| Summary: | xca ebuild (new) -- graphical cert. authority. | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | Michael Krelin <gentoodoo> |
| Component: | New packages | Assignee: | Crypto team [DISABLED] <crypto+disabled> |
| Status: | RESOLVED FIXED | ||
| Severity: | enhancement | Keywords: | EBUILD |
| Priority: | High | ||
| Version: | unspecified | ||
| Hardware: | All | ||
| OS: | All | ||
| URL: | http://www.hohnstaedt.de/xca.html | ||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
| Attachments: |
ebuild
patch to make it compile. ebuild as text/plain xca-0.5.1.ebuild with a whole kde eclass wrapped around addwrite :-) new proposal ... Another attempt on db autodiscovery - the patch. further improved configure-db.patch |
||
|
Description
Michael Krelin
2004-12-09 15:38:35 UTC
Created attachment 45638 [details]
ebuild
Created attachment 45639 [details, diff]
patch to make it compile.
please re-attach your ebuild as plain text. please see: http://www.gentoo.org/doc/en/ebuild-submit.xml Created attachment 45643 [details]
ebuild as text/plain
Sorry, I didn't know ebuild aren't recognized properly.
> addwrite ${QTDIR}/etc/settings <-- this is a no no.
please try inherit kde, then add "kde_src_compile nothing" to src_compile() to workaround access violation sandbox.
I can do that, but is it a good idea to inherit kde for an ebuild that doesn't have much to do with kde? It only depends on qt, not kde, I believe. I last tried xca in the 0.2* series where it wasn't very good. It's good to see that it's progressing still. 'inherit kde' is fine even if you don't use kde itself. Created attachment 45789 [details]
xca-0.5.1.ebuild with a whole kde eclass wrapped around addwrite :-)
I also didn't go for xca a while ago...
Here is an updated ebuild.
One more thing: the patch tagged as "qt" is likely related to gcc, not qt (not
sure, though, but I have sent the patch to xca author and he claims it compiles
well with the same qt version, but his gcc is 3.3), so it may make sense to
rename it.
had trouble compiling this. Will look again soon Strange, it works for me (gcc3.4). If you provide more feedback on what exactly fails, I can look into it. gcc -MM -I. -I.. -I/usr/include/db4 -I/usr/qt/3/include -g -I/usr/include/db3 /usr/include/db4 /usr/include/db4.1 /usr/include/db4.2 -DPREFIX=\"/usr\" -DETC=\"/etc/xca\" moc_CertDetail.cpp moc_ExportKey.cpp moc_KeyDetail.cpp moc_distname.cpp moc_CrlDetail.cpp moc_ExportTinyCA.cpp moc_MainWindow.cpp moc_ReqDetail.cpp moc_validity.cpp moc_ExportCert.cpp moc_ImportMulti.cpp moc_NewX509.cpp moc_CertExtend.cpp moc_clicklabel.cpp CertDetail.cpp ExportKey.cpp KeyDetail.cpp distname.cpp CrlDetail.cpp ExportTinyCA.cpp MainWindow.cpp ReqDetail.cpp validity.cpp ExportCert.cpp ImportMulti.cpp NewX509.cpp CertExtend.cpp clicklabel.cpp NewX509_ext.cpp MW_menu.cpp MW_help.cpp MW_database.cpp > .depend gcc: /usr/include/db4: linker input file unused because linking not done gcc: /usr/include/db4.1: linker input file unused because linking not done gcc: /usr/include/db4.2: linker input file unused because linking not done make[1]: Leaving directory `/var/tmp/portage/xca-0.5.1/work/xca-0.5.1/widgets' make[1]: Entering directory `/var/tmp/portage/xca-0.5.1/work/xca-0.5.1/widgets' gcc -I. -I.. -I/usr/include/db4 -I/usr/qt/3/include -g -I/usr/include/db3 /usr/include/db4 /usr/include/db4.1 /usr/include/db4.2 -DPREFIX=\"/usr\" -DETC=\"/etc/xca\" -c moc_CertDetail.cpp -o moc_CertDetail.o distcc[6904] ERROR: compile /var/tmp/ccache/moc_CertDe.tmp.frog.local.6897.ii on localhost failed gcc: /usr/include/db4: linker input file unused because linking not done gcc: /usr/include/db4.1: linker input file unused because linking not done gcc: /usr/include/db4.2: linker input file unused because linking not done gcc: /usr/include/db4: linker input file unused because linking not done gcc: /usr/include/db4.1: linker input file unused because linking not done gcc: /usr/include/db4.2: linker input file unused because linking not done cc1plus: /usr/include/db4: No such file or directory cc1plus: internal compiler error: Segmentation fault Please submit a full bug report, with preprocessed source if appropriate. See <URL:http://bugs.gentoo.org/> for instructions. make[1]: *** [moc_CertDetail.o] Error 1 make[1]: Leaving directory `/var/tmp/portage/xca-0.5.1/work/xca-0.5.1/widgets' make: *** [widgets/target.obj] Error 2 make: *** Waiting for unfinished jobs.... gcc: /usr/include/db4: linker input file unused because linking not done gcc: /usr/include/db4.1: linker input file unused because linking not done gcc: /usr/include/db4.2: linker input file unused because linking not done make[1]: Leaving directory `/var/tmp/portage/xca-0.5.1/work/xca-0.5.1/lib' make[1]: Entering directory `/var/tmp/portage/xca-0.5.1/work/xca-0.5.1/lib' gcc -I. -I.. -I/usr/include/db4 -I/usr/qt/3/include -g -I/usr/include/db3 /usr/include/db4 /usr/include/db4.1 /usr/include/db4.2 -DPREFIX=\"/usr\" -DETC=\"/etc/xca\" -c db_crl.cpp -o db_crl.o gcc -I. -I.. -I/usr/include/db4 -I/usr/qt/3/include -g -I/usr/include/db3 /usr/include/db4 /usr/include/db4.1 /usr/include/db4.2 -DPREFIX=\"/usr\" -DETC=\"/etc/xca\" -c db_key.cpp -o db_key.o gcc: /usr/include/db4: linker input file unused because linking not done gcc: /usr/include/db4.1: linker input file unused because linking not done gcc: /usr/include/db4.2: linker input file unused because linking not done cc1plus: /usr/include/db4: No such file or directory cc1plus: internal compiler error: Segmentation fault Please submit a full bug report, with preprocessed source if appropriate. See <URL:http://bugs.gentoo.org/> for instructions. distcc[6912] ERROR: compile db_crl.cpp on localhost failed make[1]: *** [db_crl.o] Error 1 make[1]: *** Waiting for unfinished jobs.... gcc: /usr/include/db4: linker input file unused because linking not done gcc: /usr/include/db4.1: linker input file unused because linking not done gcc: /usr/include/db4.2: linker input file unused because linking not done gcc: /usr/include/db4.2: No such file or directory gcc: /usr/include/db4: linker input file unused because linking not done gcc: /usr/include/db4.1: linker input file unused because linking not done cc1plus: /usr/include/db4: No such file or directory distcc[6916] ERROR: compile db_key.cpp on 10.10.10.10 failed make[1]: *** [db_key.o] Error 1 make[1]: Leaving directory `/var/tmp/portage/xca-0.5.1/work/xca-0.5.1/lib' make: *** [lib/target.obj] Error 2 The include paths bit add too may paths for poor cc1plus to handle without -I prefixes. Compiling it without that line at the moment generates: ld -r -o target.obj moc_About.o moc_CertDetail.o moc_CertExtend.o moc_CrlDetail.o moc_ExportCert.o moc_ExportKey.o moc_ExportTinyCA.o moc_Help.o moc_ImportMulti.o moc_KeyDetail.o moc_MainWindow.o moc_NewKey.o moc_NewX509.o moc_PassRead.o moc_PassWrite.o moc_ReqDetail.o moc_TrustState.o About.o CertDetail.o CertExtend.o CrlDetail.o ExportCert.o ExportKey.o ExportTinyCA.o Help.o ImportMulti.o KeyDetail.o MainWindow.o NewKey.o NewX509.o PassRead.o PassWrite.o ReqDetail.o TrustState.o make[1]: Leaving directory `/var/tmp/portage/xca-0.5.1/work/xca-0.5.1/ui' gcc -L/usr/qt/3/lib lib/target.obj widgets/target.obj view/target.obj ui/target.obj -lstdc++ -ldb_cxx -lqt -lcrypto -o xca /usr/lib/gcc/i686-pc-linux-gnu/3.4.3/../../../../i686-pc-linux-gnu/bin/ld: warning: libstdc++.so.5, needed by /usr/lib/gcc/i686-pc-linux-gnu/3.4.3/../../../libdb_cxx.so, may conflict with libstdc++.so.6 lib/target.obj(.text+0xcf11): In function `db_base::db_base(DbEnv*, QString, QString, DbTxn*, XcaListView*)': /var/tmp/portage/xca-0.5.1/work/xca-0.5.1/lib/db_base.cpp:71: undefined reference to `Db::open(char const*, char const*, DBTYPE, unsigned int, int)' lib/target.obj(.text+0xd1ab): In function `db_base::db_base(DbEnv*, QString, QString, DbTxn*, XcaListView*)': /var/tmp/portage/xca-0.5.1/work/xca-0.5.1/lib/db_base.cpp:71: undefined reference to `Db::open(char const*, char const*, DBTYPE, unsigned int, int)' collect2: ld returned 1 exit status distcc[10419] ERROR: compile (null) on localhost failed make: *** [xca] Error 1 There are functions to extract the latest db version and you may be able to incorporate this somehow. Created attachment 51982 [details]
new proposal ...
No one cares about xca? am I the only one interested in that package? Anyone who wants to test it download it manually: wget https://corpse.servebeer.com/pub/xca.tar.gz tar -xzvf xca.tar.gz -C /usr/portage/app-crypt emerge xca Please tell me if it works. THX fresh corpse gcc -I. -I.. -I/usr/include/db4.1 -I/usr/qt/3/include -pipe -DPREFIX=\"/usr\" -DETC=\"/etc/xca\" -c clicklabel.cpp -o clicklabel.o gcc -I. -I.. -I/usr/include/db4.1 -I/usr/qt/3/include -pipe -DPREFIX=\"/usr\" -DETC=\"/etc/xca\" -c moc_db_x509super.cpp -o moc_db_x509super.o clicklabel.cpp: In constructor `ClickLabel::ClickLabel(QWidget*, const char*, uint)': clicklabel.cpp:61: error: no match for call to `(QFont) ()' make[1]: *** [clicklabel.o] Error 1 make[1]: Leaving directory `/var/tmp/portage/xca-0.5.1/work/xca-0.5.1/widgets' make: *** [widgets/target.obj] Error 2 gcc -I. -I.. -I/usr/include/db4.1 -I/usr/qt/3/include -pipe -DPREFIX=\"/usr\" -DETC=\"/etc/xca\" -c moc_pki_base.cpp -o moc_pki_base.o make: *** Waiting for unfinished jobs.... ld -r -o target.obj db_crl.o db_key.o db_temp.o db_x509.o db_x509req.o db_x509super.o pki_base.o asn1int.o oid.o pki_pkcs7.o x509rev.o asn1time.opass_info.o pki_temp.o x509v3ext.o db_base.o pki_x509.o func.o pki_crl.o pki_x509req.o load_obj.o pki_key.o pki_x509super.o main.o pki_pkcs12.o x509name.o moc_db_crl.o moc_db_key.o moc_db_temp.o moc_db_x509.o moc_db_x509req.o moc_db_x509super.o moc_pki_base.o make[1]: Leaving directory `/var/tmp/portage/xca-0.5.1/work/xca-0.5.1/lib' !!! ERROR: app-crypt/xca-0.5.1 failed. !!! Function src_compile, Line 35, Exitcode 2 !!! emake failed !!! If you need support, post the topmost build error, NOT this status message. [ebuild R ] sys-libs/db-4.2.52_p2 -bootstrap -doc +java -nocxx +tcltk 0 [ebuild R ] x11-libs/qt-3.3.4-r2 +cups -debug -doc -firebird +gif -immqt -immqt-bc +ipv6 +mysql +nas +odbc -opengl +postgres -sqlite +xinerama +zlib 0 [ebuild R ] dev-libs/openssl-0.9.7e -debug -emacs (-uclibc) 0 kB I am interested. Just having compile troubles and a bit more focus on other bugs sorry. I appreciate your persistence here. Being the one who initially submitted ebuild I am interested in xca, but, naturally, I submit an ebuild that works for me and unfortunately had no time to play around with fixes... I am sorry. I haven't had time to look closely at the 'new proposal' tarball, but for what I gather from the compile error provided, this is the problem with gcc 3.4. I think that the 'new proposal' submitter may wish to incorporate the patch I provided. I have communicated it with the xca author and he said that the line in question was taken from some example and he doesn't mind changing it and even promised to incorporate patch into the next release. Meanwhile I think you can safely use the patch. Once again, sorry for not having time to look into the problems here... added app-crypt/xca. I'm not sure its 100% working. Can you please check for me? It did segfault for me. in one of the menus. Didn't work for me. The fix was simple: I had to add /db4.2 to the configure*patch. Isn't there a better way to detect where db is installed? My initial, somewhat funny idea with using /usr/include/db* which didn't work for more than one db* directory installed could be improved with something like '| tail -n 1' or 'head -n 1' (but which one?:)). But I'm sure gentoo should provide a better way of finding this out. Alternatively, the configure patch can be extended to look for db more thoroughly. (I think reopening the bug is a right thing to do, correct me if I'm wrong). Created attachment 52393 [details, diff]
Another attempt on db autodiscovery - the patch.
Here is the patch that extends the /usr/include/db*/ idea a bit further. The
patch is to be used instead of configure-db-4.1.patch (which is limited to the
use of db4.1 as the name suggests).
Created attachment 52394 [details, diff]
further improved configure-db.patch
Yep, the latest patch in comment #19 fixes the segfault during key-managing. crypto-herd, please apply that patch. Lars if it works for you please commit it. I'm running out of time a bit (per devaway). fixed see bug #94566 |
