| … | |
… | |
| 18 | WISH: load system-wide config file even if we don't have one |
18 | WISH: load system-wide config file even if we don't have one |
| 19 | WISH: look into XAddConnectionWatch, does anybody need that? |
19 | WISH: look into XAddConnectionWatch, does anybody need that? |
| 20 | DUMB: support tex fonts |
20 | DUMB: support tex fonts |
| 21 | |
21 | |
| 22 | TODO: /*TODO: the xlceararea/xfillrectangle below don't take scroll offste into account, ask mikachu for details */ |
22 | TODO: /*TODO: the xlceararea/xfillrectangle below don't take scroll offste into account, ask mikachu for details */ |
|
|
23 | 00:25:39 <Mikachu> put your prompt around the middle row in the terminal |
|
|
24 | and while sleep 0.1; do printf '\e[J'; done |
|
|
25 | 00:26:31 <schmorp> i assume scorllttyoutput or so wouldinterfere? |
| 23 | |
26 | |
| 24 | TODO: artificially enlargen wide characters at end to correctly cut&paste. |
27 | |
| 25 | TODO: cursor over overlays, when focus change? |
28 | TODO: cursor over overlays, when focus change? |
| 26 | TODO: focus leave == reset urgent hint? |
29 | - the aterm code now frees the as visual and image manager objects |
|
|
30 | when a window was closed. |
|
|
31 | - do not (wrongly) adjust the virtual line length inside |
|
|
32 | scr_insdel_chars/ERASE (Miroslav Lichvar). |
|
|
33 | - fix an issue where wide tab characters caused character shifts |
|
|
34 | when part of them was deleted (tracked down by Miroslav Lichvar). |
|
|
35 | - artificially enlargen the previous character at the end of a line |
|
|
36 | when trying to output a character that doesn't fit. looks ugly, but |
|
|
37 | makes copy&paste work. |
| 27 | - remove bogus "setuid/setgid security issues" from rxvt.1.pod, they |
38 | - remove bogus "setuid/setgid security issues" from rxvt.1.pod, they |
| 28 | no longer apply. |
39 | no longer apply. |
|
|
40 | - the urgency hint is now cleared on both focus in and focus out. |
|
|
41 | - cache urgency hint locally to avoid server rtt on every focus change. |
| 29 | - perl 5.10 needs PERL_SYS_INIT3 on hppa. |
42 | - perl 5.10 needs PERL_SYS_INIT3 on hppa. |
| 30 | - ignore byte-order marks and do not treat them like combining characters. |
43 | - ignore byte-order marks and do not treat them like combining characters. |
| 31 | - fix build issue when CURSOR_BLINK is not defined (Emanuele Giaquinta). |
44 | - fix build issue when CURSOR_BLINK is not defined (Emanuele Giaquinta). |
| 32 | - do some µ-optimisations to the character fast path. |
45 | - do some µ-optimisations to the character fast path. |
|
|
46 | - use a less cunning, but more robust algorithm to avoid one terminal |
|
|
47 | monopolising the whole process by outputting a lot of text. |
|
|
48 | - try to work around linux first breaking sched_yield and now breaking |
|
|
49 | the only known workaround. |
|
|
50 | - new option -icon to set _NET_WM_ICON (based on patch by Frank Schmitt). |
|
|
51 | - initialise the as visual only on demand, instead of unconditionally, |
|
|
52 | to save memory in the common case of not using visual gimmicks. |
|
|
53 | - try to work around a race between urxvt and wm's on programmatic |
|
|
54 | resizes by temporarily disabling size hints. |
|
|
55 | - implement the DECSTR soft reset sequence and use it for tput init, |
|
|
56 | and make hard RIS and DECSTR both reset more private modes, such |
|
|
57 | as mouse reporting. |
|
|
58 | - redundantly clear mouse reporting and a few other states in |
|
|
59 | tput init/tput reset, for older urxvts. |
|
|
60 | - partial/full clear screens did cause flickering and possibly pixel |
|
|
61 | droppings when urxvt viewed the scorllback buffer. |
| 33 | |
62 | |
| 34 | 9.05 Sun Jun 15 20:09:20 CEST 2008 |
63 | 9.05 Sun Jun 15 20:09:20 CEST 2008 |
| 35 | - new option --cd/chdir to set the starting working directory. |
64 | - new option --cd/chdir to set the starting working directory. |
| 36 | - tabbed extension now starts in the "correct" working directory. |
65 | - tabbed extension now starts in the "correct" working directory. |
| 37 | - work around fedora 9 providing isastream but not the relevant |
66 | - work around fedora 9 providing isastream but not the relevant |
