pdfgrep

pdfgrep [OPTIONS] PATTERN FILE...pdfgrep [OPTIONS] {-e PATTERN|-f FILE}... FILE...pdfgrep [OPTIONS] -r|-R PATTERN [FILE|DIR...]

pdfgrep searches for text patterns in PDF files using the Poppler library for text extraction. It provides a familiar grep-like interface for PDF documents.Text is extracted from each page and matched against the given regular expression. By default pdfgrep uses PCRE2 for pattern matching. Fixed-string matching is available via -F.Page number output (-n) helps locate matches within a document. Restricting the search to a page range (--page-range) speeds up searches on large files. Context lines (-C) show surrounding text to aid understanding of a match.Recursive search (-r) processes entire directory trees. Combined with --include and --exclude, this enables targeted searches across document collections. Multiple patterns can be specified with repeated -e options or read from a file with -f.The --unac option is useful when PDFs use typographic ligatures or accented characters that differ from the search term. The --cache option stores extracted text to accelerate repeated searches.

-e PATTERN, --regexp=PATTERN

Specify a search pattern. Can be used multiple times to match any of several patterns.

Read patterns from a file, one per line.

Case-insensitive matching.

Treat the pattern as a fixed string (no regular expression interpretation).

Use Perl-compatible regular expressions (PCRE2).

Prefix each match with its page number. TYPE is `index` (default) or `label`.

Print match count per file instead of matched lines.

Print match count per page (implies -n).

Print only filenames that contain a match.

Print only filenames that contain no match.

Print only the matched portion of each line.

Print the filename with each match (default when searching multiple files).

Suppress filename prefix in output.

Use a null byte instead of a colon to separate the filename from the rest of the output line. Useful for filenames containing colons or spaces.

Use SEP as the separator between the match prefix (filename, page number) and the matched line, instead of the default colon.

Search all PDF files under each directory recursively. Symlinks are followed only when specified on the command line.

Like -r, but follow all symlinks.

Only search files whose names match GLOB (default: `*.pdf`).

Skip files whose names match GLOB.

Print NUM lines of context after each match.

Print NUM lines of context before each match.

Print NUM lines of context before and after each match.

Limit the search to the specified page range (e.g., `1-10,15`).

Stop after NUM matches per file.

Use PASSWORD to decrypt a password-protected PDF.

Colorize output: `auto` (default), `always`, or `never`.

Cache rendered page text to speed up repeated searches on the same files.

Remove accents and ligatures from both the search pattern and the document text. Useful for matching words like "ae" against the ligature "æ".

Warn when a PDF contains no searchable text (e.g., scanned images without OCR).

Suppress all output. Exit status indicates whether a match was found.

Print version information.

0

One or more matches were found.

No matches were found.

An error occurred.

Text extraction quality depends on the PDF's internal structure. Scanned PDFs without embedded text require OCR preprocessing before pdfgrep can search them (use --warn-empty to detect such files). Complex multi-column layouts may not extract in reading order. Encrypted PDFs require the correct --password.

pdfgrep was written by Hans-Peter Deifel starting around 2010. It uses the Poppler library for PDF parsing and provides a grep-compatible interface for searching PDF text content.

grep(1), pdftotext(1), ripgrep(1), pdfinfo(1)