timeline_charter 0.0.1 
timeline_charter: ^0.0.1 copied to clipboard
Simple way to chart time based values based on SQL Queries.
A simple application/framework based on dart to create time based charts based on simple SQL queries. It currently has a very limited featureset, but great potential ;)
Example #
See the example project for a sample usage:
- Web: http://hpoul.github.io/timeline_charter_example/
- Code: https://github.com/hpoul/timeline_charter_example/
This is how it (should) look:
Status #
This is still a very experimental state, and it is quite hard to actually use it. In addition it currently only supports one chart format:
- 1 Year duration, 7 day intervals.
Seems quite limited, and it actually is. But the config is generic enough to improve the implementation and the charts.
Why? #
I wanted a simple way to analyze usage statistics from my database. But configuring zabbix & co is a bit annoying, slow and you get no historical data. If you know a better tool for the job, which you can configure by simply a one-liner SQL statement, let me know!
Implementation #
There are two parts:
- Analyzer: Dart application which runs the analysis and simply stores key/value pairs to a given timestamp. This has to be run from a cronjob or similar.
- Web Component: Dart polymer component (which interally uses https://github.com/shutterstock/rickshaw to display the charts)
Usage #
Create a new dart project and add polymer and timeline_charter as dependency #
Create a bin/runanalyzer.dart file to configure and run the analyzer #
Example configuration objects:
Imagine you want statistics of the user activity for your app and have a database table called 'actionlog' with a timestamp column. This is how a configuration would look like:
var analyzerConfigs = [
new AnalyzerConfig(
sql: 'SELECT COUNT(*) FROM actionlog WHERE time BETWEEN @startTime AND @endTime',
keys: const ['myapp.actionlog.count'],
labels: const ['ActionLog Count']
)
]
now we just need to use that config and run our analyzer:
import 'package:timeline_charter/analyzer.dart';
import 'package:timeline_charter/pgqueryexecutor.dart';
import 'package:timeline_charter/worktrail_config.dart';
import 'dart:io';
import 'dart:async';
int main() {
print("hello.");
var sqlurl = Platform.environment['WT_PGSQL'];
if (sqlurl == null) {
sqlurl = 'postgres://herbert:@localhost:5432/worktrail';
}
// open connection to postgresql database and open file cache ...
var pgFuture = openPgQueryExecutor(sqlurl);
var cacheFuture = SimpleFileCache.createSimpleFileCache(new Directory('cache'));
Future.wait([pgFuture, cacheFuture]).then((valueList) {
PgQueryExecutor executor = valueList[0];
SimpleFileCache cache = valueList[1];
// now that everything is set up, we can launch the analyzer.
new Analyzer()
.analyzeConfigs(executor, analyzerConfigs, cache)
.whenComplete(() {
executor.close();
Future.wait([cache.close()]);
});
});
return 0;
}
Create web/index.html to display the result #
<!DOCTYPE html>
<html>
<head>
<title>Analytics</title>
<link rel="import" href="packages/timeline_charter/tapo_graph.html" />
</head>
<body>
Action Log Count:<br />
<tapo-graph prefix="dummy" loadjson="latestdata.json"></tapo-graph>
<br /><br />
<script type='application/dart'>export 'package:polymer/init.dart';
</script>
<script src="packages/browser/dart.js"></script>
</body>
</html>
Build app with pub #
pub build
Run analyzer #
dart bin/runanalyzer.dart
Copy latestdata.json into build/ directory #
cp latestdata.json build/
Open build/index.html #
(If you are hosting on file://, make sure to open it in Firefox, because chrome won't allow http requests to file:// protocol.)
Metadata
Publisher
unverified uploader
Weekly Downloads
Metadata
Simple way to chart time based values based on SQL Queries.
Repository (GitHub)
View/report issues
License
BSD-2-Clause (license)
Dependencies
intl, logging, polymer, postgresql, route
