devmerge_2k18

Demo Presentation - https://docs.google.com/presentation/d/1A6OkhB1GI1wMIGa5h8AZoDasa4xIcA0NWPJnS6ZhJKE/edit#slide=id.p

Modules

• sudarsana-web - Serves the ranked completions and popular patterns for various methods.
• sudarsana-parser - Parses a given package and produces the AST of each expression in the code in JSON Format.
• VSCode plugin that works with the sudarsana-web is available on https://github.com/ashwanthkumar/vscode-go/tree/0.6.93-autobot.

Datasets

Stackoverflow dump of all go tagged questions and all their answers.

You can either download the CSV and have a mult-line CSV parser parse that output. It's a multi-line because of question / answer body. Or you can run the Query directly and download the query revision page (like this page, warning the page is 100M in size). The page contains the resultSet as a JSON embedded in it. Assume you've the JSON extracted from the HTML Page, you can generate the dataset using the following command

$ wget "https://data.stackexchange.com/stackoverflow/revision/913091/1136390/dump-question-answers-for-a-tag" -O dump-question-answers-for-a-tag-with-title
$ ... extract the JSON content alone in the file
$ jq .resultSets[].rows[] dump-question-answers-for-a-tag-with-title | jq -c '{"title": .[0], "qId": .[1], "aId": .[2], "tags": .[3], "qScore": .[4], "aScore": .[5], "question": .[6], "answer": .[7], "views": .[8], "nrOfCommentsInAnswer": .[9], "qDate": .[10], "aDate": .[11]}' > go-so-questions-and-all-answers-with-title.json

Authors

• Ashwanth Kumar
• Mohan