summaryrefslogtreecommitdiff
path: root/usr.bin/vi
AgeCommit message (Collapse)Author
2015-09-14Avoid .Ns right after .Pf, it's pointless.Ingo Schwarze
In some cases, do additional cleanup in the immediate vicinity.
2015-07-07Fix a regression caused by timespec changes when vi is run withoutTodd C. Miller
a file to edit. Based on a diff from Patrick Keshishian.
2015-04-24struct timespec/clock_gettime(3) conversion for vi(1).Bryan Steele
ok guenther@
2015-04-21init both fds passed to pipe as -1 instead of initing one twiceJonathan Gray
ok deraadt@ guenther@ miod@ millert@
2015-04-19Don't lock the file for "vi -R" or "view". OK deraadt@Todd C. Miller
2015-04-10This changes vi to use resizeterm(3) instead of reinitializing curses onBryan Steele
window resizes, which was leaking massive amounts of memory. Try observing vi in top(1) and while resizing the window a few times before and aftering applying this diff.. Also some more comment cleanup and another memory leak.. From github.com/lichray/nvi2 879d2ad6dd4a4343eb0a588ebfe637e1c9845bc4 a8c38480adb030a05bbb2aafec6067dd65d8c2eb ok millert@
2015-03-29Remove SA_INTERRUPT, HISTORIC_PRACTICE, and HISTORICAL_PRACTICE usingBrian Callahan
unifdef. It seems clear that no one was using these (SA_INTERRUPT didn't even build the other way). Tweak comments as appropriate. ok brynet@
2015-03-28vi was using two separate isblank functions: one defined in <ctype.h> andBrian Callahan
the other #defined in common/key.h. There is no reason to have both. For consistency use the isblank function from <ctype.h>, remove the #define in common/key.h, and add #include <ctype.h> to the files that were missing the header. ok brynet@
2015-03-27Some vi cleanup, unifdef's some signal blocking code that has never beenBryan Steele
enabled in our tree, also removes some stragglers from a global struct referencing nonexistent Tcl/TK and "IP support". And finally.. deletes an empty file missed by earlier cleanup by bentley@. Inspired by and ok bcallah@
2015-03-17Don't use the wrong escape for < and >. Tweak wording to match the page.Anthony J. Bentley
From Theo Buehler.
2015-03-13remove the first comma from constructs like ", and," and ", or,": you can useJason McIntyre
"and" and "or" to join sentence clauses, and you can use commas, but both hinders reading;
2015-03-10Display "Search wrapped" even when searching from the end of the file.Anthony J. Bentley
From Zhihao Yuan. ok millert@ miod@
2015-02-28Reduce usage of predefined strings in manpages.Anthony J. Bentley
Predefined strings are not very portable across troff implementations, and they make the source much harder to read. Usually the intended character can be written directly. No output changes, except for two instances where the incorrect escape was used in the first place. tweaks + ok schwarze@
2015-02-06Do not rely on unspecified behavior for the size_t overflow check.Todd C. Miller
OK miod@
2015-01-16Replace <sys/param.h> with <limits.h> and other less dirty headers whereTheo de Raadt
possible. Annotate <sys/param.h> lines with their current reasons. Switch to PATH_MAX, NGROUPS_MAX, HOST_NAME_MAX+1, LOGIN_NAME_MAX, etc. Change MIN() and MAX() to local definitions of MINIMUM() and MAXIMUM() where sensible to avoid pulling in the pollution. These are the files confirmed through binary verification. ok guenther, millert, doug (helped with the verification protocol)
2014-11-20Remove the vi perl api.Anthony J. Bentley
There have been various build errors for coming on two years now. It doesn't work and nobody has expressed any interest in saving it. From natano; ok millert@
2014-11-19Remove ifdef checks for LIBRARY.Anthony J. Bentley
It is undocumented and triggers the same conditional inclusions as PURIFY does. From Martin Natano.
2014-11-14from natano:Ted Unangst
_PATH_BSHELL, _PATH_SENDMAIL, _PATH_TMP and _PATH_TTY are defined in <paths.h> and _PATH_SYSV_TTY is unused. All of them can be removed from pathnames.h. The other defines can be made unconditionally.
2014-11-14from natano:Ted Unangst
The 'tcl' command in vi does nothing, except to print the message "Vi was not loaded with a Tcl interpreter". Printing the standard message for unknown commands would be equally descriptive with the benefit of reducing code size.
2014-11-14from natano:Ted Unangst
The vi editor contains code for two different file locking methods - one using flock(), the other using fcntl(). The fcntl method is unused and has severe limitations (as described in a code comment). Let's remove it for sake of readibility.
2014-11-12Remove more portability bits for older systems; from Martin NatanoTodd C. Miller
2014-11-12Ansify vi.Anthony J. Bentley
ok bcallah@ millert@
2014-11-10remove various bits of autoconf cruft. from Martin NatanoTed Unangst
2014-11-10remove ipc leftovers. from Martin NatanoTed Unangst
2014-11-10remove old, unnecessary compat code. from Martin NatanoTed Unangst
2014-11-06Clean up unused header files and docs referring to them.Anthony J. Bentley
From Martin Natano. ok nicm@
2014-11-06Remove old curses support in vi.Anthony J. Bentley
ok nicm@ millert@ plus a comment tweak from nicm@
2014-10-14create a REALLOCARRAY macro, and use it where it gives us overflowTheo de Raadt
protection for free ok guenther
2014-10-08bump max columns out to 768. screens are getting bigger...David Gwynne
ok deraadt@
2014-09-09We no longer need to convert "\<" and "\>" to "[[:<:]]" and "[[:>:]]"Todd C. Miller
respectively now that the former is natively supported. OK jsg@
2014-07-10add missing include file to bring in protosTheo de Raadt
2013-12-01Change the file reference queue from CIRCLEQ to TAILQ.Kenneth R Westerback
vi is now CIRCLEQ free! ok zhuk@
2013-12-01Change the tags queue from CIRCLEQ to TAILQ.Kenneth R Westerback
ok zhuk@
2013-12-01Change the tag queue from CIRCLEQ to TAILQ.Kenneth R Westerback
Fixes & ok zhuk@
2013-12-01Convert the ranges CIRCLEQ to TAILQ.Kenneth R Westerback
ok zhuk@
2013-11-28Convert the display screens and hidden screens CIRCLEQ's to TAILQ's.Kenneth R Westerback
ok pelikan@ zhuk@
2013-11-27Zap some pointer casts became extra (and thus dangerous) after recentVadim Zhukov
CIRCLEQ removal. okay krw@
2013-11-26fix a possible double-free/NULL deref in msg_printpelikan
While there, initialize some variables on unusual paths. Found by clang static analyzer, ok krw.
2013-11-26Tweak a tortuous manual loop into a TAILQ_FOREACH().Kenneth R Westerback
ok pelikan@
2013-11-26fix condition after CIRCLEQ -> TAILQ conversion; ok zhuk@Otto Moerbeek
2013-11-26fix incorrectly converted CIRCLEQ_END comparison to prevent NULL deref'spelikan
found by clang static analyzer, ok jca, otto
2013-11-25Replace _texth CIRCLEQ with TAILQ. One down, five to go.Kenneth R Westerback
Read, tested, fixed and ok'd zhuk@ pelikan@ millert@
2013-08-22Correct format string mismatches turned up by -Wformat=2Philip Guenther
suggestions and ok millert@
2013-06-22Tweak optimization flags on landisk until I have time to investigate further.Miod Vallat
2013-05-14When ^W (WERASE) is hit in insert mode it's possible that the lineTodd C. Miller
buffer is accessed out of bounds. If 'max' == 0 and 'tp->cno' == 1 the 'tp->cno' value is first reduced by one and then 'tp->lb' is accessed at 'tp->cno' - 1. Also remove dead (and incorrect) code in the TXT_ALTWERASE case. From Arto Jonsson; OK martynas@
2013-05-03Use open(2) / fstat(2) instead of stat(2) / open(2) for checkingMatthias Kilian
proper permissions of "local" .exrc or .nexrc files. With some hints and fixes from deraadt@ and tobias@. ok millert@ tobias@ (more tweaks suggested by millert@ and some cleanup will be done in separate diffs)
2013-04-29use FD_CLOEXEC instead of 1; from David HillOkan Demirmen
ok otto
2012-12-20use openpty() rather than hand-rolled pty opening code; ok millert@Christian Weisgerber
2012-12-03Fix hang when exiting shell in script mode. OK naddy@Todd C. Miller
2012-01-17flesh out the VI COMMANDS section somewhat;Jason McIntyre
diff from Alexis Fouilhe; help/ok sobrado