Smart dictionary is a simply data collection and analysis service that utilizes NLP to discover and categorize words and analyze their sentiment (still pending at this time). Right now, the categorization is not great, but in future with the help of data provided by other Qordoba services,
If you're running the test system, you'll want to know how to get the IPs of containers.
Once you've started running a container you'll want to run docker ps in order to find out what the container's
ID is.
Now you can run docker inspect <id> where <id> is a docker container ID. If you want a sed or awk command
you can pretty much go fuck yourself.
smart-dictionarytionary utilizes cutting edge container technology (aka Docker).
In order to run redis we'll use the qordobahub/qordoba-redis docker repository.
We use this due to a bug in more recent kernel verions of Linux affecting the AUFS layer which Docker runs on.
docker run --rm -n --expose 6379 -v `pwd`/redis-data:/volume qordobahub/qordoba-redis /usr/bin/redis-server /etc/redis/redis.conf
This command will run the docker container qordobahub/qordoba-redis and expose on the internal network that
docker uses the port 6379 from this bucket. If you wish to communicate across machines and not simply internally you'll
need to publish the port to the machine with -p 6379:6379.
At the same time, this will attach a volume in the redis-data directory relative to your curren CWD
(which should be the root of the project), which will be used by the redis configuration (which is already in
the container).
Removing the previous container (recommended since containers can be considered essentially stateless in production).
Or, you can run the Makefile command run_redis in the root directory.
Collection as a service runs via a Docker image you can either create yourself with the DockerFile or run directly
from the versioned image in qordobahub/qordoba-smart-dictionarytionary-collection.
To build the image from the Dockerfile instructions:
docker build --no-cache -t qordobahub/qordoba-smart-dictionarytionary-collector .
To run the container:
docker run --rm -n -e REDIS_HOST=172.17.0.3 qordobahub/qordoba-smart-dictionarytionary-collector
Or use the Makefile!
The data processor as a service runs via a Docker image you can either create yourself with the DockerFile or run
directly from the versioned image in qordobahub/qordoba-smart-dictionarytionary-processing.
To build the image from the Dockerfile instructions:
docker build --no-cache -t qordobahub/qordoba-smart-dictionarytionary-processing .```
To run the container:
docker run --rm -n -e REDIS_HOST=172.17.0.3 qordobahub/qordoba-smart-dictionarytionary-processing