-
Notifications
You must be signed in to change notification settings - Fork 24
3.3. Manage Working Files
Paulo Pinheiro edited this page Sep 4, 2019
·
5 revisions
Files that are under development are often stored in the working files part of HADatAc. The called Working Files is a place where files can be created, renamed, organized into folders, and edited within HADatAc. Also, the Working Files place can serve as a file repository since files can be quickly uploaded and downloaded.
The working file has a single folder hierarchy where users are allowed from any folder to create one or more subfolders. For instance, in the figure above, the user is in /CHEAR/DD folder and is allowed to go back in the hierarchy by selecting the two dots, or to create new folders using the "Create Folder" button above.
- Blank Metadata File is described in Section 3.3.1
- Create Folder
- Delete Current Folder
- Preview File as a Spreadsheet or as an image/audio/video
- Edit file inside of an Spreadsheet editor, if the file is a spreadsheet or its format is CSV
- FileType-Specific Editors
- Semantic Data Editor is described in Section 3.3.2.
- Data Dictionary Editor is described in Section 3.3.3.
- Rename file
- Move the file from its current folder to another folder
- Delete permanently removes the file from Working Files
Copyright (c) 2019, HADatAc.org
-
Installation
1.1. Installing for Linux (Production)
1.2. Installing for Linux (Development)
1.3. Installing for MacOS (Development)
1.4. Deploying with Docker (Production)
1.5. Deploying with Docker (Development)
1.6. Installing for Vagrant under Windows
1.7. Upgrading
1.8. Starting HADatAc
1.9. Stopping HADatAc -
Setting Up
2.1. Software Configuration
2.2. Knowledge Graph Bootstrap
2.2.1. Knowledge Graph
2.2.2. Bootstrap without Labkey
2.2.3. Bootstrap with Labkey
2.3. Config Verification -
Using HADatAc
3.1. Initial Page
3.1.1. Home Button
3.1.2. Sandbox Mode Button
3.2. File Ingestion
3.2.1. Ingesting Study Content
3.2.2. Manual Submission of Files
3.2.3. Automatic Submission of Files
3.2.4. Data File Operations
3.3. Manage Working Files 3.3.1. [Create Empty Semantic File from Template]
3.3.2. SDD Editor
3.3.3. DD Editor
3.4. Manage Metadata
3.4.1. Manage Instrument Infrastructure
3.4.2. Manage Deployments 3.4.3. Manage Studies
3.4.4. [Manage Object Collections]
3.4.5. Manage Streams
3.4.6. Manage Semantic Data Dictionaries
3.4.7. Manage Indicators
3.5. Data Search
3.5.1. Data Faceted Search
3.5.2. Data Spatial Search
3.6. Metadata Browser and Search
3.7. Knowledge Graph Browser
3.8. API
3.9. Data Download -
Software Architecture
4.1. Software Components
4.2. The Human-Aware Science Ontology (HAScO) -
Metadata Files
5.1. Deployment Specification (DPL)
5.2. Study Specification (STD)
5.3. Semantic Study Design (SSD)
5.4. Semantic Data Dictionary (SDD)
5.5. Stream Specification (STR) -
Content Evolution
6.1. Namespace List Update
6.2. Ontology Update
6.3. [DPL Update]
6.4. [SSD Update]
6.5. SDD Update -
Data Governance
7.1. Access Network
7.2. User Status, Categories and Access Permissions
7.3. Data and Metadata Privacy - HADatAc-Supported Projects
- Derived Products and Technologies
- Glossary