On 08 Nov 2007 14:48:12 +0100, Martin Rubey wrote: > "Bill Page" writes: > > Have you been able to compile the Axiom/Aldor interface for a recent > > version of FriCAS or OpenAxiom? As you know the instructions here: > > No, so far I have only a build with the very last wh-sandbox. I will try it as > soon as I can. But really, it's Peter who is the boss of the axiom-aldor > interaction. >
Indeed! :-) > > http://axiom-wiki.newsynthesis.org/AldorForAxiom > > > > are rather out of date and the referenced emails from Franz and Gaby > > have only brief notes on a recipe that apparently worked on an earlier > > version. I have been trying unsuccessfully to complete a build with > > FriCAS and the new version of Aldor but I am running into many > > problems. I would really like to provide the Aldor interface at the new > > web site with a more recent version/fork of Axiom. > > Yes. In fact, I wish that the aldor interface would be automatically built > when axiom is built. > You mean the FriCAs/OpenAxiom build should check for an existing installation of Aldor and just do the right thing to install the interface, right? Of course licensing issues preclude actually distributing Aldor with Axiom (unless we were to adopt APL2 for some variant of Axiom), but this does not prevent doing what you suggest. > Maybe you could just state what's the first problem you run into? > The first problem is that the scripts in src_aldor2.tgz assume the existence of intermediate directories named ./obj/xxx/bin and ./int/algebra/... On FriCAS (and OpenAxiom?) this has been replaced with directory names ./build/xxx/bin and ./src/algebra/... respectively (xxx is the host identifier. The email notes suggest building these missing directories and adding appropriate symbolic links. Both FriCAS and OpenAxiom builds are designed to work "out-of-source" but the src_aldor2 build does not work this way (yet). Of course the right way to do this would be to modify the scripts. My first attempt started this way. I got this to "almost" work by a series of incremental changes but I found that assumptions about the build environment were embedded rather deeply in the code - including one place in a lisp file. The build proceeded quite far into the process, including running the java code to sort the dependencies, but obviously I missed at least one more place somewhere since I got to a point where the pathnames were clearly incorrect. So then started again, this time using the approach suggested in the emails - adding the missing directories and symbolic links. Oddly perhaps, I actually ended up with essential the same result and at nearly the same point in the build. So I decided to ask if anyone else had tried and succeeded yet. Here is the current output of the make script: ------- [EMAIL PROTECTED]:~/fricas-test/src/aldor# echo $AXIOM /home/page/fricas-test/target/x86_64-unknown-linux [EMAIL PROTECTED]:~/fricas-test/src/aldor# echo $ALDORROOT /home/page/aldor-src/aldor/install/aldor/linux/1.1.0 [EMAIL PROTECTED]:~/fricas-test/src/aldor# make /bin/sh: line 0: cd: /home/page/fricas-test/target/x86_64-unknown-linux/../../int: No such file or directory Building libaxiom.al and associated files /home/page/fricas-test/src/aldor/aldor /bin/sh: line 0: cd: /home/page/fricas-test/target/x86_64-unknown-linux/../../int: No such file or directory make[1]: Entering directory `/home/page/fricas-test/src/aldor' /home/page/fricas-test/src/aldor/libaxiom.mk:37: AXO_AP_FILES: /home/page/fricas-test/src/aldor /home/page/fricas-test/src/aldor/libaxiom.mk:44: AXO_GEN_AP_FILES: /home/page/fricas-test/src/aldor/aldor /home/page/fricas-test/src/aldor/types.mk:156: /home/page/fricas-test/src/aldor/aldor/saxiom/spadset.mk /home/page/fricas-test/src/aldor/Make.rules:179: ALL SPADSETS sax0 saxiom /home/page/fricas-test/src/aldor/Make.rules:200: (0,0) /home/page/fricas-test/src/aldor/Make.rules:204: /home/page/fricas-test/src/aldor/aldor/sax0/spadset.mk: No such file or directory /home/page/fricas-test/src/aldor/Make.rules:200: (1,1) /home/page/fricas-test/src/aldor/Make.rules:204: /home/page/fricas-test/src/aldor/aldor/saxiom/spadset.mk: No such file or directory /home/page/fricas-test/src/aldor/Make.rules:106: W: 1 I: 1 /home/page/fricas-test/src/aldor/Make.rules:107: W: /home/page/fricas-test/src/aldor/aldor/sax0/spadset.mk ls: /home/page/fricas-test/src/aldor/aldor/../algebra/*.spad: No such file or directory /home/page/fricas-test/src/aldor/aldor/typelist unchanged make[1]: *** No rule to make target `/home/page/fricas-test/src/aldor/as//home/page/fricas-test/src/aldor/aldor/ap/axlit.as', needed by `/home/page/fricas-test/src/aldor/aldor/ap/axlit.ap'. Stop. make[1]: Leaving directory `/home/page/fricas-test/src/aldor' make: *** [all] Error 2 [EMAIL PROTECTED]:~/fricas-test/src/aldor# --------- Notice this incorrect path: /home/page/fricas-test/src/aldor/as//home/page/fricas-test/src/aldor/aldor/ap/axlit.as As I said, I get this result in both my attempts. I think what I would like to do is return to the first approach, i.e. "do it right" so that this can eventually be included in the standard make files of FriCAS and OpenAxiom. It will be later today before I can get back to this. Any help greatly appreciated. Regards, Bill Page. ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ open-axiom-devel mailing list open-axiom-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/open-axiom-devel
