how to contribute a recipe

You will be making a sentence, choose an id, create a working directory, and update related object JSON files.

example:

• How to find the business relationship between asns: markdown | see in catalog
• How to parse the ITDK: markdown | see in catalog

License

Please add the following to your recipe and related scripts, with the correct dates :

Copyright (c) 20** The Regents of the University of California
All Rights Reserved

Steps

First follow the sets in the general how to update. then add the following notes for recipes:

1. select id : select an "id" that matches your sentence using only numbers and lower case characters, and doesn't match the directory name of an existing recipe.
   • "How to make a recipe." could be how_to_make_a_recipe
2. create an issue : create an issue with the title "draft "+id with a link to your recipe's directory
   • '''draft how-to-make-a-recipe'''
   • take the issue's number and place it at the front of your branch. The names do not need to match.
3. create a branch with your issue_id and something understandable as your selected_id name.

• 345-how-to-make-a-recipe
• if this recipe is part of a wide group of changes, it should use the group's branch

1. create directory : create a Readme.md in a directory with the same name as the recipe's id in recipe
   • '''sources/recipe/(issue number)-how-to-make-a-recipe/Readme.md'''
2. add metadata : at the top of the recipe you will create a JSON file that contains the recipe's metadata in JSON. Below is the standard format we use:

{
    "id" : ,
    "name" : ,
    "description": ,
    "links": [
        
    ],
    "tags" : [
        
    ],
    "authors":[
        
    ]
}

• any value in content will be overwritten by the recipe below
• you can find examples github.com/CAIDA/catalog-data/tree/master/sources/recipe
• place a short introduction into the description
• do not add a date (it is automatically created by the server )
• Under links section, for each dataset you used, indicate the dataset id using the format {"to": "dataset:some_dataset_id"}
• Under tags section, write relevant tags that help describe this recipe.

1. add content : all the text below the metadata is stored as markdown in the recipe's content.
   • code recipes suggested structure:
     • introduction : targeted to domain expert with context for code
     • code : code snippet
       • we suggest putting the full script in the directory and only the relevant code into the Readme.md
     • background/caveats : information needed by none domain experts
     • caveats : caveats about the dataset or the recipe, such as limitations
2. commit for review : after the first draft is done a second individual will review and test the sample code
   • repeat as nessary
3. pull your branch to master : after finishing review, the recipe will be merged with master
4. pull master to v1 : once any conflicts have been resolved from merging with master, master will be merged with v1

catalog markdown != github markdown

The markdown module that we are currently using for recipes does not parse markdown the same way that github does.

• You must use ~~~ not ``` for code blocks
• There must be white space above and below the first and last ~~~
  • will not work

  something above
  ~~~
  this block will not be parsed correctly
  ~~~
  something below
  

  • this will work

  
  ~~~
  this block will be parsed correctly
  ~~~
  
  

• Tables also require white space above and below.
• use dash (-) instead of special characters (•)

 • wrong 
 - right
    - right also