search API for Noormags Server:

search a text inside of documents, with URLs like these:

http://127.0.0.1:5000/search?query=test

or

http://127.0.0.1:5000/search/test

output is documents with JSON format, like this:

{
	"query":"_query_",
	"result":[
		{
			"rank":1,
			"title":"_title1_"
			"id":"_id1_"
		},
		{
			"rank":2,
			"title":"_title2_"
			"id":"_id2_"
		}
	]
}

search a document by its ID, with URLs like these:

http://127.0.0.1:5000/content?id=1

or

http://127.0.0.1:5000/content/1

output is a document with JSON format, like this:

{
	"id":"_id_",
	"content":"_content_"
}

for installation of PyLucene, use below instruction:

sudo apt-get install pylucene

for installation of Flask, use below instruction:

sudo pip install flask

for run server with Flask, use bellow sample code:

import flask
from flask import Flask

app = Flask(__name__)

@app.route('/')
def index():
	return 'Welcome to Noormags-Server'

if __name__ == "__main__":
    app.run()

for test your code, you can use below instruction:

curl http://127.0.0.1:5000

• Note: by default flask listen on port 5000, but you can change it within your codes!

you can call index function of Indexing class, like below:

from indexing import Indexing

handler = Indexing()
handler.index()

we find all xml files in a directory that can set by passing an argument through index function of Indexing class, we called it doc_dir. after that, stem file, seperate all sections (eg, ID - Title - Content - ...) and finally index document by Lucene.

you can call retrieve function of Retrieval class, like below:

from retrieval import Retrieval

handler = Retrieval()
result = handler.retrieve(query)

we find all documents that matches with input query.