Crawls rubygems dependencies graph
First, clone the repository:
git clone https://github.com/anvaka/allgems
cd ./allgems
npm install
Now you can crawl:
- Index all gems into file:
node index.js
. This will use rubygems search api and download all packages intogems.json
file. Most likely this will take ~2-3 hours. It is slow because the search API limits results to 30 per page. - Convert it to graph:
node toGraph ./gems.json
Once the graph is ready you will have a graph.out
file in the folder. Use it
to run the layouter:
node layout.js ./graph.out
This will take around one hour and produce position files in the data
folder.
Finally you need to convert the positions into binary format, which can be
understood by visualization:
node toBinary ./graph.out
This will produce two files:
labels.json
- records all package names in a flat arraylinks.bin
- records all graph connections in a binary format.
These two files and the last positions file (./data/100.pos
) needs to be copied
into the data folder of the visualization.
MIT