[pcre-dev] Multi segment matching using pcre_dfa_exec

*

Hi, Philip!



We were wondering what do you think about adding the following pcre behviour
when multi segment matching is used with pcre_dfa_exec():



At the moment, patterns that contain alternatives at the top level are
somehow problematic as only alternatives that match at lowest point in the
subject are remembered. As a result matches might be missed for alternatives
which share the same characters.



We thought it might be useful to be able to define the number “saved” points
or at least return an indication that matches might be missed. (and then let
the caller of pcre_dfa_exec) decides if linearization should be used to find
possible lost matches.



Many thanks,

E.

*