covingto's blog

So I've been working no getting the documentation generator up and running.

So far the plan is to use sphinx to do the document generation on .rst files that we'll generate using the document parser.

Developers won't have to worry about any of this, since the .py files for each widget, signal class, and qtWidget will be parsed for documentation using our own document parsing functions.

Recently we undertook to standardize the redRCore Qt widget API. This is the API that is in the base package.

No elements of the API were removed, standard elements were pushed to be handled by the core widgetState container, which is the universal container for all redR core metawidgets.

All widgets accept the following as key word arguments;

widget - parent widget into which the widget controlArea will be placed.
label - most widgets accept and will display a label, possibly required, and possibly controlled using displayLabel

Following the developers meeting today I began to generate a set of functions for regression testing. This should increase the stability of Red-R on different OS configurations because we will be able to deploy and test all packages on each configuration, isolate problems, and report them to developers.

We are actively working on improving the documentation of widgets in Red-R. To make this easier for developers and users alike, we have begun to use the to make both developer and user documentation within the widget.py files. Doing this should greatly improve the documentation as even the default documentation may be sufficient to help the user, and entering documentation for the user by the developer should be much easier.

Today we tackled several issues in Core.

First changes were made to facilitate loading of older pipelines into the new core using the refined widget API. We are still making changes but in general hope that we will be able to allow users to have some fail-safes that will keep older schemas working. These fail-safes will likely be in place for some time, but we will encourage users to at least open all old schemas and resave to the new settings. Doing this will greatly help us to keep from being overwhelmed by legacy code.

First day with the new website and system. This is much different from the wordpress system that we used in the past. I just started looking around the new site but it seems to have lots of fantastic features to use.

Seems like we have a lot on our plates here at Red-R. Lots of packages and modules to move in from R.

I'm actually giving a talk today on Red-R and how it may be able to help scientists to better explore their data.