Hi Everyone, I'm shortly going to send a set of patches to commit the neomutt version of the sidebar (with some fixes and cleanup).
First, let me say thanks to Rich for taking a large amount of time to clean up, fix, and document the patch. Also, for working to rebase the patch on top of the mutt_window changes. This resulted in a patch that, while still large, isn't nearly as invasive as it was. The patch has quite a history, and is duly controversial. Many people say they can't use Mutt without it, while others don't want it anywhere near their screen. I can understand the latter group's feelings, so first let me note that Rich has made it a compile-time configurable option, defaulting off. While I would like to commit the current set of patches now, I am not done with the changes I'd like to make before the next release. There are still several things that are goofy with the patch, and I plan on fixing things over the next few weeks. My current todo list is: * Separate the "extended buffy" logic from the sidebar. Right now, enabling the sidebar also automatically enables logic that counts the number of new/flagged/total messages. Naturally there is a performance impact here. I want to create a new boolean option and a new separate timeout that can be enabled/disabled orthogonally, and plan on defaulting the extended buffy logic to off. * Separate this "extended buffy" logic from the "new mail" logic. The sidebar changes currently (incorrectly) conflate this in a few places. The number of unread messages is different from whether a mailbox has "new" mail. I also believe the old/new logic is not currently handled correctly, but need to look into it some more. * Create a "new mail" sidebar format string, so the sidebar can be used to just show mailboxes with "new mail", without enabling the "extended buffy" logic. Change the sidebar to use this by default. * Disentangle the sidebar from directly manipulating the BUFFY pointers. Changing the sort order in the sidebar actually rearranges the mailboxes order. Instead, the sidebar should just use an array, or create a container structure with its own pointers. * Separate the new mbox PEEK code out from the sidebar #ifdefs. * Figure out if the "realpath" code added is actually needed. I nearly pulled this out, but decided to let it in for now and review it later. I'm sure I'll find more stuff as I go, but this gives you an idea, and a promise, that I'm not done working on this, even though I'd like to get the current patches in the tree now. -- Kevin J. McCarthy GPG Fingerprint: 8975 A9B3 3AA3 7910 385C 5308 ADEF 7684 8031 6BDA
signature.asc
Description: PGP signature
