On Mon, Feb 20, 2006 at 04:41:46PM +0100, Gland Vador wrote: >Eric Blake wrote: >>>I have found a strange behaviour in the last (cygwin-1.5.19-4) >>>cygwin1.dll. >>> >>>where $(REQUIRES) are ie: module1 module2 module3 >>> $(_module*_DIR) is a folder on a nework drive, inside this folder >>>I have a optional subfolder named "inc". >> >>Try a snapshot. There are some known issues with inode calculations >>on network drives that have been addressed in snapshots, and which >>might improve your situation. >> >>>This problem appears with the oldest snapshot available at >>>http://cygwin.com/snapshots/. >> >>What about the newest one? > >I have tried almost all snapshots available on this page. I went until >the oldest in order to track on which snapshot it broke. > >The newest one doesn't work for me. When I launch a dos batch file in >the bash, it tries to execute it line by line and not calling the cmd.exe.
You might find reading http://cygwin.com/problems.html instructive. The above email has a few issues which make it difficult to figure out what you're trying to report. I've created a Makefile that looks like this: _module1_DIR=/share/tmp/module1 _module2_DIR=/share/tmp/module2 _module3_DIR=/share/tmp/module3 REQUIRES:=module1 module2 module3 all: @echo $(foreach z,$(strip $(REQUIRES)), \ $(if $(_$(z)_DIR), $(wildcard $(_$(z)_DIR)/inc),)) And created directories /share/tmp/module{1,2,3} and /share/tmp/module1/inc. When I run this makefile, /share/tmp/module1/inc is output with both 1.5.19 DLLs and with a snapshot DLL. (/share/tmp is, of course, a remote drive) So, I've gone the extra mile and tried to come up with a "simple test case" (patent pending) but apparently I didn't duplicate your scenario adequately and, so, we need more details about your environment. cgf -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/