AuthorCommitMessageCommit dateIssues
MarkMark
59ae8013bc6Translated using Weblate (Polish)Currently translated at 5.8% (8 of 137 strings)
MarkMark
2ce91ea8d73Translated using Weblate (Polish)Currently translated at 5.8% (8 of 137 strings)
MarkMark
27374499684Translated using Weblate (Polish)Currently translated at 6.5% (9 of 137 strings)
MarkMark
3ffecce3895Translated using Weblate (Polish)Currently translated at 6.5% (9 of 137 strings)
MarkMark
09fdccba024Translated using Weblate (Polish)Currently translated at 6.5% (9 of 137 strings)
MarkMark
b426e7a4555Translated using Weblate (Polish)Currently translated at 6.5% (9 of 137 strings)
MarkMark
e247a4e07a6Translated using Weblate (Polish)Currently translated at 6.5% (9 of 137 strings)
LubaLuba
1f33cc93a3aTranslated using Weblate (Czech)Currently translated at 100.0% (137 of 137 strings)
rootroot
a6a366cf8a5commit previous translations
Felix BartelsFelix Bartels
8be458c6a9fTranslated using Weblate (Chinese (Taiwan))Currently translated at 98.5% (134 of 136 strings)
Joost HopmansJoost Hopmans
515f6ef2cbeMMerge pull request #3070 in KC/kopanocore from ~JVANDERWAA/kopanocore:pkyo_tests_round1 to master* commit '755c9ceb11963c5db1140195f7bce2b2450e3824': test: add remainder of the Python-kopano testset test: introduce make test-short target
Jelle van der WaaJelle van der Waa
755c9ceb119test: add remainder of the Python-kopano testset
Jelle van der WaaJelle van der Waa
0dfc4bbad58test: introduce make test-short targetCreate a new test-short target which requires no kopano-server or other daemons and can be run on OBS, Jenkins and finish in under a 30 seconds. The python-mapi tests have been split up in server and unit tests to make it easier to run the different test sets. References: KC-1736KC-1736
Jan EngelhardtJan Engelhardt
08ca3842442MMerge branch 'kc-9.x'
Jelle van der WaaJelle van der Waa
18930def624pyko: server: handle https port not being 237The server code is very naive in that it hardcodes the only valid https port to be 237 which can be freely configured by the system administrator. Rely on urllib to parse the server_socket and obtain the valid https port when found otherwise fall back to 237. The code lacks raising an exception when the serverport is not 237 but a file:// socket.
Jelle van der WaaJelle van der Waa
8f286ea3edetest: add more python-kopano tests
Joost HopmansJoost Hopmans
646ec2fe594MMerge pull request #3062 in KC/kopanocore from ~JVANDERWAA/kopanocore:pyko_tests to master* commit 'a39f3c8594b30a76f083747fda9fd112e3869c78': test: add more pyko tests and split up test sets
Jelle van der WaaJelle van der Waa
a39f3c8594btest: add more pyko tests and split up test setsSplit up the current python-kopano tests into "unit" and "server" tests with the make test-short target for unit tests and make test for server required tests. In addition add some server tests which require a kopano-server.
Joost HopmansJoost Hopmans
5068ec4c516MMerge pull request #3063 in KC/kopanocore from ~JVANDERWAA/kopanocore:search_skip_filtered_body to kc-9.x* commit '91b19adacb95bada902d2f4f7c3bc5443ca4723f': search: do not index PR_EC_FILTERED_BODY
Jan EngelhardtJan Engelhardt
063bdfe5982build: work around "define unix"When -std=gnu++XX is in effect, "unix" is defined. In -std=c++XX, only __unix and __unix__ are, as it should be. Rename the variable in short order. References: KC-1728KC-1728
Jan EngelhardtJan Engelhardt
1d63a616af9stats: show floating point numbers with fractionsFixes SVN r34041.
Jan EngelhardtJan Engelhardt
6642d93ac4bserver: avoid losing requests with <1ms processing times in time tracking
Jan EngelhardtJan Engelhardt
cf471c1ba64server: count TLS setup towards processing/response time
Jan EngelhardtJan Engelhardt
e8678713276Revert "server: avoid some timepoint computations when request_log is not active"This reverts commit 94dbe196b5bcaada825d4e4188cf2b1a52aea8c9. StatsClient is still active is even if the request_log is not. RemoveBusyState also depends on wi_cpu.
Jelle van der WaaJelle van der Waa
91b19adacb9search: do not index PR_EC_FILTERED_BODYDo not index the PR_EC_FILTERED_BODY body property as it's just a duplicate property of PR_HTML but with sanitised HTML.
Joost HopmansJoost Hopmans
2fde1fea97fMMerge pull request #3060 in KC/kopanocore from ~JVANDERWAA/kopanocore:gitignore_updates to master* commit 'd16a3154fe3d62a98e78b5019375770d8961dd0b': test: update gitignore for new test programs
Jelle van der WaaJelle van der Waa
d0532df6fc8doc: update python cli argument man pagesIn commit c014b94fd6af5a4 the arguments where cnaged from -s to -S, -c to -C, -k to -K and -p to -Q. The man pages where not updated to reflect these changes.
Jelle van der WaaJelle van der Waa
adc7b2183ccdoc: update config_file option for kopano-searchThe configuration file argument changed from -c to -C.
Jelle van der WaaJelle van der Waa
d16a3154fe3test: update gitignore for new test programsWith the addition of `make check` test programs in e48558ed3137 and 80c4d6a2b3 the gitignore file was not updated.
Joost HopmansJoost Hopmans
8b35d04cb9bMMerge pull request #3050 in KC/kopanocore from ~JHOPMANS/kopanocore:docker-unit-test to master* commit 'b127731d4a1126dcacf468c13ddbcb8dc323f579': (21 commits) test: disable test which segfaults on Jenkins test: update gitignore for test junit files php-ext: make unit tests only run when not server available test: add markers to skip server tests test: add php-ext report to Jenkinsfile swig: add more MAPI unit tests test: add README to describe test suite runner test: sk...
Jan EngelhardtJan Engelhardt
3184bb2dbe7Kopano Groupware Core 10.0.3
Jelle van der WaaJelle van der Waa
b127731d4a1test: disable test which segfaults on Jenkinsmapi_createconversationindex("kopano"); segfaults on our Jenkins ubuntu 18.04 server.
Jelle van der WaaJelle van der Waa
8eb604583fetest: update gitignore for test junit files
Jan EngelhardtJan Engelhardt
9fa43379852MMerge branch 'kc-9.x'
Jelle van der WaaJelle van der Waa
3552b1cd362php-ext: make unit tests only run when not server available
Jelle van der WaaJelle van der Waa
1f45799c722test: add markers to skip server tests
Jelle van der WaaJelle van der Waa
c830dff25ectest: add php-ext report to Jenkinsfile
Jan EngelhardtJan Engelhardt
e28f443b3d7Kopano Groupware Core 9.0.5
Jan EngelhardtJan Engelhardt
a277fc270f2MMerge branch 'kc-8.7.x'
Jan EngelhardtJan Engelhardt
31784beb83alibserver: remove unused variableFixes: kopanocore-8.7.9-188-g20423a222
Joost HopmansJoost Hopmans
ac3ef14fd84MMerge pull request #3056 in KC/kopanocore from ~JENGELHARDT/kc:ep to kc-9.x* commit 'e113d393e69c5cf5727ed2ffd37804f8c49e7b6f': server: use EPOLLONESHOT flag
Jan EngelhardtJan Engelhardt
e113d393e69server: use EPOLLONESHOT flagSave on some syscalls and do away with the unnecessary CTL_DEL/CTL_ADD cycle between every workitem.
Jan EngelhardtJan Engelhardt
3e27f90e8aastats: relabel RTT to RTThe label is misleading, because it does not indicate the roundtrip time, but only the response time. This is very evident in source code: ECThreadManager.cpp: SCN_RESPONSE_TIME stats.cpp: mapDiffStat["response_time"].
Jan EngelhardtJan Engelhardt
704341327a5server: move SCN_{PROCESSING,REQUEST}_TIME back to millisecondsFixes: kopanocore-9.0.4-14-g02b302d57
Jan EngelhardtJan Engelhardt
5afd6b43e0cserver: avoid emitting request_log message for EOFWhen the remote side closes the writing end of a connection, poll/epoll will notify k-server with POLLIN|POLLRDHUP. Processing is deferred to a workitem as usual, because there could still be data to process before EOF. Inside WORKITEM::run, it then turns out that EOF is right upon us. soap_begin_recv returns with soap->error=SOAP_EOF set, at which point no log message should be emitted.
Jan EngelhardtJan Engelhardt
94dbe196b5bserver: avoid some timepoint computations when request_log is not active
Jan EngelhardtJan Engelhardt
cb6368dfee5server: print Tsk= and T*_CPU= with same precision
Jan EngelhardtJan Engelhardt
8168637cadflibserver: add rh1 time computation to logon/ssoLogon/logoffWas present in SOAP_ENTRY_START/END, but missing from the three manually implemented handlers.
Jan EngelhardtJan Engelhardt
dbd92b51b50server: emit tls_setup workitem into request logThe TLS handshake is its own workitem, which means there can be additional delay between a completed handshake and the first actual HTTP request. Print information about that workitem too.
Jan EngelhardtJan Engelhardt
b8c3ac9dac1doc: update text for request_log message format