Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | 50133: use read-ahead and lseek-rewind for efficient line-buffered input | Bart Schaefer | 2022-04-28 | 1 | -1/+23 |
| | |||||
* | 49792: Non-interative shell input is line buffered. | Peter Stephenson | 2022-03-03 | 1 | -7/+14 |
| | |||||
* | unposted: add 'static' to shinsavestack | Jun-ichi Takimoto | 2021-09-08 | 1 | -1/+1 |
| | |||||
* | 49290: Replace stdio for buffered shell input. | Peter Stephenson | 2021-08-27 | 1 | -11/+111 |
| | | | | | | | The previous method allowed memory management to interact with signal handlers, causing occasional crashes on some system. Instead, use a simple pre-allocated buffer and raw system calls. | ||||
* | 48504: use SEEK_ macros in fseek() calls | Oliver Kiddle | 2021-04-11 | 1 | -2/+2 |
| | |||||
* | 43511: Initialiase alias entry on input stack. | Peter Stephenson | 2018-09-21 | 1 | -0/+2 |
| | | | | This didn't happen if neither history nor alias expansion was in use. | ||||
* | 41322: reduce number of changes in signal queuing/blocking state during ↵ | Barton E. Schaefer | 2017-06-19 | 1 | -3/+5 |
| | | | | shingetline() to improve read speed | ||||
* | 40306 with doc tweaks: Change behaviour expanding alias in () function ↵ | Peter Stephenson | 2017-01-10 | 1 | -0/+27 |
| | | | | | | | definition. Now an error unless the () is part of the same error as the name. Add ALIAS_FUNC_DEF option to allow it again. | ||||
* | 40117: Revert a hunk of 40035 that changed semantics incorrectly. | Daniel Shahaf | 2016-12-09 | 1 | -1/+1 |
| | |||||
* | 40035: Cosmetic fixes for comments and documentation. | Eitan Adler | 2016-11-29 | 1 | -1/+1 |
| | | | | Mostly fixes to doubled words. | ||||
* | 36022 fix bug that some loop constructs could not be interrupted, revise ↵ | Barton E. Schaefer | 2015-08-09 | 1 | -2/+10 |
| | | | | | | | | | | | signal queueing There are two underlying ideas here: (1) Keeping signals queued around anything that's doing memory management (including push/pop of the heap) has become crucial. (2) Anytime the shell is going to run a command, be it buitin or external, it must be both safe and necessary to process any queued signals, so that the apparent order of signal arrival and command execution is preserved. | ||||
* | 35910: Don't add to raw lex buffer if lex stopped. | Peter Stephenson | 2015-07-25 | 1 | -1/+2 |
| | | | | | This was causing a segmentation violation in completion when parsing an incomplete math expression, e.g. 'echo $((3*4)<TAB>'. | ||||
* | 35668: Improved fix for command/proc subst starting in alias. | Peter Stephenson | 2015-07-02 | 1 | -1/+1 |
| | | | | | | Use input flag to suppress the unwanted backtracking. Add test for the extra case covered. | ||||
* | 34784: fix old bug with history word selection | Peter Stephenson | 2015-03-26 | 1 | -9/+18 |
| | |||||
* | 34752: another fix for history expansion in cmd subst | Peter Stephenson | 2015-03-20 | 1 | -2/+14 |
| | |||||
* | Fix up memory allocation for previous patch | Peter Stephenson | 2015-02-17 | 1 | -4/+5 |
| | |||||
* | 34560: Fix $(( that's actually a multiline cmd subst. | Peter Stephenson | 2015-02-16 | 1 | -2/+31 |
| | |||||
* | 34543: Prevent crash on garbage bytes inside $(...) | Barton E. Schaefer | 2015-02-14 | 1 | -2/+4 |
| | | | | | | Garbage input (nul bytes, etc.) can cause the $(...) parser to become confused during look-ahead and attempt to back up the input too far. This commit catches the error but does not fix the underlying cause. | ||||
* | 34220: new $(...) handling needs to back up over alias expansion | Peter Stephenson | 2015-01-10 | 1 | -0/+6 |
| | |||||
* | Fix command substitutions to parse contents as they are read in. | Peter Stephenson | 2015-01-08 | 1 | -5/+8 |
| | | | | | | | Do this by refactoring misnamed lexsave()/lexrestore() to allow continuity of history and input. Add test. | ||||
* | 33876: etc.: Separate errors and keyboards interrupts | Peter Stephenson | 2014-12-11 | 1 | -1/+2 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Combination of 12 commits from interrupt_abort branch. Basic strategy is to introduce bits to errflag and to set and reset them separately. Remove interrupt status on return to main keymap. Turn off ERRFLAG_INT for always block. Restore bit thereafter: we probably need a new variable in order to allow user interrupts to be reset in the always block. Add TRY_BLOCK_INTERRUPT This works the same as TRY_BLOCK_ERROR, but for a SIGINT, too. Ensure propagation of SIGINT from exited job. If received by foreground job, shell uses ERRFLAG_INT, not ERRFLAG_ERROR, to set the new state. Reset errflag before precmd() Add always block in _main_completion to fix ZLS_COLORS Ensures we get the right state of $ZLS_COLORS at the end of _main_complete even if there's an interrupt. However, the "right state" is a bit messy as it depends on styles. | ||||
* | 31869: reduce WINCH-twaddling in shingetline() | Barton E. Schaefer | 2013-10-21 | 1 | -2/+4 |
| | |||||
* | 31350: block SIGWINCH nearly all the time, except | Bart Schaefer | 2013-04-30 | 1 | -0/+2 |
| | | | | | when about to calculate prompts or do synchronous read, so syscalls are not interrupted by window size changes. | ||||
* | Paul 28538 / me 28540 / couple of unposted casts: | Peter Stephenson | 2010-12-20 | 1 | -1/+1 |
| | | | | use char * for pointer arithmetic rather than void * | ||||
* | 27827: fix infinite loop in recursive alias at end of parsed string | Peter Stephenson | 2010-03-25 | 1 | -9/+12 |
| | |||||
* | Joakim Rosqvist: 27591 as modified in 27594: | Peter Stephenson | 2010-01-13 | 1 | -14/+26 |
| | | | | KEYBOARD_HACK variable | ||||
* | Check the return value of all pipe(), read(), and write() calls. | Wayne Davison | 2009-12-16 | 1 | -1/+1 |
| | | | | | | Gets rid of all the remaining "ignoring return value" compiler warnings, and makes some read/write operations safer by ensuring that an EINTR is handled. | ||||
* | 25345, 25347: neaten interface from main shell to zle | Peter Stephenson | 2008-07-31 | 1 | -1/+2 |
| | |||||
* | 25002: only ZLE should update attributes resulting from prompt expansion | Peter Stephenson | 2008-05-12 | 1 | -1/+1 |
| | |||||
* | 24699: bug with ${(Q)...} on initial ">" | Peter Stephenson | 2008-03-08 | 1 | -0/+12 |
| | | | | bug with ${(z)...} on string with unterminated "(" | ||||
* | 22474: use variable argument lists to improve error message handling | Peter Stephenson | 2006-05-30 | 1 | -2/+2 |
| | |||||
* | Got rid of some unsigned-char/char pointer casts. | Wayne Davison | 2005-11-15 | 1 | -2/+1 |
| | |||||
* | c.f. 20675: improve zle as a basis for Unicode. | Peter Stephenson | 2005-01-14 | 1 | -2/+2 |
| | | | | unposted: update version to 4.2.3-dev-1 | ||||
* | 20149: improve prompt-reset code | Peter Stephenson | 2004-07-11 | 1 | -6/+7 |
| | | | | 20150: commit ancient memory leak fix(?) in completion | ||||
* | 19295: $CONTEXT zle parameter | Peter Stephenson | 2003-12-15 | 1 | -1/+4 |
| | |||||
* | a la 19209: zcalloc -> zshcalloc | Peter Stephenson | 2003-10-29 | 1 | -1/+1 |
| | |||||
* | 18941: history wrong after parse error during alias expansion | Peter Stephenson | 2003-08-11 | 1 | -0/+14 |
| | |||||
* | 18252: pass ignoreeof as flag to zleread | Peter Stephenson | 2003-02-17 | 1 | -2/+4 |
| | |||||
* | 16492: add RPROMPT2 variable for right prompts in multi-line commands | Oliver Kiddle | 2002-01-31 | 1 | -1/+4 |
| | |||||
* | 16361: include <stdio.h> to avoid implicit declarations. | Clint Adams | 2001-12-18 | 1 | -0/+4 |
| | |||||
* | remove 13108 (trap queues); replace with signal queueing to ensure that user ↵ | Sven Wischnowsky | 2001-01-16 | 1 | -3/+1 |
| | | | | signal handlers are only executed when it is safe to run them (13365) | ||||
* | Sven: 13108: Handle traps synchronously | Peter Stephenson | 2000-11-11 | 1 | -1/+3 |
| | | | | pws: 13109, 13111: clear up zle display when output produced in trap. | ||||
* | Removed the spaceflag variable. | Wayne Davison | 2000-07-18 | 1 | -29/+41 |
| | |||||
* | Initial revision | Tanaka Akira | 1999-04-15 | 1 | -0/+530 |