > else probably knows more here.
 
 That one's definitely not on the wiki.  I'll have to track down the
 canonical list of available hooks...
 -------------- next part --------------
 A non-text attachment was scrubbed...
 Name: signature.asc
 Type: application/pgp-signature
 Size: 189 bytes
 Desc: not available
 Url : http://rubyforge.org/pipermail/sup-talk/attachments/20080328/7a51264e/attachment.bin 
 
 From marcus-sup@bar-coded.net  Fri Mar 28 15:51:18 2008
 From: marcus-sup@bar-coded.net (Marcus Williams)
 Date: Fri, 28 Mar 2008 19:51:18 +0000
 Subject: [sup-talk] New User Questions
 In-Reply-To: <1206724477-sup-3121@cabinet>
 References: <1206662265-sup-7439@cabinet> <1206699001-sup-9770@tomsk>
 	<1206724477-sup-3121@cabinet>
 Message-ID: <1206733867-sup-2930@tomsk>
 
 On 28.3.2008, Marc Hartstein wrote:
 > > Theres certainly a before-edit hook, which you may be able to do
 > > things like this in. I havnt played with the pgp stuff yet so someone
 > > else probably knows more here.
 > 
 > That one's definitely not on the wiki.  I'll have to track down the
 > canonical list of available hooks...
 
 sup --list-hooks
 
 :)
 
 Marcus
 
 From marc.hartstein@alum.vassar.edu  Fri Mar 28 16:21:19 2008
 From: marc.hartstein@alum.vassar.edu (Marc Hartstein)
 Date: Fri, 28 Mar 2008 16:21:19 -0400
 Subject: [sup-talk] Crash
 Message-ID: <1206735524-sup-1255@cabinet>
 
 Attached is the exception log as requested.
 
 Sequence of events:
 
 $ sup-add -l looney maildir:/home/magus/.maildir/looney
 
 $ sup
 
 L looney while the index counter is still incrementing in the status
 line.
 
 Search page appears, a couple of lines get added (I think), then crash.
 
 Hope it helps.
 -------------- next part --------------
 An embedded and charset-unspecified text was scrubbed...
 Name: sup-exception-log.txt
 Url: http://rubyforge.org/pipermail/sup-talk/attachments/20080328/78d42c7c/attachment-0001.txt 
 -------------- next part --------------
 A non-text attachment was scrubbed...
 Name: signature.asc
 Type: application/pgp-signature
 Size: 189 bytes
 Desc: not available
 Url : http://rubyforge.org/pipermail/sup-talk/attachments/20080328/78d42c7c/attachment-0001.bin 
 
 From martindemello@gmail.com  Fri Mar 28 18:13:26 2008
 From: martindemello@gmail.com (Martin DeMello)
 Date: Fri, 28 Mar 2008 15:13:26 -0700
 Subject: [sup-talk] sup crashes on startup
 Message-ID: <f93a6bcc0803281513s429f3a3bw485c63e5d9cce818@mail.gmail.com>
 
 --- ArgumentError from thread: main
 no name for account
 /usr/lib/ruby/gems/1.8/gems/sup-0.4/lib/sup/account.rb:7:in `initialize'
 /usr/lib/ruby/gems/1.8/gems/sup-0.4/lib/sup/account.rb:44:in `new'
 /usr/lib/ruby/gems/1.8/gems/sup-0.4/lib/sup/account.rb:44:in `add_account'
 /usr/lib/ruby/gems/1.8/gems/sup-0.4/lib/sup/account.rb:26:in `initialize'
 /usr/lib/ruby/gems/1.8/gems/sup-0.4/lib/sup.rb:98:in `new'
 /usr/lib/ruby/gems/1.8/gems/sup-0.4/lib/sup.rb:98:in `start'
 /usr/lib/ruby/gems/1.8/gems/sup-0.4/bin/sup:113
 /usr/bin/sup:16:in `load'
 /usr/bin/sup:16
 
 From sup-talk@gurski.org  Fri Mar 28 23:03:56 2008
 From: sup-talk@gurski.org (Michael Gurski)
 Date: Fri, 28 Mar 2008 23:03:56 -0400
 Subject: [sup-talk] crash while sup is sitting idle
 Message-ID: <20080329030356.GA8459@gurski.org>
 
 After a 28+ hour sup-sync -a -o, on git head, sup sitting idle crashed
 with the following error:
 
 --- Ferret::StateError from thread: load threads for thread-index-mode
 State Error occured at <except.c>:93 in xraise
 Error occured in index.c:4150 - sr_get_lazy_doc
         Document 69 has already been deleted
 
 /var/lib/gems/1.8/gems/ferret-0.11.6/lib/ferret/index.rb:421:in `[]'
 /var/lib/gems/1.8/gems/ferret-0.11.6/lib/ferret/index.rb:421:in `[]'
 /usr/lib/ruby/1.8/monitor.rb:238:in `synchronize'
 /var/lib/gems/1.8/gems/ferret-0.11.6/lib/ferret/index.rb:413:in `[]'
 ./lib/sup/index.rb:228:in `each_id_by_date'
 ./lib/sup/index.rb:228:in `each'
 ./lib/sup/index.rb:228:in `each_id_by_date'
 ./lib/sup/thread.rb:326:in `load_n_threads'
 ./lib/sup/modes/thread-index-mode.rb:507:in
 `__unprotected_load_n_threads'
 (eval):12:in `load_n_threads'
 ./lib/sup/modes/thread-index-mode.rb:491:in
 `load_n_threads_background'
 ./lib/sup.rb:60:in `reporting_thread'
 ./lib/sup.rb:58:in `initialize'
 ./lib/sup.rb:58:in `new'
 ./lib/sup.rb:58:in `reporting_thread'
 ./lib/sup/modes/thread-index-mode.rb:490:in
 `load_n_threads_background'
 ./lib/sup/modes/thread-index-mode.rb:560:in
 `__unprotected_load_threads'
 (eval):12:in `load_threads'
 ./lib/sup/modes/thread-index-mode.rb:70:in `initialize'
 ./lib/sup/modes/line-cursor-mode.rb:177:in `call'
 ./lib/sup/modes/line-cursor-mode.rb:177:in `call_load_more_callbacks'
 ./lib/sup/modes/line-cursor-mode.rb:177:in `each'
 ./lib/sup/modes/line-cursor-mode.rb:177:in `call_load_more_callbacks'
 ./lib/sup/modes/line-cursor-mode.rb:80:in `cursor_down'
 ./lib/sup/mode.rb:49:in `send'
 ./lib/sup/mode.rb:49:in `handle_input'
 ./lib/sup/buffer.rb:231:in `handle_input'
 bin/sup:223
 --- SystemExit from thread: main
 State Error occured at <except.c>:93 in xraise
 Error occured in index.c:4150 - sr_get_lazy_doc
         Document 69 has already been deleted
 
 ./lib/sup.rb:64:in `select'
 ./lib/sup/buffer.rb:31:in `nonblocking_getch'
 bin/sup:217
 
 -- 
 Michael A. Gurski (opt. [first].)[last]@pobox.com  http://www.pobox.com/~[last]
 1024R/39B5BADD PGP: 34 93 A9 94 B1 59 48 B7  17 57 1E 4E 62 56 45 70
 1024D/1166213E GPG: 628F 37A4 62AF 1475 45DB  AD81 ADC9 E606 1166 213E
 4096R/C0B4F04B GPG: 5B3E 75D7 43CF CF34 4042  7788 1DCE B5EE C0B4 F04B
 Views expressed by the host do not reflect the staff, management or sponsors.
 
 
 From chrisw@rice.edu  Sat Mar 29 03:57:51 2008
 From: chrisw@rice.edu (Christopher Warrington)
 Date: Sat, 29 Mar 2008 02:57:51 -0500
 Subject: [sup-talk] [PATCH] polling is now done per source
 In-Reply-To: <1206724477-sup-3121@cabinet>
 References: <1206724477-sup-3121@cabinet>
 Message-ID: <1206777471-1360-1-git-send-email-chrisw@rice.edu>
 
 Each source has a poll_interval property. This property is used to
 determine whether the source should be polled. The user can still for a
 poll of all sources.
 ---
  bin/sup                    |    4 +-
  bin/sup-add                |   12 ++-
  lib/sup.rb                 |    4 +-
  lib/sup/imap.rb            |    6 +-
  lib/sup/maildir.rb         |    6 +-
  lib/sup/mbox/loader.rb     |    4 +-
  lib/sup/modes/poll-mode.rb |    4 +-
  lib/sup/poll.rb            |  211 +++++++++++++++++++++++++++++++++-----------
  lib/sup/source.rb          |    8 +-
  9 files changed, 189 insertions(+), 70 deletions(-)
  mode change 100755 => 100644 lib/sup/buffer.rb
 
 diff --git a/bin/sup b/bin/sup
 index 84fd77c..86a2d9f 100644
 --- a/bin/sup
 +++ b/bin/sup
 @@ -197,7 +197,7 @@ begin
      end
    end unless $opts[:no_initial_poll]
    
 -  imode.load_threads :num => ibuf.content_height, :when_done => lambda { reporting_thread("poll after loading inbox") { sleep 1; PollManager.poll } unless $opts[:no_threads] || $opts[:no_initial_poll] }
 +  imode.load_threads :num => ibuf.content_height, :when_done => lambda { reporting_thread("poll after loading inbox") { sleep 1; PollManager.auto_poll } unless $opts[:no_threads] || $opts[:no_initial_poll] }
  
    if $opts[:compose]
      ComposeMode.spawn_nicely :to_default => $opts[:compose]
 @@ -263,7 +263,7 @@ begin
      when :compose
        ComposeMode.spawn_nicely
      when :poll
 -      reporting_thread("user-invoked poll") { PollManager.poll }
 +      reporting_thread("user-invoked poll") { PollManager.forced_poll }
      when :recall_draft
        case Index.num_results_for :label => :draft
        when 0
 diff --git a/bin/sup-add b/bin/sup-add
 index 50bbb29..88705e1 100644
 --- a/bin/sup-add
 +++ b/bin/sup-add
 @@ -39,6 +39,7 @@ EOS
    opt :unusual, "Do not automatically poll these sources for new messages."
    opt :labels, "A comma-separated set of labels to apply to all messages from this source", :type => String
    opt :force_new, "Create a new account for this source, even if one already exists."
 +  opt :poll_interval, "The interval (in seconds) between new message polls. The default is #{Redwood::DEFAULT_POLL_INTERVAL}.", :type => :int
  end
  
  Trollop::die "require one or more sources" if ARGV.empty?
 @@ -84,6 +85,9 @@ index.lock_or_die
  begin
    index.load_sources
  
 +  Trollop::die "The poll interval must be a positive integer." if $opts[:poll_interval] <= 0
 +  poll_interval = $opts[:poll_interval] || Redwood::DEFAULT_POLL_INTERVAL
 +
    ARGV.each do |uri|
      labels = $opts[:labels] ? $opts[:labels].split(/\s*,\s*/).uniq : []
  
 @@ -100,14 +104,14 @@ begin
          say "For SSH connections, if you will use public key authentication, you may leave the username and password blank."
          say ""
          username, password = get_login_info uri, index.sources
 -        Redwood::MBox::SSHLoader.new uri, username, password, nil, !$opts[:unusual], $opts[:archive], nil, labels
 +        Redwood::MBox::SSHLoader.new uri, username, password, nil, !$opts[:unusual], $opts[:archive], nil, poll_interval, labels
        when "imap", "imaps"
          username, password = get_login_info uri, index.sources
 -        Redwood::IMAP.new uri, username, password, nil, !$opts[:unusual], $opts[:archive], nil, labels
 +        Redwood::IMAP.new uri, username, password, nil, !$opts[:unusual], $opts[:archive], nil, poll_interval, labels
        when "maildir"
 -        Redwood::Maildir.new uri, nil, !$opts[:unusual], $opts[:archive], nil, labels
 +        Redwood::Maildir.new uri, nil, !$opts[:unusual], $opts[:archive], nil, poll_interval, labels
        when "mbox"
 -        Redwood::MBox::Loader.new uri, nil, !$opts[:unusual], $opts[:archive], nil, labels
 +        Redwood::MBox::Loader.new uri, nil, !$opts[:unusual], $opts[:archive], nil, poll_interval, labels
        when nil
          Trollop::die "Sources must be specified with an URI"
        else
 diff --git a/lib/sup.rb b/lib/sup.rb
 index 1946f3c..c27a4bc 100644
 --- a/lib/sup.rb
 +++ b/lib/sup.rb
 @@ -50,6 +50,8 @@ module Redwood
    YAML_DOMAIN = "masanjin.net"
    YAML_DATE = "2006-10-01"
  
 +  DEFAULT_POLL_INTERVAL = 300
 +
  ## record exceptions thrown in threads nicely
    def reporting_thread name
      if $opts[:no_threads]
 @@ -72,7 +74,7 @@ module Redwood
    def save_yaml_obj object, fn, safe=false
      if safe
        safe_fn = "#{File.dirname fn}/safe_#{File.basename fn}"
 -      mode = File.stat(fn) if File.exists? fn
 +      mode = File.stat(fn).mode if File.exists? fn
        File.open(safe_fn, "w", mode) { |f| f.puts object.to_yaml }
        FileUtils.mv safe_fn, fn
      else
 diff --git a/lib/sup/buffer.rb b/lib/sup/buffer.rb
 old mode 100755
 new mode 100644
 diff --git a/lib/sup/imap.rb b/lib/sup/imap.rb
 index 1d36976..8b58cba 100644
 --- a/lib/sup/imap.rb
 +++ b/lib/sup/imap.rb
 @@ -51,13 +51,13 @@ class IMAP < Source
  
    attr_accessor :username, :password
    yaml_properties :uri, :username, :password, :cur_offset, :usual,
 -                  :archived, :id, :labels
 +                  :archived, :id, :poll_interval, :labels
  
 -  def initialize uri, username, password, last_idate=nil, usual=true, archived=false, id=nil, labels=[]
 +  def initialize uri, username, password, last_idate=nil, usual=true, archived=false, id=nil, poll_interval=nil, labels=[]
      raise ArgumentError, "username and password must be specified" unless username && password
      raise ArgumentError, "not an imap uri" unless uri =~ %r!imaps?://!
  
 -    super uri, last_idate, usual, archived, id
 +    super uri, last_idate, usual, archived, id, [poll_interval, SCAN_INTERVAL].max
  
      @parsed_uri = URI(uri)
      @username = username
 diff --git a/lib/sup/maildir.rb b/lib/sup/maildir.rb
 index 584e657..ef71e0b 100644
 --- a/lib/sup/maildir.rb
 +++ b/lib/sup/maildir.rb
 @@ -12,9 +12,9 @@ class Maildir < Source
    SCAN_INTERVAL = 30 # seconds
  
    ## remind me never to use inheritance again.
 -  yaml_properties :uri, :cur_offset, :usual, :archived, :id, :labels
 -  def initialize uri, last_date=nil, usual=true, archived=false, id=nil, labels=[]
 -    super uri, last_date, usual, archived, id
 +  yaml_properties :uri, :cur_offset, :usual, :archived, :id, :poll_interval, :labels
 +  def initialize uri, last_date=nil, usual=true, archived=false, id=nil, poll_interval=nil, labels=[]
 +    super uri, last_date, usual, archived, id, [poll_interval, SCAN_INTERVAL].max
      uri = URI(Source.expand_filesystem_uri(uri))
  
      raise ArgumentError, "not a maildir URI" unless uri.scheme == "maildir"
 diff --git a/lib/sup/mbox/loader.rb b/lib/sup/mbox/loader.rb
 index 7fe9129..44317d5 100644
 --- a/lib/sup/mbox/loader.rb
 +++ b/lib/sup/mbox/loader.rb
 @@ -9,7 +9,7 @@ class Loader < Source
    attr_accessor :labels
  
    ## uri_or_fp is horrific. need to refactor.
 -  def initialize uri_or_fp, start_offset=nil, usual=true, archived=false, id=nil, labels=[]
 +  def initialize uri_or_fp, start_offset=nil, usual=true, archived=false, id=nil, poll_interval=nil, labels=[]
      @mutex = Mutex.new
      @labels = ((labels || []) - LabelManager::RESERVED_LABELS).uniq.freeze
  
 @@ -26,7 +26,7 @@ class Loader < Source
        @path = uri_or_fp.path
      end
  
 -    super uri_or_fp, start_offset, usual, archived, id
 +    super uri_or_fp, start_offset, usual, archived, id, poll_interval
    end
  
    def file_path; @path end
 diff --git a/lib/sup/modes/poll-mode.rb b/lib/sup/modes/poll-mode.rb
 index 5849f3e..5521cdc 100644
 --- a/lib/sup/modes/poll-mode.rb
 +++ b/lib/sup/modes/poll-mode.rb
 @@ -10,11 +10,11 @@ class PollMode < LogMode
      self << s + "\n"
    end
  
 -  def poll
 +  def poll sources
      puts unless @new
      @new = false
      puts "Poll started at #{Time.now}"
 -    PollManager.do_poll { |s| puts s }
 +    PollManager.do_poll_sources(sources) { |s| puts s }
    end
  end
  
 diff --git a/lib/sup/poll.rb b/lib/sup/poll.rb
 index d32c893..01db8e2 100644
 --- a/lib/sup/poll.rb
 +++ b/lib/sup/poll.rb
 @@ -12,13 +12,22 @@ Variables:
  EOS
  
    HookManager.register "before-poll", <<EOS
 +Executes immediately before any poll for new messages commences.
 +Variables:
 +                 accts: an array of source URIs that will be polled
 +EOS
 +
 +  HookManager.register "before-acct-poll", <<EOS
  Executes immediately before a poll for new messages commences.
 -No variables.
 +Variables:
 +                  acct: the URI of the account being polled
  EOS
  
 -  HookManager.register "after-poll", <<EOS
 -Executes immediately after a poll for new messages completes.
 +  HookManager.register "after-acct-poll", <<EOS
 +Executes immediately after a poll for new messages completes. The statistics are
 +only for the account specified.
  Variables:
 +                  acct: the URI of the account being polled
                     num: the total number of new messages added in this poll
               num_inbox: the number of new messages added in this poll which
                          appear in the inbox (i.e. were not auto-archived).
 @@ -28,14 +37,25 @@ num_inbox_total_unread: the total number of unread messages in the inbox
                          only those messages appearing in the inbox
  EOS
  
 -  DELAY = 300
 +  HookManager.register "after-poll", <<EOS
 +Executes immediately after any poll for new messages completes. The statistics
 +are for all sources polled.
 +Variables:
 +                 accts: an array of source URIs that were polled
 +                   num: the total number of new messages added in this poll
 +             num_inbox: the number of new messages added in this poll which
 +                        appear in the inbox (i.e. were not auto-archived).
 +num_inbox_total_unread: the total number of unread messages in the inbox
 +         from_and_subj: an array of (from email address, subject) pairs
 +   from_and_subj_inbox: an array of (from email address, subject) pairs for
 +                        only those messages appearing in the inbox
 +EOS
  
    def initialize
      @mutex = Mutex.new
      @thread = nil
 -    @last_poll = nil
      @polling = false
 -    
 +
      self.class.i_am_the_instance self
    end
  
 @@ -44,47 +64,113 @@ EOS
      b
    end
  
 -  def poll
 +  def start
 +    ## Periodically calls auto_poll, effectively automatically polling
 +    ## in the background.
 +    ##
 +    ## auto_poll is called more frequently than the smallest poll
 +    ## interval. auto_poll is not guaranteed to be called in multiples
 +    ## of poll intervals. I.e., don't rely on auto_poll to be called
 +    ## every x seconds: when auto_poll is called assume an
 +    ## indeterminate amount of time has passed.
 +    @min_poll_interval = Index.usual_sources.collect{|s| s.poll_interval}.select{|pi| !pi.nil?}.min || Redwood::DEFAULT_POLL_INTERVAL
 +
 +    Redwood::log "Background poll interval is #{@min_poll_interval} seconds."
 +
 +    @thread = Redwood::reporting_thread("periodic poll") do
 +      while true
 +        Redwood::log "Sleeping for #{@min_poll_interval / 2} seconds."
 +
 +        sleep @min_poll_interval / 2
 +        auto_poll
 +      end
 +    end
 +  end
 +
 +  def stop
 +    @thread.kill if @thread
 +    @thread = nil
 +  end
 +
 +  def auto_poll
 +    ## This method is called by the thread spawned in start. It
 +    ## collects the sources that should be polled
 +    ## (source.poll_interval has expired or the source has never been
 +    ## polled) and polls them.
 +    ##
 +    ## Returns an array [# of new messages, # of new messages to
 +    ## inbox, new message subject/name array pairs, subject/name array
 +    ## pairs loaded to inbox]
 +
 +    sources_to_poll = []
 +    @mutex.synchronize do # William, do we need to synchronize here?
 +      begin
 +        sources_to_poll = Index.usual_sources.select do |source|
 +          begin
 +            source.last_poll.nil? || (Time.now - source.last_poll) >= source.poll_interval
 +          end
 +        end
 +      end
 +    end
 +
 +    poll_sources sources_to_poll
 +  end
 +
 +  def forced_poll
 +    ## This method is called when the user explicitly requests a poll.
 +    ##
 +    ## Returns an array [# of new messages, # of new messages to
 +    ## inbox, new message subject/name array pairs, subject/name array
 +    ## pairs loaded to inbox]
 +    poll_sources Index.usual_sources
 +  end
 +
 +  def poll_sources sources
 +    ## Polls the given sources. Clients of PollManager should call
 +    ## this method, not do_poll_sources or poll_source. (Well, only
 +    ## PollMode should call do_poll_sources)
 +    ##
 +    ## Returns an array [# of new messages, # of new messages to
 +    ## inbox, new message subject/name array pairs, subject/name array
 +    ## pairs loaded to inbox]
      return if @polling
      @polling = true
 -    HookManager.run "before-poll"
 +
 +    source_uris = sources.map{|s| s.uri}
 +
 +    HookManager.run "before-acct-poll", :accts => source_uris
  
      BufferManager.flash "Polling for new messages..."
 -    num, numi, from_and_subj, from_and_subj_inbox = buffer.mode.poll
 +    num, numi, from_subj, from_subj_inbox = buffer.mode.poll sources
      if num > 0
        BufferManager.flash "Loaded #{num.pluralize 'new message'}, #{numi} to inbox." 
      else
        BufferManager.flash "No new messages." 
      end
  
 -    HookManager.run "after-poll", :num => num, :num_inbox => numi, :from_and_subj => from_and_subj, :from_and_subj_inbox => from_and_subj_inbox, :num_inbox_total_unread => lambda { Index.num_results_for :labels => [:inbox, :unread] }
 +    HookManager.run "after-poll", :accts => source_uris, :num => num, :num_inbox => numi, :from_and_subj => from_and_subj, :from_and_subj_inbox => from_and_subj_inbox, :num_inbox_total_unread => lambda { Index.num_results_for :labels => [:inbox, :unread] }
  
      @polling = false
 -    [num, numi]
 -  end
 -
 -  def start
 -    @thread = Redwood::reporting_thread("periodic poll") do
 -      while true
 -        sleep DELAY / 2
 -        poll if @last_poll.nil? || (Time.now - @last_poll) >= DELAY
 -      end
 -    end
 +    [num, numi, from_subj, from_subj_inbox]
    end
  
 -  def stop
 -    @thread.kill if @thread
 -    @thread = nil
 -  end
 +  def do_poll_sources sources, &block
 +    ## Polls each source, keeping track of vital statistics (number
 +    ## loaded, name/e-mail pairs, &c.) about the poll results.
 +    ##
 +    ## We need explicit access to the block so that we can pass it to
 +    ## poll_source.
 +    ##
 +    ## Returns an array [total # of new messages, total # of new
 +    ## messages to inbox, all new message subject/name array pairs,
 +    ## all subject/name array pairs loaded to inbox]
  
 -  def do_poll
      total_num = total_numi = 0
 -    from_and_subj = []
 -    from_and_subj_inbox = []
 +    total_from_and_subj = []
 +    total_from_and_subj_inbox = []
  
      @mutex.synchronize do
 -      Index.usual_sources.each do |source|
 -#        yield "source #{source} is done? #{source.done?} (cur_offset #{source.cur_offset} >= #{source.end_offset})"
 +      sources.each do |source|
          begin
            yield "Loading from #{source}... " unless source.done? || source.has_errors?
          rescue SourceError => e
 @@ -93,32 +179,57 @@ EOS
            next
          end
  
 -        num = 0
 -        numi = 0
 -        add_messages_from source do |m, offset, entry|
 -          ## always preserve the labels on disk.
 -          m.labels = entry[:label].split(/\s+/).map { |x| x.intern } if entry
 -          yield "Found message at #{offset} with labels {#{m.labels * ', '}}"
 -          unless entry
 -            num += 1
 -            from_and_subj << [m.from.longname, m.subj]
 -            if m.has_label?(:inbox) && ([:spam, :deleted, :killed] & m.labels).empty?
 -              from_and_subj_inbox << [m.from.longname, m.subj]
 -              numi += 1 
 -            end
 -          end
 -          m
 -        end
 -        yield "Found #{num} messages, #{numi} to inbox." unless num == 0
 +        HookManager.run "before-acct-poll", :acct => source.uri
 +
 +        BufferManager.flash "Polling #{source.uri} for new messages..."
 +        num, numi, from_and_subj, from_and_subj_inbox = poll_source source, &block
 +
          total_num += num
          total_numi += numi
 +        total_from_and_subj += from_and_subj
 +        total_from_and_subj_inbox += from_and_subj_inbox
 +
 +        HookManager.run "after-acct-poll", :acct => source.uri, :num => num, :num_inbox => numi, :from_and_subj => from_and_subj, :from_and_subj_inbox => from_and_subj_inbox, :num_inbox_total_unread => lambda { Index.num_results_for :labels => [:inbox, :unread] }
        end
 +    end
 +
 +    [total_num, total_numi, total_from_and_subj, total_from_and_subj_inbox]
 +  end
  
 -      yield "Done polling; loaded #{total_num} new messages total"
 -      @last_poll = Time.now
 -      @polling = false
 +  def poll_source source
 +    ## Polls the given source for new messages.
 +    ##
 +    ## @mutex must be held before calling! See do_poll_sources.
 +    ##
 +    ## Returns an array [# of new messages, # of new messages to
 +    ## inbox, new message subject/name array pairs, subject/name array
 +    ## pairs loaded to inbox]
 +
 +    num = 0
 +    numi = 0
 +    from_and_subj = []
 +    from_and_subj_inbox = []
 +
 +    add_messages_from source do |m, offset, entry|
 +      ## always preserve the labels on disk.
 +      m.labels = entry[:label].split(/\s+/).map { |x| x.intern } if entry
 +      yield "Found message at #{offset} with labels {#{m.labels * ', '}}"
 +      unless entry
 +        num += 1
 +        from_and_subj << [m.from.longname, m.subj]
 +        if m.has_label?(:inbox) && ([:spam, :deleted, :killed] & m.labels).empty?
 +          from_and_subj_inbox << [m.from.longname, m.subj]
 +          numi += 1
 +        end
 +      end
 +      m
      end
 -    [total_num, total_numi, from_and_subj, from_and_subj_inbox]
 +
 +    yield "For source #{source.uri}, found #{num} messages, #{numi} to inbox." unless num == 0
 +
 +    source.last_poll = Time.now
 +
 +    [num, numi, from_and_subj, from_and_subj_inbox]
    end
  
    ## this is the main mechanism for adding new messages to the
 diff --git a/lib/sup/source.rb b/lib/sup/source.rb
 index 6510aae..fd8d381 100644
 --- a/lib/sup/source.rb
 +++ b/lib/sup/source.rb
 @@ -62,10 +62,10 @@ class Source
    ## dirty? means cur_offset has changed, so the source info needs to
    ## be re-saved to sources.yaml.
    bool_reader :usual, :archived, :dirty
 -  attr_reader :uri, :cur_offset
 -  attr_accessor :id
 +  attr_reader :uri, :cur_offset, :poll_interval
 +  attr_accessor :id, :last_poll
  
 -  def initialize uri, initial_offset=nil, usual=true, archived=false, id=nil
 +  def initialize uri, initial_offset=nil, usual=true, archived=false, id=nil, poll_interval=nil
      raise ArgumentError, "id must be an integer: #{id.inspect}" unless id.is_a? Fixnum if id
  
      @uri = uri
 @@ -73,6 +73,8 @@ class Source
      @usual = usual
      @archived = archived
      @id = id
 +    @poll_interval = poll_interval || Redwood::DEFAULT_POLL_INTERVAL #seconds
 +    @last_poll = nil
      @dirty = false
    end
  
 -- 
 1.5.4
 
 
 From chrisw@rice.edu  Sat Mar 29 05:59:31 2008
 From: chrisw@rice.edu (Christopher Warrington)
 Date: Sat, 29 Mar 2008 04:59:31 -0500
 Subject: [sup-talk] [PATCH] polling is now done per source
 In-Reply-To: <1206777471-1360-1-git-send-email-chrisw@rice.edu>
 References: <1206724477-sup-3121@cabinet>
 	<1206777471-1360-1-git-send-email-chrisw@rice.edu>
 Message-ID: <1206784680-sup-3052@chris-tablet>
 
 Excerpts from Christopher Warrington's message of Sat Mar 29 02:57:51 -0500 2008:
 > Each source has a poll_interval property. This property is used to
 > determine whether the source should be polled. The user can still
 > for a poll of all sources.
 
 There is a bug in this patch. Watch this space for an updated patch.
 
 -- 
 Christopher Warrington <chrisw at rice.edu>
 
 From chrisw@rice.edu  Sat Mar 29 06:02:31 2008
 From: chrisw@rice.edu (Christopher Warrington)
 Date: Sat, 29 Mar 2008 05:02:31 -0500
 Subject: [sup-talk] Newline Issues
 In-Reply-To: <1206549226-sup-8056@south>
 References: <1204536487-sup-9413@chris-tablet> <1206549226-sup-8056@south>
 Message-ID: <1206784905-sup-8496@chris-tablet>
 
 Excerpts from William Morgan's message of Wed Mar 26 11:40:24 -0500 2008:
 > Reformatted excerpts from Christopher Warrington's message of 2008-03-03:
 > > When I use offlineimap under Cgywin, the messages are delivered with
 > > CRLF. Sup cannot parse these: it thinks that the body is empty. I know
 > > that sup is seeing the entire message. If I press H, I see all of the
 > > message.
 > 
 > I've finally had a chance to look into this. I've just patched Sup to
 > handle carriage returns in mbox files, but things are still broken
 > because of RubyMail.
 > 
 > It looks like there's an outstanding RubyMail patch here:
 > 
 > http://rubyforge.org/tracker/index.php?func=detail&aid=2821&group_id=446&atid=17
 > 56
 > 
 > If you apply that and get the latest git, I think things should work.
 
 That appears to have done it. All is well in Cygwin land, for now.
 
 -- 
 Christopher Warrington <chrisw at rice.edu>
 
 
 From chrisw@rice.edu  Sat Mar 29 06:32:05 2008
 From: chrisw@rice.edu (Christopher Warrington)
 Date: Sat, 29 Mar 2008 05:32:05 -0500
 Subject: [sup-talk] [PATCH] polling is now done per source
 In-Reply-To: <1206784680-sup-3052@chris-tablet>
 References: <1206784680-sup-3052@chris-tablet>
 Message-ID: <1206786725-5456-1-git-send-email-chrisw@rice.edu>
 
 Each source has a poll_interval property. This property is used to
 determine whether the source should be polled. The user can still for a
 poll of all sources.
 ---
  bin/sup                    |    4 +-
  bin/sup-add                |   12 ++-
  lib/sup.rb                 |    4 +-
  lib/sup/imap.rb            |    6 +-
  lib/sup/maildir.rb         |    6 +-
  lib/sup/mbox/loader.rb     |    4 +-
  lib/sup/modes/poll-mode.rb |    4 +-
  lib/sup/poll.rb            |  211 +++++++++++++++++++++++++++++++++-----------
  lib/sup/source.rb          |    8 +-
  9 files changed, 189 insertions(+), 70 deletions(-)
  mode change 100755 => 100644 lib/sup/buffer.rb
 
 diff --git a/bin/sup b/bin/sup
 index 84fd77c..86a2d9f 100644
 --- a/bin/sup
 +++ b/bin/sup
 @@ -197,7 +197,7 @@ begin
      end
    end unless $opts[:no_initial_poll]
    
 -  imode.load_threads :num => ibuf.content_height, :when_done => lambda { reporting_thread("poll after loading inbox") { sleep 1; PollManager.poll } unless $opts[:no_threads] || $opts[:no_initial_poll] }
 +  imode.load_threads :num => ibuf.content_height, :when_done => lambda { reporting_thread("poll after loading inbox") { sleep 1; PollManager.auto_poll } unless $opts[:no_threads] || $opts[:no_initial_poll] }
  
    if $opts[:compose]
      ComposeMode.spawn_nicely :to_default => $opts[:compose]
 @@ -263,7 +263,7 @@ begin
      when :compose
        ComposeMode.spawn_nicely
      when :poll
 -      reporting_thread("user-invoked poll") { PollManager.poll }
 +      reporting_thread("user-invoked poll") { PollManager.forced_poll }
      when :recall_draft
        case Index.num_results_for :label => :draft
        when 0
 diff --git a/bin/sup-add b/bin/sup-add
 index 50bbb29..88705e1 100644
 --- a/bin/sup-add
 +++ b/bin/sup-add
 @@ -39,6 +39,7 @@ EOS
    opt :unusual, "Do not automatically poll these sources for new messages."
    opt :labels, "A comma-separated set of labels to apply to all messages from this source", :type => String
    opt :force_new, "Create a new account for this source, even if one already exists."
 +  opt :poll_interval, "The interval (in seconds) between new message polls. The default is #{Redwood::DEFAULT_POLL_INTERVAL}.", :type => :int
  end
  
  Trollop::die "require one or more sources" if ARGV.empty?
 @@ -84,6 +85,9 @@ index.lock_or_die
  begin
    index.load_sources
  
 +  Trollop::die "The poll interval must be a positive integer." if $opts[:poll_interval] <= 0
 +  poll_interval = $opts[:poll_interval] || Redwood::DEFAULT_POLL_INTERVAL
 +
    ARGV.each do |uri|
      labels = $opts[:labels] ? $opts[:labels].split(/\s*,\s*/).uniq : []
  
 @@ -100,14 +104,14 @@ begin
          say "For SSH connections, if you will use public key authentication, you may leave the username and password blank."
          say ""
          username, password = get_login_info uri, index.sources
 -        Redwood::MBox::SSHLoader.new uri, username, password, nil, !$opts[:unusual], $opts[:archive], nil, labels
 +        Redwood::MBox::SSHLoader.new uri, username, password, nil, !$opts[:unusual], $opts[:archive], nil, poll_interval, labels
        when "imap", "imaps"
          username, password = get_login_info uri, index.sources
 -        Redwood::IMAP.new uri, username, password, nil, !$opts[:unusual], $opts[:archive], nil, labels
 +        Redwood::IMAP.new uri, username, password, nil, !$opts[:unusual], $opts[:archive], nil, poll_interval, labels
        when "maildir"
 -        Redwood::Maildir.new uri, nil, !$opts[:unusual], $opts[:archive], nil, labels
 +        Redwood::Maildir.new uri, nil, !$opts[:unusual], $opts[:archive], nil, poll_interval, labels
        when "mbox"
 -        Redwood::MBox::Loader.new uri, nil, !$opts[:unusual], $opts[:archive], nil, labels
 +        Redwood::MBox::Loader.new uri, nil, !$opts[:unusual], $opts[:archive], nil, poll_interval, labels
        when nil
          Trollop::die "Sources must be specified with an URI"
        else
 diff --git a/lib/sup.rb b/lib/sup.rb
 index 1946f3c..c27a4bc 100644
 --- a/lib/sup.rb
 +++ b/lib/sup.rb
 @@ -50,6 +50,8 @@ module Redwood
    YAML_DOMAIN = "masanjin.net"
    YAML_DATE = "2006-10-01"
  
 +  DEFAULT_POLL_INTERVAL = 300
 +
  ## record exceptions thrown in threads nicely
    def reporting_thread name
      if $opts[:no_threads]
 @@ -72,7 +74,7 @@ module Redwood
    def save_yaml_obj object, fn, safe=false
      if safe
        safe_fn = "#{File.dirname fn}/safe_#{File.basename fn}"
 -      mode = File.stat(fn) if File.exists? fn
 +      mode = File.stat(fn).mode if File.exists? fn
        File.open(safe_fn, "w", mode) { |f| f.puts object.to_yaml }
        FileUtils.mv safe_fn, fn
      else
 diff --git a/lib/sup/buffer.rb b/lib/sup/buffer.rb
 old mode 100755
 new mode 100644
 diff --git a/lib/sup/imap.rb b/lib/sup/imap.rb
 index 1d36976..8b58cba 100644
 --- a/lib/sup/imap.rb
 +++ b/lib/sup/imap.rb
 @@ -51,13 +51,13 @@ class IMAP < Source
  
    attr_accessor :username, :password
    yaml_properties :uri, :username, :password, :cur_offset, :usual,
 -                  :archived, :id, :labels
 +                  :archived, :id, :poll_interval, :labels
  
 -  def initialize uri, username, password, last_idate=nil, usual=true, archived=false, id=nil, labels=[]
 +  def initialize uri, username, password, last_idate=nil, usual=true, archived=false, id=nil, poll_interval=nil, labels=[]
      raise ArgumentError, "username and password must be specified" unless username && password
      raise ArgumentError, "not an imap uri" unless uri =~ %r!imaps?://!
  
 -    super uri, last_idate, usual, archived, id
 +    super uri, last_idate, usual, archived, id, [poll_interval, SCAN_INTERVAL].max
  
      @parsed_uri = URI(uri)
      @username = username
 diff --git a/lib/sup/maildir.rb b/lib/sup/maildir.rb
 index 584e657..ef71e0b 100644
 --- a/lib/sup/maildir.rb
 +++ b/lib/sup/maildir.rb
 @@ -12,9 +12,9 @@ class Maildir < Source
    SCAN_INTERVAL = 30 # seconds
  
    ## remind me never to use inheritance again.
 -  yaml_properties :uri, :cur_offset, :usual, :archived, :id, :labels
 -  def initialize uri, last_date=nil, usual=true, archived=false, id=nil, labels=[]
 -    super uri, last_date, usual, archived, id
 +  yaml_properties :uri, :cur_offset, :usual, :archived, :id, :poll_interval, :labels
 +  def initialize uri, last_date=nil, usual=true, archived=false, id=nil, poll_interval=nil, labels=[]
 +    super uri, last_date, usual, archived, id, [poll_interval, SCAN_INTERVAL].max
      uri = URI(Source.expand_filesystem_uri(uri))
  
      raise ArgumentError, "not a maildir URI" unless uri.scheme == "maildir"
 diff --git a/lib/sup/mbox/loader.rb b/lib/sup/mbox/loader.rb
 index 7fe9129..44317d5 100644
 --- a/lib/sup/mbox/loader.rb
 +++ b/lib/sup/mbox/loader.rb
 @@ -9,7 +9,7 @@ class Loader < Source
    attr_accessor :labels
  
    ## uri_or_fp is horrific. need to refactor.
 -  def initialize uri_or_fp, start_offset=nil, usual=true, archived=false, id=nil, labels=[]
 +  def initialize uri_or_fp, start_offset=nil, usual=true, archived=false, id=nil, poll_interval=nil, labels=[]
      @mutex = Mutex.new
      @labels = ((labels || []) - LabelManager::RESERVED_LABELS).uniq.freeze
  
 @@ -26,7 +26,7 @@ class Loader < Source
        @path = uri_or_fp.path
      end
  
 -    super uri_or_fp, start_offset, usual, archived, id
 +    super uri_or_fp, start_offset, usual, archived, id, poll_interval
    end
  
    def file_path; @path end
 diff --git a/lib/sup/modes/poll-mode.rb b/lib/sup/modes/poll-mode.rb
 index 5849f3e..5521cdc 100644
 --- a/lib/sup/modes/poll-mode.rb
 +++ b/lib/sup/modes/poll-mode.rb
 @@ -10,11 +10,11 @@ class PollMode < LogMode
      self << s + "\n"
    end
  
 -  def poll
 +  def poll sources
      puts unless @new
      @new = false
      puts "Poll started at #{Time.now}"
 -    PollManager.do_poll { |s| puts s }
 +    PollManager.do_poll_sources(sources) { |s| puts s }
    end
  end
  
 diff --git a/lib/sup/poll.rb b/lib/sup/poll.rb
 index d32c893..1ff8014 100644
 --- a/lib/sup/poll.rb
 +++ b/lib/sup/poll.rb
 @@ -12,13 +12,22 @@ Variables:
  EOS
  
    HookManager.register "before-poll", <<EOS
 +Executes immediately before any poll for new messages commences.
 +Variables:
 +                 accts: an array of source URIs that will be polled
 +EOS
 +
 +  HookManager.register "before-acct-poll", <<EOS
  Executes immediately before a poll for new messages commences.
 -No variables.
 +Variables:
 +                  acct: the URI of the account being polled
  EOS
  
 -  HookManager.register "after-poll", <<EOS
 -Executes immediately after a poll for new messages completes.
 +  HookManager.register "after-acct-poll", <<EOS
 +Executes immediately after a poll for new messages completes. The statistics are
 +only for the account specified.
  Variables:
 +                  acct: the URI of the account being polled
                     num: the total number of new messages added in this poll
               num_inbox: the number of new messages added in this poll which
                          appear in the inbox (i.e. were not auto-archived).
 @@ -28,14 +37,25 @@ num_inbox_total_unread: the total number of unread messages in the inbox
                          only those messages appearing in the inbox
  EOS
  
 -  DELAY = 300
 +  HookManager.register "after-poll", <<EOS
 +Executes immediately after any poll for new messages completes. The statistics
 +are for all sources polled.
 +Variables:
 +                 accts: an array of source URIs that were polled
 +                   num: the total number of new messages added in this poll
 +             num_inbox: the number of new messages added in this poll which
 +                        appear in the inbox (i.e. were not auto-archived).
 +num_inbox_total_unread: the total number of unread messages in the inbox
 +         from_and_subj: an array of (from email address, subject) pairs
 +   from_and_subj_inbox: an array of (from email address, subject) pairs for
 +                        only those messages appearing in the inbox
 +EOS
  
    def initialize
      @mutex = Mutex.new
      @thread = nil
 -    @last_poll = nil
      @polling = false
 -    
 +
      self.class.i_am_the_instance self
    end
  
 @@ -44,47 +64,113 @@ EOS
      b
    end
  
 -  def poll
 +  def start
 +    ## Periodically calls auto_poll, effectively automatically polling
 +    ## in the background.
 +    ##
 +    ## auto_poll is called more frequently than the smallest poll
 +    ## interval. auto_poll is not guaranteed to be called in multiples
 +    ## of poll intervals. I.e., don't rely on auto_poll to be called
 +    ## every x seconds: when auto_poll is called assume an
 +    ## indeterminate amount of time has passed.
 +    @min_poll_interval = Index.usual_sources.collect{|s| s.poll_interval}.select{|pi| !pi.nil?}.min || Redwood::DEFAULT_POLL_INTERVAL
 +
 +    Redwood::log "Background poll interval is #{@min_poll_interval} seconds."
 +
 +    @thread = Redwood::reporting_thread("periodic poll") do
 +      while true
 +        Redwood::log "Sleeping for #{@min_poll_interval / 2} seconds."
 +
 +        sleep @min_poll_interval / 2
 +        auto_poll
 +      end
 +    end
 +  end
 +
 +  def stop
 +    @thread.kill if @thread
 +    @thread = nil
 +  end
 +
 +  def auto_poll
 +    ## This method is called by the thread spawned in start. It
 +    ## collects the sources that should be polled
 +    ## (source.poll_interval has expired or the source has never been
 +    ## polled) and polls them.
 +    ##
 +    ## Returns an array [# of new messages, # of new messages to
 +    ## inbox, new message subject/name array pairs, subject/name array
 +    ## pairs loaded to inbox]
 +
 +    sources_to_poll = []
 +    @mutex.synchronize do # William, do we need to synchronize here?
 +      begin
 +        sources_to_poll = Index.usual_sources.select do |source|
 +          begin
 +            source.last_poll.nil? || (Time.now - source.last_poll) >= source.poll_interval
 +          end
 +        end
 +      end
 +    end
 +
 +    poll_sources sources_to_poll
 +  end
 +
 +  def forced_poll
 +    ## This method is called when the user explicitly requests a poll.
 +    ##
 +    ## Returns an array [# of new messages, # of new messages to
 +    ## inbox, new message subject/name array pairs, subject/name array
 +    ## pairs loaded to inbox]
 +    poll_sources Index.usual_sources
 +  end
 +
 +  def poll_sources sources
 +    ## Polls the given sources. Clients of PollManager should call
 +    ## this method, not do_poll_sources or poll_source. (Well, only
 +    ## PollMode should call do_poll_sources)
 +    ##
 +    ## Returns an array [# of new messages, # of new messages to
 +    ## inbox, new message subject/name array pairs, subject/name array
 +    ## pairs loaded to inbox]
      return if @polling
      @polling = true
 -    HookManager.run "before-poll"
 +
 +    source_uris = sources.map{|s| s.uri}
 +
 +    HookManager.run "before-acct-poll", :accts => source_uris
  
      BufferManager.flash "Polling for new messages..."
 -    num, numi, from_and_subj, from_and_subj_inbox = buffer.mode.poll
 +    num, numi, from_subj, from_subj_inbox = buffer.mode.poll sources
      if num > 0
        BufferManager.flash "Loaded #{num.pluralize 'new message'}, #{numi} to inbox." 
      else
        BufferManager.flash "No new messages." 
      end
  
 -    HookManager.run "after-poll", :num => num, :num_inbox => numi, :from_and_subj => from_and_subj, :from_and_subj_inbox => from_and_subj_inbox, :num_inbox_total_unread => lambda { Index.num_results_for :labels => [:inbox, :unread] }
 +    HookManager.run "after-poll", :accts => source_uris, :num => num, :num_inbox => numi, :from_and_subj => from_subj, :from_and_subj_inbox => from_subj_inbox, :num_inbox_total_unread => lambda { Index.num_results_for :labels => [:inbox, :unread] }
  
      @polling = false
 -    [num, numi]
 -  end
 -
 -  def start
 -    @thread = Redwood::reporting_thread("periodic poll") do
 -      while true
 -        sleep DELAY / 2
 -        poll if @last_poll.nil? || (Time.now - @last_poll) >= DELAY
 -      end
 -    end
 +    [num, numi, from_subj, from_subj_inbox]
    end
  
 -  def stop
 -    @thread.kill if @thread
 -    @thread = nil
 -  end
 +  def do_poll_sources sources, &block
 +    ## Polls each source, keeping track of vital statistics (number
 +    ## loaded, name/e-mail pairs, &c.) about the poll results.
 +    ##
 +    ## We need explicit access to the block so that we can pass it to
 +    ## poll_source.
 +    ##
 +    ## Returns an array [total # of new messages, total # of new
 +    ## messages to inbox, all new message subject/name array pairs,
 +    ## all subject/name array pairs loaded to inbox]
  
 -  def do_poll
      total_num = total_numi = 0
 -    from_and_subj = []
 -    from_and_subj_inbox = []
 +    total_from_and_subj = []
 +    total_from_and_subj_inbox = []
  
      @mutex.synchronize do
 -      Index.usual_sources.each do |source|
 -#        yield "source #{source} is done? #{source.done?} (cur_offset #{source.cur_offset} >= #{source.end_offset})"
 +      sources.each do |source|
          begin
            yield "Loading from #{source}... " unless source.done? || source.has_errors?
          rescue SourceError => e
 @@ -93,32 +179,57 @@ EOS
            next
          end
  
 -        num = 0
 -        numi = 0
 -        add_messages_from source do |m, offset, entry|
 -          ## always preserve the labels on disk.
 -          m.labels = entry[:label].split(/\s+/).map { |x| x.intern } if entry
 -          yield "Found message at #{offset} with labels {#{m.labels * ', '}}"
 -          unless entry
 -            num += 1
 -            from_and_subj << [m.from.longname, m.subj]
 -            if m.has_label?(:inbox) && ([:spam, :deleted, :killed] & m.labels).empty?
 -              from_and_subj_inbox << [m.from.longname, m.subj]
 -              numi += 1 
 -            end
 -          end
 -          m
 -        end
 -        yield "Found #{num} messages, #{numi} to inbox." unless num == 0
 +        HookManager.run "before-acct-poll", :acct => source.uri
 +
 +        BufferManager.flash "Polling #{source.uri} for new messages..."
 +        num, numi, from_and_subj, from_and_subj_inbox = poll_source source, &block
 +
          total_num += num
          total_numi += numi
 +        total_from_and_subj += from_and_subj
 +        total_from_and_subj_inbox += from_and_subj_inbox
 +
 +        HookManager.run "after-acct-poll", :acct => source.uri, :num => num, :num_inbox => numi, :from_and_subj => from_and_subj, :from_and_subj_inbox => from_and_subj_inbox, :num_inbox_total_unread => lambda { Index.num_results_for :labels => [:inbox, :unread] }
        end
 +    end
 +
 +    [total_num, total_numi, total_from_and_subj, total_from_and_subj_inbox]
 +  end
  
 -      yield "Done polling; loaded #{total_num} new messages total"
 -      @last_poll = Time.now
 -      @polling = false
 +  def poll_source source
 +    ## Polls the given source for new messages.
 +    ##
 +    ## @mutex must be held before calling! See do_poll_sources.
 +    ##
 +    ## Returns an array [# of new messages, # of new messages to
 +    ## inbox, new message subject/name array pairs, subject/name array
 +    ## pairs loaded to inbox]
 +
 +    num = 0
 +    numi = 0
 +    from_and_subj = []
 +    from_and_subj_inbox = []
 +
 +    add_messages_from source do |m, offset, entry|
 +      ## always preserve the labels on disk.
 +      m.labels = entry[:label].split(/\s+/).map { |x| x.intern } if entry
 +      yield "Found message at #{offset} with labels {#{m.labels * ', '}}"
 +      unless entry
 +        num += 1
 +        from_and_subj << [m.from.longname, m.subj]
 +        if m.has_label?(:inbox) && ([:spam, :deleted, :killed] & m.labels).empty?
 +          from_and_subj_inbox << [m.from.longname, m.subj]
 +          numi += 1
 +        end
 +      end
 +      m
      end
 -    [total_num, total_numi, from_and_subj, from_and_subj_inbox]
 +
 +    yield "For source #{source.uri}, found #{num} messages, #{numi} to inbox." unless num == 0
 +
 +    source.last_poll = Time.now
 +
 +    [num, numi, from_and_subj, from_and_subj_inbox]
    end
  
    ## this is the main mechanism for adding new messages to the
 diff --git a/lib/sup/source.rb b/lib/sup/source.rb
 index 6510aae..fd8d381 100644
 --- a/lib/sup/source.rb
 +++ b/lib/sup/source.rb
 @@ -62,10 +62,10 @@ class Source
    ## dirty? means cur_offset has changed, so the source info needs to
    ## be re-saved to sources.yaml.
    bool_reader :usual, :archived, :dirty
 -  attr_reader :uri, :cur_offset
 -  attr_accessor :id
 +  attr_reader :uri, :cur_offset, :poll_interval
 +  attr_accessor :id, :last_poll
  
 -  def initialize uri, initial_offset=nil, usual=true, archived=false, id=nil
 +  def initialize uri, initial_offset=nil, usual=true, archived=false, id=nil, poll_interval=nil
      raise ArgumentError, "id must be an integer: #{id.inspect}" unless id.is_a? Fixnum if id
  
      @uri = uri
 @@ -73,6 +73,8 @@ class Source
      @usual = usual
      @archived = archived
      @id = id
 +    @poll_interval = poll_interval || Redwood::DEFAULT_POLL_INTERVAL #seconds
 +    @last_poll = nil
      @dirty = false
    end
  
 -- 
 1.5.4
 
 
 From 5srmspw02@sneakemail.com  Sat Mar 29 12:41:14 2008
 From: 5srmspw02@sneakemail.com (Guarded Identity)
 Date: Sat, 29 Mar 2008 11:41:14 -0500
 Subject: [sup-talk] Different Key Bindings (was: New User Questions)
 In-Reply-To: <1206699001-sup-9770@tomsk>
 References: <1206662265-sup-7439@cabinet> <1206699001-sup-9770@tomsk>
 Message-ID: <27382-65971@sneakemail.com>
 
 Excerpts from Marcus Williams marcus-sup-at-bar-coded.net |Sup_Talk|'s message of Fri Mar 28 05:33:45
 > > 8. Is there any way to specify custom keybindings in a configuration
 > > file, or would it require hacking the source?
 >
 > Hack the source afaik
 
 I know I'd /definitely/ be interested in this kind of thing.  For the most
 part, I'm really happy with current key bindings, but there's a few that bug
 me.  For instance.  I wish 'l' and 'L' were reversed.  My rationale is that I
 don't like the unshifted characters to do persistent things (just in-case I hit
 the wrong key).  Maybe it's just me, but I hit 'l' sometimes, briefly believe
 that Sup is hung, and then realize that I'm editing labels, and then Ctrl-G'd
 out before accidentally committing the junk labels.
 
 -Sukant
 
 From chrisw@rice.edu  Sat Mar 29 23:15:37 2008
 From: chrisw@rice.edu (Christopher Warrington)
 Date: Sat, 29 Mar 2008 22:15:37 -0500
 Subject: [sup-talk] [PATCH] polling is now done per source
 In-Reply-To: <1206786725-5456-1-git-send-email-chrisw@rice.edu>
 References: <1206784680-sup-3052@chris-tablet>
 	<1206786725-5456-1-git-send-email-chrisw@rice.edu>
 Message-ID: <1206846791-sup-7998@chris-tablet>
 
 Excerpts from Christopher Warrington's message of Sat Mar 29 05:32:05 -0500 2008:
 > Each source has a poll_interval property. This property is used to
 > determine whether the source should be polled. The user can still
 > for a poll of all sources.
 
 This is the fixed patch. I've been running it all day now with Maildir
 sources and it seems to be working. I'd love some feedback, especially
 from those who use mbox or IMAP.
 
 -- 
 Christopher Warrington <chrisw at rice.edu>