Workflow

Sample Sourcerer Workflow

From running the crawler to setting up code search

STEP 1: Run Crawler

STEP 2: Set up Repository

STEP 3: Run fact extractor

STEP 4: Run DB Importer

STEP 5: Usage Calculation

• Required Services: DB

• TODO
• run run-usage-writer.sh
• also, create
  popular-fqn.txt – list of highly ised fqns, to be used as ‘stop-words’ (ie exclude them while computing similarity)

STEP 6: Compute Similarity Data Model

STEP 7: Index, first pass just usage fields

• Required Services: DB, File Server

STEP 8: Index, second pass (full index)

• Required Services: DB, File Server, Solr server serving index from first pass. Similarity Server
• refer to http://github.com/sourcerer/Sourcerer/raw/master/infrastructure/services/solr-config/deployment-base/README.txt

STEP 9: Run search server (Solr) serving second pass index

STEP 10: Deploy and run one of the code search tool that confirms to the index schema served by the search server