Loading info files



>>>>> "Rupert" == Rupert Swarbrick <rswarbrick at gmail.com> writes:

[snip]
    Rupert>   GCL gives:
    Rupert>   " should be skipped (I prefer algebra)" 

    Rupert>   SBCL gives:
    Rupert>   " should be skipped (I prefer algebra)" 

    Rupert>   SCL gives:
    Rupert>   Warning: the kernel parameter max_map_count=65530 is less than the 189440+
    Rupert>   pages usable by Scieneer Common Lisp.  It is recommended that max_map_count be
    Rupert>   increased to at least 189440.  This can be done with either:
    Rupert>      sysctl -w vm.max_map_count=189440
    Rupert>      echo 189440 > /proc/sys/vm/max_map_count
    Rupert>   Scieneer Common Lisp 1.3.9
    Rupert>   Copyright (c) 2000-2008, Scieneer Pty Ltd. All Rights Reserved.
    Rupert>   Restricted noncommercial license. Licensed to Rupert Swarbrick.
    Rupert>   Loaded subsystems:
    Rupert>       Compiler 1.0, target Intel x86
    Rupert>   * 
    Rupert>   "hould be skipped (I prefer algebra)" 

I'm kind of surprised scl uses characters.  Makes (file-position s
posn) more difficult to implement, I think.  And if it really does use
characters, I think to be fair you need to set the external format for
the file.  Maybe scl makes a guess at the kind of encoding in the
file?  Or the default is utf8?

This test with cmucl produces the expected results, assuming cmucl
uses byte offsets.

    Rupert> Looks like SBCL, GCL and ACL use byte offsets and SCL uses character
    Rupert> offsets. I didn't test the other lisps we support.

I don't think gcl supports unicode, so it's not surprising.  It works
fine for displaying info files, though.

Ray