summaryrefslogtreecommitdiff
AgeCommit message (Expand)Author
5 daysFix: Forgot to add the actual keep alive flagDirk Engling
5 daysSort info_hashes in a scrape so that the reply forms a valid bencoded dictDirk Engling
5 daysAdd new test for connection keep aliveDirk Engling
5 daysRevive http keep alive to save tcp connection setup costsDirk Engling
5 daysretrieve the address of the sending socket for comparison. In jails it might be forced to its primary addressDirk Engling
5 daysAdding the actual tests to do aforementioned 'Massively increase coverage by pulling fullscrapes and sending udp packets'Dirk Engling
5 daysMassively increase coverage by pulling fullscrapes and sending udp packetsDirk Engling
5 daysFix some sanitizer findings, mostly UB if no accesslist is configured and thread join problemsDirk Engling
5 daysadd more checkers to the sanitizer and integration check runsDirk Engling
5 daysDrop -lz as Makefile.gzip already carries itDirk Engling
5 daysAvoid an UB of adding NULL + 0 by keeping an offset instead of advancing the pointer into a vector structDirk Engling
5 daysAdd integration test and its runnerDirk Engling
5 daysreplace all calls to sprintf with a format helper that advances a cursor. Also convert the meta-formatter (i.e. results printer) to this schemeDirk Engling
5 daysPotentially overlapping areas need memmove, not memcpyDirk Engling
5 daysMake sure to not accept mere prefixes of addresses from query stringDirk Engling
5 daysWith libowfat gone, we don't need to disable the auto free feature for older versionsDirk Engling
5 daysReplace libowfat's event waiter with our own abstractionDirk Engling
5 daysAdd poll/epoll/kqueue wrapper to wait for eventsDirk Engling
5 daysThird run of making variable and function names more readable, integer scans and ip parsersDirk Engling
5 daysSecond run of making variable and function names more readable, urlencoder now matches overall naming schemeDirk Engling
5 daysFirst run of making variable names more readableDirk Engling
5 daysUse new fdstate subsystem to store per fd cookie and timeoutDirk Engling
5 daysIntroduce a sub system that can store io cookies for file descriptors, thread safely soDirk Engling
6 daysFactor out the timeout setter, will probably go later once we do the io loop ourselfDirk Engling
6 daysReplace libowfat's iobatch with our own smaller (and probably only temporary) ot_batchDirk Engling
6 daysPrevent UAF when resizing iobatchesDirk Engling
6 daysAdd mutex unit testDirk Engling
6 daysBe stricter about const char*, Wwrite-strings will warn in the future and we don't want to try to overwrite string literalsDirk Engling
6 daysFor workqueue tasks, class has to be in the tasktype or no worker will ever claim the task, forgot oneDirk Engling
6 daysFor workqueue tasks, class has to be in the tasktype or no worker will ever claim the taskDirk Engling
6 dayswhite space fixesDirk Engling
6 daysSplit options parsing and config code from opentracker.c, it has grown humongousDirk Engling
6 daysFix build. Last commit was too huge ;)Dirk Engling
6 daysDecouple config parsing from binding and worker spawningDirk Engling
6 daysClean up peer re-distribution between single bucket vectors and bucket lists, add testsDirk Engling
7 daysDrop dead test helperDirk Engling
7 daysRemove dead code left over from refactoringDirk Engling
7 daysRemove test for length only mode of the url escaperDirk Engling
7 daysRemove unused length calc code from urlencoderDirk Engling
7 daysa suffix rule takes no prerequisites, so make the deps explicitDirk Engling
7 daysmake all headers self containingDirk Engling
7 daysDon't let the new iovec allocation dangle in case of a malloc failure for its new dataDirk Engling
7 daysUse the new ot_socket abstraction for socket creation, read and writeDirk Engling
7 daysAdd sub system to abstract low level multi platform socket handling ourselvesDirk Engling
7 daysProperly handle socket_recv errorsDirk Engling
7 daysReplace custom int64 with its standardized int64_t typeDirk Engling
7 daysPrevent a task from outliving its connectionDirk Engling
7 daysFree the iovec when canceling tasks and not only the data chunks it holdsDirk Engling
7 daysfree_peerlist now checks peer_list for NULL, so on alloc failure we don't dereference NULLDirk Engling
7 daysReplace libowfat's ip scanner and formatters with own ownDirk Engling