README.md in reports_kit-0.2.0 vs README.md in reports_kit-0.3.0
- old
+ new
@@ -1,7 +1,9 @@
ReportsKit
=====
+[](https://travis-ci.org/tombenner/reports_kit)
+
ReportsKit lets you easily create beautiful charts with customizable, interactive filters.
For interactive examples, see [reportskit.co](https://www.reportskit.co/).
---
@@ -61,40 +63,32 @@
```
Quick Start
-----------
-After installation, you can create your first chart with a single line!
+After installation, you can create your first chart with just a YAML file and a single line in any view.
-In any view, create a chart that shows the number of records of a model (e.g. `user`) created over time:
+Configure the chart in the YAML file:
-`app/views/users/index.html.haml`
-```haml
-= render_report measure: { key: 'user', dimensions: ['created_at'] }
-```
-
-You're done! `render_report` will render the following chart:
-
-[<img src="docs/images/users_by_created_at.png?raw=true" width="500" />](docs/images/users_by_created_at.png?raw=true)
-
-Instead of passing a hash to `render_report`, you can alternatively configure your charts using YAML and then pass the filename to `render_report`:
-
`config/reports_kit/reports/my_users.yml`
```yaml
-measure:
- key: user
- dimensions:
- - created_at
+measure: user
+dimensions:
+- created_at
```
+Then pass that filename to `render_report` in a view:
+
`app/views/users/index.html.haml`
```haml
= render_report 'my_users'
```
-The YAML approach is more maintainable and readable, so we'll use it in the rest of the documentation.
+You're done! `render_report` will render the following chart:
+[<img src="docs/images/users_by_created_at.png?raw=true" width="500" />](docs/images/users_by_created_at.png?raw=true)
+
### Form Controls
You can add a date range form control to the above chart with a single line, using one of ReportsKit's form helpers:
`app/views/users/index.html.haml`
@@ -107,10 +101,14 @@
Many other form controls are available; see [Filters](docs/filters.md) for more.
### How It Works
-In the Quick Start chart, `key: 'user'` tells ReportsKit to count the number of `User` records, and `dimensions: ['created_at']` tells it to group by the week of the `created_at` column. Since `created_at` is a `datetime` column, ReportsKit knows that it should sort the results chronologically.
+In the Quick Start chart, `measure: user` tells ReportsKit to count the number of `User` records, and `dimensions: ['created_at']` tells it to group by the week of the `created_at` column. Since `created_at` is a `datetime` column, ReportsKit knows that it should group the counts by week (the granularity is configurable), sort them chronologically, and add in zeros for any missing weeks.
+
+ReportsKit infers sane defaults from your ActiveRecord model configurations. If there was a `belongs_to :company` association on `User` and you used `dimensions: ['company']`, then ReportsKit would count users grouped by the `company_id` column and show company names on the x-axis.
+
+If you need more customization (e.g. custom filters, custom dimensions, custom aggregation functions, custom orders, aggregations of aggregations, etc), ReportsKit is very flexible and powerful and supports all of these with a simple syntax. It lets you use SQL, too.
To learn how to use more of ReportsKit's features, check out the following resources:
* [Examples](https://www.reportskit.co/)
* [Documentation](docs)