1. Home
  2. Documents
  3. Search Text in Multiple Files

Search Text in Multiple Files

Find words across many PDF, Word, Hangul and TXT files at once.

Runs in your browser · No upload
Add files or a folder and their text is read once. After that you can change the search and results update instantly.
Separate several terms with commas or new lines to find any of them. In regex mode each line is one pattern.

How to use

  1. Drop or choose files. To search a whole folder, press Choose folder. Up to 500 files and 1 GB in total.
  2. As soon as files are added, their text is read one by one and kept in memory. A progress bar shows, and Stop reading halts it.
  3. Type words in Search for. Separate several with commas or new lines to find any of them (OR).
  4. Results are grouped by file, most matches first. Each match shows its page, slide or sheet, 40 characters of context on each side, and the matched text highlighted.
  5. Files without matches and files that could not be read (password-protected, damaged) are collected in collapsed lists. Results CSV saves file, location and context.

A file is read only once, so changing the search or options updates the results immediately.

Search options

Option Behavior
Default Finds exactly what you typed. Symbols such as ., + and ( count as plain characters
Whole word Matches only when not next to letters, digits or underscores. In Korean, a particle attached to a word counts as more letters
Match case Off by default
Regex One regular expression per line. An invalid pattern is reported with the reason, and ^ / $ work per line

Example: searching 사과 as a whole word in 사과 사과를 사과, 사과1 matches only the first and third occurrences (사과를 and 사과1 are excluded). The regex \d{2,3}-\d{3,4}-\d{4} finds phone numbers shaped like 010-1234-5678 and 02-123-4567.

Hangul jamo are not decomposed, so initial-consonant search (finding 사과 with ㅅㄱ) is not supported.

Location labels and limits

  • Location is a page for PDF, a slide for PPTX and the sheet name for XLSX. For HWP and HWPX the page is estimated by matching the same phrase in page text in order, so it may be empty for table-only pages. DOCX, TXT and similar have no page information.
  • Regexes run in a separate worker, and a file is skipped if its search takes longer than 4 seconds. At most 10,000 matches are collected in total.
  • A scanned PDF has no text layer, so nothing is found in it. Old .doc, .xls and .ppt files cannot be read.
  • Each file may be up to 200 MB, and only the first 50 million characters of text are searched.

Related: Document Word Count, Compare Documents, Redact PDF, HWP Personal Data Finder.

Your files are processed only in your browser and never sent to a server.

Can't find the tool you need?

Tell us what you want to do. We build the most requested tools first.