Bugs item #1082938, was opened at 2004-12-10 15:23
Message generated for change (Settings changed) made by zmoelnig
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=478070&aid=1082938&group_…
Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: puredata
Group: None
>Status: Closed
Resolution: Fixed
Priority: 5
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Nobody/Anonymous (nobody)
Summary: sys_load_lib() on OSX too limited?
Initial Comment:
...externals are loaded and linked as NSModules, but it
seems that we may be limiting ourselves by using the
NSLINKMODULE_OPTION_PRIVATE flag? This flag came to my
attention while I was trying to compile
pdp2gem.pd_darwin, which acts as a bridge between pdp
and gem: ie. it requires linking to symbols from both
libraries, which should be done at module load
time...this is not possible with the PRIVATE flag, but
it does work with out it...
...so, I'm wondering if there's a general rational
behind this that isn't obvious, or if we can just go
ahead and remove it? Preliminary testing has not
showed any obvious problems with multiple libraries...
...here's the relevent docs from xcode:
options
An unsigned long value. Pass one or more of the
following bit masks or NSLINKMODULE_OPTION_NONE to
specify no options.
NSLINKMODULE_OPTION_BINDNOW
The dynamic linker binds all of the undefined
references immediately, rather than waiting until the
references are actually used. All dependent libraries
are also be bound.
NSLINKMODULE_OPTION_PRIVATE
Do not add the global symbols from the module to the
global symbol list. Instead, you must use the
NSLookupSymbolInModule function to obtain symbols from
this module.
NSLINKMODULE_OPTION_RETURN_ON_ERROR
If an error occurs while binding the module, return
NULL. You can then use the function NSLinkEditError to
retrieve information about the error.
Without this option, NSLinkModule calls the linkEdit
error handler you have installed using the
NSInstallLinkEditErrorHandlers function. If you have
not installed a link edit error handler, NSLinkModule
prints a message to the standard error stream and
causes a breakpoint trap to end the program.
----------------------------------------------------------------------
Comment By: IOhannes m zmölnig (zmoelnig)
Date: 2006-10-18 10:37
Message:
Logged In: YES
user_id=564396
since libraries on os-x are now loaded using dlopen(), i
think we can close this.
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=478070&aid=1082938&group_…
Bugs item #1082938, was opened at 2004-12-10 15:23
Message generated for change (Comment added) made by zmoelnig
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=478070&aid=1082938&group_…
Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: puredata
Group: None
Status: Open
>Resolution: Fixed
Priority: 5
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Nobody/Anonymous (nobody)
Summary: sys_load_lib() on OSX too limited?
Initial Comment:
...externals are loaded and linked as NSModules, but it
seems that we may be limiting ourselves by using the
NSLINKMODULE_OPTION_PRIVATE flag? This flag came to my
attention while I was trying to compile
pdp2gem.pd_darwin, which acts as a bridge between pdp
and gem: ie. it requires linking to symbols from both
libraries, which should be done at module load
time...this is not possible with the PRIVATE flag, but
it does work with out it...
...so, I'm wondering if there's a general rational
behind this that isn't obvious, or if we can just go
ahead and remove it? Preliminary testing has not
showed any obvious problems with multiple libraries...
...here's the relevent docs from xcode:
options
An unsigned long value. Pass one or more of the
following bit masks or NSLINKMODULE_OPTION_NONE to
specify no options.
NSLINKMODULE_OPTION_BINDNOW
The dynamic linker binds all of the undefined
references immediately, rather than waiting until the
references are actually used. All dependent libraries
are also be bound.
NSLINKMODULE_OPTION_PRIVATE
Do not add the global symbols from the module to the
global symbol list. Instead, you must use the
NSLookupSymbolInModule function to obtain symbols from
this module.
NSLINKMODULE_OPTION_RETURN_ON_ERROR
If an error occurs while binding the module, return
NULL. You can then use the function NSLinkEditError to
retrieve information about the error.
Without this option, NSLinkModule calls the linkEdit
error handler you have installed using the
NSInstallLinkEditErrorHandlers function. If you have
not installed a link edit error handler, NSLinkModule
prints a message to the standard error stream and
causes a breakpoint trap to end the program.
----------------------------------------------------------------------
>Comment By: IOhannes m zmölnig (zmoelnig)
Date: 2006-10-18 10:37
Message:
Logged In: YES
user_id=564396
since libraries on os-x are now loaded using dlopen(), i
think we can close this.
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=478070&aid=1082938&group_…
Bugs item #1371511, was opened at 2005-12-02 08:41
Message generated for change (Comment added) made by zmoelnig
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=478070&aid=1371511&group_…
Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: puredata
Group: None
>Status: Pending
Resolution: None
Priority: 5
Submitted By: Hans-Christoph Steiner (eighthave)
Assigned to: Miller Puckette (millerpuckette)
Summary: declaration of `y1' shadows a global declaration
Initial Comment:
When using math.h on Darwin, the y1 variable causes
lots of warnings.
../../pd/src/g_canvas.h:247: warning: declaration of
`y1' shadows a global declaration
/usr/include/architecture/ppc/math.h:483: warning:
shadowed declaration is here
Renaming y1 in g_canvas.h would fix this problem.
----------------------------------------------------------------------
>Comment By: IOhannes m zmölnig (zmoelnig)
Date: 2006-10-18 10:33
Message:
Logged In: YES
user_id=564396
a) has this been fixed already? (i cannot reproduce this
anymore; also the pd(-extended) buildlogs do not show this
warning)
b) i don't think that this is a bug in pd; no global include
file (/usr/include/architecture/ppc/math.h) should use
global variable names like "i", "y1" and the like.
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=478070&aid=1371511&group_…
Bugs item #1531757, was opened at 2006-07-31 15:46
Message generated for change (Comment added) made by zmoelnig
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=478070&aid=1531757&group_…
Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: puredata
Group: None
>Status: Closed
>Resolution: Fixed
Priority: 5
Submitted By: Jamie Bullock (postlude)
Assigned to: Nobody/Anonymous (nobody)
Summary: Opening curly brace causes freeze until a closing brace
Initial Comment:
Typing a curly brace '{' anywhere in pd causes a freeze
until a closing brace is given.
----------------------------------------------------------------------
>Comment By: IOhannes m zmölnig (zmoelnig)
Date: 2006-10-18 10:13
Message:
Logged In: YES
user_id=564396
seems like this is fixed with pd-0.40-1 (probably earlier):
typing opening curly brace results in "keycode 123: dropped".
(the better solution would of course be to properly escape
curly brackets so that they could be used)
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=478070&aid=1531757&group_…
Patches item #1579049, was opened at 2006-10-17 11:49
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1579049&group_…
Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: externals
Group: bugfix
Status: Open
Resolution: None
Priority: 5
Submitted By: Hans-Christoph Steiner (eighthave)
Assigned to: Hans-Christoph Steiner (eighthave)
Summary: build flag for cyclone to fix Windows issues
Initial Comment:
>From an email from Krzysztof:
Btw, I learned, while scanning pd-list messages, that
some people complain about miXed externals instability
in pd-extended for Windows (which I do not use). Do
you confirm, that is the case? If so, I would like to
ask you for help, since I would rather avoid installing
mingw + doing cvs checkout on windows, etc. just for
the sake of flipping some weird little flag. So, when
you have time, would you check if things improve after
applying this patch to miXed/Makefile.common:
@@ -50,8 +50,8 @@
CC = gcc
X_SUFFIX = dll
DEFINES = -DNT -DMSW
-# assuming pd itself was either vc-compiled or
declared -mms-bitfields too...
-OPT_CFLAGS = -funroll-loops -mms-bitfields
+# assuming pd itself was mingw-compiled, and
-mms-bitfields was not set
+OPT_CFLAGS = -funroll-loops
LFLAGS = -shared $(PD_DIR)/../bin/pd.dll
-Wl,--enable-runtime-pseudo-reloc
endif
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1579049&group_…
Bugs item #1523433, was opened at 2006-07-16 17:48
Message generated for change (Comment added) made by zmoelnig
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=478070&aid=1523433&group_…
Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: pd-extended
Group: v0.39.2
>Status: Pending
>Resolution: Fixed
Priority: 5
Submitted By: Hans-Christoph Steiner (eighthave)
Assigned to: Hans-Christoph Steiner (eighthave)
Summary: zexy's regex compiled without regex support
Initial Comment:
In Pd-0.39.2-extended-test4, instantiating [regex]
gives this message:
[regex]
part of zexy-2.1
Copyright (l) IOhannes m zmölnig, 1999-2006
forum::fÌr::umlÀute
IEM
compiled: May 25 2006
error: [regex] non-functional: compiled without
regex-support!
----------------------------------------------------------------------
>Comment By: IOhannes m zmölnig (zmoelnig)
Date: 2006-10-17 10:55
Message:
Logged In: YES
user_id=564396
since pd-extended now uses zexy's build system for building
as single externals, this should be fixed (in theory).
please confirm.
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=478070&aid=1523433&group_…
Bugs item #1563095, was opened at 2006-09-21 21:13
Message generated for change (Comment added) made by zmoelnig
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=478070&aid=1563095&group_…
Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: pd-extended
Group: v0.39.2
Status: Open
Resolution: None
Priority: 5
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Nobody/Anonymous (nobody)
Summary: bug in iemlib's for++
Initial Comment:
There is a bug with the iemlib for++ object in PD-
extended:
I discovered a workaround to load for++ for now:
1. try to make an object [for++] -> PD will say that
there is no object like for++
2. try to make an object [forpp] -> PD will say that
there is no object like forpp
3. try again to make an object [for++] -> et viola ,
there is thefor++ I was looking for
cheers,
Nils
----------------------------------------------------------------------
>Comment By: IOhannes m zmölnig (zmoelnig)
Date: 2006-10-17 09:38
Message:
Logged In: YES
user_id=564396
i have assigned this bugto you, since i thought that this
bug is related to pd-extended only (and you are the one who
takes care of this).
the bug does not appear when using iemlib as it is thought
to be used (as a library).
it only happens when it is compiled as single externals
(which is not supported upstream)
probably i'll find time to talk the author into fixing it
(or at least accept a fix)
----------------------------------------------------------------------
Comment By: Hans-Christoph Steiner (eighthave)
Date: 2006-10-13 00:57
Message:
Logged In: YES
user_id=27104
I never use this object, and its not my lib, so please stop
assigning this bug to me. Its not something I am going to fix.
The problem is that the source file name doesn't match the
class name, therefore the resulting binary doesn't load the
right class.
----------------------------------------------------------------------
Comment By: Nobody/Anonymous (nobody)
Date: 2006-09-26 08:40
Message:
Logged In: NO
Not a fix, but another workaround: I've checked in a
for++.pd abstraction, which is compatible with the
IEMLIB-external. It's in the CVS at
http://pure-data.cvs.sourceforge.net/pure-data/abstractions/purepd/
ciao
-- fbar
----------------------------------------------------------------------
Comment By: Hans-Christoph Steiner (eighthave)
Date: 2006-09-22 16:01
Message:
Logged In: YES
user_id=27104
This is most likely because the class name is "for++", but
the source filename is "forpp.c". I think + is a valid
character for the filesystem, otherwise it could be renamed
using the hex loader.
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=478070&aid=1563095&group_…
Patches item #1370034, was opened at 2005-11-30 08:25
Message generated for change (Comment added) made by eighthave
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1370034&group_…
Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: puredata-dev
Group: wishlist
>Status: Pending
Resolution: None
Priority: 5
Submitted By: Thomas Grill (xovo)
Assigned to: Nobody/Anonymous (nobody)
Summary: Matju's beautifications (untested)
Initial Comment:
Hey all,
here's Matju's changes to the devel_0_39 codebase.
There's a bug inside m_class.c, function pd_typedmess,
where the break is missing at the end of the A_DEFSYM
case, just before default.
greetings,
Thomas
----------------------------------------------------------------------
>Comment By: Hans-Christoph Steiner (eighthave)
Date: 2006-10-17 01:38
Message:
Logged In: YES
user_id=27104
this has been here for a long while with no activity, so I
am setting it to Pending. It'll be automatically set to
closed after a while if no one tends to it.
----------------------------------------------------------------------
Comment By: Thomas Grill (xovo)
Date: 2005-11-30 08:55
Message:
Logged In: YES
user_id=350252
there are also changes to extra. see the matju-extra.patch
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1370034&group_…
Patches item #1564617, was opened at 2006-09-24 13:20
Message generated for change (Comment added) made by eighthave
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1564617&group_…
Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: puredata
Group: bugfix
>Status: Closed
>Resolution: Accepted
Priority: 8
Submitted By: Hans-Christoph Steiner (eighthave)
Assigned to: Miller Puckette (millerpuckette)
Summary: remove doubleclick as rightclick on Mac OS X
Initial Comment:
>From an email from Luke Iannini:
I don't see the point of this feature; Control-Click is
quite standardized across OS X applications (and works
quite happily in Pd as well). Considering that anyone
using Pd probably has a 2 button mouse (or has learned
to control click), this is basically just a huge PITA.
The example that finally brought me here was changing
arguments to 64 instances of an abstraction in a patch
I'm working on: one click selects the entire text of an
object box, and another should place the cursor - oh,
that's unless you click too fast, in which case you get a
@#$% contextual menu. Ok, I'm going to go have a beer : ).
----------------------------------------------------------------------
>Comment By: Hans-Christoph Steiner (eighthave)
Date: 2006-10-17 01:18
Message:
Logged In: YES
user_id=27104
miller took it for 0.40.0
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1564617&group_…
Patches item #1562296, was opened at 2006-09-20 11:37
Message generated for change (Comment added) made by eighthave
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1562296&group_…
Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: puredata
Group: bugfix
>Status: Closed
>Resolution: Accepted
Priority: 9
Submitted By: Hans-Christoph Steiner (eighthave)
Assigned to: Miller Puckette (millerpuckette)
Summary: add sensible double-click actions to help browser
Initial Comment:
to open files, you need to double-click them now. You
can still browse using single-clicks.
----------------------------------------------------------------------
>Comment By: Hans-Christoph Steiner (eighthave)
Date: 2006-10-17 01:16
Message:
Logged In: YES
user_id=27104
miller took it for 0.40.0
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=478072&aid=1562296&group_…