- 15 Dec, 2015 2 commits
-
-
Joseph Morrow authored
The /oic/d and /oic/p resources must have oic.if.r and oic.if.baseline interfaces specified. This is also for IOT-874. Ordering as follows. - oic.if.baseline - oic.if.r Change-Id: Ia6a666c03e954634d4589c2bbe8a3a4d72ad8a86 Signed-off-by:
Joseph Morrow <joseph.l.morrow@intel.com> Reviewed-on: https://gerrit.iotivity.org/gerrit/4607Reviewed-by:
Uze Choi <uzchoi@samsung.com> Tested-by:
-
Joseph Morrow authored
Prior to this, Clients could only perform a GET request directly on /oic/d and /oic/p. This commit also enables performing DISCOVERY requests on /oic/d and /oic/p as well. This aligns with OIC spec compliancy. This fixes Jira Ticket IOT-874. Change-Id: I29693a67bb245b3fae8d1ed54833eacd1397884e Signed-off-by:
jenkins-iotivity <jenkins-iotivity@opendaylight.org> Reviewed-by:
Markus Jung <markus.jung85@gmail.com> Reviewed-by:
Jon A. Cruz <jonc@osg.samsung.com> (cherry picked from commit 30c25f54) Reviewed-on: https://gerrit.iotivity.org/gerrit/4585
-
- 10 Dec, 2015 1 commit
-
-
hyuna0213.jo authored
since there is a preprocessor for TCP_ADAPTER in RI. when user is using a interface file(octype.h) user application have to add -DTCP_ADAPTER flag in their App. since CA already check TCP_ADAPTER flag, it is better to remove this preprocessor in RI. Change-Id: I8dd2998831d9744e25e68c024b2b3e3fbbbc1633 Signed-off-by:
jihwan.seo <jihwan.seo@samsung.com> Signed-off-by:
hyuna0213.jo <hyuna0213.jo@samsung.com> Reviewed-on: https://gerrit.iotivity.org/gerrit/3349Tested-by:
Ashok Babu Channa <ashok.channa@samsung.com> Reviewed-by:
-
- 09 Dec, 2015 1 commit
-
-
koushik.girijala authored
Moved OC random implementation to c_common folder to make it available for CA as well. As Seed Random should be called at common place for RI and CA. Change-Id: I0a7cd674b9c4212b3a1d1e382aa1f455bec1a6d2 Signed-off-by:
koushik.girijala <g.koushik@samsung.com> Reviewed-on: https://gerrit.iotivity.org/gerrit/4101Tested-by:
-
- 08 Dec, 2015 1 commit
-
-
koushik.girijala authored
Modified code to make more heap available in arduino MEGA Change-Id: I62ebc1bcf08331d7d90871bb2a6dd201f4ae7e4b Signed-off-by:
vimala.v <vimala.v@samsung.com> Signed-off-by:
Jaehong Jo <jaehong.jo@samsung.com> Reviewed-by:
-
- 07 Oct, 2015 1 commit
-
-
jihwan.seo authored
- There is no way to select ALL option without TCP in previous build option. - CoAP/TCP should be also used WITH_TCP option like WITH_RA. - and CoAP/TCP can be supported optionally more easy for Non-IP Transfer. Change-Id: I173205871d44a4e85fdfe5626841f74c27206c7a Signed-off-by:
Patrick Lankswert <patrick.lankswert@intel.com>
-
- 06 Oct, 2015 4 commits
-
-
Woochul Shim authored
- PMDeviceDiscovery() add discovered devices in the list whenever callback is hit. If timeout happens the pointer to a list could be freed by the app but still there is a registered discovery callback. If the callback is hit after defined timeout period, callback try to add a discovered device to the invalid pointer (to freed memory). It will cause segmentation fault. Fix: After defined timeout, PMDeviceDiscovery() explicitly remove registered callback using OCCancel. Change-Id: Ic713cefb75c3495cfc9ad6688df44eb124d88f82 Signed-off-by:Woochul Shim <woochul.shim@samsung.com> Reviewed-on: https://gerrit.iotivity.org/gerrit/3545Tested-by:
Sachin Agrawal <sachin.agrawal@intel.com>
-
daeken.kwon authored
[IOT-762] Something wrong with handling unicast presence response - client always get "PRESENCE_TIMEOUT" < Cause > ocstack don't do ResetPresenceTTL() after getting presence response from server because there was no change in sequnceNumber < Solution > To resolve this case, it is just needed to put ResetPresenceTTL() before exit in this caseTo resolve this case, it seems that we just put ResetPresenceTTL() before exit in this case Change-Id: If1e32692d3b859cea84382bded0cceecf42874d5 Signed-off-by:
daeken.kwon <daeken.kwon@samsung.com> Reviewed-on: https://gerrit.iotivity.org/gerrit/3595Reviewed-by:
-
vimala.v authored
As cherry-pick from master branch failed, following commits are pushed as a standalone commit to 1.0.0-dev branch. 1) https://gerrit.iotivity.org/gerrit/#/c/2867/ 2) https://gerrit.iotivity.org/gerrit/#/c/3277/ 3) https://gerrit.iotivity.org/gerrit/#/c/2807/ Change-Id: I3a66dae476509f0f57995d8792e49ed5a5b6fe94 Signed-off-by:
Abhishek Pandey <abhi.siso@samsung.com> Reviewed-by:
-
Habib Virji authored
Corrects the format as per the specification. Issue: IOT-717 Change-Id: Ic03d67541ef5afdfd6acc3820dc224b61484bd83 Signed-off-by:
Habib Virji <habib.virji@samsung.com> Reviewed-on: https://gerrit.iotivity.org/gerrit/3283Tested-by:
Madan Lanka <lanka.madan@samsung.com>
-
- 29 Sep, 2015 2 commits
-
-
Habib Virji authored
APIs to start and stop multicast traffic. For stopping and staring multicast traffic it manipulates multicast fd and IP_DROP_MEMBERSHIP. OCStopMulticastTraffic interface is called from rd_client. It stops the multicast traffic. OCStartMulticastTraffic is implemented but it is not called as this functionality is not yet implemented. Change-Id: If56c7af074cbba097b7b23d1abd787093872418a Signed-off-by:
Abhishek Sharma <ce.abhishek@samsung.com> Tested-by:
-
Habib Virji authored
This is based on the new strucutre of the /oic/res as currently being defined by Ravi. This is useful for the resource directory as it needs to include multiple links maps inside links array. It differs with the current /oic/res response as it includes links array with multiple links map object. The strucutre of the payload appears in this format: Array TagsObject di baseURI bitmap ins LinksArray LinkMapObject href rt itf ins Change-Id: Ifc61a70ede1fc44e9b2321604907d99961eea13b Signed-off-by:
-
- 28 Sep, 2015 2 commits
-
-
koushik.girijala authored
Fixed double free corruptions and errors in freeing static memory If accepted, this changeset should be cherrypicked to 1.0.0-dev. Change-Id: I5b37f5090b68b45156b5965f715b68daebfe9e37 Signed-off-by:
Gabriel Schulhof <gabriel.schulhof@intel.com> Reviewed-by:
-
Mandeep Shetty authored
This fixes the client side crash in IOT-733. OCCancel () allocates a pointer that is free'd in routingutility.c. Although the pointer was passed in as reference into routingutility and is re-assigned with a new malloc'ed pointer, OCCancel() free's the old pointer. This is because there two structs that are being passed around. The pointer that is double free'd is free'd twice as part of two DIFFERENT structs. Got rid of the extra struct to keep things consistent and resolve the crash. If accepted, this changeset should be cherrypicked to 1.0.0-dev. Change-Id: I8c961776406c02dbe7706a787bb19db53ba83853 Signed-off-by:
Mandeep Shetty <mandeep.shetty@intel.com> Reviewed-on: https://gerrit.iotivity.org/gerrit/3055Tested-by:
-
- 21 Sep, 2015 1 commit
-
-
Habib Virji authored
RD server response for the RD Client Publish. Include change to adapt recent changes in cbor payloadType handling. Change-Id: I55e9cacaa1a81c87c8bcc0e80c3b57f6fd98f8fc Signed-off-by:
-
- 20 Sep, 2015 1 commit
-
-
Mandeep Shetty authored
args to some functions were validated for NULL to return error codes even though NULLs were legal for the args. For eg. it is legal for options to be NULL. These incorrect validations caused early returns and observation actions were not set to OC_OBSERVE_NO_OPTION which is 2. The field in the struct instead defaulted to OC_OBSERVE_REGISTER = 0 due to default inits. Fixed both by removing or replacing the incorrect validations with more appropriate ones. Change-Id: Ia7453227a668bb26971b2b7d84381f64151b88f9 Signed-off-by:
-
- 19 Sep, 2015 1 commit
-
-
Erich Keane authored
The functions that add resource type/interfaces to the resources were not correctly validating them to the RFC. This commit adds validation to these such that RFC6690 is enforced, and gives the developer immediate/early warning of invalid values. Change-Id: If04f72c1aa69ec2f9fbc83d849a503c2ee86f502 Signed-off-by:
Erich Keane <erich.keane@intel.com> Reviewed-on: https://gerrit.iotivity.org/gerrit/2605Tested-by:
-
- 17 Sep, 2015 1 commit
-
-
Jaehong Jo authored
Implementation for Routing Manager. 1) Gateway resource Hosting and exchange of routing tables. 2) Linklist implemenation. 3) Routing table manager to add, remove, get next hop etc. 4) Routing message parser using cbor parser funtions. 5) Dynamic route updates on addition/removal of gateways. 6) Support for Endpoint Please refer to below links for wiki and jira updates. https://wiki.iotivity.org/routing_through_heterogeneous_transports https://jira.iotivity.org/browse/IOT-565 Change-Id: I228b59ef922f55552386fb0b0e1b2bb2343b1adb Signed-off-by:
-
- 16 Sep, 2015 3 commits
-
-
jihwan.seo authored
since there is no code in CoAP Spec. I have removed CA_SUCCESS code from CA. According to CoAP Spec, 2.05 Content code is like HTTP 200 "OK" thur 2.05 will be used instead of 2.00. --------------------------------------------------------- 5.9.1.5. 2.05 Content This Response Code is like HTTP 200 "OK" but only used in response to GET requests. --------------------------------------------------------- Change-Id: I9f4d3c09661c4dec967f2bb91f51941b93084a93 Signed-off-by:
-
Sakthivel Samidurai authored
After removing the OCPayloadType from the payload, the stack has to fill the appropriate payload type for security and non-security payload type. Added PAYLOAD_TYPE_SECURITY as a payload type for security resource request in HandleResourceWithEntityHandler and response in HandleCAResponses. Change-Id: I3007a1e79ad25b4e557ab15edb038a22cb9e85c3 Signed-off-by:
Sakthivel Samidurai <sakthivel.samidurai@intel.com> Reviewed-on: https://gerrit.iotivity.org/gerrit/2539Reviewed-by:
-
Erich Keane authored
Previously, the CBOR container included the type of the message to make parsing of the CBOR items easier. This is unfortunately against the spec, so this patch removes it and requires the caller to ensure that the expected message type is sent. Incorrectly formatted messages will result in the message not passed back via the callback. Change-Id: If1ca9eed65b5274239b670e8a2bd6cadce763c19 Signed-off-by:
-
- 15 Sep, 2015 2 commits
-
-
Stephane Lejeune authored
Since draft-ietf-core-observe-02 (September 16, 2011), the result value "2.00" was depricated and replaced with "2.05" for GET a method and "2.04" for POST and PUT methods. Change-Id: I8061089469609d5ad801fa0456fda326cbc1fc33 Signed-off-by:
Stephane Lejeune <stlejeun@cisco.com> Reviewed-on: https://gerrit.iotivity.org/gerrit/2396Tested-by:
Hauke Mehrtens <hauke.mehrtens@lantiq.com> Reviewed-by:
-
Stephane Lejeune authored
This commit adds the CBOR accept option to all requests. When a request comes in with an accept option, the option is evaluated against the supported encoding formats. If the format is undefined or when CBOR is requested, CBOR will be used to encod the payload. When the response does not carry a payload, the accept option is ignored and the response will proceed irrespective whether the accept format was supported or not. This allowsi for example a device that performs a delet operation proceed first working out which format is supported by the server. Change-Id: I7ae3430d11f1481a91413088a959f105b216ffea Signed-off-by:
-
- 12 Sep, 2015 1 commit
-
-
Mandeep Shetty authored
The resource uri was moved from the endpoint to the uri response info struct in IPv6. The function sendDirectStackResponses responsible for sending out ACKs, RESETs and error codes back to the client was not updated to take in the resource uri. As a result, the CA layer ate this error with a uri is NULL log and the client did not receive any response. Added resource Uri to the responseInfo struct to fix the problem. Change-Id: Ic3b964c646244698c1621860922fc06b40842bc3 Signed-off-by:
-
- 11 Sep, 2015 1 commit
-
-
hyuna0213.jo authored
- added new transport adapter in RI - linux platform can be supported - https://wiki.iotivity.org/proposal_for_cloud_interface_in_iotivity Change-Id: Iaf245c3bbf30e76708545d136435a371b86fda7e Signed-off-by:
-
- 08 Sep, 2015 2 commits
-
-
koushik.girijala authored
OC log and OC log V used almost 3500bytes of RAM in Mega board because of improper log usage. Updated code using PCF properly in OC logging to save almost 3500 bytes of RAM memory. Verfied changes on both Arduino Due and Mega boards and found Working fine. Change-Id: I9ca134d5cff9c8a91707a18bd9279089533a9b99 Signed-off-by:
-
Stephane Lejeune authored
Added a new option defining content format for CBOR encoded payloads. This will allow a more robust interpretation of incomming payloads and will also permit generic stacks that have no apriori knowledge about OIC to interpret the payloads going accross the network. Change-Id: I34feb4e305458fcef4e7343cda313d9a52d12046 Signed-off-by:
-
- 04 Sep, 2015 2 commits
-
-
jihwan.seo authored
it can be gaven some confusion. Change-Id: If3b228b404e830fad23ce793334723d0e0f92e89 Signed-off-by:
-
Stephane Lejeune authored
The options that have variable byte length are now being encoded to their minimal value. This patch also truncates the option data to their maximum value allowed per the spec. One specific example is the observer data option which receives 4 bytes of data from the OC stack. When encoding in CoAP, the maximum value is 3 bytes. Consequently, the CA layer which is aware of the CoAP encoding truncates the value back to 3 bytes. Also when registering an observe, the data length is now set to 0 using the CoAP var byte encoding which saves the 4 bytes that were just carying zero's before thsi fix. Change-Id: I033969b407aa3426df8c85c3836206c9366fb0a2 Signed-off-by:
-
- 27 Aug, 2015 1 commit
-
-
Sakthivel Samidurai authored
SRM should generate the device ID for the resource server Change-Id: Ic4f83f63a59827bf5cfb412afc34e1d955ee69e2 Signed-off-by:
-
- 25 Aug, 2015 1 commit
-
-
minus2k authored
Change-Id: I5bb08b94bf3cf4cab3ff5878260c845bee5592a5 Signed-off-by:
Myungchul Kwak <minus2k@hotmail.com> Reviewed-on: https://gerrit.iotivity.org/gerrit/2256Tested-by:
Naga Ashok Jampani <jn.ashok@samsung.com> Reviewed-by:
-
- 24 Aug, 2015 1 commit
-
-
Shilpa Sodani authored
Resolved bug IOT-685. Due to unintialized value of field "interface" of struct CAEndpoint_t and changed return value type of method u_arraylist_add() broke the security apps. The unintailzed value of interface gave it random value which inturn set the wrong value for ipv6 sockaddr_in6 struct field "sin6_scope_id" causing "Network unreachable" error. The return value of methode "u_arraylist_add()" was recently changed from CAResult_t to bool. This caused seg fault. Change-Id: I23fea80a0d6e12dd4122f2e032497a72496a4745 Signed-off-by:
Shilpa Sodani <shilpa.a.sodani@intel.com> Reviewed-on: https://gerrit.iotivity.org/gerrit/2249Tested-by:
John Light <john.j.light@intel.com> Reviewed-by:
-
- 21 Aug, 2015 1 commit
-
-
Omkar Hegde authored
Previously, the CA layer code was assuming that all responses are always sent as unicast messages. But, in the case of presence, it is always multicast. As a result, the server sent presence notifications only to the ipv4 address and the clients subscribed to server's ipv6 address ignored the notification. Adding isMulticast flag to the CAResponseInfo_t structure and changing camessagehandler to handle multicast responses. Change-Id: I4a0991aca6cf60b4b1ad3f0dfcda683284680adc Signed-off-by:
Omkar Hegde <omkar.m.hegde@intel.com> Reviewed-on: https://gerrit.iotivity.org/gerrit/2252Tested-by:
-
- 14 Aug, 2015 2 commits
-
-
Sakthivel Samidurai authored
Move security identity to CAInfo_t. Propagate security to OCClientResponse for application. Change-Id: I535577bedf0ed4f46a97249add56aec6928d2d84 Signed-off-by:
-
Erich Keane authored
RA_ADAPTER check was merged improperly, this removes the second instance. Change-Id: Id2d02251d55f6917804eb943ab6cbe7ed27bbf83 Signed-off-by:
Ossama Othman <ossama.othman@intel.com> Tested-by:
-
- 13 Aug, 2015 2 commits
-
-
Erich Keane authored
Update filtering in camessagehandler.c Filter duplicates based on last 4 messages. Filter both requests and responses, separately. Change-Id: I191436f91c733893bcbf12c9ffdb05bc841bfa2f Signed-off-by:
-
John Light authored
Some testing Change-Id: I77b152b1e0c088b61b8f217313d70d62eb9f7383 Signed-off-by:
Doug Hudson <douglas.hudson@intel.com> Reviewed-by:
-
- 12 Aug, 2015 1 commit
-
-
sandipan.p authored
Change-Id: I93d2c10760d60f614588703dc52496943f446d56 Signed-off-by:
sandipan.p <sandipan.p@samsung.com> Reviewed-on: https://gerrit.iotivity.org/gerrit/2168Tested-by:
-
- 10 Aug, 2015 1 commit
-
-
John Light authored
Change-Id: Ia08ce526ffa759a5749a90d0d31ff5dbca60b072 Signed-off-by:
-
- 05 Aug, 2015 1 commit
-
-
Jon A. Cruz authored
These fixes clear a bit over 200 more of the basic warnings seen. A few signed/unsigned issues that might require thought were left for later. Change-Id: Ie04411927cef8a73f17f16e329f6546229c2c3d1 Signed-off-by:
-
