aboutsummaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAge
...
* Status line fixes: don't truncate status-right now the length calculation isNicholas Marriott2009-06-26
| | | | | | done for UTF-8, limit to the maximum length correctly when printing, and always print a space even if the left string is longer than the width available.
* Remove some dead assignments, found by sthen with clang.no_author2009-06-26
|
* #ifndef nitems.Nicholas Marriott2009-06-26
|
* Update TODO.Nicholas Marriott2009-06-25
|
* vis compat stuff.Nicholas Marriott2009-06-25
|
* Start of portability update: bitstring.h.Nicholas Marriott2009-06-25
|
* Lose -V.Nicholas Marriott2009-06-25
|
* Skip 8 line diffs ($Id$ change only).Nicholas Marriott2009-06-25
|
* Comparison script.Nicholas Marriott2009-06-25
|
* libgen.h.Nicholas Marriott2009-06-25
|
* More diff-to-OpenBSD reduction. Move a lot of compat stuff into compat.h.Nicholas Marriott2009-06-25
|
* Whitespace and more syncing.Nicholas Marriott2009-06-25
|
* Don't need linker magic anymore.Nicholas Marriott2009-06-25
|
* No more xmalloc-debug.Nicholas Marriott2009-06-25
|
* Fix rejects.Nicholas Marriott2009-06-25
|
* Sync this properly.Nicholas Marriott2009-06-25
|
* Restore $Id$ and add script to do so.Nicholas Marriott2009-06-25
|
* If getcwd() fails, use the user's home directory, or /, instead of failing withNicholas Marriott2009-06-25
| | | | | an error.
* lines variable can be -1 (to be ignored), so should be signed. Found by lint.Nicholas Marriott2009-06-25
|
* Remove error about using -L and -S together which was never displayed asNicholas Marriott2009-06-25
| | | | | | logging wasn't yet enabled, was unnecessary, and contradicted the man page which says using -S will cause -L to be ignored.
* tmux doesn't and won't need syslog logging, so remove it and some other unusedNicholas Marriott2009-06-25
| | | | | functions found by lint.
* Miscellaneous unused functions, including one which was basically aNicholas Marriott2009-06-25
| | | | | duplicate. Found by lint.
* Nuke unused buffer functions. Found by lint.Nicholas Marriott2009-06-25
| | | | | Also remove some old debug output which was #if 0.
* Unused prototypes. Found by lint, no binary change.Nicholas Marriott2009-06-25
|
* Unused variables. Found by lint, no binary change.Nicholas Marriott2009-06-25
|
* Now that a UTF-8-capable puts function exists, use it for printing strings inNicholas Marriott2009-06-25
| | | | | | choice/more modes - lines with UTF-8 now display properly in find-window results.
* Fix a type mismatch warning in assignment.Nicholas Marriott2009-06-25
|
* Change find-window and monitor-content to use fnmatch(3). For convenience andNicholas Marriott2009-06-25
| | | | | compatibility, *s are implicitly added at the start and end of the pattern.
* Add a dedicated function to convert a line into a string and use it to ↵Nicholas Marriott2009-06-25
| | | | simplify the search window function.
* Change to match xterm's resizing behaviour, including pushing lines into theNicholas Marriott2009-06-25
| | | | | | history from the top when reducing vertical size and pulling them again when increasing size.
* Make remain-on-exit work again when there is only one pane left, which wasNicholas Marriott2009-06-25
| | | | | | broken sometime during the pane/layout changes. Reported/tested by Iain Morgan, thanks.
* Trying to predict the cursor position for UTF-8 output in the same way as forNicholas Marriott2009-06-25
| | | | | | | normal eight-bit output is wrong, separate it into a different function. Fixes spacing when mixing UTF-8 with some escape sequences, notably the way w3m does it.
* Constify utf8_width() function argument.Nicholas Marriott2009-06-25
|
* If a pane is "zombified" (remain-on-exit flag), don't continue to queue key andNicholas Marriott2009-06-25
| | | | | mouse input for it (otherwise they are processed after respawn).
* LC_ALL overrides LC_CTYPE and LANG. Comment was correct but the codeNicholas Marriott2009-06-25
| | | | | wrong. Pointed out by Hannah Schroeter, thanks.
* Terminate cwd buffer before running xstrdup on it.Nicholas Marriott2009-06-25
|
* Check the first of LC_CTYPE, LC_ALL and LANG, rather than just the last, whenNicholas Marriott2009-06-25
| | | | | trying to decide about UTF-8, and use strcasestr. Reported by Geert Hendrickx.
* strdup the input to putenv to avoid in one case passing a string that is laterNicholas Marriott2009-06-25
| | | | | freed and in the other const strings.
* Call setproctitle earlier in the client, and include the socket name. Makes itNicholas Marriott2009-06-25
| | | | | easier to match client to server in ps/pgrep when using several servers.
* Print a better message than '(null)' if no command is specified ("tmux \;").Nicholas Marriott2009-06-25
|
* Zero the password given to -U in the client as well.Nicholas Marriott2009-06-25
|
* If the prompt is hidden or a password is sent with -U, zero it before freeingNicholas Marriott2009-06-25
| | | | | it.
* Implement the CBT (backward tab) sequence (\033[Z).Nicholas Marriott2009-06-25
|
* Proper support for tab stops (\033H etc), using a bitstring(3). Makes anotherNicholas Marriott2009-06-25
| | | | | vttest test happy.
* Okay, so I screwed up when testing this, doh. Unbreak so that CAN/SUB actuallyNicholas Marriott2009-06-25
| | | | | do cancel the sequence, and tweak to make the code more clear.
* Oops, CAN and SUB should abort the sequence (return to first state), not remainNicholas Marriott2009-06-25
| | | | | in the same state.
* More input compatibility love. Support C0 within escape sequences, and the C0Nicholas Marriott2009-06-25
| | | | | | control character VT (vertical tab, \013), which is treated as LF like VT102. Makes another vttest happy.
* Support insert mode by using insert character to shift the cells before writingNicholas Marriott2009-06-25
| | | | | as normal.
* Implement the DEC alignment test. With the last change this is enough for theNicholas Marriott2009-06-25
| | | | | | first cursor test in vttest (in ports) to pass; it still shops a few more problems though.
* Fix some miscalculations when clearing to start of screen: the number of linesNicholas Marriott2009-06-25
| | | | | to the cursor is cy not cy - 1, and the current cursor cell should be included.
generated by cgit (git 2.34.1) at 2026-02-08 15:10:53 +0000