Home

Welcome to the CrosswalksWorkflow wiki! You should have both the Crosswalks and CrosswalksWorkflow repositories in the

/Users/yourAccountName/ directory.

The pages here facilitate setting up a local machine to succeed at the background processes the analysis works on. Once a user has accomplished setup the pages detailing the usage of scripts should be employed. A basic overview of the workflow is as follows:

Step Zero: Follow the instructions on the following pages to prepare your computer.

• How to Add Saxon Home Edition
• How To Install Anaconda

Step One: Create Branches for the Analysis

• Create branches in both repositories entitled "organizationName_recTag" eg: DataOne_LTER or NASA_UMM-C. This will create a snapshot of all of the data that goes into an analysis.
• Extract dialects used at data center, compare resultant definitions with concepts the recommendation contains to prepare AllCrosswalks.xml to create accurate rubrics.

Step Two: AllCrosswalks.xml Improvements

• Enhancing and creating dialects using the Dialect Extractor and Merge Dialect tools.
• Ingesting and testing of concept references in a recommendation component. Since the set of concepts that are contained in the different parts of the recommendation are the most important concepts for the analysis, compare a list of references in the recommendation

Step Three: Rubric Creation

• Create rubrics using the crosswalks tool
• This can be facilitated through the use of the batch script writeAllRubricTransforms.sh which runs specific instances of rubricTransform.sh
• Test rubrics

Step Four: Ingest Metadata

• Ingest the organization's collections into the directory "rawMetadata". Then copy these collection sub-directories into a sub-directory of CrosswalksWorkflow/collections e.g. "CrosswalksWorkflow/collections/NASA" and begin to clean the directory structure and ensure namespace agreement for the prefix used in AllCrosswalks.

Step Five: Create json

• Using the batch script, jsonCreator.sh, leverage runTransform.sh to create the json needed for your analysis.

Step Six: Create xlsx

• Use the generateSS.sh script to run the python scripts needed to combine the json. The resultant file is deposited into Reports under the recommendation tag.

Step Seven: Create Visualizations