Skip to content

An interactive visualization of temporal data from a software development process

License

Notifications You must be signed in to change notification settings

grip-on-software/timeline

Repository files navigation

Timeline

This visualization produces a timeline of temporal data from a software development process. The data may include different types of events and regularly reoccurring cycles, such as Scrum sprints.

Installation and running

You need to have installed node and npm to build the project, see Node.js for details. Within the directory of this project, use npm install to retrieve all dependencies. Then, either execute npm run watch to start a development server that also refreshes browsers upon code changes, or npm run production to create a minimized bundle. The resulting HTML, CSS and JavaScript is made available in the public directory.

In order to build the timeline web-based interface, you need a directory public/data containing a main file public/data/data.json, as well as other JSON files named after the event they describe and possibly split out into different project directories and Scrum sprints. See the data format for more details.

Another option for installation and deployment is to use a Docker container image. Use docker build -t gros-timeline . to create the image with all dependencies. If the directory already contains a prepopulated public/data directory, then you can directly start a web server in the Docker container using docker run -P --rm gros-timeline. This opens a port on the localhost where you can connect to in order to find the timeline.

You may want to use Docker but still break up the installation, building (pre-publishing) and publishing steps. This can be done for example on a Jenkins server. In this case, at the end of the installation job which runs the docker build command, push the container image to a (private) repository. In the build job, provide the data in the workspace public/data directory, pull the image into the build job, and then run the container as follows:

docker pull gros-timeline
docker run --rm -u `id -u`:`id -g` -v `pwd`:/work -w /work gros-timeline \
  /bin/bash -ce 'rm -rf /work/node_modules; \
  ln -s /usr/src/app/node_modules/ /work; npm run production'

The Jenkinsfile contains further example steps for performing the data collection upon a Grip on Software database using the data-analysis repository (specifically, the timeline.r script and sprint_burndown analysis report), as well as the visualization building.

Configuration

In order to configure the visualization, including options for setting the default language and the link to the sprint details, copy lib/config.json to config.json and alter its contents there. The following configuration items are known:

  • language: The language code of the default language of the timeline.
  • jira_url: The URL pointing to a Jira instance in order to link to sprints. This is only used as a fallback in case the Jira instance is not defined in the source data. If this is also empty, then no link to the Jira instance is generated in the subchart title.
  • visualization_url: The URL to the visualization hub. This may include a protocol and domain name, but does not need to in case all the visualizations and the timeline are hosted on the same domain (for example in a development environment). The remainder is a path to the root of the visualizations, where the dashboard is found and every other visualization has sub-paths below it.
  • path: The relative path at which the timeline is made available on the server. This can remain the default . to work just fine.

Features

  • Interactive zoom and scroll to restrict the timeline view to certain ranges or expand the view
  • Different time scales: normal, weekday (mapping weekends to workdays)
  • Responsive display, scales when resizing
  • Dynamic type filter to simplify the timeline for only showing certain events. Initial type filter is configurable in data/types.json.
  • Subchart showing additional events during a specified range of a development cycle, with chunked loading of data
  • Fully translatable interface

Format

In order to build the web interface, certain files are required during the build step. This section describes the format of these files. The visualization-site repository contains JSON schema files with specific type descriptions of all the nested fields.

  • public/data/data.json: Contents are a JSON object with summary data. It contains at least the following keys and values:
    • min_date: The earliest date from all other data sources
    • max_date: The latest date from all other data sources, may be later than the current date
    • update_date: The date when the data was generated, used for cache invalidation
    • projects: A list of project names that are used in the other data sources
  • public/data/[type].json: Contents are a JSON object with project names as keys and a list of event data as values. Each event object has the following keys and values:
    • project_name: Same as the key of the overall object
    • sprint_id: Unique identifier of a range defined by a sprint_start event
    • sprint_name: Name of the range defined by a sprint_start event
    • date: The date and time that the event took place, in YYYY-MM-DDTHH:MM:SS format 1
    • type: Same as the type specified in the file name
    • end_date (Optional): The date and time that the event ends. For sprint_start events, this is the end of the range defined by it.
    • board_id (Optional): The Jira board ID that the sprint comes from.
  • public/data/[project_name]/[type].[sprint_id].json: Contents is a JSON list with entries of the same format as the event objects used for other event types. The events belong to a specific project name and occur during or around a range defined by the sprint_start event with the sprint_id identifier. Currently, this format is only used by the commits type.
  • public/data/features.json: Contents is an object where keys are project names, and values are objects with sprint identifiers as keys. For each sprint, there is an object which has key-value pairs describing attributes of that time range.
  • public/data/boards.json: JSON object of project names as keys and Jira board IDs as values. A board ID is either an integer or null to indicate that no board is known. The board ID may then also come from feature data of a sprint.
  • public/data/links.json: JSON object of project names as keys and objects with source links as values. The nested objects should have keys that end in _url, but in particular the jira_url key is used to obtain the Jira instance URL of the project.
  • public/data/types.json: Registry of available types and their initial filtering settings. Contents is a JSON list with object entries, each containing at least a name whose value is the type used in the various data files. For non-subchart types, enabled determines the initial filter state. Subchart types are configured as such with a subchart key. Each type may have a locales key with an object containing language codes as keys and names as values, in order to localize the type name.
  • public/data/locales.json: Contents is an object containing language codes as keys, whose values are objects of feature keys and localized names.
  • public/data/sprint_burndown/[project_name]/sprint_burndown.[sprint_id].json: Contents is a JSON list with object entires, each containing date, points and type keys with changes to the burndown chart.

License

Timeline is released under the MIT License. Some parts are based on EventDrops (fork), also released under the MIT License.