Commit message (Collapse) | Author | Age | Files | Lines | ||
---|---|---|---|---|---|---|
... | ||||||
* | bfstd: New xwcwidth() wrapper | Tavian Barnes | 2023-12-16 | 3 | -2/+7 | |
| | ||||||
* | main: Warn if setlocale() fails | Tavian Barnes | 2023-12-16 | 1 | -1/+12 | |
| | | | | This should help users understand why issues like #128 happen. | |||||
* | parse: Reject integers that start with whitespace | Tavian Barnes | 2023-12-09 | 2 | -1/+6 | |
| | ||||||
* | bfstd: Wrap is[w]{alpha,digit,...}() | Tavian Barnes | 2023-12-09 | 2 | -31/+62 | |
| | ||||||
* | Revert "README: Realign badges" | Tavian Barnes | 2023-11-28 | 1 | -3/+4 | |
| | | | | | | Looks like GitHub fixed their rendering. This reverts commit 3f15559a82484d46942a68e18ba18c7e1c31aab3. | |||||
* | color: Make sure errno is non-zero if parse_colors() fails | Tavian Barnes | 2023-11-28 | 1 | -43/+43 | |
| | ||||||
* | bfstd: Work around more instances of llvm/llvm-project#65532 | Tavian Barnes | 2023-11-23 | 1 | -14/+19 | |
| | ||||||
* | darray: Remove | Tavian Barnes | 2023-11-23 | 4 | -189/+0 | |
| | | | | | The new RESERVE() macro from alloc.h is nicer, and handles alignment properly. | |||||
* | printf: Switch from darray to RESERVE() | Tavian Barnes | 2023-11-23 | 1 | -175/+195 | |
| | ||||||
* | mtab: Switch from darray to RESERVE() | Tavian Barnes | 2023-11-23 | 1 | -23/+23 | |
| | ||||||
* | ctx: Switch paths from darray to RESERVE() | Tavian Barnes | 2023-11-23 | 4 | -15/+16 | |
| | ||||||
* | alloc: New helpers for growing dynamic arrays | Tavian Barnes | 2023-11-23 | 2 | -22/+80 | |
| | ||||||
* | exec: Use the new REALLOC_ARRAY() macro | Tavian Barnes | 2023-11-23 | 1 | -1/+1 | |
| | ||||||
* | dstring: Use the new REALLOC_FLEX() macro | Tavian Barnes | 2023-11-23 | 2 | -51/+65 | |
| | | | | | | While I'm at it, switch the capacity to include the NUL byte, so that it could be used by __counted_by() for example. And fix a theoretical provenance issue. | |||||
* | alloc: New helpers for aligned reallocation | Tavian Barnes | 2023-11-23 | 2 | -4/+62 | |
| | | | | | | There is no aligned_realloc(), so the new xrealloc() function emulates it by manually reallocating and copying for over-aligned types. The new REALLOC_ARRAY() and REALLOC_FLEX() macros wrap xrealloc(). | |||||
* | alloc: New ALLOC_MAX macro | Tavian Barnes | 2023-11-23 | 1 | -2/+10 | |
| | ||||||
* | alloc: New is_aligned() helper | Tavian Barnes | 2023-11-23 | 2 | -5/+10 | |
| | ||||||
* | bftw: Add some whitespace to struct bftw_args | Tavian Barnes | 2023-11-23 | 1 | -0/+4 | |
| | ||||||
* | tests: Set abort_on_error for the sanitizers | Tavian Barnes | 2023-11-23 | 1 | -1/+1 | |
| | ||||||
* | tests: Fix --stop | Tavian Barnes | 2023-11-23 | 1 | -0/+6 | |
| | ||||||
* | xspawn: Detect posix_spawn() support | Tavian Barnes | 2023-11-19 | 2 | -10/+38 | |
| | ||||||
* | ioq: Don't crash on allocation failures | Tavian Barnes | 2023-11-15 | 1 | -0/+4 | |
| | ||||||
* | config: New attr_maybe_unused macro | Tavian Barnes | 2023-11-15 | 4 | -7/+19 | |
| | ||||||
* | tests: Fix a possible infinite loop | Tavian Barnes | 2023-11-15 | 1 | -2/+4 | |
| | | | | | | | | | | | | | | | | | | | The POSIX spec for readdir() [1] says: > If a file is removed from or added to the directory after the most > recent call to opendir() or rewinddir(), whether a subsequent call > to readdir() returns an entry for that file is unspecified. which implies that a loop of readdir()/unlink()/creat() may continue to return new files unendingly. This was even observed on a Linux 6.6 tmpfs mount [2]. It's not clear whether find(1) is also permitted to loop endlessly in this case, but in case it is, let's avoid the whole problem by limiting the -exec to happen at most once. [1]: https://pubs.opengroup.org/onlinepubs/9699919799/functions/readdir.html [2]: https://lore.kernel.org/linux-fsdevel/20231113180616.2831430-1-tavianator@tavianator.com/ | |||||
* | xspawn: API tweaks | Tavian Barnes | 2023-11-15 | 3 | -43/+48 | |
| | ||||||
* | bit: Implement a branchless has_single_bit() | Tavian Barnes | 2023-11-13 | 1 | -1/+2 | |
| | ||||||
* | eval: Remove Cosmopolitan-specific workaround | Tavian Barnes | 2023-11-13 | 1 | -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/ | |||||
* | tests: Fix --verbose=commands with --bfs="wrapper bfs" | Tavian Barnes | 2023-11-13 | 1 | -1/+1 | |
| | ||||||
* | Revert "bit: Implement a branchless has_single_bit()" | Tavian Barnes | 2023-11-11 | 2 | -2/+2 | |
| | | | | | | Doesn't work for the highest bit due to overflow. This reverts commit 208376ef99da243545efcd6fb02d3469b4c068ed. | |||||
* | Initial support for Cosmopolitan Libc | Tavian Barnes | 2023-11-10 | 5 | -3/+33 | |
| | ||||||
* | exec: Don't do setrlimit() in the parent | Tavian Barnes | 2023-11-10 | 1 | -22/+4 | |
| | | | | | | | | | This was hacky, but it's also broken because it can make posix_spawn() fail with EMFILE, as happens on at least musl. Fixes: 7d69fef6a0b80ad57acde7bac8a22b83531fec0d Link: https://www.austingroupbugs.net/view.php?id=603 Link: https://sourceware.org/bugzilla/show_bug.cgi?id=31049 | |||||
* | bit: Implement a branchless has_single_bit() | Tavian Barnes | 2023-11-10 | 2 | -1/+5 | |
| | ||||||
* | config: Add constants for C standard versions | Tavian Barnes | 2023-11-10 | 4 | -8/+16 | |
| | ||||||
* | config: Add (de)allocator attributes | Tavian Barnes | 2023-11-09 | 4 | -25/+89 | |
| | ||||||
* | config: New attr_noinline and attr_cold macros | Tavian Barnes | 2023-11-09 | 4 | -0/+36 | |
| | ||||||
* | config: New attr_target_clones() macro | Tavian Barnes | 2023-11-09 | 2 | -8/+17 | |
| | ||||||
* | config: s/BFS_FORMATTER/attr_format/ | Tavian Barnes | 2023-11-09 | 10 | -29/+29 | |
| | ||||||
* | config: Remove BFS_SUPPRESS() | Tavian Barnes | 2023-11-09 | 2 | -19/+14 | |
| | ||||||
* | list: Simplify slist_remove_impl() | Tavian Barnes | 2023-11-09 | 1 | -4/+4 | |
| | | | | We now assume that all-bits-zero is a null pointer, so memset is fine. | |||||
* | ioq: Implement a better non-blocking pop | Tavian Barnes | 2023-11-09 | 3 | -102/+189 | |
| | ||||||
* | parse: Reject -{exec,ok}dir if $PATH contains a relative path | Tavian Barnes | 2023-11-07 | 7 | -0/+32 | |
| | | | | This matches the behaviour of GNU find. | |||||
* | tests: Fix output interleaving with -j1 | Tavian Barnes | 2023-11-07 | 1 | -5/+5 | |
| | ||||||
* | xspawn: Do $PATH resolution up-front | Tavian Barnes | 2023-11-07 | 1 | -21/+19 | |
| | | | | | | | | | posix_spawnp() is typically implemented like execvp(), i.e., by repeatedly trying execv() with each $PATH component until it succeeds. This is much slower than resolving the executable path up-front and then calling execv() once, so do that. Fixes: https://github.com/tavianator/bfs/pull/127#issuecomment-1795095126 | |||||
* | Treat NO_COLOR="" the same as unset | Tavian Barnes | 2023-11-06 | 3 | -1/+34 | |
| | | | | | | | | | | | | | | The docs say > Command-line software which adds ANSI color to its output by default > should check for a NO_COLOR environment variable that, when present > and not an empty string (regardless of its value), prevents the > addition of ANSI color. but we were not checking for the empty string. Link: https://no-color.org/ Link: https://github.com/sharkdp/fd/pull/1421 | |||||
* | exec: Try harder to use posix_spawn() | Tavian Barnes | 2023-11-06 | 1 | -5/+26 | |
| | ||||||
* | xspawn: Wrap the real posix_spawn() if possible | Tavian Barnes | 2023-11-06 | 3 | -48/+136 | |
| | | | | Fixes #47. | |||||
* | ctx: Store the original and current RLIMIT_NOFILE | Tavian Barnes | 2023-11-06 | 5 | -37/+38 | |
| | ||||||
* | bfstd: Expose rlim_cmp() | Tavian Barnes | 2023-11-06 | 3 | -33/+40 | |
| | ||||||
* | bench: New -exec benchmarks | Tavian Barnes | 2023-11-02 | 1 | -3/+89 | |
| | ||||||
* | tests: Add tests for ulimit -n after -exec | Tavian Barnes | 2023-11-02 | 4 | -0/+36 | |
| |