[Bug 185262] Review Request: kchmviewer - CHM viewer
by Red Hat Bugzilla
Please do not reply directly to this email. All additional
comments should be made in the comments box of this bug report.
Summary: Review Request: kchmviewer - CHM viewer
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=185262
------- Additional Comments From rdieter(a)math.unl.edu 2006-05-01 07:44 EST -------
Oops, better yet, change:
desktop-file-install --vendor fedora \
--dir ${RPM_BUILD_ROOT}%{_datadir}/applications \
--add-category X-Fedora \
kio-msits/kchmviewer.desktop
to
desktop-file-install --vendor fedora \
--dir ${RPM_BUILD_ROOT}%{_datadir}/applications \
--add-category X-Fedora \
--delete-original \
$RPM_BUILD_ROOT%{_datadir}/applnk/*/kchmviewer.desktop
--
Configure bugmail: https://bugzilla.redhat.com/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.
18 years
[Bug 188369] Review Request: ctapi-cyberjack
by Red Hat Bugzilla
Please do not reply directly to this email. All additional
comments should be made in the comments box of this bug report.
Summary: Review Request: ctapi-cyberjack
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=188369
------- Additional Comments From j.w.r.degoede(a)hhs.nl 2006-05-01 07:41 EST -------
I see,
In that case the libctapi-xx.so shouldn't be versioned again, but then we have
the problem of the ifd handler linking to it, or we could once again make the
ifd-handler staticly linked to the ctapi part, but that would mean having
essentially the same code on your HD twice. Still I believe this is the best,
which effectivly means that you can undo all the makefile changes as upstream's
makefile then does seem todo the right thing after all, sorry about all this.
This might actually all workout nicely since with the ifd handler staitcly
linked it will not auto require and has no reason to manually require the main
package allowing for it to be installed seperatly.
Did you already drop the versioning in your next try (and link the ifd handler
staticly against ctapi-xxx?) ifso let me know and I'll do a full review as /
when time permits. If not, please post yet another version.
Also I can acutally read (and speak a bit) German, but your German is no doubt
better, could you copy the relevants parts avout the libctapi-xxxx.so being the
standard convention from the standard to a txt file and attach that?
Then I can review the standard and file a bug against openct. I personally
believe that since these .so files are unversioned and dlopened they should be
put under /usr/lib/ctapi/ instead of in plain /usr/lib does the standard say
anything about this?
Sorry about all this ping - pong you've picked a hard one to start with and I
want to get this right, currently the towitoko drivers aren't packaged and if
we're going to set a precedent on how to handle this I would like to set a good
precedent.
--
Configure bugmail: https://bugzilla.redhat.com/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.
18 years
[Bug 188369] Review Request: ctapi-cyberjack
by Red Hat Bugzilla
Please do not reply directly to this email. All additional
comments should be made in the comments box of this bug report.
Summary: Review Request: ctapi-cyberjack
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=188369
------- Additional Comments From frank-buettner(a)gmx.net 2006-05-01 06:53 EST -------
The CT-API system is complete diferent from PC/SC
When an app will use an reader via CT-API then it load's first the 3 function's
CT_init CT_data and CT_close via dl_load.
So for every reader you need an extra lib.
On Windows this files called ctXX32.dll wehre XX is the name of the reader.
So on Linux the files must be called libctapi-XX.so where XX is the name of the
reader. And as an result of this the package for the towitoko reader must
changed, because the wrong name. The complete description of the CT-API system
is done in german because it was primaly develop for the german helth system.
But niw in germany nearliy all application that need access to SmartCards use
this system, because it is mutch simpler than PC/SC. There exits also an english
description of the CT-API but can not tell if all correct in this description.
When you will have an look to this vist http://www.teletrust.de
here the next try
http://prdownloads.sourceforge.net/qsmartcard/ctapi-cyberjack-2.0.8-7.src...
--
Configure bugmail: https://bugzilla.redhat.com/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.
18 years
[Bug 188369] Review Request: ctapi-cyberjack
by Red Hat Bugzilla
Please do not reply directly to this email. All additional
comments should be made in the comments box of this bug report.
Summary: Review Request: ctapi-cyberjack
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=188369
------- Additional Comments From j.w.r.degoede(a)hhs.nl 2006-05-01 06:38 EST -------
(In reply to comment #33)
> The lib for the CT-API dos not put in the ld cache, because all apps the use
> CT-API call must use the lib via dl_open().
Hmm, is this written / documented somewhere? Since there are multiple possible
providers for the ctapi (libopenctapi.so , libtowitoko.so,
libctapi-cyberjack.so) this doesn't sound completly unlogical, but I wonder if
its documented somewhere. Also how is a user supposed to find out what ctapi
providers there are with all these non consistent names?
--
Configure bugmail: https://bugzilla.redhat.com/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.
18 years
[Bug 188369] Review Request: ctapi-cyberjack
by Red Hat Bugzilla
Please do not reply directly to this email. All additional
comments should be made in the comments box of this bug report.
Summary: Review Request: ctapi-cyberjack
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=188369
------- Additional Comments From frank-buettner(a)gmx.net 2006-05-01 06:16 EST -------
I have done some little error.
The lib for the CT-API dos not put in the ld cache, because all apps the use
CT-API call must use the lib via dl_open().
The the error about ctapi-cyberjack library-without-ldconfig-postin is in this
case not an error, because of the structure of the CT-API system.
--
Configure bugmail: https://bugzilla.redhat.com/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.
18 years
[Bug 188369] Review Request: ctapi-cyberjack
by Red Hat Bugzilla
Please do not reply directly to this email. All additional
comments should be made in the comments box of this bug report.
Summary: Review Request: ctapi-cyberjack
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=188369
------- Additional Comments From j.w.r.degoede(a)hhs.nl 2006-05-01 06:01 EST -------
Great that you've almost all rpmlint errors and warnings fixed. About the last one:
Hmm,
With normal dyn linked libs the lniker will search for the -soname so xxx.so.0
and the plain .so symlink is only used during compile to find the real .so.x
file. What apps are using dlopen, the included ones? Couldn't you fix them to
use dlopen .so.0 ?
The .so symlink really belongs in the -devel package. We could make an exception
but thats rather ugly. For example there apps that ldopen libGL.so, but
libGL.so still is in the -devel package and the apps are supposed to open
libGL.so.1 The whole idea behind versioned .so is that an open will open the
versioned lib so that if an ABI change happens ther app will keep working as
long as the old version is not uninstalled.
--
Configure bugmail: https://bugzilla.redhat.com/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.
18 years
[Bug 188369] Review Request: ctapi-cyberjack
by Red Hat Bugzilla
Please do not reply directly to this email. All additional
comments should be made in the comments box of this bug report.
Summary: Review Request: ctapi-cyberjack
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=188369
------- Additional Comments From frank-buettner(a)gmx.net 2006-05-01 05:00 EST -------
so all is fixed. only the problem with the link to the lib exists:(
W: ctapi-cyberjack devel-file-in-non-devel-package /usr/lib64/libctapi-cyberjack.so
I have try to remove the file, but then the apps that use the lib will not run:(
Because they call libctapi-cyberjack.so via dl_load()
--
Configure bugmail: https://bugzilla.redhat.com/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.
18 years