I've been happily using the Linux beta with 32 bit Linux, but thought Wine
might be a better option for a 64 OS. So far, so good: I have 3.23 running
well and life is once again good (ish), but one minor problem is the absence
of Help. It opens fine and displays the left-hand pane, but the right-hand
pane is blank and also the icons all appear to be a standard default.
My guess is that this is a Wine issue, but I thought it worth asking here in
case there is an obvious solution?
Thanks to anyone.
Peter HB
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Wine's hh tool doesn't fully support chm format (you can see "fixme" messages
when running wine hh ./pwsafe.chm in console. There are some discussions
about it in winehq forum).
Empty right pane possibly means, that you doesn't have Gecko installed (you can see message about it
in console output of wine hh ./pwsafe.chm)
Gecko can be installed manually, or with winetricks).
After installing Gecko, text on the right pane is displayed, but there are no
pictures.
This can be solved by:
using external chm viewer (xchm, okular, ...)
installing ie6 using winetricks
Sorry for my English
PS: I've checked it under wine 1.3.7.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Many thanks for your help, which led to a solution. I upgraded wine to 1.3.8
and used winetricks to install gecko and ie8. Lots of "fixme"s, but something
in that process has done the trick and I now have text and pictures in
Password Safe help.
I'll try to be a little more scientific on another machine and refine what the
fix actually was.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I went ahead and built this from source per the instructions in the tarball. I
upgraded it on my Lucid 64bit from 0.6.0 and tested it. I will share it with
those who would like a copy unless the developer would rather I not.
No changes to the source were made. I just followed your instructions in the
README.LINUX.DEVELOPERS.txt file and built it on a vanilla 10.04.3 64bit box.
Thank you for making this available.
Brian
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thanks for the patches. I've a problem with the second one, though, as C++
guarantees that delete-ing a null vary is safe and a no-op, so I don't see how
the if statement you've added protects against a double-free. Perhaps the
memory corruption's occurring elsewhere?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I reverted the second patch and can no longer repro the segfault which is a
bit odd as I was getting it constantly on startup. Even tried deleting my
.pwsafe directory. Well, nevermind then.
Cheers,
Micha.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
No, you can't cast a pointer to an integer shorter than a pointer, even though
you can cast or assign integers to shorter integers (C++ is weird). This is
why I used uintptr_t as that is always guaranteed to be the same size as a
pointer. But you can assign the result of the cast directly to the short :
What was wrong with reinterpret_casting to an intptr_t and then assigning to
retval? That keeps it C++ and type-safe and is also the smallest change from
the existing code.
Anyway, I don't care as long as it compiles and runs :)
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Please forgive the noob question in advance, but is it expected to have the
amd64 deb package request an additional 240 packages (all of which are i386)?
I'm sure they are dependencies, but some packages it is requesting seem
strange. Like "bluez-alsa:i386" and "libgstreamer-plugins:i386".
Also, this happens whether I compile 0.7.0BETA or 0.8.0BETA and on two
different machines.
Thanks
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I've been happily using the Linux beta with 32 bit Linux, but thought Wine
might be a better option for a 64 OS. So far, so good: I have 3.23 running
well and life is once again good (ish), but one minor problem is the absence
of Help. It opens fine and displays the left-hand pane, but the right-hand
pane is blank and also the icons all appear to be a standard default.
My guess is that this is a Wine issue, but I thought it worth asking here in
case there is an obvious solution?
Thanks to anyone.
Peter HB
Wine's hh tool doesn't fully support chm format (you can see "fixme" messages
when running
wine hh ./pwsafe.chm
in console. There are some discussionsabout it in winehq forum).
Empty right pane possibly means, that you doesn't have
Gecko installed (you can see message about it
in console output of
wine hh ./pwsafe.chm
)Gecko can be installed manually, or with
winetricks).
After installing Gecko, text on the right pane is displayed, but there are no
pictures.
This can be solved by:
Sorry for my English
PS: I've checked it under wine 1.3.7.
Many thanks for your help, which led to a solution. I upgraded wine to 1.3.8
and used winetricks to install gecko and ie8. Lots of "fixme"s, but something
in that process has done the trick and I now have text and pictures in
Password Safe help.
I'll try to be a little more scientific on another machine and refine what the
fix actually was.
Same here I'm running PWS v.3.26 in wine ubuntu 11.04 x64. It works ok, but
would love a native 64 bit compile.
Just a question... why not go ahead and compile it for yourself?
You might actually have fun learning a new skill :~)
https://sourceforge.net/projects/passwordsafe/files/Linux-
BETA/0.6.0/
Please find on the above link both 32 and 64 bit debs.
to https://www.google.com/accounts :-)
What about version 0.7 for ubuntu x64?
Hi Scad18,
I did write to developer about this. He said he would update very soon. I
guess he is busy with other things.
I went ahead and built this from source per the instructions in the tarball. I
upgraded it on my Lucid 64bit from 0.6.0 and tested it. I will share it with
those who would like a copy unless the developer would rather I not.
http://dl.dropbox.com/u/39638663/ubuntu/passwordsafe-
ubuntu-0.7.0BETA.amd64.deb
Thanks for sharing your build.
Did you make any changes to the source to get it to compile? If so, please
send me the patch, so that I can add the changes to the repository.
Rony
Hi Rony,
No changes to the source were made. I just followed your instructions in the
README.LINUX.DEVELOPERS.txt file and built it on a vanilla 10.04.3 64bit box.
Thank you for making this available.
Brian
Just tried compiling trunk (svn4885) under 64-bit Debian and had to make the
two following changes.
$ svn diff src/ui/wxWidgets/addeditpropsheet.cpp
$ svn diff src/core/PWSfile.cpp
Thanks for the patches. I've a problem with the second one, though, as C++
guarantees that delete-ing a null vary is safe and a no-op, so I don't see how
the if statement you've added protects against a double-free. Perhaps the
memory corruption's occurring elsewhere?
Hi Ronys,
Wow, I'm rustier than I thought with my C++.
I reverted the second patch and can no longer repro the segfault which is a
bit odd as I was getting it constantly on startup. Even tried deleting my
.pwsafe directory. Well, nevermind then.
Cheers,
Hi,
With your first patch, would this one-line change be good enough? (i.e.
changing <int> to <short>)
From:
to:
David
No, you can't cast a pointer to an integer shorter than a pointer, even though
you can cast or assign integers to shorter integers (C++ is weird). This is
why I used uintptr_t as that is always guaranteed to be the same size as a
pointer. But you can assign the result of the cast directly to the short :
Since you're controlling what goes into the combo-box, you're sure to only
ever get shorts out, otherwise you might also want to check for overflows.
Ie,
How about the old-fashioned way (and yes - we know the value is only a short -
in fact way less than 16 (decimal) in the current version?
Sorry - formatting issue:
David
What was wrong with reinterpret_casting to an intptr_t and then assigning to
retval? That keeps it C++ and type-safe and is also the smallest change from
the existing code.
Anyway, I don't care as long as it compiles and runs :)
Simplest version of patch that compiles cleanly & runs under 64bit has been
committed (4886).
Thanks,
Rony
Cool, thanks Rony :)
I successfully built passwordsafe-debian-0.7.0BETA.amd64.deb on debian 6.05.
Seemed to require no modifications. works great. thanks.
Please forgive the noob question in advance, but is it expected to have the
amd64 deb package request an additional 240 packages (all of which are i386)?
I'm sure they are dependencies, but some packages it is requesting seem
strange. Like "bluez-alsa:i386" and "libgstreamer-plugins:i386".
Also, this happens whether I compile 0.7.0BETA or 0.8.0BETA and on two
different machines.
Thanks
For those who want to compile on LinuxMint, here are the two items I changed
to get past the compile errors:
In the Makefile inside ./pwsafe-0.8BETA/install/deb change line 7 from:
to
and
Regarding my noob question - I may have answered this myself and am a bit
embarrassed.