…es there. thanks Zi Lin for the report.
…d the exact return value of the ngx_http_filter_finalize_request() calls." This reverts commit e9fa313.
…act return value of the ngx_http_filter_finalize_request() calls.
…r_filter_by_lua* might lead to Lua stack overflow.
… systems to save a little memory.
…() might get overwritten by nginx's header filter to the fully qualified form (with the scheme and host parts).
…er.HEADER API might get overwritten by nginx's header filter to the fully qualified form (with the scheme and host parts).
…ct might lead to unaligned accesses which could lead to crashes on architectures sensitive to address alignment (like ARMv6). thanks Shuxin Yang for the fix and thanks Stefan Parvu and Brandon B for the report in openresty/ngx_openresty#84.
…nks Zi Lin for the report.
…le path in content_by_lua_file/rewrite_by_lua_file/access_by_lua_file.
…ters properly. thanks Yang Yue for the report.
…"\Z" is expected. thanks laodouya for the original patch in #447.
…ctions for ngx.re.gsub's replacement. these tests exposed a bug in lua-resty-core.
…G macro is set to 1 or a C compiler without variadic macro support (like the Microsoft Visual C++ compiler) is used. thanks itpp16 for the patch in #443.
…t of memory (which should be extremely rare though).
…sten on, which is not true when a local DNS caching server is enabled.
…ght after a 0-width match that happens to be the last match. thanks Guanlan Dai for the patch in #442.
…lding the current ngx_cycle_t pointer.
…monduty's request in #440.