Search integration
I am working on a website that currently has several scattered search functions, for example:
- Scanning through the front door of the website
- Search that links to a web service
- etc...
What would be the best way to tie them together and provide what appears to be a single search function?
I found the following list on wikipedia
Free open source enterprise search software
- Lucene and Solr
- Xapian
Proprietary business search software providers
- AskMeNow
- Autonomy Corporation
- Search concept is limited
- Coveo
- Dieselpoint, Inc.
- dtSearch Corp.
- Endeca Technologies Inc.
- Exalead
- Expert system SpA
- Funnelback
- Google Search Appliance
- IBM
- ISYS software
- Microsoft (includes Microsoft Search Server, Quick Search and Transfer):
- Open text corporation
- Oracle Corporation
- Queplix Universal Finder
- SAP
- TeraText
- Vivisimo
- X1 Technologies, Inc.
- ZyLAB Technologies
Thanks for any advice regarding this.
source to share
Solr is an incredibly flexible search solution. Just last year I coded 2 solr-based sites and worked on a third existing one, each one worked in a completely different way.
Solr just eats XML queries to add something to the index, and XML queries to find something inside the index. It doesn't scan or extract text for you, but most of the time it's easy to do. There are many existing add-ons to Solr / Lucene glass, so maybe something already exists for you.
I would steer clear of proprietary software unless you're sure Solr isn't enough. This is one of the nicest programs I have worked with, very flexible when you need it and at the same time you can start in a few minutes without reading a long manual.
source to share
Note that no matter which search solution you use, your search setup is "scattered" in nature. You will still have a pointer, search user interface, or "framework".
You will BE SURE if you marry a specific search technology. You really want the UI to be separate from the search backend. The backend may stop scaling, or there may be a better search engine tomorrow.
Search engine switching is very common, so never - never - write your interface with a specific search engine. Always abstract it so the UI is unaware of the search technology used.
Keep it modular and you will thank yourself later.
Using a standard web services interface, you can also let third parties create material for you, and they don't have to "learn" any search engine you use on the backend.
source to share
Take a look at these similar questions:
- Best text search engine to integrate with a custom web application
- How to implement a search function on a website?
My personal recommendation: Solr .
source to share
All of these companies offer different Universal Search features. Smaller companies have carved themselves into highly functional and highly desirable niches. For example, Queplix allows any search engine to work with structured data and enterprise applications by retrieving data, business objects, roles, and permissions from all indexed applications. It provides ranking criteria as well as compliance alerts.
Two other solutions that were not as well known and available or were available at the time of the original question:
- Google Custom Search - especially as the Disable Public URL option was recently added
- YaCy - You can join the web, or download and flip your own independent servers.
source to share