Oracle Visual Builder: Applications That Outlive Their First Release · Module 17 · Quarterly updates and VBCS
A regression pack for an application
Lesson 167 of 176 · 2 min
WHICH SCREENS to re-run, with WHAT DATA, and KEEPING THE PACK SMALL ENOUGH THAT SOMEBODY ACTUALLY RUNS IT. The constraint in the title is the design goal. A comprehensive regression pack that takes a day is a pack that gets skipped, and a skipped pack is worth nothing. SO SELECT BY CONSEQUENCE RATHER THAN BY COVERAGE. Which screens, if they broke, would matter most? Anything that writes data, anything money touches, anything a lot of people use daily — and the answer is usually four or five screens rather than thirty. THEN ADD THE SCREENS THAT ARE MOST EXPOSED, WHICH IS A DIFFERENT LIST. Anything with an unusual component, anything reading a Fusion field that Lesson 2's scan flagged, anything with a custom component — Module 9 Lesson 4. THE DATA MATTERS AS MUCH AS THE SCREENS, AND IT IS THE PART PEOPLE GET WRONG. A regression run on a…
The full lesson is part of the course
The video, the complete written lesson and the module quiz are included in Oracle Visual Builder: Applications That Outlive Their First Release, with a certificate on completion and a fourteen-day refund window.
