Bug 670915
Summary: | Review Request: aprsg - Amateur Radio APRS Gateway | ||
---|---|---|---|
Product: | [Fedora] Fedora | Reporter: | Andrew Elwell <andrew.elwell> |
Component: | Package Review | Assignee: | Nobody's working on this, feel free to take it <nobody> |
Status: | CLOSED NOTABUG | QA Contact: | Fedora Extras Quality Assurance <extras-qa> |
Severity: | medium | Docs Contact: | |
Priority: | low | ||
Version: | rawhide | CC: | fedora-package-review, gwync, misc, notting, randyn3lrx, steve.traylen |
Target Milestone: | --- | ||
Target Release: | --- | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Fixed In Version: | Doc Type: | Bug Fix | |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | 2013-04-10 18:07:34 UTC | Type: | --- |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: | |||
Bug Depends On: | 669010 | ||
Bug Blocks: | 201449 |
Description
Andrew Elwell
2011-01-19 16:45:24 UTC
Please note, I am still seeking a sponsor. Bug #633102 has links to some of my other submissions and informal reviews. Some first issues: Assuming you want to provide for no older releases than EPEL6 (according to libfap), you may drop the BuildRoot definition, the %clean section, "rm -rf $RPM_BUILD_ROOT" from %install and %defattr from %files. See http://fedoraproject.org/wiki/EPEL/GuidelinesAndPolicies#EL6. For EPEL 6. there are currently no differences to the common Fedora packaging guidelines. Please remove the gz extension from the manpage in %files and let the build system choose the compression format. Well, it is gz now anyway, but could be changed in the future. Koji scratch build: http://koji.fedoraproject.org/koji/taskinfo?taskID=3307866 The build fails for Rawhide. See the build.log: http://koji.fedoraproject.org/koji/getfile?taskID=3307867&name=build.log aprspacket.cpp: In constructor 'AprsPacket::AprsPacket(const string&, const string&, const std::vector<std::basic_string<char> >&, const string&, const bool&)': aprspacket.cpp:131:55: error: too few arguments to function 'void fap_explain_error(fap_error_code_t, char*)' /usr/include/fap.h:381:6: note: declared here aprspacket.cpp: In copy constructor 'AprsPacket::AprsPacket(const AprsPacket&, const std::vector<std::basic_string<char> >&)': aprspacket.cpp:184:57: error: too few arguments to function 'void fap_explain_error(fap_error_code_t, char*)' /usr/include/fap.h:381:6: note: declared here aprspacket.cpp: In constructor 'AprsPacket::AprsPacket(fap_packet_t*, const bool&)': aprspacket.cpp:506:54: error: too few arguments to function 'void fap_explain_error(fap_error_code_t, char*)' /usr/include/fap.h:381:6: note: declared here Don't know how to interpret this. The libfap-devel dependency is present in Rawhide (v.1.1.1). Any news about this? (In reply to comment #2) > Some first issues: > > Assuming you want to provide for no older releases than EPEL6 (according to > libfap), you may drop the BuildRoot definition, the %clean section, "rm -rf > $RPM_BUILD_ROOT" from %install and %defattr from %files. See > http://fedoraproject.org/wiki/EPEL/GuidelinesAndPolicies#EL6. For EPEL 6. there > are currently no differences to the common Fedora packaging guidelines. > > Please remove the gz extension from the manpage in %files and let the build > system choose the compression format. Well, it is gz now anyway, but could be > changed in the future. > > Koji scratch build: > http://koji.fedoraproject.org/koji/taskinfo?taskID=3307866 > > The build fails for Rawhide. See the build.log: > http://koji.fedoraproject.org/koji/getfile?taskID=3307867&name=build.log > > aprspacket.cpp: In constructor 'AprsPacket::AprsPacket(const string&, const > string&, const std::vector<std::basic_string<char> >&, const string&, const > bool&)': > aprspacket.cpp:131:55: error: too few arguments to function 'void > fap_explain_error(fap_error_code_t, char*)' > /usr/include/fap.h:381:6: note: declared here > aprspacket.cpp: In copy constructor 'AprsPacket::AprsPacket(const AprsPacket&, > const std::vector<std::basic_string<char> >&)': > aprspacket.cpp:184:57: error: too few arguments to function 'void > fap_explain_error(fap_error_code_t, char*)' > /usr/include/fap.h:381:6: note: declared here > aprspacket.cpp: In constructor 'AprsPacket::AprsPacket(fap_packet_t*, const > bool&)': > aprspacket.cpp:506:54: error: too few arguments to function 'void > fap_explain_error(fap_error_code_t, char*)' > /usr/include/fap.h:381:6: note: declared here > > Don't know how to interpret this. The libfap-devel dependency is present in > Rawhide (v.1.1.1). Same errors in my attempts to build. For F14,15 and rawhide. OK, upstream has released 1.4 so I'm having another attempt to repackage. still failing to build under rawhide (but is OK in F16 and older) g++ -DHAVE_CONFIG_H -I. -I.. -Wall -DSYSCONFDIR=\""/etc"\" -I/usr/lib64/wx/include/gtk2-unicode-release-2.8 -I/usr/includ e/wx-2.8 -D_FILE_OFFSET_BITS=64 -D_LARGE_FILES -D__WXGTK__ -DwxUSE_GUI=0 -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexc eptions -fstack-protector --param=ssp-buffer-size=4 -m64 -mtune=generic -c -o serialport.o `test -f 'linux/serialport.cp p' || echo './'`linux/serialport.cpp g++ -DHAVE_CONFIG_H -I. -I.. -Wall -DSYSCONFDIR=\""/etc"\" -I/usr/lib64/wx/include/gtk2-unicode-release-2.8 -I/usr/includ e/wx-2.8 -D_FILE_OFFSET_BITS=64 -D_LARGE_FILES -D__WXGTK__ -DwxUSE_GUI=0 -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexc eptions -fstack-protector --param=ssp-buffer-size=4 -m64 -mtune=generic -c -o connection.o connection.cpp linux/serialport.cpp: In constructor 'Serialport::Serialport(const string&, const string&, const unsigned int&, const unsi gned int&, const unsigned int&, const char&, const unsigned int&, const bool&)': linux/serialport.cpp:68:19: error: 'getpid' was not declared in this scope linux/serialport.cpp: In destructor 'Serialport::~Serialport()': linux/serialport.cpp:240:20: error: 'close' was not declared in this scope linux/serialport.cpp: In member function 'bool Serialport::transmit(const string&)': linux/serialport.cpp:255:54: error: 'write' was not declared in this scope linux/serialport.cpp: In member function 'bool Serialport::receive(std::string&, unsigned int)': linux/serialport.cpp:286:39: error: 'read' was not declared in this scope make[2]: *** [serialport.o] Error 1 ... ongoing OK - with the gcc-4.7 cleanup I had to explicitly #include <unistd.h> in serialport.cpp -- patch submitted upstream to author, together with discussions on logrotate etc (I have a logrotate.d/aprsg ready to go, but haven't included it in this as it doesn't log by default). Files available at http://dl.dropbox.com/u/6594808/Fedora/aprsg-1.4-1.fc16.src.rpm http://dl.dropbox.com/u/6594808/Fedora/aprsg.logrotate http://dl.dropbox.com/u/6594808/Fedora/aprsg.spec RPMlint output: [aelwell@pcitgtelwell aprsg]$ rpmlint aprsg.spec 0 packages and 1 specfiles checked; 0 errors, 0 warnings. [aelwell@pcitgtelwell aprsg]$ rpmlint ~/rpmbuild/SRPMS/aprsg-1.4-1.fc16.src.rpm aprsg.src: W: spelling-error %description -l en_US soundmodem -> sound modem, sound-modem, soundness 1 packages and 0 specfiles checked; 0 errors, 1 warnings. (False positive - soundmodem is the name of the package) On a built binary (mock, F16-x86_64) I also get wrong-file-end-of-line-encoding for COPYING -- Text is identical to the GPL -- Should I fix with dos2unix or get upstream to change? A few comment : - Requires: libfap seems to be wrong, is there a reason, is it not auto detected ? - there is no startup script, and I think one is needed ( either sysv init, or systemd, sysv for older fedora and EPEL ). Agreed, BuildRequiring libfap-devel should do it. And since this is a new package, the init script needs to by systemd, not sysvinit. Thanks for feedback - will throw out an update in the next 48h or so. Hi Andrew, no update ? Hi - Got some time to work on this (I know, I know) -- however I'm also likely to put aprx as a new package -- seems a bit simpler. Any news here? No response in ages; I'm going to go ahead and close this. |