Search TXT, Markdown, PDF, DOCX and PNG/JPG/WebP/BMP images.
Enter one keyword or phrase per line. All keywords must occur in the same text line, PDF page or DOCX paragraph when you choose All. Spaces within a line remain part of the phrase.
Text PDFs are searched directly. Pages without selectable text and images are recognized when enabled. Completed recognition stays on this page for retry or a new search.
Search across documents and images with one keyword or phrase per line. Match any keyword or require all within the same text section or image.
Results include the source filename, matching excerpt and line, page or paragraph number. Export every completed match to CSV, including results beyond the visible preview. Cancel while keeping completed files, or retry a file that could not be read.
Document Search FAQ
Can I search scanned PDFs?
Yes. Enable text recognition to search scanned pages and PNG/JPG/WebP/BMP images, then choose the text language. Recognition can miss or misread text; check important results against the source.
How are result positions determined?
PDF results use the actual source page number. DOCX results count paragraphs in the main document in file order, including empty paragraphs; these are paragraph positions, not Word page numbers. Text and Markdown results use source line numbers.
Are all matches counted?
Yes. Every matching occurrence is counted and kept for the CSV report; the preview displays results in batches. Matching is literal, with optional case sensitivity. All keywords must be present within the same PDF page, DOCX paragraph or text line when All is selected. Matches do not span separate sections or files. Images are matched as a whole.
Are my files uploaded?
No. Files and recognized text are processed in your browser. Recognition resources may need downloading first. Closing the page clears its temporary search results.
Code Formatters
Format and beautify Java, Python, JavaScript, JSON, HTML and more; adjust indentation, spacing and line breaks.