Commit Graph
31396 Commits
Author SHA1 Message Date
Pádraig Brady 1a44a25808 cut: -f: fix handling of multi-byte delimiters that span buffers
* src/cut.c (cut_fields_bytesearch): Ensure up to delim_bytes -1
is left for the next refill.
* tests/cut/cut.pl: Add a test case.
2026-04-06 15:52:58 +01:00
Pádraig Brady 57c87043f6 cut,fold,expand,unexpand: ensure we process all available characters
* gl/lib/mbbuf.h: Adjust mbbuf_fill() to process full characters
in the slop at the end of a read().  Previously valid characters
in the last MCEL_LEN_MAX bytes were ignored until the next read().
* src/cut.c (cut_fields_bytesearch): Adjust to the new naming.
* NEWS: Mention the fold(1) responsiveness fix, which was
improved with the change from fread() to read(),
and completed with this patch.
2026-04-06 15:52:56 +01:00
Pádraig Brady 1b4b8104d3 cut: -b: avoid function calls in hot loop
$ time LC_ALL=C src/cut-before -b1 sl.in >/dev/null
  real	0m0.115s

  $ time LC_ALL=C src/cut-after -b1 sl.in >/dev/null
  real	0m0.076s

* src/cut.c (cut_bytes): Hoist the fileno() invariant outside the loop.
Avoid memchr for very short lines.
(search_bytes): Similar to copy_bytes() and write_bytes() helpers.
Note adding code to probe 3 or 4 bytes resulted in worse register
allocation.  I.e. slower operation even if the input was only 2 bytes.
2026-04-05 13:15:56 +01:00
Pádraig Brady 374ff00c36 cut: fix logic issue with field delim in last byte of buffer
With field delimiter = line delimiter we need to know
if there is any more data to be read, as field delimiter
in the last byte of the file is treated differently.
So reiterate the loop to ensure enough read()s to make
the appropriate determination.
2026-04-05 13:15:56 +01:00
Pádraig Brady d757d32f86 cut: ensure responsive input processing
* gl/lib/mbbuf.h (fill_buf): Switch from fread() to read()
as the former retries read() internally to fill the buffer.
* src/cut.c: Adjust accordingly, and avoid getc() interface entirely.
* bootstrap.h: Depend explicitly on fseterr.  This is already depended
on transitively, so should not introduce new build portability issues.
2026-04-05 13:15:56 +01:00
Pádraig Brady 8daf2a91be maint: cut: rename line_in to bytes_in
* src/cut.c: We're not reading a line, rather a buffer of bytes.
Suggested by Collin Funk.
2026-04-05 13:15:56 +01:00
Pádraig Brady 1c9e223298 tests: cut: add more multi-byte tests
* tests/cut/cut.pl: Add more multi-byte combinations.
2026-04-05 13:15:56 +01:00
Pádraig Brady 76edef14d8 cut: make the dependency on memchr2 explicit
* bootstrap.conf: Remove now unused getndelim2, add memchr2.
* src/cut.c: Remove now unused getndelim2.h.
2026-04-05 13:15:56 +01:00
Pádraig Brady 890cf82593 cut: combine cut_bytes_no_split and cut_characters
per character based so merge.
2026-04-05 13:15:56 +01:00
Pádraig Brady fe00823330 doc: cut: clarify that combining characters are not treated specially
This is for consistency with other implementations and since the
interface separates -b and -c it might in future support -g (graphemes).
Normalizing content with a filter seems like the most appropriate
approach anyway, as there are various normalizations possible including
case etc. rather than baking that into every tool
2026-04-05 13:15:56 +01:00
Pádraig Brady 25f0702eaa maint: cut: various code cleanups and comments
* src/cut.c: Document some functions, and remove extraneous
 abstractions.
2026-04-05 13:15:56 +01:00
Pádraig Brady 5d339d583d cut: support no delimiter match fast path with -s
* src/cut.c (cut_fields_bytesearch): Just skip the data with -s.
2026-04-05 13:15:56 +01:00
Pádraig Brady c3e819fadc doc: cut: resintate and expand -d info
* doc/coreutils.texi (cut invocation): Add back the -d description,
and adjust for multi-byte support, and expand on specifying a NUL
delimitier, and detail the behavior when the delimiter matches
the line delimiter.
2026-04-05 13:15:56 +01:00
Pádraig Brady 032ecdee9b maint: cut: cleanup context management for byte search
* src/cut.c: Hoist at_eof into context so we're not
querying it multiple times.  Also add a helper
to explicitly init bytesearch_context.
2026-04-05 13:15:56 +01:00
Pádraig Brady 24571c41f3 cut: optimize UTF-8 input with 0xF5-0xFF delimiters
* src/cut.c (bytesearch_field_delim_ok): Expand the range
of bytes that can be simply searched for. 0xF5-0xFF can't
appear in valid UTF-8 characters, and so may be used as
delimiters in UTF-8 input, so it's worth optimizing for.
* tests/cut/cut.pl: Add a test case (mainly as documentation).
2026-04-05 13:15:56 +01:00
Pádraig Brady 03a686a456 doc: cut: clarify that -s suppressed lines with only trimmed spaces
* doc/coreutils.texi (cut invocation): State explicitly that
-s --whitespace-delimited=trimmed will suppress lines that
do not have field separating blanks.
2026-04-05 13:15:56 +01:00
Pádraig Brady 2c1ea231ca doc: cut: mention the default -O used with -w
* doc/coreutils.texi (cut invocation): Mention the default
--output-delimiter is a TAB when matching runs of blanks in the input.
2026-04-05 13:15:56 +01:00
Pádraig Brady 7ceb4c348b maint: cut: simplify mbbuf_fill
We can only byte search with uni-byte or utf-8.
utf-8 implicitly can't false match a delimiter at buffer boundary.
So don't worry about finding the exact utf8 boundary at end of buffer,
rather just ensuring the buffer always starts with a valid character
(by ensuring MCEL_LEN_MAX-1 moved to start of buffer on each refill).
2026-04-05 13:15:56 +01:00
Pádraig Brady 5dd1a0bc77 maint: refactor is_utf8_charset helper to system.h
* src/cut.c: Move from here.
* src/numfmt.c: Likeise.
* src/system.h: To here.
2026-04-05 13:15:56 +01:00
Pádraig Brady 36751ab993 cut: optimize per character memcpy
$ time src/cut-before -f10 -w ll.in >/dev/null
real	0m4.309s

$ time src/cut-after -f10 -w ll.in >/dev/null
real	0m3.136s

* src/cut.c (cut_bytes): Add a new helper that avoids
the memcpy call in the common case of adding characters to a buffer.
2026-04-05 13:15:56 +01:00
Pádraig Brady 4ae2275952 cut: refactor skip_line_remainder logic
Simplify and optimize field exhaustion logic:

$ time LC_ALL=C src/cut-before -f1 -dc as.in >/dev/null
real	0m0.057s

$ time LC_ALL=C src/cut-after -f1 -dc as.in >/dev/null
real	0m0.023s

* src/cut.c (cut_fields_bytesearch): Refactor.
2026-04-05 13:15:56 +01:00
Pádraig Brady 30c19a1279 maint: cut simplify cut_fields_bytesearch
refactor line delimiter output,
and resetting of parse record state.
2026-04-05 13:15:56 +01:00
Pádraig Brady 1da3ccced2 cut: enable fast path for all delimiter lengths
1. Removed !have_pending_line from the fast path condition.
This is safe because:
  - field_1_n_bytes == 0 already ensures we haven't started
    buffering field 1 content
  - The fast path correctly continues any pending partial line
    by writing raw bytes including the completing \n
2. Added have_pending_line = false after the fast path write,
since all lines up to the last \n are now complete.

$ time src/cut.before -f1 -dç sl.in >/dev/null
real	0m0.081s
$ time src/cut.after -f1 -dç sl.in >/dev/null
real	0m0.012s

$ time src/cut.before -f10 -dç sl.in >/dev/null
real	0m0.081s
$ time src/cut.after -f10 -dç sl.in >/dev/null
real	0m0.012s
2026-04-05 13:15:56 +01:00
Pádraig Brady 7a124eb585 cut: optimize -f with -d longer than lines
$ time src/cut.before -f1 -dç sl.in >/dev/null
real	0m0.157s

$ time src/cut.after -f1 -dç sl.in >/dev/null
real	0m0.084s
2026-04-05 13:15:56 +01:00
Pádraig Brady e73dbeb5ab cut: optimize -b for short lines
For a 40% performance increase it's worth reinstating the simple
original cut_bytes() which avoids data copying and function calls.
Once a longer line is encountered we defer to the buffered variant.

$ time src/cut.before -b2 sl.in  >/dev/null
real	0m0.101s

$ time src/cut.after -b2 sl.in  >/dev/null
real	0m0.060s
2026-04-05 13:15:56 +01:00
Pádraig Brady f5cab8e30c cut: optimize per character field scanning
Use specialized loops rather than branching per character,
giving a 28% increase.

$ time src/cut -f1 -w ll.in >/dev/null
real	0m7.199s

$ time src/cut -f1 -w ll.in >/dev/null
real	0m5.204s
2026-04-05 13:15:56 +01:00
Pádraig Brady a78ef2b65c cut: prefer c_isblank() to c32issep()
12% perf increase with:

$ time src/cut -f2 -w ll.in >/dev/null
real	0m6.469s

$ time src/cut -f2 -w ll.in >/dev/null
real	0m5.689s
2026-04-05 13:15:56 +01:00
Pádraig Brady 9817599526 cut: avoid fwrite calls for smaller amounts of data
This is quite significant:

yes abcdfeg | head -n1MB > big-file

$ time src/cut-before -b1,3 big-file >/dev/null
real	0m0.050s

$ time src/cut-after -b1,3 big-file >/dev/null
real	0m0.029s
2026-04-05 13:15:56 +01:00
Pádraig Brady a16d56d60c cut: optimize -b by avoiding per byte iteration
Always memchr(line_delim) which is fast and allows:

- skipping whole segments when the next selected byte is beyond them
- skipping unselected prefixes in bulk
- writing contiguous selected spans in bulk

This wins for lines >= 4 characters,
but is slower lines <= 3 characters, especially if selecting bytes 1-3.
That is unusual though.
2026-04-05 13:15:56 +01:00
Pádraig Brady ea6a7ba547 cut: optimize when no delimiter in input
This is about 20x faster.
Note we only do the delimiter search once per chunk,
and it's usually quick as delimiters wouldn't be too far
into the a chunk if present, so we don't bother
to cache the found delimiter.
2026-04-05 13:15:56 +01:00
Pádraig Brady 7d017f83bc tests: cut: ensure multi-byte delimiter is rejected in uni-byte locales
tests/cut/cut.pl: Check the appropriate diagnostic is presented.
2026-04-05 13:15:56 +01:00
Pádraig Brady 700ffc51a1 cut: optimize -w for uni-byte case
* src/cut.c: Limit search to SPACE and TAB
2026-04-05 13:15:56 +01:00
Pádraig Brady f5b7d38d13 doc: cut: reorder -s in texi
Keep in alphabetical order.
2026-04-05 13:15:56 +01:00
Pádraig Brady c1d7b492c6 doc: cut: document the -w option
* src/cut.c (usage): Mention blank characters are used to separate.
* doc/coreutils.texi (cut invocation): Likewise.  Also describe
the 'trimmed' argument and the relation to -F.
2026-04-05 13:15:56 +01:00
Pádraig Brady cf25ef286a cut: refactor find_bytesearch_field_terminator to be stateful
Allows better/simpler avoidance of repeated line/delim scans

TODO: speed up our really slow cut_fields_mb_any.
Compare for example:
  time src/cut -w  -f1 ll.in >/dev/null #14s
  time src/cut -d, -f1 ll.in >/dev/null #.1s
Could adjust so that LC_ALL=C does memchr2(space,tab) ?
2026-04-05 13:15:56 +01:00
Pádraig Brady 0adb7c6edd cut: avoid repeated searchs for line_delim in the multi-byte delim case
TODO: Refactor all this into find_bytesearch_field_terminator.
Also handle in the delim_length==1 case.
2026-04-05 13:15:56 +01:00
Pádraig Brady d9825aa9b1 cut: refactor all byte search to find_bytesearch_field_terminator
TODO: Perhaps also add search only fields mode
to avoid rescans of very long lines
2026-04-05 13:15:56 +01:00
Pádraig Brady 6250b59ef9 cut: optimize -f when finished processing fields for a line
TODO: simplify and compare perf
2026-04-05 13:15:56 +01:00
Pádraig Brady 352a396a16 cut: optimize -f for fhe common case of single byte delimiters
* TODO: perf comparison
2026-04-05 13:15:56 +01:00
Pádraig Brady 2a6b36ff5b cut: optimize -d '?' in UTF-8 case
ensure all ascii delims are processed with byte search in UTF-8
2026-04-05 13:15:56 +01:00
Pádraig Brady f6b3055f74 cut: merge cut_fields and cut_fields_bytesearch
TODO: See why this is much slower:
time LC_ALL=C.UTF-8 src/cut -f1 -dc as.in > /dev/null
2026-04-05 13:15:56 +01:00
Pádraig Brady b3ef6231bd cut: refactor -f to byte search and character processing
Not sure about this at all.
Only worthwhile if can also remove cut_fields_line_delim.

Refactored src/cut.c so the old UTF-8 byte-search path is now the
general byte-search field engine for all safe byte-search cases:
ordinary single-byte delimiters and valid UTF-8 delimiters. The old
cut_fields path did not go away completely; it is now
cut_fields_line_delim and is used only when the field delimiter equals
the record delimiter, because -d $'\n' and -z -d '' have different
semantics from normal line-based field splitting.

As part of that, I also folded the duplicated “start selected field”
logic into a shared helper, and renamed the byte-search helpers to match
their broader use. The current dispatcher in src/cut.c is now:
whitespace parser, then line- delimiter field mode, then byte-search
field mode, then the decoded multibyte parser.
2026-04-05 13:15:56 +01:00
Pádraig Brady 16b1ff40ae cut: fix 25% perf regression mentioned in previous change
* src/cut.c: Prefer strstr() over memmem() as the former is
optimized per platforma on GLIBC, where as the latter is only
optimized for S390 as of glibc-2.42

TODO: look at merging cut_fields and cut_fields_mb_utf8
as they're both byte search
2026-04-05 13:15:56 +01:00
Pádraig Brady 26028fb2c6 cut: use bounded memory in utf8 mode when possible
TODO: See why a bit slower than old code

$ time src/cut.old -f1 -dç mb.in >/dev/null
real	0m0.136s
user	0m0.096s
sys	0m0.039s

$ time src/cut.new -f1 -dç mb.in >/dev/null
real	0m0.170s
user	0m0.139s
sys	0m0.030s
2026-04-05 13:15:56 +01:00
Pádraig Brady ba7c1fbadb cut: add utf8 helper to mbbuf
* gl/lib/mbbuf.h: To safely search a bounded buffer,
without needed to have unbounded memory with getndelim2.

TODO: rename mbbuf_utf8_safe_prefix to mbbuf_fill_utf8
2026-04-05 13:15:56 +01:00
Pádraig Brady 801686242e cut: faster utf8 processing
TODO: improve to use bounded memory where possible
2026-04-05 13:15:56 +01:00
Pádraig Brady a14ac29629 cut: support -F as an alias for -f -w -O ' '
To improve compatibility with toybox/busybox scripts.
2026-04-05 13:15:56 +01:00
Pádraig Brady 74f6692aa0 maint: cut: refactor buffered and ordinary field scanning
* src/cut.c: Merge scan_mb_field and read_mb_field_to_buffer
2026-04-05 13:15:56 +01:00
Pádraig Brady eb1f057746 cut: support --whitespace-delimited=trimmed
Support ignoring leading and trailing whitespace.
E.g. this matches awk's default field splitting mode.

* src/cut.c
* tests/cut/cut.pl: Add test cases.
2026-04-05 13:15:56 +01:00
Pádraig Brady 77ccacb9a7 cut: support -O as an alias for --output-delimiter
To improve compatibility with toybox/busybox scripts.

* doc/coreutils.texi (cut invocation): Add -O description.
* src/cut.c: Support -O as well as --output-delimiter
* tests/cut/cut.pl: Adjust one case to use -O.
2026-04-05 13:15:56 +01:00