Re: [PATCHES] Re: (call-process ...) to return correct exit-status on NT

At 07:02 PM 10/17/99 +0200, Adrian Aichner wrote: I used the one for 21.2-b19. No deal. The patch applied cleanly to a newly updated source tree, but during the build, temacs died trying to make the DOC file. The relevent output is: rm -f ../lib-src/DOC; \ EMACSBOOTSTRAPLOADPATH="/XEmacs/xemacs/src/../lisp/;/XEmacs/xemacs" EMACSBOOTSTR APMODULEPATH="/XEmacs/xemacs/src/../modules/;/XEmacs/xemacs" ./temacs -batch -l /XEmacs/xemacs/src/../lisp/make-docfile.el -- \ -o ../lib-src/DOC -d /XEmacs/xemacs/src -i ../lib-src/../site-packages \ abbrev.c alloc.c blocktype.c buffer.c bytecode.c callint.c callproc.c casefiddle.c casetab.c chartab.c cmdloop.c cmds.c console.c console-stream.c data.c device.c dired.c doc.c doprnt.c dynarr.c editfns.c elhash.c emacs.c eval.c events.c debug.c ntplay.c nt.c ntheap.c ntproc.c dired-msw.c unexcw.c scrollbar-msw.c menubar-msw.c toolbar-msw.c dialog-msw.c console-msw.c device-msw.c event-msw.c frame-msw.c objects-msw.c select-msw.c redisplay-msw.c glyphs-msw.c gui-msw.c dragdrop.c dgif_lib.c gif_io.c menubar.c scrollbar.c dialog.c toolbar.c file-coding.c realpath.c getloadavg.c inline.c process-nt.c event-stream.c extents.c faces.c fileio.c filemode.c floatfns.c fns.c font-lock.c frame.c general.c glyphs.c glyphs-eimage.c glyphs-widget.c gui.c gutter.c hash.c imgproc.c indent.c insdel.c intl.c keymap.c line-number.c lread.c lstream.c macros.c marker.c md5.c minibuf.c objects.c opaque.c print.c process.c profile.c rangetab.c redisplay.c redisplay-output.c regex.c search.c select.c sheap.c signal.c sound.c specifier.c strftime.c symbols.c syntax.c sysdep.c undo.c widget.c window.c free-hook.c Loading /XEmacs/xemacs/src/../lisp/make-docfile.el... Loading very-early-lisp... Loading find-paths.el... Loading packages.el... Loading setup-paths.el... Loading dump-paths.el... Loading c:\XEmacs\xemacs\lisp\dumped-lisp.el... Spawning make-docfile ...Fatal error: assertion failed, file sysdep.c, line 353, pid != NULL After I said OK to the "temacs died" dialog, it continued on with the build process and finally produced an xemacs.exe file. I installed it and put the packages and extra EXE's in place. I can use the Tools -> Grep to search files in a directory, but if I try to update the packages it still complains about an exec format error. I set the debug on error flag an got the following lisp backtrace: Signaling: (file-error "Spawning child process" "Exec format error") call-process-internal("minitar" nil #<buffer "*Package Output*"> t "c:\\WIN98\\TEMP\\scheme-1.09-pkg.tar.gz") apply(call-process-internal "minitar" nil #<buffer "*Package Output*"> t "c:\\WIN98\\TEMP\\scheme-1.09-pkg.tar.gz") call-process("minitar" nil #<buffer "*Package Output*"> t "c:\\WIN98\\TEMP\\scheme-1.09-pkg.tar.gz") "c:\\WIN98\\TEMP\\scheme-1.09-pkg.tar.gz") package-admin-install-function-mswindows("c:\\WIN98\\TEMP\\scheme-1.09-pkg.t ar.gz" "c:\\PROGRAM FILES\\XEMACS\\21-2-B19\\lib\\xemacs\\xemacs-packages\\" #<buffer "*Package Output*">) Output*">) package-admin-add-binary-package("c:\\WIN98\\TEMP\\scheme-1.09-pkg.tar.gz" "c:\\PROGRAM FILES\\XEMACS\\21-2-B19\\lib\\xemacs\\xemacs-packages\\") byte-code("..." [package-get-info-version package-get-info-find-package package-get-base package version this-package package-get-info-prop latest package-get-key :version installed requires this-requires nil found package-get-remote search-dirs filename base-filename t package-status filenames full-package-filename error "Couldn't find package %s with version %s" "No download sites or local package locations specified." "No filename associated with package %s, version %s" package-admin-get-install-dir install-dir mule-base string-to-number warn "Installing %s package version %s, you had a newer version %s" "Skipping %s package, you have a newer version %s" throw skip-update string-match "^\\(..*\\).tar.gz$" append match-string 1 ".tgz" conflict never package-get-installedp done (byte-code "À‰‰‰‰‰ƒ— I was trying to install the scheme package at the time. If you're using OpenProcess() with PROCESS_ALL_ACCESS, I would think that you should be able to do TerminateProcess() on the process. Does that not work? Craig