At 12:31 AM 1/21/2002 -0500, Michael G Schwern wrote: >On Sun, Jan 20, 2002 at 10:49:42PM -0600, Craig A. Berry wrote: >> Something recently broke our ability to build extensions that run a >> Makefile.PL in a subdirectory, such as SDBM_File does. > >Was this before or after I botched ExtUtils::MakeMaker::check_hints()? >ie. what patchlevel is this against? This was against 14331 but the problem had been there for a week or two, apparently showing up somewhere between 14191 and 14256 by my logs. The repository browser is several months out of date for the "show all patches to affect a file" feature, which contributed to my failure to pinpoint exactly where the breakage occurred. I did see some list traffic about SDBM_File being broken on Win32 recently but as far as I can tell this was unrelated. Paul Green may be right that something he did exposed a latent bug, but I'm not sure that matters since the code I ended up changing was definitely inadequate. The symptom was that $self->{PERL}, which looks like C<MCR [--]miniperl.exe> in the parent, ended up looking like C<[-.MCR]> in the child because the command verb was being treated as a directory spec and getting relativized.Thread Previous | Thread Next