summaryrefslogtreecommitdiffstats
path: root/src/eval.c
Commit message (Collapse)AuthorAgeFilesLines
* prelude: s/fallthru/_fallthrough/ for consistency with other attributesTavian Barnes2024-08-091-2/+2
|
* prelude: Simplify attributesTavian Barnes2024-07-281-1/+1
|
* Merge branch 'revert-eloop'Tavian Barnes2024-06-031-0/+7
|\
| * Make ELOOP an error again, except for -xtype.3.3.1Tavian Barnes2024-06-031-0/+7
| | | | | | | | | | | | | | | | | | | | POSIX requires an error if (for example) -L encounters a symlink loop. The GNU find change was restricted to -xtype, so add a manual ELOOP test to eval_xtype() for compatibility. This reverts commit 470589cbd9ca3e73d8c01ac3a96cbc065179dcc5. Link: https://savannah.gnu.org/bugs/?19605
* | eval: Fix -D stat printed return valueTavian Barnes2024-06-031-1/+1
|/
* eval: Don't turn on NOKERNINFOTavian Barnes2024-05-271-95/+8
| | | | | Other utilities on FreeBSD like dd, sleep, etc. that respond to SIGINFO still print the kernel info line too.
* eval: Use a flag instead of a counter for SIGINFOTavian Barnes2024-05-271-16/+13
|
* eval: Print more information on filesystem loopsTavian Barnes2024-05-271-3/+13
|
* eval: Show/hide the bar on SIGINFO/SIGUSR1Tavian Barnes2024-05-251-1/+125
|
* opt: Don't raise RLIMIT_NOFILE if it would prevent using posix_spawn()Tavian Barnes2024-05-231-0/+3
| | | | | | | | | | If we raise RLIMIT_NOFILE, we have to lower it before calling exec() for compatibility with select(). If posix_spawn() doesn't support that, we fall back to fork(), which is quite a bit slower. Therefore, if we're going to exec() on most files, it's better to keep RLIMIT_NOFILE the same to avoid the fork() cost, even though it makes bftw() somewhat slower.
* expr: New for_expr macroTavian Barnes2024-05-231-5/+5
|
* Stop using %mTavian Barnes2024-05-201-11/+9
|
* Cast AT_FDCWD to int for comparisonsTavian Barnes2024-05-171-1/+1
| | | | | Some platforms define AT_FDCWD to a constant like 0xFFFAFDCD that gets typed as an unsigned int.
* eval: Plug memory leak if bfs_opendir() failsTavian Barnes2024-04-241-18/+22
|
* Rename config.h to prelude.hTavian Barnes2024-04-191-1/+1
|
* Implement -contextTavian Barnes2024-04-101-14/+29
| | | | Closes: https://github.com/tavianator/bfs/issues/27
* Implement -limit NTavian Barnes2024-03-201-0/+13
| | | | Closes: https://github.com/tavianator/bfs/issues/133
* Re-run include-what-you-useTavian Barnes2024-03-111-2/+2
|
* xtime: Call tzset() from main() instead of lazilyTavian Barnes2024-03-071-1/+1
| | | | | | | | | | | | | POSIX specifies[1] that If a thread accesses tzname, daylight, or timezone directly while another thread is in a call to tzset(), or to any function that is required or allowed to set timezone information as if by calling tzset(), the behavior is undefined. So calling it lazily from arbitrary threads is risky. [1]: https://pubs.opengroup.org/onlinepubs/9699919799/functions/tzset.html
* eval: Tweak status bar punctuationTavian Barnes2024-03-061-1/+1
|
* eval: Add thousands separators to -statusTavian Barnes2024-02-121-2/+2
|
* sanity: Don't use self-init for uninit()Tavian Barnes2024-02-121-1/+1
| | | | | | | | | | | Self-initialization like bool ret = ret; is a GCC trick to suppress uninitialized variable warnings, but it's not actually well-defined, and will trip a recent enough MemorySanitizer: src/eval.c:1088:13: runtime error: load of value 128, which is not a valid value for type 'bool'
* ctx: Fill in ctx->threads earlierTavian Barnes2024-02-061-19/+2
|
* bftw: stat() files asynchronouslyTavian Barnes2024-02-061-10/+10
|
* eval: Squelch an uninitialized variable warningTavian Barnes2024-01-221-1/+3
|
* bfstd: New {error,errno}_is_like() functionsTavian Barnes2024-01-131-1/+1
| | | | | | | We used to have is_nonexistence_error() to consistently treat ENOENT and ENOTDIR the same. Recently, we started considering EFAULT the same as ENAMETOOLONG on DragonFly BSD to work around a kernel bug. Unify both of these behind a more generic interface.
* eval: Check for xbasename() allocation failureTavian Barnes2024-01-071-1/+8
|
* expr: Make expressions variadicTavian Barnes2024-01-071-31/+24
| | | | | | | | | Rather than only unary/binary expressions, we now support an arbitrary number of children. The optimizer has been re-written almost completely and now supports optimal reordering of longer expression chains, rather than just arm-swapping. Fixes #85.
* config: s/attr_format/attr_printf/Tavian Barnes2023-12-181-1/+1
|
* config: New variadic attr(...) macroTavian Barnes2023-12-181-1/+1
|
* bfstd: New xwcwidth() wrapperTavian Barnes2023-12-161-1/+1
|
* ctx: Switch paths from darray to RESERVE()Tavian Barnes2023-11-231-2/+1
|
* eval: Remove Cosmopolitan-specific workaroundTavian Barnes2023-11-131-3/+1
| | | | | | | Turns out this is not specific to Cosmopolitan, it's a recent Linux behaviour change. Link: https://lore.kernel.org/linux-fsdevel/20231113180616.2831430-1-tavianator@tavianator.com/
* Initial support for Cosmopolitan LibcTavian Barnes2023-11-101-1/+3
|
* config: s/BFS_FORMATTER/attr_format/Tavian Barnes2023-11-091-1/+1
|
* ctx: Store the original and current RLIMIT_NOFILETavian Barnes2023-11-061-14/+19
|
* bfstd: Expose rlim_cmp()Tavian Barnes2023-11-061-33/+0
|
* Use {0} to initialize mbstate_t rather than memset()Tavian Barnes2023-10-271-2/+1
|
* eval: TTY-escape paths in the status barTavian Barnes2023-10-271-4/+8
|
* bfstd: Expose xmbrtowc() and use it in eval_status()Tavian Barnes2023-10-271-24/+15
|
* bftw: New flag to control whiteout visibilityTavian Barnes2023-10-171-0/+1
|
* dir: Add a flags parameter to bfs_opendir()Tavian Barnes2023-10-171-3/+3
|
* bfstd: Add a thread-safe wrapper for strerror()Tavian Barnes2023-10-051-1/+1
|
* Formatting fixesTavian Barnes2023-09-271-8/+7
|
* dstring: New dchar typedef for dynamic stringsTavian Barnes2023-09-261-2/+2
|
* eval: Don't oversubscribe the CPU by defaultTavian Barnes2023-07-111-3/+3
|
* Use strcmp() instead of fnmatch() if possibleTavian Barnes2023-07-061-4/+17
|
* dir: Arena-allocate directoriesTavian Barnes2023-06-201-12/+22
|
* eval: Pre-allocate the highest fdTavian Barnes2023-06-131-0/+18
| | | | | | This avoids the need to grow the fd table during the search, significantly reducing kernel contention when opening directories in parallel.
* bftw: Use an I/O queue to open directoriesTavian Barnes2023-06-131-0/+22
| | | | Parallelism is controlled by the new -j flag.