1 | TODO: mkv |
1 | TODO: mkv |
2 | TODO/FIXME: pixmap-bg-trick sometimes doesn't work (race?) |
2 | TODO/FIXME: pixmap-bg-trick sometimes doesn't work (race?) |
3 | TODO/FIXME: remove event does not recalculate schanuzer content size |
3 | TODO/FIXME: remove event does not recalculate schanuzer content size |
4 | TODO: with manual placement, initial contents aren't displayed |
4 | TODO: with manual placement, initial contents aren't displayed |
5 | TODO: update a file that has been removed etc. |
5 | TODO: update a file that has been removed etc. |
|
|
6 | TODO: note to self, the reason mplayer hangs during long copies is that execve hangs, because it closes an fd to the target file. |
6 | |
7 | |
|
|
8 | TODO: metacluster |
7 | TODO: mädhcne utf-8 bug move to ERROR is a directory |
9 | TODO: "mädhcne" utf-8 bug move to ERROR is a directory |
|
|
10 | Pango-WARNING **: Invalid UTF-8 string passed to pango_layout_set_text() at /opt/bin/cv line 224. |
|
|
11 | - work around parser bug in 5.18.1 (-a). |
|
|
12 | - use ffprobe instead of mplayer to probe video size/type. |
|
|
13 | - switch to using AnyEvent::Fork as provider for the processpool, which ought |
|
|
14 | to save gobs of memory. |
|
|
15 | - much improved async communications with worker processes. |
|
|
16 | - batch expose events together, in an attempt to reduce the amount of |
|
|
17 | expose races in gtk+ (does not fix scrolling races). |
|
|
18 | - create #cpu + 1 worker jobs now, not 1.5 * #cpu + 1 as before. |
|
|
19 | - better shell-quoting: use ""-style for utf-8 filenames and paste as unicode, |
|
|
20 | use bash's $'' syntax for other filenames. |
8 | - no longer pass a-z from viewer to schnauzer. |
21 | - no longer pass a-z from viewer to schnauzer. |
9 | - properly serialise jobs by priority, instead of executing |
22 | - properly serialise jobs by priority, instead of executing |
10 | jobs in parallel on the same path *sometimes*. |
23 | jobs in parallel on the same path *sometimes*. |
11 | - escape select-by-prefix menu item labels. |
24 | - escape select-by-prefix menu item labels. |
12 | - work around more perl unicode bugs. |
25 | - work around more perl unicode bugs. |
… | |
… | |
21 | - do not enter the last directory on the commandline, if more |
34 | - do not enter the last directory on the commandline, if more |
22 | than one argument is specified. |
35 | than one argument is specified. |
23 | - started using AnyEvent. |
36 | - started using AnyEvent. |
24 | - space by default activates first file, or first directory, if no files |
37 | - space by default activates first file, or first directory, if no files |
25 | (as opposed to first file or last directory in previous versions). |
38 | (as opposed to first file or last directory in previous versions). |
|
|
39 | - fix race condition where the schnauzer might stay non-sensitive |
|
|
40 | on fast directory changes. |
|
|
41 | - allow reversal of rotation via .cvrc $REVERSE_ROTATION = 1. |
|
|
42 | - delay initialisation of Gtk2::CV::Schnauzer until first use. |
|
|
43 | - fix a runtime error when starting namecluster with newer gtk's. |
|
|
44 | - do not rely on /dev/shm being world-writable, use |
|
|
45 | /run/shm, /dev/shm and /tmp, in this order. |
|
|
46 | - expect mplayer to sometimes generate more than 2 frames when 1 is |
|
|
47 | requested. |
|
|
48 | - tune video thumbnail generation a bit. |
|
|
49 | - more fixups for "file -i" failing. |
|
|
50 | - properly follows symlinks instead of bailing out. |
|
|
51 | - support perl multicore specification (http://perlmulticore.schmorp.de/) |
|
|
52 | for jpeg loading and image transformations, although nothing takes |
|
|
53 | advantage of this *yet*. |
|
|
54 | - shift-up/down in schnauzer tries to move cluster cursor. |
26 | |
55 | |
27 | 1.56 Thu Oct 28 14:13:36 CEST 2010 |
56 | 1.56 Thu Oct 28 14:13:36 CEST 2010 |
28 | - try an ugly workaround against Gtk's CONTINUED brokenness |
57 | - try an ugly workaround against Gtk's CONTINUED brokenness |
29 | with regards to non-ascii filenames :(. |
58 | with regards to non-ascii filenames :(. |
30 | - REALLY hide paths that are in the process of being removed |
59 | - REALLY hide paths that are in the process of being removed |