I wish my demand be echoed from the Designers.

Jean

Bug 6179 ln(0) & var=0.sm (10,24 KiB) scaricato 1252 volta(e).

The problem here is not ln(0), neither the numerical engine. Logarithm works in the same way in both the SS versions.

The different behavior is due to a difference in the symbolic engine; in SS 0.97 0*anything=0 while in SS 0.98 0*anything depends on what is anything (in this case, undefined). Since hatch(...) is designed for continuous functions in the [u;v] domain (both the bounds included), your fix seems to me the most reasonable and consistent way to proceed.