This application has been designed to compare the speed between the different css selector-engines existing in gwtquery
The html document and selectors used for the benchmark have been taken from the SlickSpeed
Click 'start the race' to test the selectors against most popular frameworks.
If you wanted to experiment with other frameworks or different gquery engine implementations, append the parameter ask=true to the url and select the options in the popup window after clicking 'start'.
Because normally css selectors takes less than 1 or 2 millisecons, each selector is run during a minimum of 200ms for each engine, then we divide the total time between the number of iterations and put this value in the cell,
marking in green the faster engine.
If an engine does not support a selector we put the word 'Error' in the corresponding cell and penalize the engine setting a value of 200ms for that iteration.
The last row represents the acumulated time of the test for each engine.
Some times, you could obtain different results when running the test with the same browser.
This is due to random browser or system events which makes the current selector running delay some milliseconds.
To avoid that, try to run the test in a browser running just one window and one tab, and in a computer without activity: close all applications, do not move the mouse etc.
- gwt_compiled: run css selectors which were optimized during the gwt compilation.
- gwt_dynamic: evaluates selectors in runtime using the most suitable engine for this browser.
- jquery, dojo, prototype ... : runs the selectors using those external libraries.
- gwt_*: different runtime engines, some are experimental, and they could fail depending on the browser.
Source code of this application
is available in the samples module of gwtquery.
- The GwtQuery team.