From senatorzergling at gmail.com Fri Aug 14 09:32:05 2009 From: senatorzergling at gmail.com (szergling) Date: Fri, 14 Aug 2009 21:32:05 +1200 Subject: [xcvb-devel] Installation problem Message-ID: <8a4b88390908140232x5bfc95dco86e3fb1eb3428423@mail.gmail.com> Hi all, I've just tried to install xcvb to try it out using the following (apologies for the inadvertent line wrap, if any): tyc20 at binks:~/hda/code/lisp/libs/xcvb-0.349$ make install-using-asdf PREFIX=/home/tyc20/hda/code/lisp/libs/xcvb-0.349-install LISP=/home/tyc20/hda/sbcl/sbcl-1.0.15-x86-linux/src/runtime/sbcl ./dependencies/cl-launch/cl-launch.sh \ -l /home/tyc20/hda/sbcl/sbcl-1.0.15-x86-linux/src/runtime/sbcl \ -I /home/tyc20/hda/code/lisp/libs/xcvb-0.349-install/share/common-lisp/source/cl-launch \ -o /home/tyc20/hda/code/lisp/libs/xcvb-0.349-install/bin/cl-launch \ -B install mkdir -p /home/tyc20/hda/code/lisp/libs/xcvb-0.349-install/share/common-lisp/source/cl-launch create_file 644 /home/tyc20/hda/code/lisp/libs/xcvb-0.349-install/share/common-lisp/source/cl-launch/wrapper.sh print_shell_wrapper create_file 644 /home/tyc20/hda/code/lisp/libs/xcvb-0.349-install/share/common-lisp/source/cl-launch/launcher.lisp print_lisp_launcher create_file 644 /home/tyc20/hda/code/lisp/libs/xcvb-0.349-install/share/common-lisp/source/cl-launch/cl-launch.asd print_cl_launch_asd create_file 644 /home/tyc20/hda/code/lisp/libs/xcvb-0.349-install/share/common-lisp/source/cl-launch/build.xcvb print_build_xcvb create_file 755 /home/tyc20/hda/code/lisp/libs/xcvb-0.349-install/bin/cl-launch print_configured_launcher ./dependencies/cl-launch/cl-launch.sh: line 886: /home/tyc20/hda/code/lisp/libs/xcvb-0.349-install/bin/cl-launch.tmp5418~: No such file or directory make: *** [install-cl-launch] Error 1 In that install folder, this is all there is so far: tyc20 at binks:~/hda/code/lisp/libs/xcvb-0.349$ ls -aR /home/tyc20/hda/code/lisp/libs/xcvb-0.349-install/ /home/tyc20/hda/code/lisp/libs/xcvb-0.349-install/: . .. share /home/tyc20/hda/code/lisp/libs/xcvb-0.349-install/share: . .. common-lisp /home/tyc20/hda/code/lisp/libs/xcvb-0.349-install/share/common-lisp: . .. source /home/tyc20/hda/code/lisp/libs/xcvb-0.349-install/share/common-lisp/source: . .. cl-launch /home/tyc20/hda/code/lisp/libs/xcvb-0.349-install/share/common-lisp/source/cl-launch: . .. build.xcvb cl-launch.asd launcher.lisp wrapper.sh Either I've done something real dumb, or there's a bug somewhere... I get a similar sort of problem if I tried 'make install' instead. All these options and gazillion Unix languages are quite overwhelming to me. Any chance of success if I just do this? (mapc (lambda (x) (load (compile-file (append-path x)))) '("driver" "pkgdcl" "macros" ...)) Could be automated somewhat by direct extraction of the form in xcvb.asd. While I'm here, does xcvb even support plain ol' Lisp style REPL and system construction 'within 1 image'? If you can do distributed multiple images, surely we can also do it all within the same image? 99% of use case today is like that. Cheers everyone. From fahree at gmail.com Fri Aug 14 20:16:59 2009 From: fahree at gmail.com (=?ISO-8859-1?Q?Far=E9?=) Date: Fri, 14 Aug 2009 16:16:59 -0400 Subject: [xcvb-devel] Installation problem In-Reply-To: <8a4b88390908140232x5bfc95dco86e3fb1eb3428423@mail.gmail.com> References: <8a4b88390908140232x5bfc95dco86e3fb1eb3428423@mail.gmail.com> Message-ID: <653bea160908141316w1d6a114dna5c4c603e0e807a6@mail.gmail.com> Dear Senator Zergling, 2009/8/14 szergling : > I've just tried to install xcvb to try it out using the following > (apologies for the inadvertent line wrap, if any): > > tyc20 at binks:~/hda/code/lisp/libs/xcvb-0.349$ make install-using-asdf [...] > ./dependencies/cl-launch/cl-launch.sh: line 886: > /home/tyc20/hda/code/lisp/libs/xcvb-0.349-install/bin/cl-launch.tmp5418~: > No such file or directory > make: *** [install-cl-launch] Error 1 > OK. The release tarball's Makefile wasn't creating the output directory ${INSTALL_BIN} early enough Also, it wasn't correctly finding the .asd's to install. > In that install folder, this is all there is so far: > > tyc20 at binks:~/hda/code/lisp/libs/xcvb-0.349$ ls -aR > /home/tyc20/hda/code/lisp/libs/xcvb-0.349-install/ I assume you chose that install directory. Then don't forget to include /home/tyc20/hda/code/lisp/libs/xcvb-0.349-install/bin in your PATH afterwards so you will be able to call xcvb and cl-launch. > Either I've done something real dumb, or there's a bug somewhere... There was a bug in my packaging code indeed. It should be fixed, as I've managed to install-using-asdf on one of my machines. Note that if you are using either sbcl or clisp, I recommend to install using the bootstrapped xcvb files. > I get a similar sort of problem if I tried 'make install' instead. Yup - I was trying to install cl-launch before having created the INSTALL_BIN directory. > All these options and gazillion Unix languages are quite overwhelming > to me. Any chance of success if I just do this? > > (mapc (lambda (x) (load (compile-file (append-path x)))) > ? ? ?'("driver" "pkgdcl" "macros" ...)) > > Could be automated somewhat by direct extraction of the form in > xcvb.asd. > Nope. XCVB is a binary that has to be created by cl-launch somehow and be in your $PATH. > While I'm here, does xcvb even support plain ol' Lisp style REPL and > system construction 'within 1 image'? If you can do distributed > multiple images, surely we can also do it all within the same image? > 99% of use case today is like that. > It is in our plans to add this feature, but this hasn't been implemented yet. We know how to do it, it isn't even *THAT* hard, but someone has to just do it. [ Fran?ois-Ren? ?VB Rideau | Reflection&Cybernethics | http://fare.tunes.org ] You say you couldn't live if you thought the world had no purpose. You're saying that you can't form purposes of your own -- that you need someone to tell you what to do. The average child has more gumption than that. -- John McCarthy From senatorzergling at gmail.com Mon Aug 24 11:58:53 2009 From: senatorzergling at gmail.com (szergling) Date: Mon, 24 Aug 2009 23:58:53 +1200 Subject: [xcvb-devel] Installation problem In-Reply-To: <653bea160908141316w1d6a114dna5c4c603e0e807a6@mail.gmail.com> References: <8a4b88390908140232x5bfc95dco86e3fb1eb3428423@mail.gmail.com> <653bea160908141316w1d6a114dna5c4c603e0e807a6@mail.gmail.com> Message-ID: <8a4b88390908240458x3b4c5e53xf91297d30dac2ff4@mail.gmail.com> On Sat, Aug 15, 2009 at 8:16 AM, Far? wrote: > Dear Senator Zergling, > > 2009/8/14 szergling : Hi Far?, Sorry for the long delay -- I'm only hacking on this whenever I find time and motivation etc... It looks like I've fallen into the cracks here, so I'm not sure if any changes would be required to xcvb to support my configuration/setup, so take the following steps or hacks as just something to be aware of. So in spite of the long rambling email (as I tend to do), I only have one question about the issue with make-makefile. > I assume you chose that install directory. Then don't forget to include > /home/tyc20/hda/code/lisp/libs/xcvb-0.349-install/bin > in your PATH afterwards so you will be able to call xcvb and cl-launch. > >> Either I've done something real dumb, or there's a bug somewhere... > There was a bug in my packaging code indeed. It should be fixed, as > I've managed to install-using-asdf on one of my machines. Note that if > you are using either sbcl or clisp, I recommend to install using the > bootstrapped xcvb files. Well the bootstrap files worked for clisp, but not for my SBCL 1.0.15. These symbols were not present: sb-kernel:uninteresting-redefinition sb-kernel:undefined-alien-style-warning sb-kernel:lexical-environment-too-complex I encountered another problem with (require :sb-posix) which convinced me to get the latest SBCL. At the same time, I also downloaded the xcvb-0.357 tarball. The new SBCL also had that same issue with (require :sb-posix). Running (require :sb-posix) means we need to look into SBCL's contrib directory. Somehow, I had to set SBCL_HOME to the contrib folder directly, eg $ export SBCL_HOME=/home/tyc20/hda/sbcl/sbcl-1.0.29-x86-linux/contrib/ where sbcl-1.0.29-x86-linux is the contents of the binary tarball downloaded from sbcl.org/Sourceforge. SBCL's INSTALL file talked about SBCL_HOME being the path to sbcl.core too, so this sounds like a conflict/overloaded environment variable. XCVB is quite consistent in the --core option though, so perhaps SBCL_HOME could be used exclusively for the /contrib folder, and not the core file? I think there are valid reasons to having the SBCL binary in non-default locations. In my case, I needed to keep multiple versions around. Someone else may need to keep the binary local to his homedir, for example. -------------------- Anyway, the SBCL bootstrap works after this. After modifying the Makefile (the main one) further, adding/hardcoding the SBCL_HOME option to OPTIONS, if I remember correctly, "xcvb-bootstrapped-install" seems to run. I said "seems" because I kept getting fatal errors, but make just kept going and finished creating an xcvb executable. See attachment. xcvb seems to work, as I can get a 'xcvb repl', and after patching a small bug, also tried an asdf conversion. Another example: tyc20 at binks:~/hda/code/lisp/libs/xcvb-0.357/xcvb/test/mock/a/c$ xcvb show-search-path Discarding invalid path element "/home/tyc20/.local/share/common-lisp/source/" Discarding invalid path element "/usr/share/common-lisp/source/xcvb/" Discarding invalid path element "/usr/local/share/common-lisp/source/" Registered search paths: #P"/home/tyc20/hda/code/lisp/libs/xcvb-0.357/xcvb/test/mock/a/c/" #P"/usr/share/common-lisp/source/" Builds found in the search paths: (:BUILD "/c") in "/home/tyc20/hda/code/lisp/libs/xcvb-0.357/xcvb/test/mock/a/c/build.xcvb" (:BUILD "/c/x") in "/home/tyc20/hda/code/lisp/libs/xcvb-0.357/xcvb/test/mock/a/c/x/build.xcvb" (:BUILD "/c/y") in "/home/tyc20/hda/code/lisp/libs/xcvb-0.357/xcvb/test/mock/a/c/y/build.xcvb" However, at the moment, I still couldn't get make-makefile to run: tyc20 at binks:~/hda/code/lisp/my-first-xcvb-test$ xcvb make-makefile --build . end of file on # 0: unhandled PRINT-NOT-READABLE: # cannot be printed readably. Argh! error within --disable-debugger error handling tyc20 at binks:~/hda/code/lisp/my-first-xcvb-test$ ls build.xcvb module1.lisp module2.fas module2.lisp build.xcvb~ module1.lisp~ module2.lib obj tyc20 at binks:~/hda/code/lisp/my-first-xcvb-test$ cat obj/target-properties.lisp-expr tyc20 at binks:~/hda/code/lisp/my-first-xcvb-test$ ls obj/ target-properties.lisp-expr Ignore the .fas .lib stuff. I think they were leftover from a previous attempt (that's clisp, not SBCL). Is target-properties.lisp-expr supposed to ever be empty? What is read-first-file-form supposed to return for empty files (instead of read with errorp t), nil? In both places that use it, there are some checks, eg read-target-properties: (let* ((file (target-properties-file)) (form (read-first-file-form file))) (unless (and (consp form) (eq 'setf (car form))) (error "Malformed target properties file ~S" file)) ...) so (read in nil nil) could work. Or have I not got the right things in the target file? >> (mapc (lambda (x) (load (compile-file (append-path x)))) >> '("driver" "pkgdcl" "macros" ...)) >> >> Could be automated somewhat by direct extraction of the form in >> xcvb.asd. >> > Nope. XCVB is a binary that has to be created by cl-launch somehow and > be in your $PATH. So do that mapc thing (and asdf load dependencies), then save-image and resave image with cl-launch? Although if xcvb can do some of the building instead of using make, that could be easier too. >> While I'm here, does xcvb even support plain ol' Lisp style REPL and >> system construction 'within 1 image'? If you can do distributed >> multiple images, surely we can also do it all within the same image? >> 99% of use case today is like that. >> > It is in our plans to add this feature, but this hasn't been implemented yet. > We know how to do it, it isn't even *THAT* hard, but someone has to just do it. Next time, I'm gonna try and run MAKE-MAKEFILE, REMOVE-XCVB-COMMAND (in main.lisp) etc on the REPL. Is that how you are all using it at the moment? Thanks for the help, I will continue to explore when I find time. Yong. -------------- next part -------------- tyc20 at binks:~/hda/code/lisp/libs/xcvb-0.357$ make install mkdir -p /home/tyc20/hda/code/lisp/libs/xcvb-0.357-install/share/common-lisp/source/cl-launch /home/tyc20/hda/code/lisp/libs/xcvb-0.357-install/bin ./dependencies/cl-launch/cl-launch.sh \ -l /home/tyc20/hda/sbcl/sbcl-1.0.29-x86-linux/src/runtime/sbcl \ -I /home/tyc20/hda/code/lisp/libs/xcvb-0.357-install/share/common-lisp/source/cl-launch \ -o /home/tyc20/hda/code/lisp/libs/xcvb-0.357-install/bin/cl-launch \ -B install mkdir -p /home/tyc20/hda/code/lisp/libs/xcvb-0.357-install/share/common-lisp/source/cl-launch create_file 644 /home/tyc20/hda/code/lisp/libs/xcvb-0.357-install/share/common-lisp/source/cl-launch/wrapper.sh print_shell_wrapper create_file 644 /home/tyc20/hda/code/lisp/libs/xcvb-0.357-install/share/common-lisp/source/cl-launch/launcher.lisp print_lisp_launcher create_file 644 /home/tyc20/hda/code/lisp/libs/xcvb-0.357-install/share/common-lisp/source/cl-launch/cl-launch.asd print_cl_launch_asd create_file 644 /home/tyc20/hda/code/lisp/libs/xcvb-0.357-install/share/common-lisp/source/cl-launch/build.xcvb print_build_xcvb create_file 755 /home/tyc20/hda/code/lisp/libs/xcvb-0.357-install/bin/cl-launch print_configured_launcher mkdir -p /home/tyc20/hda/code/lisp/libs/xcvb-0.357-install/bin # make -f xcvb.mk./home/tyc20/hda/sbcl/sbcl-1.0.29-x86-linux/src/runtime/sbcl obj/xcvb.image ln -s ../obj xcvb/ make -C xcvb xcvb-bootstrapped-install PREFIX=/home/tyc20/hda/code/lisp/libs/xcvb-0.357-install LISP=/home/tyc20/hda/sbcl/sbcl-1.0.29-x86-linux/src/runtime/sbcl INSTALL_BIN=/home/tyc20/hda/code/lisp/libs/xcvb-0.357-install/bin INSTALL_LISP=/home/tyc20/hda/code/lisp/libs/xcvb-0.357-install/share/common-lisp LISP_SOURCE=/home/tyc20/hda/code/lisp/libs/xcvb-0.357-install/share/common-lisp/source LISP_SYSTEMS=/home/tyc20/hda/code/lisp/libs/xcvb-0.357-install/share/common-lisp/systems SBCL_HOME=/home/tyc20/hda/sbcl/sbcl-1.0.29-x86-linux/contrib/ /home/tyc20/hda/code/lisp/libs/xcvb-0.357-install/bin/cl-launch: line 2407: implementation_/home/tyc20/hda/sbcl/sbcl-1.0.29-x86-linux/src/runtime/sbcl: No such file or directory /home/tyc20/hda/code/lisp/libs/xcvb-0.357-install/bin/cl-launch: line 2399: implementation_/home/tyc20/hda/sbcl/sbcl-1.0.29-x86-linux/src/runtime/sbcl: No such file or directory fatal error encountered in SBCL pid 7412(tid 1075367040): can't find core file make[1]: Entering directory `/home/tyc20/hda/code/lisp/libs/xcvb-0.357/xcvb' mkdir -p /home/tyc20/hda/code/lisp/libs/xcvb-0.357-install/bin cl-launch --lisp '/home/tyc20/hda/sbcl/sbcl-1.0.29-x86-linux/src/runtime/sbcl sbcl clisp ccl' --path /home/tyc20/hda/code/lisp/libs/xcvb-0.357-install/share/common-lisp/systems --no-include --image $PWD/obj/xcvb.image \ --output /home/tyc20/hda/code/lisp/libs/xcvb-0.357-install/bin/xcvb --dump ! --init '(xcvb::main)' /home/tyc20/hda/code/lisp/libs/xcvb-0.357-install/bin/cl-launch: line 1548: implementation_/home/tyc20/hda/sbcl/sbcl-1.0.29-x86-linux/src/runtime/sbcl: No such file or directory /home/tyc20/hda/code/lisp/libs/xcvb-0.357-install/bin/cl-launch: line 1548: implementation_/home/tyc20/hda/sbcl/sbcl-1.0.29-x86-linux/src/runtime/sbcl: No such file or directory [undoing binding stack and other enclosing state... done] [saving current Lisp image into /home/tyc20/hda/code/lisp/libs/xcvb-0.357-install/bin/xcvb: writing 2752 bytes from the read-only space at 0x01000000 writing 1712 bytes from the static space at 0x01100000 writing 28524544 bytes from the dynamic space at 0x09000000 done] make[1]: Leaving directory `/home/tyc20/hda/code/lisp/libs/xcvb-0.357/xcvb' -------------- next part -------------- A non-text attachment was scrubbed... Name: 0001-Minor-fix-to-verbosity-arg.patch Type: text/x-patch Size: 706 bytes Desc: not available URL: From fahree at gmail.com Mon Aug 24 12:50:59 2009 From: fahree at gmail.com (=?ISO-8859-1?Q?Far=E9?=) Date: Mon, 24 Aug 2009 08:50:59 -0400 Subject: [xcvb-devel] Installation problem In-Reply-To: <8a4b88390908240458x3b4c5e53xf91297d30dac2ff4@mail.gmail.com> References: <8a4b88390908140232x5bfc95dco86e3fb1eb3428423@mail.gmail.com> <653bea160908141316w1d6a114dna5c4c603e0e807a6@mail.gmail.com> <8a4b88390908240458x3b4c5e53xf91297d30dac2ff4@mail.gmail.com> Message-ID: <653bea160908240550t63a307dbvab942ad623b30ebb@mail.gmail.com> Honorable Senator Zergling, thanks a lot for taking the time and trouble to beta-test our software. After your previous email, I've tried my best to address your legitimate concerns about being able to use an SBCL at a non-standard location, and the latest release of XCVB (0.360) should include provisions to let you use whichever SBCL is properly installed to a directory in your $PATH. So make sure that "sbcl" works at your shell prompt (don't try to run it uninstalled from its source directory -- that's painful). If the autodetection fails, please report a bug again. I recommend using SBCL 1.0.30.4 or later (1.0.31.0 is due RSN), which will have the advantage of providing your with CFASL support in addition to being something with which we have tested our software. .360 also includes the same verbosity patch as you wrote (as I found the bug and its fix independently). It was tested on a multi-system build (exscribe and all dependencies but disabling cl-typesetting), but not yet on QRes. So hopefully it will work for you. > However, at the moment, I still couldn't get make-makefile to run: > > tyc20 at binks:~/hda/code/lisp/my-first-xcvb-test$ xcvb make-makefile --build . > end of file on # /home/tyc20/hda/code/lisp/my-first-xcvb-test/obj/target-properties.lisp-expr" > {AB591F1}> > 0: > unhandled PRINT-NOT-READABLE: # cannot be > printed readably. > > Argh! error within --disable-debugger error handling Ouch. Which target lisp implementation are you using? If that still happens with the latest SBCL, can you trace how xcvb invokes the target lisp to extract its properties, and what makes the target lisp choke? > Ignore the .fas .lib stuff. I think they were leftover from a previous > attempt (that's clisp, not SBCL). Is target-properties.lisp-expr > supposed to ever be empty? What is read-first-file-form supposed to > return for empty files (instead of read with errorp t), nil? ?In both > places that use it, there are some checks, eg read-target-properties: > It is indeed a bug that target-properties.lisp-expr should ever be empty, and a bug that read-first-file-form should not handle a file without a form (I will look for a fix RSN). If the problem persists when you remove this file, then there's an important bug in xcvb. > (let* ((file (target-properties-file)) > ? ? ? (form (read-first-file-form file))) > ?(unless (and (consp form) (eq 'setf (car form))) > ? ?(error "Malformed target properties file ~S" file)) > ?...) > > so (read in nil nil) could work. Or have I not got the right things in > the target file? > Yes. .361 will have (read in nil nil). Thanks for this fix. >>> (mapc (lambda (x) (load (compile-file (append-path x)))) >>> ? ? ?'("driver" "pkgdcl" "macros" ...)) >>> >>> Could be automated somewhat by direct extraction of the form in >>> xcvb.asd. >>> >> Nope. XCVB is a binary that has to be created by cl-launch somehow and >> be in your $PATH. > > So do that mapc thing (and asdf load dependencies), then save-image > and resave image with cl-launch? Although if xcvb can do some of the > building instead of using make, that could be easier too. > Yes, you could do that. But indeed the Makefile should do it for you already. If it doesn't that's a bug. Note that you may bootstrap with: make xcvb-using-asdf > Next time, I'm gonna try and run MAKE-MAKEFILE, REMOVE-XCVB-COMMAND > (in main.lisp) etc on the REPL. Is that how you are all using it at > the moment? At the moment, I am using it at the shell prompt, though sometimes indeed I use the repl for debugging -- particularly with C-u M-x slime env LISP_FASL_CACHE=NIL XCVB_PATH=... xcvb repl > Thanks for the help, I will continue to explore when I find time. Thank you so much for your testing. My apologies for the trouble you're having. I'm definitely committed to make XCVB a usable tool for everyone and not just something that merely works for me in my particular setting. [ Fran?ois-Ren? ?VB Rideau | Reflection&Cybernethics | http://fare.tunes.org ] The trouble with opportunity is that it always comes disguised as hard work. -- Herbert V. Prochnow