phyllo

PHilologicallY Linguistic LegwOrk.

To run the application:

Initial build:

1. docker for linux (https://docs.docker.com/engine/installation/linux/docker-ce/ubuntu/) OR docker-toolbox for windows(https://www.docker.com/products/docker-toolbox) is required
2. clone or download the files from the github repository

3. unzip the zip file (if downloaded the file)

4. In linux: open terminal and navigate to the unzipped directory with cd command OR   In Windows: open up the docker quickstart terminal and navigate to the unzipped directory with cd command

5. run docker-compose up command

6. A new or initial build is gonna take the longest of around 20 - 30 mins, since the build downloads the data from the http://www.thelatinlibrary.com/ and then generate FTS tables for them
7. The build will be done once the terminal says debugger is active!

To access the search engine:

8. open a browser and enter the url: http://192.168.99.100:5000/

Type of searches that can be performed:

1. The search engine works for any number of words and sentences
2. word*: in the search gets all the documents where the 'word' is a prefix. * can be used multiple times like word1* word2* word3* in the search bar.

3. ^word: this search gets all the documents with sentences where 'word' is the first word. can only be used once in a search and ^word*: this gets all the documents where is 'word' is a prefix of the first word in a sentence

4. to search author names or book titles: author:name or title:name should be used

5. the search author:name word searches for the documents where the authored by the specified author and contains the specified word

6. author:name word* can also be used

7. Also author:name ^word* can also be searched

8. word1 OR word2: this kind of search results all the documents which contain either word1 or word2

9. The NEAR clause can be used to search for non continous words. For example, two words can have any number of words and characters between them an estimate of that number can be used to perform searches as well. word1 NEAR/x word2 where x is the estimated number of characters between word1 and word2.

NEAR clause can be used multiple times through the query.