grapheekdb 0.0.25

GrapheekDB is a pure Python graph database which is fast and lightweight

Getting started :

• Documentation index

• Quick install guide

• GrapheekDB features

• Intro : What is a graph database ?

Tutorial :

• Part 1 : Adding and removing nodes and edges

• Part 2 : Lookup nodes and edges

• Part 3 : Path traversal, collecting and aggregating data

• Part 4 : Scaling our app : indexes, performance tips

• Part 5 : Production use : client/server configuration

Links:

• Home Page : https://bitbucket.org/nidusfr/grapheekdb

Tests and documentation :

code base test coverage is currently : 99 %

documentation is a work in progress

Backends :

It can use various KVS (Key/Value Store) backends :

• Local memory (default backend - not transactionnal)

• Kyoto Cabinet : http://fallabs.com/kyotocabinet/

• Symas LMDB : http://symas.com/mdb/

• Sqlite3 - not transactionnal - graph can become inconsistent

• Shelve - not transactionnal - graph can become inconsistent

Known issues :

There’s an issue when running all tests as it opens a lot of files

You may need to raise limits before running tests.

If you want to raise those limits permanently, on Linux, you’ll need to edit /etc/security/limits.conf and add those lines :

username soft nofile 8192 username hard nofile 8192

• Change username to fit your needs

• 8192 appeared to be enough for tests to run

Then, depending on your linux distribution, you may need to edit /etc/pam.d/login to add :

session required /lib/security/pam_limits.so

About me :

I’m a freelance working with Python and Django since 2000. You can find more information on my Website : Nidus (In French)

If you have interesting projects (involving GrapheekDB or not), don’t hesitate to contact me :)

See you

Contributors :

• Edwin Cox (sqlite3 and shelve backends)

• Pasdavoine (bug tracking and fixing)