fquery

SQL-like queries on files and their contents. Fast, multi-threaded.

# Find TODOs in source files
fquery 'SELECT * FROM "src/" WHERE LINE CONTAINS "TODO"'

# Count lines per file type with size
fquery 'SELECT FILEEXT, SUM(LINECOUNT) AS Lines, "{SUM(FILESIZE)/1024/1024} MB" AS Size FROM "." GROUP BY FILEEXT ORDER BY SUM(LINECOUNT) DESC'

# Export as JSON
fquery --json 'SELECT FILENAME AS file, LINECOUNT AS lines FROM "src/" WHERE FILEEXT = "rs"'

Install

cargo build --release
# Binary at target/release/fquery

Features

SQL Syntax

Full query language: SELECT, FROM, WHERE, GROUP BY, HAVING, ORDER BY, LIMIT, OFFSET, DISTINCT, SEPARATOR.

17+ Columns

Column	Description
LINE	Line content
LINENO	Line number
FILENAME	File name (e.g. main.rs)
FILEPATH	Full path
FILEDIR	Directory part
FILEEXT	Extension (e.g. rs)
FILESIZE	Size in bytes
FILETYPE	text or binary
MODIFIED	Last modified (UTC)
CREATED	Creation time (UTC)
LINECOUNT	Total lines in file
WORDCOUNT	Total words in file
CHARCOUNT	Total characters in file
CONTENT	Entire file content

Aggregates

COUNT(*), COUNT(col), SUM(col), AVG(col), MIN(col), MAX(col) -- freely mixable with columns.

String Functions

TRIM, TRIMSTART, TRIMEND (with optional char arg), UPPER, LOWER, SPLIT, REPLACE, LEN.

Format Strings

fquery 'SELECT "{FILENAME}: {FILESIZE/1024} KB" FROM "src/"'
fquery 'SELECT "{FILENAME}: {SUM(LINECOUNT)} lines" FROM "src/" GROUP BY FILENAME'

Variables, aggregates, and arithmetic (+-*/%) inside {}. Escape with \{.

WHERE Conditions

=, !=, CONTAINS, MATCHES (regex), LIKE (SQL wildcards %/_), BETWEEN, IN (...), LEN(), parentheses (A OR B) AND C, FROMFILE("path"), NOW().

# Files modified today
fquery 'SELECT FILENAME FROM "src/" WHERE MODIFIED > "2026-05-20 00:00:00"'

# Files larger than 1 MB
fquery 'SELECT FILENAME FROM "." WHERE FILESIZE BETWEEN 1024*1024 AND 1024*1024*10'

# Only Rust and Python
fquery 'SELECT * FROM "src/" WHERE FILEEXT IN ("rs", "py")'

Output Formats

fquery --json 'SELECT FILENAME AS file, LINECOUNT AS lines FROM "src/"'
fquery --csv 'SELECT FILENAME AS file, FILESIZE AS bytes FROM "src/"'
fquery --excel -o data.csv 'SELECT FILENAME AS file, FILESIZE AS bytes FROM "src/"'
fquery --table 'SELECT FILENAME AS File, FILESIZE AS Bytes, LINECOUNT AS Lines FROM "src/"'

Column aliases with AS. Headers auto-shown when aliases are present. --no-header to suppress.

Performance

• Multi-threaded file scanning
• File-level query optimization: skips reading file contents when only metadata is needed
• Binary file detection (512-byte null check) -- skipped automatically
• FILESIZE/FILETYPE pre-filtering before any I/O
• Competitive with hand-written bash pipelines on large directory trees

Piping

# Read from stdin
echo "data" | fquery 'SELECT * FROM - WHERE LINE CONTAINS "data"'

# Chain queries
fquery 'SELECT FILEPATH FROM "src/" WHERE LINE CONTAINS "error"' | fquery 'SELECT * FROM -'

Full Documentation

fquery docs

License

MIT