19

Looking for a Windows program (similar to "Everything") that can:

  • Scan a very large (2-3TB, 10,000s or even 100,000s of files in 1,000s of folders) directory recursively
  • For each "text" file, index ALL the text in it fully
  • Offers ability to find which files a given search string is in.
  • "text" files at the minimum include .txt, .pdf, .epub, .mobi. Ideally other known ebook formats (.fb2, .doc, .docx)
  • Advanced search ability desired (search for all/any of a set of search terms, negate search term. Regex, ideally. PCRE really ideally).

Optional desired features:

  • Support for non-English text, both for indexing and searching, in other formats; specifically Russian text (KOI-8, Windows-1251)
  • Support for Unicode (indexing and searching).
  • Can look through archives (at least, .zip and .rar)
  • Good GUI (think "Everything" app) to display search results. Fast filtered grid, action menu for every found file including copying location, opening containing folder, copying/cutting file as if in Windows Explorer menu.
  • Automatically keeps the index refreshed as files are added/removed/changed on filesystem, the way Everything does.
  • Ideally, gratis but not required as long as the price is sensible.
  • Windows XP required. Windows8 extra bonus points.
  • 1
    Nice sci-fi collection! – Franck Dernoncourt Mar 9 '14 at 3:23
  • Next time, please tag the specific operating system in your question. Also you said WindowsXP is required, but you put it in your optional features. Please clarify. I missed that point and wrote an answer up for Windows 7 (which will work with Win8 I believe), which I will still post. – Yos233 Mar 9 '14 at 23:05
  • @Yos233 - I would rather have a good software that's 8-only, than none (I can move the collection from XP to 8 if required). But currently it's on XP system. – DVK Mar 9 '14 at 23:28
  • 1
    Keep in mind that Regexes/PCRE are NOT compatible with indexing approach. Fulltext indexes cannot be utilized by regex searches, there is not enough information in them to serve a regex. For regex-based approach, you need to perform classic search. – miroxlav Apr 14 '14 at 0:20
2

As far as I am aware, Windows 7 Explorer has all the basic features you need, and some of the optional features as well.


You can index a file directory in Windows (7+) in two ways. Directly index the directory, or make it a library.

Direct Index: See here: wikiHow: How to Add a Folder to the Windows 7 File Index

Create Library: In Explorer, go to "Libraries", and click on "New Library".


Advanced Search in Windows is something I had to look up just for this, but How-To Geek has a very informative article on that. Article

Also make sure you have file content searching enabled: wikiHow: How to Make Windows 7 Search File Contents


Addendum: I noticed after writing this that the OP asked for Windows XP. I'm keeping this up even if it isn't accepted for someone else who comes along (and so I didn't waste 30 minutes).

  • 2
    How well does it scale? Specifically can it effectively handle the size stated in the question without bogging down the system both indexing and searching? Can it index contents of text PDFs? EPUBs? – DVK Mar 9 '14 at 23:32
  • I don't think Explorer would bog down the system (at the expense of possibly slower search). PDF indexing is possible through the Adobe IFilter .dll, adobe.com/support/downloads/detail.jsp?ftpID=4025. I couldn't find a similar one for EPUB, so potentially not. – Yos233 Mar 10 '14 at 0:23
  • Wouldn't bog down the system? It bogs down opening (forget search) a directory with several thousand files in it!!! – DVK Mar 10 '14 at 0:41
  • You asked the question, and i gave you my best answer. To me, bogging the system down refers to slowing down other processes (ie CPU hog). Yes of course searching thousands of files will take some time. I don't think you will find a single program that can search your sizable library with any great amount of speed. – Yos233 Mar 10 '14 at 0:58
  • I wasn't referring to indexing. I was referring to displaying results. Everything can display thousands of results snappyly. Windows Explorer will get frozen – DVK Mar 10 '14 at 1:37
2

Regain (desktop search)

 

You can find a list of desktop search managers on Wikipedia but I find the Regain opensource project as being a sensible choice, beside that is free(as in libre) and also opensource and still in development which means new features will appear(full features list here).

Short Description

Regain is a Java search engine based on Jakarta Lucene. It provides indexing and searching files for plenty of formats (HTML,XML,doc(x),xls(x),ppt(x),oo,PDF,RTF,mp3,mp4,Java). A TagLibrary eases integrating search results in your JSP based web page.

Main features I find very useful:

  • web server (so it can be accessed in LAN on all the devices in your LAN)
  • tray icon (fast access)
  • cached version of the indexed file (sometimes loading a big PDF doesn't pay)
  • search keywords (powerful Lucent keyword set)
  • can support additional formats (regain supports I-Filter)
  • supports API

Just a few snapshots:

enter image description here

enter image description here

  • 1
    I would love to use this, but documentation is lacking... – Matt Aug 24 '17 at 17:09
1

Copernic Desktop Search

For fulltext-based approach take either Windows Search or Copernic Desktop Search ($50 for full version, free version ("Lite") for non-commercial use is limited to 75,000 files).

Especially Copernic Desktop Search has all capabilities you need. I have tested it with 4,000,000 documents, searching was still very fast. It recognizes operators like AND, OR, NOT, NEAR to support your searches.

1

Maybe DocFetcher could help.

From the home page:

The application runs on Windows, Linux and Mac OS X, and is made available under the Eclipse Public License.

Notable Features

  • A portable version: There is a portable version of DocFetcher that runs on Windows, Linux and Mac OS X. How this is useful is described in more detail further down this page.
  • 64-bit support: Both 32-bit and 64-bit operating systems are supported.
  • Unicode support: DocFetcher comes with rock-solid Unicode support for all major formats, including Microsoft Office, OpenOffice.org, PDF, HTML, RTF and plain text files. The only exception is CHM, for which we don't have Unicode support yet.
  • Archive support: DocFetcher supports the following archive formats: zip, 7z, rar, and the whole tar.* family. The file extensions for zip archives can be customized, allowing you to add more zip-based archive formats as needed. Also, DocFetcher can handle an unlimited nesting of archives (e.g. a zip archive containing a 7z archive containing a rar archive... and so on).
  • Search in source code files: The file extensions by which DocFetcher recognizes plain text files can be customized, so you can use DocFetcher for searching in any kind of source code and other text-based file formats. (This works quite well in combination with the customizable zip extensions, e.g. for searching in Java source code inside Jar files.)
  • Outlook PST files: DocFetcher allows searching for Outlook emails, which Microsoft Outlook typically stores in PST files.
  • Detection of HTML pairs: By default, DocFetcher detects pairs of HTML files (e.g. a file named "foo.html" and a folder named "foo_files"), and treats the pair as a single document. This feature may seem rather useless at first, but it turned out that this dramatically increases the quality of the search results when you're dealing with HTML files, since all the "clutter" inside the HTML folders disappears from the results.
  • Regex-based exclusion of files from indexing: You can use regular expressions to exclude certain files from indexing. For example, to exclude Microsoft Excel files, you can use a regular expression like this: .*.xls
  • Mime-type detection: You can use regular expressions to turn on "mime-type detection" for certain files, meaning that DocFetcher will try to detect their actual file types not just by looking at the filename, but also by peeking into the file contents. This comes in handy for files that have the wrong file extension.
  • Powerful query syntax: In addition to basic constructs like OR, AND and NOT DocFetcher also supports, among other things: Wildcards, phrase search, fuzzy search ("find words that are similar to..."), proximity search ("these two words should be at most 10 words away from each other"), boosting ("increase the score of documents containing...")

Supported Document Formats

  • Microsoft Office (doc, xls, ppt)
  • Microsoft Office 2007 and newer (docx, xlsx, pptx, docm, xlsm, pptm)
  • Microsoft Outlook (pst)
  • OpenOffice.org (odt, ods, odg, odp, ott, ots, otg, otp)
  • Portable Document Format (pdf)
  • EPUB (epub)
  • HTML (html, xhtml, ...)
  • TXT and other plain text formats (customizable)
  • Rich Text Format (rtf)
  • AbiWord (abw, abw.gz, zabw)
  • Microsoft Compiled HTML Help (chm)
  • MP3 Metadata (mp3)
  • FLAC Metadata (flac)
  • JPEG Exif Metadata (jpg, jpeg)
  • Microsoft Visio (vsd)
  • Scalable Vector Graphics (svg)
0

WinGrep

You can search on demand for given words even in binary files, (n.b. this will not work well for some PDFs e.g. those from scanners). with wingrep - it is free and will search even inside .zip files. I does not slow the PC down all the time nor does it use a lot of disc space as it doesn't build indexes but as a result it doesn't run that fast. It is Free from Micro$oft so will probably work on most flavours of Windows.

Calibre eBook Manager

Doesn't search inside of files AFAIK but searches eBook metadata and you can edit the metadata but calibre has the following features:

  • It is specifically for maintaining libraries of eBooks,
  • can convert between formats for you,
  • includes viewers for many formats,
  • can manage eBooks on most device(s).

It is Free and Open Source and will run just about everywhere.

I would strongly recommend getting it whatever you do.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.