HistoryBook is a browser history search engine that enables you to perform searches against the content of web pages you have previously visited.
We have all had this problem - we are trying to find a page we were looking at a few weeks/months ago. We try Google, and can't seem to find the right page. We try searching our browser history, but it only searches page titles. HistoryBook is the solution you have been looking for! It searches the actual content of those pages so you can find that page again.
Note that HistoryBook is in alpha state. There are numerous features that need to be implemented before you would want to use it in "production".
For example:
- security/privacy concerns have not been addressed yet
- App is a Minimum Viable Product (with an emphasis on minimum)
Still here?
First, start the service:
prompt> ./gradlew run
To set up your browser, you have two options:
Setup Option 1: If you are using Chrome, you can use the extension at historybook-chrome-extension. This has a number of other advantages (OmniBox integration and full page indexing).
Setup Option 2: Use the proxy server by setting your proxy configuration to localhost:8082
On a Mac, you can do this at:
System Preferences -> Network -> Advanced -> Proxies -> Web Proxy (HTTP)
Note that Option 2 will only index the content of the initial page fetch, not any subsequent content loaded via JavaScript.
Use: Start browsing. As you browse, the pages you visit will be added to the search index in real time.
To perform a search, point your browser to http://localhost:8080/search
Note that you can get search engine debug information (useful for development) by appending "&debug=true" to the URL.
If you are using Chrome and want to bulk load your browser history, take a look at the historybook-import project.
HistoryBook consists of a single process that exposes two services.
One is web service that exposes a search index allowing searching and adding pages to the index. It also has an endpoint that provides a web user interface to the search. Lucene is used as the backend search engine.
The second service is a web proxy that transparently adds viewed pages to the search index.
Copyright 2016 Andrew W. Buchanan (buchanan@difference.com)
Licensed under the Apache License, Version 2.0