Timeline for What's the best way to index PDF or Word document for full text search?
Current License: CC BY-SA 3.0
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 8, 2015 at 9:10 | comment | added | Naou | Thanks, could you share your architecture set up in github please to use it as a starter point? | |
| Nov 8, 2015 at 9:02 | vote | accept | Naou | ||
| Nov 8, 2015 at 9:01 | vote | accept | Naou | ||
| Nov 8, 2015 at 9:01 | |||||
| Nov 8, 2015 at 8:41 | comment | added | Markus W Mahlberg | @Naou I am afraid, but you have to dig for this yourself. There is no tutorial for this. And you link ElasticSearch to MongoDB's GridFS (where you store the files). You can use Solr or Lucene, but the you'll have to integrate it yourself. With the stacks I suggested, the effort is relatively small ;). Scaling problems? No, if you know how to scale properly. Scaling is a very hard thing to do properly, and far out of scope. But in general, yes, the setup I suggested is almost infinitely scalable. Most useful resource for starters: university.mongodb.com/courses/M102/about | |
| Nov 8, 2015 at 8:28 | comment | added | Naou | thanks for your answer, so you suggest a 3-Tier web application with MongoDB, Elastic and Spring MVC, Spring Security for authentication, However I still don't know how to map my documents(Word, PDF, ...) with my application, should I link mongodb to the file system or elasticsearch? Why don't use just Solr or lucene since it's not a distributed application, I will have any scale problems. At least can you share with me some usefull resources or tutorials where I can get up and runing with this 3-tier solution please? Regards | |
| Nov 7, 2015 at 14:42 | review | First posts | |||
| Nov 7, 2015 at 18:44 | |||||
| Nov 7, 2015 at 14:38 | history | answered | Markus W Mahlberg | CC BY-SA 3.0 |