> On Jul 24, 2015, at 07:51, [email protected] wrote:
>
> To Whom It May Concern:
>
> I fear that I am a bit stumped after spending about four days attempting to
> install the latest available version of Octave on a new machine running OSX
> 10.10.3 using fink.
>
> fink installed fine (I’ve used it on different machines for the past decade
> starting in PPC land). XCode 6.4 and its associated Command Line Tools 6.4
> also are fine. FinkCommander was also installed and works although for this
> project I have been directly using the command line in bash.
>
> I am not new to the process. A year and a half ago I successfully installed
> Octave on an old Intel PowerMac (8 cores) running OSX 10.7.5 currently
> running Octave 3.6.4.
>
> I used the following command (with the understanding that this will be a
> 64bit installation using Apple’s native Quartz):
>
> fink install octave-atlas-qtx11
>
> It attempted to install Octave 3.8.2-6. During the build, after about 28
> hours, it failed with the hint that I should switch from using all 4 cores to
> using only 1. I assume this implies that something may be attempting to
> compile or whatever before a dependent part being built on another core is
> completed so it could not find it and thus the failure.
>
>
Yup. It also makes the logs more intelligible, since there aren’t multiple
threads echoing back to the terminal in arbitrary order.
> Switching to a single core, the above error did not recur and the process
> continued until it seemed to be trying to find gawk. Well, indeed, to my
> surprise, I didn’t find it in the system either, so under FinkCommander I
> installed gawk without difficulty and started again.
>
>
It would have been nice to have had that as a _separate_ bug report with some
output. Was the failure even in Octave, or some dependency? If I knew what
the package was, I could check and see if there is a missing build dependency.
> Having gotten past that problem, it finally failed again for reasons I cannot
> fathom. Towards the end, there was a series of hundreds of “Overfull /
> Underfull” messages (which sounds to my ignorance something like a stack
> problem).
Nope. Those are from TeX—generating documentation--and have to do with its
pickiness. They look like warnings to me.
> I have attached with this message a TextEdit version of the Terminal session.
> Probably the only thing useful to look at is the part highlighted in red
> towards the top of the file and the system-capture information generated by
> fink at the bottom. However, just to be safe, I have included everything.
>
>
> Before reaching out to you, I carefully read the entire Fink FAQ but nothing
> seems to capture what I am experiencing.
>
> I am a physicist in Hong Kong and Octave has now advanced to the point where
> there is a GUI for it in the most recent versions of OS X and I am hoping to
> play with it on a portable machine (MacBook Air).
>
> Thanks in advance for any and all help in this matter.
>
> Regards,
>
> John Babson
>
> 1 attachment, as stated (file: “Overfull Underfull”)
>
> <Overfull Underfull.rtf>
What looks to be causing the failure to me is :
(/sw/src/fink.build/octave-atlas-qtx11-3.8.2-6/octave-3.8.2/doc/interpreter/cp-
idx.texi (Concept Index) [867] [868] (./octave.cps
./octave.cps:56: Argument of \\ has an extra }.
<inserted text>
\par
<to be read again>
}
\initial ... .5\baselineskip \leftline {\secbf #1}
\nobreak \vskip .33\baseli...
l.56 \initial {\\}
?
Runaway argument?
\hss
./octave.cps:56: Paragraph ended before \\ was complete.
<to be read again>
\par
<to be read again>
}
\initial ... .5\baselineskip \leftline {\secbf #1}
\nobreak \vskip .33\baseli...
l.56 \initial {\\}
?
[869] [870] [871] [872] [873]))
(/sw/src/fink.build/octave-atlas-qtx11-3.8.2-6/octave-3.8.2/doc/interpreter/fn-
idx.texi (Function Index) [874] (./octave.fns [875] [876] [877] [878] [879]
[880] [881] [882] [883] [884] [885] [886] [887]))
(/sw/src/fink.build/octave-atlas-qtx11-3.8.2-6/octave-3.8.2/doc/interpreter/op-
idx.texi (Operator Index) [888] (./octave.ops
…
(and similar further instances)
That being said, I don’t reproduce your error (10.10.4/Xcode 6.4).
Could you please set your build jobs to 1 and use “fink -l build
octave382-atlas-qtmac” to generate a plain text log? Then send the file to me
off list (it will be named
/private/tmp/fink-build-log-octave-atlas-qtmac-3.8.1-6-<timestamp>.log). I’ll
generate the corresponding file on my own system and I can try to see what’s
going on.
--
Alexander Hansen, Ph.D.
Fink User Liaison
------------------------------------------------------------------------------
_______________________________________________
Fink-users mailing list
[email protected]
List archive:
http://news.gmane.org/gmane.os.macosx.fink.user
Subscription management:
https://lists.sourceforge.net/lists/listinfo/fink-users
