.TH UGREP-INDEXER "1" "September 15, 2026" "ugrep-indexer 7.8.5" "User Commands"
.SH NAME
\fBugrep-indexer\fR -- file indexer to accelerate recursive searching
.SH SYNOPSIS
.B ugrep-indexer [\fB-0\fR...\fB9\fR] [\fB-c\fR|\fB-d\fR|\fB-f\fR] [\fB-I\fR] [\fB-q\fR] [\fB-S\fR] [\fB-s\fR] [\fB-X\fR] [\fB-z\fR] [\fIPATH\fR]
.SH DESCRIPTION
The \fBugrep-indexer\fR utility recursively indexes files to accelerate
recursive searching with the \fBug --index\fR \fIPATTERN\fR commands:
.IP
$ \fBugrep-indexer\fR [\fB-I\fR] [\fB-z\fR]
.IP
  ...
.IP
$ \fBug\fR \fB--index\fR [\fB-I\fR] [\fB-z\fR] [\fB-r\fR|\fB-R\fR] \fIOPTIONS\fR \fIPATTERN\fR
.IP
$ \fBugrep\fR \fB--index\fR [\fB-I\fR] [\fB-z\fR] [\fB-r\fR|\fB-R\fR] \fIOPTIONS\fR \fIPATTERN\fR
.PP
where option \fB-I\fR or \fB--ignore-binary\fR ignores binary files, which is
recommended to limit indexing storage overhead and to reduce search time.
Option \fB-z\fR or \fB--decompress\fR indexes and searches archives and
compressed files.
.PP
Indexing speeds up searching file systems that are large and cold (not recently
cached in RAM) and file systems that are generally slow to search.  Note that
indexing may not speed up searching few files or recursively searching fast
file systems.
.PP
Searching with \fBug --index\fR is safe and never skips modified files that may
match after indexing; the \fBug --index\fR \fIPATTERN\fR command always
searches files and directories that were added or modified after indexing.
When option \fB--stats\fR is used with \fBug --index\fR, a search report is
produced showing the number of files skipped not matching any indexes and the
number of files and directories that were added or modified after indexing.
Note that searching with \fBug --index\fR may significantly increase the
start-up time when complex regex patterns are specified that contain large
Unicode character classes combined with `*' or `+' repeats, which should be
avoided.
.PP
\fBugrep-indexer\fR stores a hidden index file in each directory indexed.  The
size of an index file depends on the number of files indexed and the specified
indexing accuracy.  Higher accuracy produces larger index files to improve
search performance by reducing false positives (a false positive is a match
prediction for a file when the file does not match the regex pattern.)
.PP
\fBugrep-indexer\fR accepts an optional \fIPATH\fR to the root of the directory
tree to index.  The default is to index the working directory tree.
.PP
\fBugrep-indexer\fR incrementally updates indexes.  To force reindexing,
specify option \fB-f\fR or \fB--force\fR.  Indexes are deleted with option
\fB-d\fR or \fB--delete\fR.
.PP
\fBugrep-indexer\fR may be stopped and restarted to continue indexing at any
time.  Incomplete index files do not cause errors.
.PP
ASCII, UTF-8, UTF-16 and UTF-32 files are indexed and searched as text files
unless their UTF encoding is invalid.  Files with other encodings are indexed
as binary files and can be searched with non-Unicode regex patterns using
\fBug --index \fB-U\fR.
.PP
When \fBugrep-indexer\fR option \fB-I\fR or \fB--ignore-binary\fR is specified,
binary files are ignored and not indexed.  Avoid searching these non-indexed
binary files with \fBug --index -I\fR using option \fB-I\fR.
.PP
\fBugrep-indexer\fR option \fB-X\fR or \fB--ignore-files\fR respects gitignore
rules.  Likewise, avoid searching non-indexed ignored files with \fBug --index
--ignore-files\fR using option \fB--ignore-files\fR.
.PP
Archives and compressed files are indexed with \fBugrep-indexer\fR option
\fB-z\fR or \fB--decompress\fR.  Otherwise, archives and compressed files are
indexed as binary files or are ignored with option \fB-I\fR or
\fB--ignore-binary\fR.  Note that once an archive or compressed file is indexed
as a binary file, it will not be reindexed with option \fB-z\fR to index the
contents of the archive or compressed file.  Only files that are modified after
indexing are reindexed, which is determined by comparing time stamps.
.PP
Symlinked files are indexed with \fBugrep-indexer\fR option \fB-S\fR or
\fB--dereference-files\fR.  Symlinks to directories are never followed.  
.PP
To save a log file of the indexing process, specify option \fB-v\fR or
\fB--verbose\fR and redirect standard output to a log file.  All messages and
warnings are sent to standard output and captured by the log file.
.PP
A .ugrep-indexer configuration file with configuration options is loaded when
present in the working directory or in the home directory.  A configuration
option consists of the name of a long option and its argument when applicable.
.PP
The following options are available:
.TP
\fB\-0\fR, \fB\-1\fR, \fB\-2\fR, \fB\-3\fR, ..., \fB\-9\fR, \fB\-\-accuracy\fR=\fIDIGIT\fR
Specifies indexing accuracy.  A low accuracy reduces the indexing
storage overhead at the cost of a higher rate of false positive
pattern matches (more noise).  A high accuracy reduces the rate of
false positive regex pattern matches (less noise) at the cost of an
increased indexing storage overhead.  An accuracy between 2 and 7
is recommended.  The default accuracy is 4.
.TP
\fB\-.\fR, \fB\-\-hidden\fR
Index hidden files and directories.
.TP
\fB\-?\fR, \fB\-\-help\fR
Display a help message and exit.
.TP
\fB\-c\fR, \fB\-\-check\fR
Recursively check and report indexes without reindexing files.
.TP
\fB\-d\fR, \fB\-\-delete\fR
Recursively remove index files.
.TP
\fB\-f\fR, \fB\-\-force\fR
Force reindexing of files, even those that are already indexed.
.TP
\fB\-I\fR, \fB\-\-ignore\-binary\fR
Do not index binary files.
.TP
\fB\-q\fR, \fB\-\-quiet\fR, \fB\-\-silent\fR
Quiet mode: do not display indexing statistics.
.TP
\fB\-S\fR, \fB\-\-dereference\-files\fR
Follow symbolic links to files.  Symbolic links to directories are
never followed.
.TP
\fB\-s\fR, \fB\-\-no\-messages\fR
Silent mode: nonexistent and unreadable files are ignored, i.e.
their error messages and warnings are suppressed.
.TP
\fB\-V\fR, \fB\-\-version\fR
Display version and exit.
.TP
\fB\-v\fR, \fB\-\-verbose\fR
Produce verbose output.  Files are marked A for archive, C for
compressed, and B for binary or I for ignored binary.  Deletions
are marked D.
.TP
\fB\-X\fR, \fB\-\-ignore\-files\fR, \fB\-\-ignore\-files\fR=\fIFILE\fR
Do not index files and directories matching the globs in FILE
encountered during indexing.  The default FILE is `.gitignore'.
This option may be repeated to specify additional files.
.TP
\fB\-z\fR, \fB\-\-decompress\fR
Index the contents of compressed files and archives.  Hidden files
in archives are ignored unless option \fB\-.\fR or \fB\-\-hidden\fR is specified.
Option \fB\-I\fR or \fB\-\-ignore\-binary\fR ignores compressed binary files.  When
used with option \fB\-\-zmax\fR=\fINUM\fR, indexes the contents of compressed
files and archives stored within archives up to NUM levels deep.
Supported compression formats: gzip (.gz), compress (.Z), zip, 7z,
bzip2 (requires suffix .bz, .bz2, .bzip2, .tbz, .tbz2, .tb2, .tz2),
lzma and xz (requires suffix .lzma, .tlz, .xz, .txz),
lz4 (requires suffix .lz4),
zstd (requires suffix .zst, .zstd, .tzst),
brotli (requires suffix .br),
bzip3 (requires suffix .bz3).
.TP
\fB\-\-zmax\fR=\fINUM\fR
When used with option \fB\-z\fR (\fB\-\-decompress\fR), indexes the contents of
compressed files and archives stored within archives by up to NUM
expansion levels deep.  The default \fB\-\-zmax\fR=1 only permits indexing
uncompressed files stored in cpio, pax, tar, zip and 7z archives;
compressed files and archives are detected as binary files and are
effectively ignored.  Specify \fB\-\-zmax\fR=2 to index compressed files
and archives stored in cpio, pax, tar, zip and 7z archives.  NUM
may range from 1 to 99 for up to 99 decompression and de\-archiving
steps.  Increasing NUM values gradually degrades performance.
.SH "EXIT STATUS"
The \fBugrep-indexer\fR utility exits with one of the following values:
.IP 0
Indexes are up to date.
.IP 1
Indexing check \fB-c\fR detected missing and outdated index files.
.SH EXAMPLES
Recursively and incrementally index all non-binary files showing progress:
.IP
$ ugrep-indexer -I -v
.PP
Recursively and incrementally index all non-binary files, including non-binary
files stored in archives and in compressed files, showing progress:
.IP
$ ugrep-indexer -z -I -v
.PP
Incrementally index all non-binary files, including archives and compressed
files, show progress, follow symbolic links to files (but not to directories),
but do not index files and directories matching the globs in .gitignore:
.IP
$ ugrep-indexer -z -I -v -S -X
.PP
Force re-indexing of all non-binary files, including archives and compressed
files, follow symbolic links to files (but not to directories), but do not
index files and directories matching the globs in .gitignore:
.IP
$ ugrep-indexer -f -z -I -v -S -X
.PP
Same, but decrease index file storage to a minimum by decreasing indexing
accuracy from 4 (the default) to 0:
.IP
$ ugrep-indexer -f -0 -z -I -v -S -X
.PP
Increase search performance by increasing the indexing accuracy from 4
(the default) to 7 at a cost of larger index files:
.IP
$ ugrep-indexer -f7zIvSX
.PP
Recursively delete all hidden ._UG#_Store index files to restore the directory
tree to non-indexed:
.IP
$ ugrep-indexer -d
.SH COPYRIGHT
Copyright (c) 2021-2025 Robert A. van Engelen <engelen@acm.org>
.PP
\fBugrep-indexer\fR is released under the BSD\-3 license.  All parts of the
software have reasonable copyright terms permitting free redistribution.  This
includes the ability to reuse all or parts of the ugrep source tree.
.SH "SEE ALSO"
ug(1), ugrep(1).
.SH BUGS
Report bugs at:
.IP
https://github.com/Genivia/ugrep-indexer/issues