https://bz.apache.org/bugzilla/show_bug.cgi?id=66243
--- Comment #6 from Tj <[email protected]> --- Going further; attempting to use the presence of my new env-var SSI_INCLUDE_LEVEL >= 1 to trigger the nested path in includes_filter() causes a SIGSEGV because r->main is always NULL. I think this code must have been broken for some time; as a newcomer I'm finding it difficult to figure out where to look in both code and commit history to determine if this once did work (was tested and proven) and some subsequent change broke it (or if the choice of MPM affects it). Regardless of that, it looks like there needs to be a new mechanism to replace this self-admitted kludge. Adding/removing an env-var may indeed be the way. Right now the kludge is only there for finfo.mtime correction but my additions to track and report include level and counts also needs it. Recommendations on best path forward? My instinct is to totally remove the kludge and use my new variables to detect the nesting, plus possibly an temporary env-var that is set/removed internally to carry finfo.mtime, but that might be seen by some as inelegant! -- You are receiving this mail because: You are the assignee for the bug. --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
