As a user, I want to be able to view and access data that is residing in an HDFS cluster.
When I click the HDFS link in the side rail menu a table view of the directories is displayed.
- The user can search files and directories within the list displayed. Acceptable queries are either full or partial match. E.g.; for the query "Name" files that are named "Name" or contained "Name" are displayed.
- The user can navigate back to previous directories by clicking the breadcrumb path.
• Only previous three levels are surfaced to the user. If there are more than three levels the breadcrumbs will be presented as follow: … > Directory3 > Directory4
• Clicking on the […] displays a dropdown with the list directories names (e.g. Directory1; Directory2)
- The user can view the number of files and Directories displayed in that page
- The Directories and files are organized around the following columns
• Name (name may be truncated depending on screen size?)
• Type (directory, type of file - parquet, csv, etc.)
• Size of directories and files. The size will be converted to the following units Bytes, KB, MB, GB, etc. A conversion will be calculated based on the size of the file or a directory. Sizes are displayed with one decimal place
• If size cannot be calculated the value is displayed as "- -"
• Last Modified (MM/DD/YYYY - HH:MM)
• Owner (name of owner)
• Group (name of group)
• Permission (surface string)
- The user can sort the list by clicking the columns header.
- Icons visually differentiate Directories from Files
- In the list view, each row is clickable.
- Row background changes color on hover.
- Clicking on a directory takes on a different screen and displays the directories and files inside that directory. The user can navigate back by clicking on the breadcrumb or by clicking the browser back button
- Clicking on a file
1. System checks if file format is recognizable
2. Files open in Data Prep (system applies a default parse)