From tioga at thirlmere.plus.com  Sun Jan 27 17:52:47 2008
From: tioga at thirlmere.plus.com (Andrew Cooper)
Date: Sun, 27 Jan 2008 22:52:47 +0000
Subject: [Tioga-users] Stack Smashing
Message-ID: <479D0B3F.7070103@thirlmere.plus.com>

Hello all,

I've just installed my favourite plotting tools, Tioga 1.6 and Ctioga 
1.6.2, under Ubuntu 7.10 ('Gutsy'). The default install of Tioga seems 
to work fine (well, the tests ran OK). Ctioga's tests, however, all died 
with the following opaque statement:

*** stack smashing detected ***: ruby terminated
Aborted (core dumped)

It turns out that gcc has SSP turned on by default, as packaged 
currently in Ubuntu. See:

https://wiki.ubuntu.com/GccSsp

So I recompiled Tioga with -fno-stack-protector and the problem "went 
away", of course. I had a very brief stab at trying to figure out where 
exactly the overrun was, but with no luck. After recompiling one file at 
a time without SSP, I think the problem might be in axes.c, but that's 
as far as I got.

Hope this is of some use.

Regards,

--
Andrew
email: a.p.cooper at durham.ac.uk

From fourmond at gmail.com  Mon Jan 28 03:19:47 2008
From: fourmond at gmail.com (Vincent Fourmond)
Date: Mon, 28 Jan 2008 09:19:47 +0100
Subject: [Tioga-users] Stack Smashing
In-Reply-To: <479D0B3F.7070103@thirlmere.plus.com>
References: <479D0B3F.7070103@thirlmere.plus.com>
Message-ID: <2e474d6f0801280019w1906e32av2f486bae924fc89f@mail.gmail.com>

  Hello,

On Jan 27, 2008 11:52 PM, Andrew Cooper <tioga at thirlmere.plus.com> wrote:
> I've just installed my favourite plotting tools, Tioga 1.6 and Ctioga
> 1.6.2, under Ubuntu 7.10 ('Gutsy'). The default install of Tioga seems
> to work fine (well, the tests ran OK). Ctioga's tests, however, all died
> with the following opaque statement:
>
> *** stack smashing detected ***: ruby terminated
> Aborted (core dumped)
>
> It turns out that gcc has SSP turned on by default, as packaged
> currently in Ubuntu. See:
>
> https://wiki.ubuntu.com/GccSsp
>
> So I recompiled Tioga with -fno-stack-protector and the problem "went
> away", of course. I had a very brief stab at trying to figure out where
> exactly the overrun was, but with no luck. After recompiling one file at
> a time without SSP, I think the problem might be in axes.c, but that's
> as far as I got.

  Another user also encountered the problem, but we couldn't track it
to its source. I thought I had tried every single gcc option
available, but I missed this one. I'll have a look at that, and I'll
forward a patch to the mailing list -- we might actually even release
a new version (there has been lots of things going on since the last
one).

  Thanks !

      Vincent

From fourmond at gmail.com  Tue Jan 29 14:18:26 2008
From: fourmond at gmail.com (Vincent Fourmond)
Date: Tue, 29 Jan 2008 20:18:26 +0100
Subject: [Tioga-users] Tioga 1.7 release
Message-ID: <479F7C02.7000108@gmail.com>


  Hello dear Tioga users !

  I'm glad to announce the 1.7 release of Tioga, the first in quite a
long while. While this may not have shown on this list, a lot has been
going on in Tioga's development over the past six month, as Taro Sato is
preparing a Python port of Tioga, and we had to adapt the code so it
would be easier.

  But that wasn't the only change. Among other things, let's name a few:

  * it is now possible to know the exact size of TeX text, using
get_text_size. I suggest you have a look at the documentation, and at
the text_size examples in samples/figures/figures.rb. This came somehow
as a shock, as we never thought it would be possible.

  * legend information without lines, just call save_legend_info with
'line_type' => 'None' in addition to the marker information. You can
also give a 'legend' argument to show_marker.

  * some bug fixes, which I hope will solve the various stack overflows
we've seen recently on this list (I'm not entirely sure, as I could
never reproduce them on my machine).

  * some improvements in Dvector.fancy_read (around three times faster
now), and new methods to Function.

  * the release should be Ruby 1.9-ready. Please, check it out and let
us know.

  The release is, as usual, available at

http://rubyforge.org/frs/?group_id=701

  Enjoy !

	Vincent

From kenshen at physics.ucsb.edu  Tue Jan 29 20:35:02 2008
From: kenshen at physics.ucsb.edu (Ken Shen)
Date: Tue, 29 Jan 2008 17:35:02 -0800 (PST)
Subject: [Tioga-users] Tioga 1.7 installation error
In-Reply-To: <479F7C02.7000108@gmail.com>
References: <479F7C02.7000108@gmail.com>
Message-ID: <Pine.LNX.4.64.0801291732580.5951@mrwilson.physics.ucsb.edu>

Hi Vincent, thanks for the update.  Unfortunately, no luck with the 
installation!  This is what it spat out when I ran QUICK_INSTALL:

checking for zlib.h... yes
checking for compress() in -lz... yes
checking for ieee754.h... no
You lack the ieee754.h header file, which might mean lower reliability 
when Marshalling Dvectors and Dtables
checking for isnan() ... no
MacOS specific installation
Writing Makefile
gcc -g -O2 -pipe -fno-common -O2 -Wall    -I/usr/local/lib/ruby/1.8 
-I/usr/local/lib/ruby/1.8/i686-darwin9.1.0 -I. -I./include -IDvector 
-IDvector/include -IDtable -IDtable/include -IFlate -IFlate/include 
-IFunction -IFunction/include -ITioga -D_HAVE_ZLIB_H -D_HAVE_Z -c 
Dtable/dtable.c -o  Dtable/dtable.o
In file included from Dtable/dtable_intern.h:27,
                 from Dtable/dtable.c:23:
Dvector/include/dvector.h:34: error: expected '=', ',', ';', 'asm' or 
'__attribute__' before 'rb_export_Dvector_Data_for_Read_type'
Dvector/include/dvector.h:36: error: expected '=', ',', ';', 'asm' or 
'__attribute__' before 'rb_export_Dvector_Data_Copy_type'
Dvector/include/dvector.h:38: error: expected '=', ',', ';', 'asm' or 
'__attribute__' before 'rb_export_Dvector_Data_for_Write_type'
Dvector/include/dvector.h:40: error: expected '=', ',', ';', 'asm' or 
'__attribute__' before 'rb_export_Dvector_Data_Resize_type'
Dvector/include/dvector.h:41: error: expected '=', ',', ';', 'asm' or 
'__attribute__' before 'rb_export_Dvector_Data_Replace_type'
Dvector/include/dvector.h:44: error: expected '=', ',', ';', 'asm' or 
'__attribute__' before 'rb_export_Dvector_Create_type'
Dvector/include/dvector.h:45: error: expected '=', ',', ';', 'asm' or 
'__attribute__' before 'rb_export_Dvector_Store_Double_type'
Dvector/include/dvector.h:47: error: expected '=', ',', ';', 'asm' or 
'__attribute__' before 'rb_export_Dvector_Push_Double_type'
Dvector/include/dvector.h:50: error: expected '=', ',', ';', 'asm' or 
'__attribute__' before 'rb_export_c_dvector_spline_interpolate_type'
Dvector/include/dvector.h:53: error: expected '=', ',', ';', 'asm' or 
'__attribute__' before 'rb_export_c_dvector_linear_interpolate_type'
Dvector/include/dvector.h:55: error: expected '=', ',', ';', 'asm' or 
'__attribute__' before 
'rb_export_c_dvector_create_spline_interpolant_type'
Dtable/dtable.c: In function 'dtable_row':
Dtable/dtable.c:410: warning: implicit declaration of function 
'Dvector_Create'
Dtable/dtable.c:411: warning: implicit declaration of function 
'Dvector_Data_Replace'
Dtable/dtable.c: In function 'dtable_set_row':
Dtable/dtable.c:425: warning: implicit declaration of function 
'Dvector_Data_for_Read'
Dtable/dtable.c:425: warning: initialization makes pointer from integer 
without a cast
Dtable/dtable.c: In function 'dtable_set_column':
Dtable/dtable.c:447: warning: initialization makes pointer from integer 
without a cast
Dtable/dtable.c: In function 'dtable_column':
Dtable/dtable.c:473: warning: implicit declaration of function 
'Dvector_Data_Resize'
Dtable/dtable.c:475: warning: implicit declaration of function 
'Dvector_Store_Double'
Dtable/dtable.c: In function 'Init_Dtable':
Dtable/dtable.c:1919: error: 'Dvector_Create' undeclared (first use in 
this function)
Dtable/dtable.c:1919: error: (Each undeclared identifier is reported only 
once
Dtable/dtable.c:1919: error: for each function it appears in.)
Dtable/dtable.c:1920: error: 'Dvector_Data_Resize' undeclared (first use 
in this function)
Dtable/dtable.c:1921: error: 'Dvector_Data_Replace' undeclared (first use 
in this function)
Dtable/dtable.c:1922: error: 'Dvector_Data_for_Read' undeclared (first use 
in this function)
Dtable/dtable.c:1923: error: 'Dvector_Store_Double' undeclared (first use 
in this function)
Dtable/dtable.c: At top level:
Dtable/dtable.c:1930: error: expected '=', ',', ';', 'asm' or 
'__attribute__' before 'rb_export_Dvector_Create_type'
Dtable/dtable.c:1931: error: expected '=', ',', ';', 'asm' or 
'__attribute__' before 'rb_export_Dvector_Data_Resize_type'
Dtable/dtable.c:1932: error: expected '=', ',', ';', 'asm' or 
'__attribute__' before 'rb_export_Dvector_Data_Replace_type'
Dtable/dtable.c:1933: error: expected '=', ',', ';', 'asm' or 
'__attribute__' before 'rb_export_Dvector_Data_for_Read_type'
Dtable/dtable.c:1934: error: expected '=', ',', ';', 'asm' or 
'__attribute__' before 'rb_export_Dvector_Store_Double_type'
make: *** [Dtable/dtable.o] Error 1



-Ken


On Tue, 29 Jan 2008, Vincent Fourmond wrote:

> 
>   Hello dear Tioga users !
> 
>   I'm glad to announce the 1.7 release of Tioga, the first in quite a
> long while. While this may not have shown on this list, a lot has been
> going on in Tioga's development over the past six month, as Taro Sato is
> preparing a Python port of Tioga, and we had to adapt the code so it
> would be easier.
> 
>   But that wasn't the only change. Among other things, let's name a few:
> 
>   * it is now possible to know the exact size of TeX text, using
> get_text_size. I suggest you have a look at the documentation, and at
> the text_size examples in samples/figures/figures.rb. This came somehow
> as a shock, as we never thought it would be possible.
> 
>   * legend information without lines, just call save_legend_info with
> 'line_type' => 'None' in addition to the marker information. You can
> also give a 'legend' argument to show_marker.
> 
>   * some bug fixes, which I hope will solve the various stack overflows
> we've seen recently on this list (I'm not entirely sure, as I could
> never reproduce them on my machine).
> 
>   * some improvements in Dvector.fancy_read (around three times faster
> now), and new methods to Function.
> 
>   * the release should be Ruby 1.9-ready. Please, check it out and let
> us know.
> 
>   The release is, as usual, available at
> 
> http://rubyforge.org/frs/?group_id=701
> 
>   Enjoy !
> 
> 	Vincent
> _______________________________________________
> Tioga-users mailing list
> Tioga-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/tioga-users
> 

From fourmond at gmail.com  Wed Jan 30 03:49:15 2008
From: fourmond at gmail.com (Vincent Fourmond)
Date: Wed, 30 Jan 2008 09:49:15 +0100
Subject: [Tioga-users] Tioga 1.7 installation error
In-Reply-To: <Pine.LNX.4.64.0801291732580.5951@mrwilson.physics.ucsb.edu>
References: <479F7C02.7000108@gmail.com>
	<Pine.LNX.4.64.0801291732580.5951@mrwilson.physics.ucsb.edu>
Message-ID: <2e474d6f0801300049i23258304od14b57f719637df0@mail.gmail.com>

  Hello,

On Jan 30, 2008 2:35 AM, Ken Shen <kenshen at physics.ucsb.edu> wrote:
> Hi Vincent, thanks for the update.  Unfortunately, no luck with the
> installation!  This is what it spat out when I ran QUICK_INSTALL:
>
> checking for zlib.h... yes
> checking for compress() in -lz... yes
> checking for ieee754.h... no
> You lack the ieee754.h header file, which might mean lower reliability
> when Marshalling Dvectors and Dtables
> checking for isnan() ... no
> MacOS specific installation
> Writing Makefile
> gcc -g -O2 -pipe -fno-common -O2 -Wall    -I/usr/local/lib/ruby/1.8
> -I/usr/local/lib/ruby/1.8/i686-darwin9.1.0 -I. -I./include -IDvector
> -IDvector/include -IDtable -IDtable/include -IFlate -IFlate/include
> -IFunction -IFunction/include -ITioga -D_HAVE_ZLIB_H -D_HAVE_Z -c
> Dtable/dtable.c -o  Dtable/dtable.o
> In file included from Dtable/dtable_intern.h:27,
>                  from Dtable/dtable.c:23:
> Dvector/include/dvector.h:34: error: expected '=', ',', ';', 'asm' or
> '__attribute__' before 'rb_export_Dvector_Data_for_Read_type'
> [...]

  This is very puzzling. Did you manage to install Tioga 1.6 this way
? Because the bit of code causing the problem didn't change since
Tioga 1.1.

  Could you send Dtable/dtable.i (compressed) created by in split/:

gcc -g -O2 -E -pipe -fno-common -O2 -Wall    -I/usr/local/lib/ruby/1.8
-I/usr/local/lib/ruby/1.8/i686-darwin9.1.0 -I. -I./include -IDvector
-IDvector/include -IDtable -IDtable/include -IFlate -IFlate/include
-IFunction -IFunction/include -ITioga -D_HAVE_ZLIB_H -D_HAVE_Z -c
Dtable/dtable.c -o  Dtable/dtable.o

  I really don't understand what is happening here.

  Cheers,

      Vincent

From jdrs at physics.ucsb.edu  Wed Jan 30 13:11:32 2008
From: jdrs at physics.ucsb.edu (Justin D. R. Steinfadt)
Date: Wed, 30 Jan 2008 10:11:32 -0800
Subject: [Tioga-users] Restricting Line Plotting Domain/Range?
In-Reply-To: <2e474d6f0801300049i23258304od14b57f719637df0@mail.gmail.com>
References: <479F7C02.7000108@gmail.com>	<Pine.LNX.4.64.0801291732580.5951@mrwilson.physics.ucsb.edu>
	<2e474d6f0801300049i23258304od14b57f719637df0@mail.gmail.com>
Message-ID: <47A0BDD4.8020501@physics.ucsb.edu>

Hi all, I'm hoping there is an easy way to do this in Tioga that I'm 
just not seeing.

Here's the situation... I have two vectors of data defining the X and Y 
locations for points in my plot that define a function and say the 
domain runs from Xmin to Xmax in the X vector.  However, I only want to 
plot (using something like show_polyline) from Xbegin to Xend within my 
plot where Xbegin>Xmin and/or Xend<Xmax. (This also could be extended to 
the range of the data.)

I don't want to have to truncate my data to accommodate this and I would 
rather not have to look inside my data at all.

Is there perhaps some way to create a "Bounding Box" of sorts within the 
plot frame that restricts where a method like show_polyline might be 
able to draw a line?

Thanks for any help!
-Justin

From kenshen at physics.ucsb.edu  Wed Jan 30 14:03:58 2008
From: kenshen at physics.ucsb.edu (Ken Shen)
Date: Wed, 30 Jan 2008 11:03:58 -0800 (PST)
Subject: [Tioga-users] Tioga 1.7 installation error
In-Reply-To: <2e474d6f0801300049i23258304od14b57f719637df0@mail.gmail.com>
References: <479F7C02.7000108@gmail.com>
	<Pine.LNX.4.64.0801291732580.5951@mrwilson.physics.ucsb.edu>
	<2e474d6f0801300049i23258304od14b57f719637df0@mail.gmail.com>
Message-ID: <64646.128.111.23.180.1201719838.squirrel@webmail.physics.ucsb.edu>

QUICK_INSTALL seems to work fine for version 1.6.  Here is a compressed
version of dtable.o (I assume that's what you meant instead of dtable.i?)

Thanks,
Ken


>   Hello,
>
> On Jan 30, 2008 2:35 AM, Ken Shen <kenshen at physics.ucsb.edu> wrote:
>> Hi Vincent, thanks for the update.  Unfortunately, no luck with the
>> installation!  This is what it spat out when I ran QUICK_INSTALL:
>>
>> checking for zlib.h... yes
>> checking for compress() in -lz... yes
>> checking for ieee754.h... no
>> You lack the ieee754.h header file, which might mean lower reliability
>> when Marshalling Dvectors and Dtables
>> checking for isnan() ... no
>> MacOS specific installation
>> Writing Makefile
>> gcc -g -O2 -pipe -fno-common -O2 -Wall    -I/usr/local/lib/ruby/1.8
>> -I/usr/local/lib/ruby/1.8/i686-darwin9.1.0 -I. -I./include -IDvector
>> -IDvector/include -IDtable -IDtable/include -IFlate -IFlate/include
>> -IFunction -IFunction/include -ITioga -D_HAVE_ZLIB_H -D_HAVE_Z -c
>> Dtable/dtable.c -o  Dtable/dtable.o
>> In file included from Dtable/dtable_intern.h:27,
>>                  from Dtable/dtable.c:23:
>> Dvector/include/dvector.h:34: error: expected '=', ',', ';', 'asm' or
>> '__attribute__' before 'rb_export_Dvector_Data_for_Read_type'
>> [...]
>
>   This is very puzzling. Did you manage to install Tioga 1.6 this way
> ? Because the bit of code causing the problem didn't change since
> Tioga 1.1.
>
>   Could you send Dtable/dtable.i (compressed) created by in split/:
>
> gcc -g -O2 -E -pipe -fno-common -O2 -Wall    -I/usr/local/lib/ruby/1.8
> -I/usr/local/lib/ruby/1.8/i686-darwin9.1.0 -I. -I./include -IDvector
> -IDvector/include -IDtable -IDtable/include -IFlate -IFlate/include
> -IFunction -IFunction/include -ITioga -D_HAVE_ZLIB_H -D_HAVE_Z -c
> Dtable/dtable.c -o  Dtable/dtable.o
>
>   I really don't understand what is happening here.
>
>   Cheers,
>
>       Vincent
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: dtable.o.zip
Type: application/zip
Size: 26760 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/tioga-users/attachments/20080130/418e0108/attachment-0001.zip 

From nomo17k at gmail.com  Wed Jan 30 14:17:40 2008
From: nomo17k at gmail.com (Taro Sato)
Date: Wed, 30 Jan 2008 11:17:40 -0800
Subject: [Tioga-users] Restricting Line Plotting Domain/Range?
In-Reply-To: <47A0BDD4.8020501@physics.ucsb.edu>
References: <479F7C02.7000108@gmail.com>
	<Pine.LNX.4.64.0801291732580.5951@mrwilson.physics.ucsb.edu>
	<2e474d6f0801300049i23258304od14b57f719637df0@mail.gmail.com>
	<47A0BDD4.8020501@physics.ucsb.edu>
Message-ID: <ebdf6c730801301117h22ebde5s43c1a6374fe2f41@mail.gmail.com>

On Jan 30, 2008 10:11 AM, Justin D. R. Steinfadt <jdrs at physics.ucsb.edu> wrote:
>
> Is there perhaps some way to create a "Bounding Box" of sorts within the
> plot frame that restricts where a method like show_polyline might be
> able to draw a line?


Justin -- I think you should be able to do this by manually specifying
your boundary in show_plot:

 t.show_plot(bounds)

where you probably are using something like plot_boundaries( ... ) to
make "bounds" on fly now.  Perhaps you can get the return value of
plot_boundaries into an array and replace the elements corresponding
to the the limits for x axis.  (Unfortunately it's been a while since
I used Ruby so I cannot give you a sample code.)

Taro

From fourmond at gmail.com  Wed Jan 30 14:53:51 2008
From: fourmond at gmail.com (Vincent Fourmond)
Date: Wed, 30 Jan 2008 20:53:51 +0100
Subject: [Tioga-users] Tioga 1.7 installation error
In-Reply-To: <64646.128.111.23.180.1201719838.squirrel@webmail.physics.ucsb.edu>
References: <479F7C02.7000108@gmail.com>
	<Pine.LNX.4.64.0801291732580.5951@mrwilson.physics.ucsb.edu>
	<2e474d6f0801300049i23258304od14b57f719637df0@mail.gmail.com>
	<64646.128.111.23.180.1201719838.squirrel@webmail.physics.ucsb.edu>
Message-ID: <47A0D5CF.6060902@gmail.com>


  Hello,

Ken Shen wrote:
> QUICK_INSTALL seems to work fine for version 1.6.

  I'm puzzled... Could you post the output of QUICK_INSTALL for Tioga 1.6 ?

> Here is a compressed version of dtable.o (I assume that's what you
meant instead of dtable.i?)

  Yes - it should have been dtable.i if I had removed the -o dtable.o
command-line argument ;-)...

  Try replacing in split/namespace.h

# define INTERN __private_extern__

by something like

# define INTERN extern

  Does that change the output message ?

  Cheers,

	Vincent

From kenshen at physics.ucsb.edu  Wed Jan 30 16:19:33 2008
From: kenshen at physics.ucsb.edu (Ken Shen)
Date: Wed, 30 Jan 2008 13:19:33 -0800 (PST)
Subject: [Tioga-users] Tioga 1.7 installation error
In-Reply-To: <47A0D5CF.6060902@gmail.com>
References: <479F7C02.7000108@gmail.com>
	<Pine.LNX.4.64.0801291732580.5951@mrwilson.physics.ucsb.edu>
	<2e474d6f0801300049i23258304od14b57f719637df0@mail.gmail.com>
	<64646.128.111.23.180.1201719838.squirrel@webmail.physics.ucsb.edu>
	<47A0D5CF.6060902@gmail.com>
Message-ID: <Pine.LNX.4.64.0801301315010.25886@mrwilson.physics.ucsb.edu>

Changing "# define INTERN __private_extern__" to "# define INTERN extern" 
in namespace.h definitely changes the output, and it finishes the 
installation, but with lots of warnings.  Running Tioga after the install 
gives:

-----------------
/usr/local/lib/ruby/site_ruby/1.8/i686-darwin9.1.0/Dobjects/Dtable.bundle: 
Failed to lookup Init function 
/usr/local/lib/ruby/site_ruby/1.8/i686-darwin9.1.0/Dobjects/Dtable.bundle 
(LoadError)
	from /usr/local/lib/ruby/site_ruby/1.8/Tioga/tioga.rb:1
	from 
/usr/local/lib/ruby/site_ruby/1.8/Tioga/tioga_ui_cmds.rb:22:in `require'
	from /usr/local/lib/ruby/site_ruby/1.8/Tioga/tioga_ui_cmds.rb:22
	from /usr/local/lib/ruby/site_ruby/1.8/Tioga/tioga_ui.rb:2:in 
`require'
	from /usr/local/lib/ruby/site_ruby/1.8/Tioga/tioga_ui.rb:2
	from /usr/local/bin/tioga:2:in `require'
	from /usr/local/bin/tioga:2
------------------


Also, here is the output for Tioga 1.6's QUICK_INSTALL:

------------------
checking for zlib.h... yes
checking for compress() in -lz... yes
checking for ieee754.h... no
You lack the ieee754.h header file, which might mean lower reliability 
when Marshalling Dvectors and Dtables
checking for isnan() ... no
MacOS specific installation
Writing Makefile
make: Nothing to be done for `build'.
/usr/bin/install -c -m 644 Dvector/lib/Numeric_extras.rb 
/usr/local/lib/ruby/site_ruby/1.8/Dobjects/Numeric_extras.rb
/usr/bin/install -c -m 644 Dvector/lib/Dvector_extras.rb 
/usr/local/lib/ruby/site_ruby/1.8/Dobjects/Dvector_extras.rb
/usr/bin/install -c Dvector.bundle 
/usr/local/lib/ruby/site_ruby/1.8/i686-darwin9.1.0/Dobjects/Dvector.bundle
/usr/bin/install -c -m 644 Dtable/lib/Dtable_extras.rb 
/usr/local/lib/ruby/site_ruby/1.8/Dobjects/Dtable_extras.rb
/usr/bin/install -c Dtable.bundle 
/usr/local/lib/ruby/site_ruby/1.8/i686-darwin9.1.0/Dobjects/Dtable.bundle
/usr/bin/install -c Flate.bundle 
/usr/local/lib/ruby/site_ruby/1.8/i686-darwin9.1.0/Flate.bundle
/usr/bin/install -c -m 644 Function/lib/Function_extras.rb 
/usr/local/lib/ruby/site_ruby/1.8/Dobjects/Function_extras.rb
/usr/bin/install -c Function.bundle 
/usr/local/lib/ruby/site_ruby/1.8/i686-darwin9.1.0/Dobjects/Function.bundle
/usr/bin/install -c -m 644 Tioga/lib/Utils.rb 
/usr/local/lib/ruby/site_ruby/1.8/Tioga/Utils.rb
/usr/bin/install -c -m 644 Tioga/lib/FigMkr.rb 
/usr/local/lib/ruby/site_ruby/1.8/Tioga/FigMkr.rb
/usr/bin/install -c -m 644 Tioga/lib/FigureConstants.rb 
/usr/local/lib/ruby/site_ruby/1.8/Tioga/FigureConstants.rb
/usr/bin/install -c -m 644 Tioga/lib/MarkerConstants.rb 
/usr/local/lib/ruby/site_ruby/1.8/Tioga/MarkerConstants.rb
/usr/bin/install -c -m 644 Tioga/lib/irb_tioga.rb 
/usr/local/lib/ruby/site_ruby/1.8/Tioga/irb_tioga.rb
/usr/bin/install -c -m 644 Tioga/lib/tioga_ui_cmds.rb 
/usr/local/lib/ruby/site_ruby/1.8/Tioga/tioga_ui_cmds.rb
/usr/bin/install -c -m 644 Tioga/lib/tioga.rb 
/usr/local/lib/ruby/site_ruby/1.8/Tioga/tioga.rb
/usr/bin/install -c -m 644 Tioga/lib/TexPreamble.rb 
/usr/local/lib/ruby/site_ruby/1.8/Tioga/TexPreamble.rb
/usr/bin/install -c -m 644 Tioga/lib/ColorConstants.rb 
/usr/local/lib/ruby/site_ruby/1.8/Tioga/ColorConstants.rb
/usr/bin/install -c -m 644 Tioga/lib/tioga_ui.rb 
/usr/local/lib/ruby/site_ruby/1.8/Tioga/tioga_ui.rb
/usr/bin/install -c -m 644 Tioga/lib/maker.rb 
/usr/local/lib/ruby/site_ruby/1.8/Tioga/maker.rb
/usr/bin/install -c FigureMaker.bundle 
/usr/local/lib/ruby/site_ruby/1.8/i686-darwin9.1.0/Tioga/FigureMaker.bundle
ruby -r fileutils -e 'FileUtils.install(ARGV[0],ARGV[1],:mode => 0755)' 
scripts/tioga /usr/local/bin/tioga
ruby -r fileutils -e 'FileUtils.install(ARGV[0],ARGV[1],:mode => 0755)' 
scripts/irb_tioga /usr/local/bin/irb_tioga
ruby -r fileutils -e 'FileUtils.install(ARGV[0],ARGV[1],:mode => 0755)' 
repreview /usr/local/bin/repreview
ruby -r fileutils -e 'FileUtils.install(ARGV[0],ARGV[1],:mode => 0755)' 
scripts/Reload_Preview_Document.scpt 
/usr/local/bin/Reload_Preview_Document.scpt
This is the test suite for Tioga 1.6
Loaded suite ts_Tioga
Started
....................................................................
compressed from 0 to 8
compressed from 13 to 21
compressed from 70 to 69
compressed from 3631 to 1642
. 
 
 
>>> NOTE: please look at tests/Icon_Test.pdf and compare it to 
samples/Icon.pdf
...........
Finished in 1.121682 seconds.

80 tests, 726 assertions, 0 failures, 0 errors
-----------


Thanks,
Ken


On Wed, 30 Jan 2008, Vincent Fourmond wrote:

> 
>   Hello,
> 
> Ken Shen wrote:
> > QUICK_INSTALL seems to work fine for version 1.6.
> 
>   I'm puzzled... Could you post the output of QUICK_INSTALL for Tioga 1.6 ?
> 
> > Here is a compressed version of dtable.o (I assume that's what you
> meant instead of dtable.i?)
> 
>   Yes - it should have been dtable.i if I had removed the -o dtable.o
> command-line argument ;-)...
> 
>   Try replacing in split/namespace.h
> 
> # define INTERN __private_extern__
> 
> by something like
> 
> # define INTERN extern
> 
>   Does that change the output message ?
> 
>   Cheers,
> 
> 	Vincent
> 

From fourmond at gmail.com  Thu Jan 31 03:48:28 2008
From: fourmond at gmail.com (Vincent Fourmond)
Date: Thu, 31 Jan 2008 09:48:28 +0100
Subject: [Tioga-users] Tioga 1.7 installation error
In-Reply-To: <Pine.LNX.4.64.0801301315010.25886@mrwilson.physics.ucsb.edu>
References: <479F7C02.7000108@gmail.com>
	<Pine.LNX.4.64.0801291732580.5951@mrwilson.physics.ucsb.edu>
	<2e474d6f0801300049i23258304od14b57f719637df0@mail.gmail.com>
	<64646.128.111.23.180.1201719838.squirrel@webmail.physics.ucsb.edu>
	<47A0D5CF.6060902@gmail.com>
	<Pine.LNX.4.64.0801301315010.25886@mrwilson.physics.ucsb.edu>
Message-ID: <2e474d6f0801310048h7c0d6e5andad33a9023f7e1f6@mail.gmail.com>

  Hello,

On Jan 30, 2008 10:19 PM, Ken Shen <kenshen at physics.ucsb.edu> wrote:
> Changing "# define INTERN __private_extern__" to "# define INTERN extern"
> in namespace.h definitely changes the output, and it finishes the
> installation, but with lots of warnings.  Running Tioga after the install
> gives:
>
> -----------------
> /usr/local/lib/ruby/site_ruby/1.8/i686-darwin9.1.0/Dobjects/Dtable.bundle:
> Failed to lookup Init function
> /usr/local/lib/ruby/site_ruby/1.8/i686-darwin9.1.0/Dobjects/Dtable.bundle
> (LoadError)
>         from /usr/local/lib/ruby/site_ruby/1.8/Tioga/tioga.rb:1
>         from
> /usr/local/lib/ruby/site_ruby/1.8/Tioga/tioga_ui_cmds.rb:22:in `require'
>         from /usr/local/lib/ruby/site_ruby/1.8/Tioga/tioga_ui_cmds.rb:22
>         from /usr/local/lib/ruby/site_ruby/1.8/Tioga/tioga_ui.rb:2:in
> `require'
>         from /usr/local/lib/ruby/site_ruby/1.8/Tioga/tioga_ui.rb:2
>         from /usr/local/bin/tioga:2:in `require'
>         from /usr/local/bin/tioga:2
> ------------------

  Oops... Try

# define INTERN __attribute__ ((visibility ("hidden")))


> Also, here is the output for Tioga 1.6's QUICK_INSTALL:
>
> ------------------
> checking for zlib.h... yes
> checking for compress() in -lz... yes
> checking for ieee754.h... no
> You lack the ieee754.h header file, which might mean lower reliability
> when Marshalling Dvectors and Dtables
> checking for isnan() ... no
> MacOS specific installation
> Writing Makefile
> make: Nothing to be done for `build'.

  This is what I thought ;-)... Please, the output from QUICK_INSTALL
(you don't have to be root, it will fail for installation but not for
building) from a *fresh* tioga-1.6 tarball. It appears that you
changed compiler since the last time you installed Tioga, which is the
likely cause for your problems. What gives

gcc --version

  Thanks !

      Vincent, suddenly much less puzzled