{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":116033455,"defaultBranch":"master","name":"cyclonedds","ownerLogin":"eclipse-cyclonedds","currentUserCanPush":false,"isFork":false,"isEmpty":false,"createdAt":"2018-01-02T16:13:04.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/45015330?v=4","public":true,"private":false,"isOrgOwned":true},"refInfo":{"name":"","listCacheKey":"v0:1693214593.0","currentOid":""},"activityList":{"items":[{"before":"7c253ad3c4461b10dc4cac36a257b097802cd043","after":"460851ace3417b63eee5b093471be3a3461220f6","ref":"refs/heads/master","pushedAt":"2024-05-01T05:12:39.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Exported the ddsrt_getifaddrs and ddsrt_freeifaddrs symbols which are required by the Iceoryx PSMX plugin.","shortMessageHtmlLink":"Exported the ddsrt_getifaddrs and ddsrt_freeifaddrs symbols which are…"}},{"before":"d52f5440826a4321b654e0bd647b2cc79e07e39d","after":"7c253ad3c4461b10dc4cac36a257b097802cd043","ref":"refs/heads/master","pushedAt":"2024-04-29T12:23:24.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Fix building for Zephyr V3.6, QNX and other platforms that don't support single-source multicast and/or dynamic library loading\n\nSigned-off-by: Patrick Masselink ","shortMessageHtmlLink":"Fix building for Zephyr V3.6, QNX and other platforms that don't supp…"}},{"before":"6b998f5e5e506f720cace134428b3897703ed369","after":"843ae6d099f70bdeb7c5dd3f4f2cc3897d9614d2","ref":"refs/heads/releases/0.10.x","pushedAt":"2024-04-26T10:17:34.000Z","pushType":"pr_merge","commitsCount":7,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Fix handling of empty second copy of PropertySeq\n\nThis changes the handling of PropertySeq and BinaryPropertySeq to always return the latest\none in the message. Without this change a second (or later) empty sequence would be\nignored, but a second (or later) non-empty sequence would be returned. The memory is\ninitialised to an empty sequence on allocation, so it doesn't affect the behaviour for the\nfirst copy. This matters insofar as being consistent with some other deserialization\nfunctions matters.\n\nSigned-off-by: Erik Boasson ","shortMessageHtmlLink":"Fix handling of empty second copy of PropertySeq"}},{"before":"f2ec5247bd7a56e70f07c9b74950f3dbb159fd6d","after":"6b998f5e5e506f720cace134428b3897703ed369","ref":"refs/heads/releases/0.10.x","pushedAt":"2024-04-26T09:43:14.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Removed clang warnings\n\nSigned-off-by: Alejandro Hernández Cordero ","shortMessageHtmlLink":"Removed clang warnings"}},{"before":"4d5b6f325fc15e88987bcd60e335c5accfb749c2","after":"d52f5440826a4321b654e0bd647b2cc79e07e39d","ref":"refs/heads/master","pushedAt":"2024-04-25T11:32:29.000Z","pushType":"pr_merge","commitsCount":3,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Use hash of network config as Iceoryx locator\n\nFor Iceoryx (or PSMX more generally) we need a unique identifier for Iceoryx instance.\nIdeally, Iceoryx would provide that, but it doesn't ...\n\nThe plug-in used to try to come up with an identifier by fiddling with service discovery,\nbut there is a race condition in that logic that can't be solved in the absence of an\nupper limit to the number of processes while not having any form of synchronization or\ntime bounds.\n\nFortunately, Iceoryx refuses to have multiple instances of itself on a single machine, and\ntherefore something that can uniquely identify a machine can also be used. I haven't been\nable to find a good way to get a machine id from the kernel. That leaves writing stuff to\nthe filesystem (which I don't like doing), allocating and retaining some form of shared\nmemory to store it (also fraught with complications), or using some proxy for a proper\nmachine id. The network configuration provides a reasonable option, as the MAC addresses\nare supposed to be unique.\n\nThen we get into trouble again: the network configuration can change, our getifaddrs()\ndoesn't return the MAC address on Windows, etc. The code here is therefore flawed, but\npresumed good enough as replacement for the previous code that definitely did not work.\n\nSigned-off-by: Erik Boasson ","shortMessageHtmlLink":"Use hash of network config as Iceoryx locator"}},{"before":"b17c425a0c0f1fa0c12cb82b899094c7075277d5","after":"4d5b6f325fc15e88987bcd60e335c5accfb749c2","ref":"refs/heads/master","pushedAt":"2024-04-25T07:48:17.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Fix leading underscore handling in IDLC\n\nIDL identifiers start with a letter, and a (single) leading underscore is allowed as an\nescape character to distinguish them keywords. That is, `Gloeiworm` and `_Gloeiworm` are\nthe same identifier. This it implemented correctly.\n\nHowever, it follows that where an identifier is expect, a single underscore is forbidden,\nand so is anything with a leading double underscore. This detects those cases and raises\nan error.\n\nSigned-off-by: Erik Boasson ","shortMessageHtmlLink":"Fix leading underscore handling in IDLC"}},{"before":"2ad37f4cc3d545149f680d6a8b0ff31e0ea84edc","after":"b17c425a0c0f1fa0c12cb82b899094c7075277d5","ref":"refs/heads/master","pushedAt":"2024-04-22T11:34:14.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Initialize the ret variable to fix compiler warnings.","shortMessageHtmlLink":"Initialize the ret variable to fix compiler warnings."}},{"before":"31a4843667830be87a38bd8bb22268061f79b42b","after":"2ad37f4cc3d545149f680d6a8b0ff31e0ea84edc","ref":"refs/heads/master","pushedAt":"2024-04-16T07:56:16.000Z","pushType":"pr_merge","commitsCount":3,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Fix handling of empty second copy of PropertySeq\n\nThis changes the handling of PropertySeq and BinaryPropertySeq to always return the latest\none in the message. Without this change a second (or later) empty sequence would be\nignored, but a second (or later) non-empty sequence would be returned. The memory is\ninitialised to an empty sequence on allocation, so it doesn't affect the behaviour for the\nfirst copy. This matters insofar as being consistent with some other deserialization\nfunctions matters.\n\nSigned-off-by: Erik Boasson ","shortMessageHtmlLink":"Fix handling of empty second copy of PropertySeq"}},{"before":"ac3c75003a1eeda38fa8d51368b11fcec9730b6b","after":"31a4843667830be87a38bd8bb22268061f79b42b","ref":"refs/heads/master","pushedAt":"2024-04-12T07:02:44.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Add fuzzer for security deserializer (#1967)\n\n* fuzz: add fuzzer targeting deserializer in security plugin\r\n\r\n* fuzz_security_deser: set linker language to cxx","shortMessageHtmlLink":"Add fuzzer for security deserializer (#1967)"}},{"before":"6498bf992059d3f491f5b351da8c7f0eb15f04a7","after":"ac3c75003a1eeda38fa8d51368b11fcec9730b6b","ref":"refs/heads/master","pushedAt":"2024-04-11T14:29:37.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Add EXPORT_ALL_SYMBOLS option to build system\n\nSigned-off-by: Erik Boasson ","shortMessageHtmlLink":"Add EXPORT_ALL_SYMBOLS option to build system"}},{"before":"826a070cfc6e3470ce26b7e186d4265d7ddf3e24","after":"6498bf992059d3f491f5b351da8c7f0eb15f04a7","ref":"refs/heads/master","pushedAt":"2024-04-10T11:53:31.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Add TOPIC_DATA to ddsi_default_qos_topic\n\nNo externally visible effect for applications, but not setting is wrong and affects the\nC++ QoS provider implementation because it directly looks at the default QoS objects.\n\nSigned-off-by: Erik Boasson ","shortMessageHtmlLink":"Add TOPIC_DATA to ddsi_default_qos_topic"}},{"before":"186f3efb2e1a9b527397425b15baa7c0cb48562c","after":"826a070cfc6e3470ce26b7e186d4265d7ddf3e24","ref":"refs/heads/master","pushedAt":"2024-04-10T09:52:17.000Z","pushType":"pr_merge","commitsCount":7,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"dds_qset_entity_name should free old value if set\n\nSigned-off-by: Erik Boasson ","shortMessageHtmlLink":"dds_qset_entity_name should free old value if set"}},{"before":"44ca08e0c200ee6188373c8bc5c17edcf06c0b54","after":"186f3efb2e1a9b527397425b15baa7c0cb48562c","ref":"refs/heads/master","pushedAt":"2024-04-05T12:27:17.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Delete some left-over code\n\nHasn't been necessary since 0845337, so for 4 years.\n\nSigned-off-by: Erik Boasson ","shortMessageHtmlLink":"Delete some left-over code"}},{"before":"4765392c690a3a5248e261b56a3a1fe33db1f3fe","after":"44ca08e0c200ee6188373c8bc5c17edcf06c0b54","ref":"refs/heads/master","pushedAt":"2024-04-04T15:00:56.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Fix padding issue in authentication plugin\n\nThe discovery data is to be serialized as a parameter list, serialising each parameter in\nturn as a 4-byte aligned, 32-bit header followed by arbitrary payload, itself serialised\naccording to the rules of the original CDR spec (i.e., XCDR1 in XTypes' Newspeak) as-if\nthat payload starts at offset 0. There is never any padding other than that required by\nthese rules.\n\nThe code for handling strings and octet sequences used by the authentication plugin got\nthis wrong by (inconsistently) padding out strings and octet sequences to the next\nmultiple of 4. In the deserialization this also interferes with the bounds checking.\n\nIn principle this can lead to a failure to read/write well-formed discovery data. In\npractice, the types happen to be such that that padding always ends up being present\nbecause of the next field.\n\nSigned-off-by: Erik Boasson ","shortMessageHtmlLink":"Fix padding issue in authentication plugin"}},{"before":"0bf9627036a93de0bf74afb5d082da44ac07d3e8","after":"4765392c690a3a5248e261b56a3a1fe33db1f3fe","ref":"refs/heads/master","pushedAt":"2024-04-03T12:57:33.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"qos_provider support (#1955)\n\n* qos_provider support added\r\n* qos_provider tests\r\n* qos_provider documentation\r\n* qos_provider build flag\r\n* system definition parsing support\r\n\r\nSigned-off-by: Splinter1984 \r\n\r\n* fix #1955 remarks\r\n* CI analyzer comliant fixes\r\n\r\nSigned-off-by: Splinter1984 \r\n\r\n---------\r\n\r\nSigned-off-by: Splinter1984 ","shortMessageHtmlLink":"qos_provider support (#1955)"}},{"before":"b47d42fd976e96cf0833b5cbb2371d82970bebaf","after":"f2ec5247bd7a56e70f07c9b74950f3dbb159fd6d","ref":"refs/heads/releases/0.10.x","pushedAt":"2024-04-02T15:18:55.000Z","pushType":"pr_merge","commitsCount":3,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Resolve `snprintf` buffer overflow under GCC 12.3.0 (#1817)\n\n* ddsi_config: Fix buffer overflow for snprintf under GCC 12.3.0\n\n* ddsi_config: Update generated files","shortMessageHtmlLink":"Resolve snprintf buffer overflow under GCC 12.3.0 (#1817)"}},{"before":"085bd8753df15fc57e0240926d0023b083b5df63","after":"0bf9627036a93de0bf74afb5d082da44ac07d3e8","ref":"refs/heads/master","pushedAt":"2024-03-28T16:22:24.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Always make idlc_generate available in install\n\nWith cross-builds, and now therefore also static builds, one does need IDLC. Installing\nGenerate.cmake unconditionally means one can always write\n\n idlc_generate(TARGET konijn FILES wortel.idl)\n\nand one only has to make sure that CMake can find \"idlc\" and the backend library. Adding\nthe host build to the CMAKE_PREFIX_PATH as in:\n\n cmake -DCMAKE_PREFIX_PATH=.../target-install\\;.../host-install\n\nresults in CMake finding the CycloneDDS package for the target in the target-install\ndirectory while also looking (and finding) idlc and the backend somewhere in the\nhost-install directory.\n\nSigned-off-by: Erik Boasson ","shortMessageHtmlLink":"Always make idlc_generate available in install"}},{"before":"0118042cb454303a6dbcc3b9e27e45932fa67f41","after":"085bd8753df15fc57e0240926d0023b083b5df63","ref":"refs/heads/master","pushedAt":"2024-03-28T15:28:47.000Z","pushType":"pr_merge","commitsCount":2,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Test initializing defaulted external sequence\n\nThis test relies on the combination of the address sanitizer and the undefined behaviour\nsanitizer: the former for initializing freshly allocated memory, the latter for faulting\non trying to load `bool` from memory with a value other than 0 or 1.\n\nSigned-off-by: Erik Boasson ","shortMessageHtmlLink":"Test initializing defaulted external sequence"}},{"before":"dfdbaf1578756dd63046ad76ed7c7d725034e7f1","after":"0118042cb454303a6dbcc3b9e27e45932fa67f41","ref":"refs/heads/master","pushedAt":"2024-03-27T10:54:17.000Z","pushType":"pr_merge","commitsCount":5,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Fix race condition in security FSM unit test\n\nSigned-off-by: Erik Boasson ","shortMessageHtmlLink":"Fix race condition in security FSM unit test"}},{"before":"840c1ffc5225460ba9c4965a69c49ca37a0489f7","after":"dfdbaf1578756dd63046ad76ed7c7d725034e7f1","ref":"refs/heads/master","pushedAt":"2024-03-27T09:09:07.000Z","pushType":"pr_merge","commitsCount":7,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Add -rpath to macOS idlc/xtests build flags","shortMessageHtmlLink":"Add -rpath to macOS idlc/xtests build flags"}},{"before":"6f5ce8647814cbf339893ec5547335bcfd3f5a7c","after":"840c1ffc5225460ba9c4965a69c49ca37a0489f7","ref":"refs/heads/master","pushedAt":"2024-03-26T12:21:24.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Add build-time option for TCP+TLS support\n\nENABLE_SSL did two things:\n\n* Enable the built-in plugins (if ENABLE_SECURITY)\n* Enable TCP+TLS support\n\nTCP support is rarely used and has some problems, TCP+TLS support is even more rarely used\nand we recommend using DDS Security instead. If TCP+TLS is not recommended, it makes sense\nto support builds that leave it out. This also has the advantage of being able to include\nsupport for DDS Security with the built-in plugins but without introducing a dependency on\nOpenSSL for the core library.\n\nSigned-off-by: Erik Boasson ","shortMessageHtmlLink":"Add build-time option for TCP+TLS support"}},{"before":"84b8713f57e54780a93883b647074b41fdfe1ced","after":"b47d42fd976e96cf0833b5cbb2371d82970bebaf","ref":"refs/heads/releases/0.10.x","pushedAt":"2024-03-26T09:50:23.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Fix topic name array sizes in \"find topic\" tests\n\nSigned-off-by: Erik Boasson ","shortMessageHtmlLink":"Fix topic name array sizes in \"find topic\" tests"}},{"before":"648a898d686e6664ab3267bc9b3e86bac2d06992","after":"6f5ce8647814cbf339893ec5547335bcfd3f5a7c","ref":"refs/heads/master","pushedAt":"2024-03-25T13:30:46.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Drop support for OpenSSL pre-1.1.1\n\nOpenSSL versions older than 1.1.1 have all been dead for over 4 years. I don't see why an\nCyclone would have to continue supporting the bad practice of not updating EOL'd security\nsensitive libraries full of known vulnerabilities.\n\nOf course nobody should be using OpenSSL 1.1.1 anymore (it has been EOL'd about half a\nyear ago), but I know there are still plenty of systems in the field that rely on it and\neven the CI on Azure gets the latest Linux images with it pre-installed.\n\nSigned-off-by: Erik Boasson ","shortMessageHtmlLink":"Drop support for OpenSSL pre-1.1.1"}},{"before":"53a204fa67f045704451471b5b7195eb9daca330","after":"648a898d686e6664ab3267bc9b3e86bac2d06992","ref":"refs/heads/master","pushedAt":"2024-03-25T10:55:59.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"remove deprecate warnings when using openssl v3 (#1826)\n\n* update the security plugins remove deprecate warning when using openssl v3\r\n\r\nSigned-off-by: Marcel Jordense \r\n\r\n* resolve memory leak in security test cases and process review comments\r\n\r\nSigned-off-by: Marcel Jordense \r\n\r\n---------\r\n\r\nSigned-off-by: Marcel Jordense ","shortMessageHtmlLink":"remove deprecate warnings when using openssl v3 (#1826)"}},{"before":"2246c005a0b5ada7f22da1e68f9c841006f29363","after":"53a204fa67f045704451471b5b7195eb9daca330","ref":"refs/heads/master","pushedAt":"2024-03-25T10:09:55.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Allow multipathing between PSMX and DDSI\n\nOnce upon a time sending a sample to a single destination via PSMX and DDSI at the same\ntime would lead to stuttering because of insufficient precise filtering in the delivery\npaths, but that problem has been fixed a long time ago (I think in commit\nef91c275e5b6649279fff7d29462d491a0d08b28). For a long time, no-one realised that the bit\nthat prevented this from happening was left in.\n\nSigned-off-by: Erik Boasson ","shortMessageHtmlLink":"Allow multipathing between PSMX and DDSI"}},{"before":"81c1d448d77b83eeb306be3113b6e6f5c2f614c8","after":"2246c005a0b5ada7f22da1e68f9c841006f29363","ref":"refs/heads/master","pushedAt":"2024-03-22T09:35:12.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Fix typo in getting started guide\n\nSigned-off-by: Patrick Masselink ","shortMessageHtmlLink":"Fix typo in getting started guide"}},{"before":"a4937a96f7e8f9b935cf7433542e14725b90a0b7","after":"81c1d448d77b83eeb306be3113b6e6f5c2f614c8","ref":"refs/heads/master","pushedAt":"2024-03-22T09:33:26.000Z","pushType":"pr_merge","commitsCount":2,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Fix fuzz_sample_deser build for honggfuzz engine\n\nIt built successfully, but would not run. This also gets reported as a build error by\nOSS-Fuzz.\n\nSigned-off-by: Erik Boasson ","shortMessageHtmlLink":"Fix fuzz_sample_deser build for honggfuzz engine"}},{"before":"f724b4fbdaba6939c71e3412bdeec29d1fa94946","after":"a4937a96f7e8f9b935cf7433542e14725b90a0b7","ref":"refs/heads/master","pushedAt":"2024-03-20T12:19:39.000Z","pushType":"pr_merge","commitsCount":7,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Ignore SPDP from non-selected interfaces\n\nIgnore SPDP messages that:\n\n* Provably arrived over a network interface we didn't enable\n\n* Were multicast over a network interface on which did not allow\n multicast discovery.\n\nFor an example why this is useful: on Linux, running two proceses one allowing multicast\nover loopback but not Ethernet, and one allowing it over Ethernet but not over loopback,\nboth processes do receive the multicast message despite not having joined the multicast\ngroup on the interface it is sent on. With this commit, those SPDP messages are ignored\nand the two processes do not discover each other.\n\nThe direct reason for this change is to make it possible to have discovery over Ethernet\nonly with unicast messages, while using relying on multicast over loopback. This turns\nout to be a rather important set-up, and is even the new default in ROS 2.\n\nSigned-off-by: Erik Boasson ","shortMessageHtmlLink":"Ignore SPDP from non-selected interfaces"}},{"before":"79680a162494355a6991f474c956320a076b321a","after":"f724b4fbdaba6939c71e3412bdeec29d1fa94946","ref":"refs/heads/master","pushedAt":"2024-03-15T13:34:54.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Fix AFL build for new deserialization fuzzer\n\nSigned-off-by: Erik Boasson ","shortMessageHtmlLink":"Fix AFL build for new deserialization fuzzer"}},{"before":"bc80a19e7b2a43d5d126cc226406c76899fbd763","after":"79680a162494355a6991f474c956320a076b321a","ref":"refs/heads/master","pushedAt":"2024-03-13T13:57:57.000Z","pushType":"pr_merge","commitsCount":4,"pusher":{"login":"eboasson","name":null,"path":"/eboasson","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/16984005?s=80&v=4"},"commit":{"message":"Windows + choco OpenSSL 3.2.1 + CMake woes\n\nInstalling openssl 3.2.1 with Chocolatey if openssl 1.x is already present gives\ninteresting results (it uses the include files from 3.x but the libraries of 1.x). You'd\nthink that uninstalling openssl 1.x, then installing openssl 3.2.1 would solve all\nproblems, but then CMake says it can't find OpenSSL at all, and indeed it can't find the\nlibraries in the locations where it is looking for them.\n\nSigned-off-by: Erik Boasson ","shortMessageHtmlLink":"Windows + choco OpenSSL 3.2.1 + CMake woes"}}],"hasNextPage":true,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAAEPqN_lgA","startCursor":null,"endCursor":null}},"title":"Activity · eclipse-cyclonedds/cyclonedds"}