1. Moved the multi_vendor approach under common directory as a library.
So, this multi_vendor library can be used for agent as well as controller.
2. Under common, created a multi_vendor directory.
this directory will contain the vendors directory.
as of now, multi_vendor directory contains a example vendor directory.
and this example vendor will contain two dir, first is include that contains the header files specific to vendors
and second as src dir contains the cpp files.
3. Refactor the structure like this, to avoid duplication. at first level, we have a map with the VENDOR_OUI as key and
a map as value, and this nested map has message type as key and handler function as value
static std::map<uint32_t, std::map<ieee1905_1::eMessageType, tlv_function_t>>
4. rename the files and implementation of sample vendor to example.
5. Removed the vendors oui vector and hence avoided the loop through the registered vendors oui every time in each message.
Created a function add_vs_tlv() in the multi_vendor framework which will loop through the table 'tlv_function_table' and
invoke the handler to add the TLV to the message. This one line function call can be added in each message for example,
ap search message, capability report message etc to add the Vendor specific TLVs to each message.
Closes PPM-2974
Signed-off-by: Sangeetha.Selvaraj <sangeetha.s1@tataelxsi.co.in>
Currently we are seeing the Intel based Vendor Specific TLV and
provision for other vendor TLV to be included is not available in
prplMesh. So, we are proposing the Generic Approach where
different Vendor's can add their multi-Vendor TLV's as part of
existing 1905 Messages.
By this the Agent will be compatible to work with another vendor
Controller's and enable specific functionalities which are
contributed back to prplMesh community.
Will be adding files yaml, multivendor.cpp, multivendor.h and
specific files for sample vendors functionalities to showcase
generic approach.
Closes PPM-2974
Signed-off-by: Sangeetha.Selvaraj <sangeetha.s1@tataelxsi.co.in>
Co-Authored-by: Sundram <sundram.p@tataelxsi.co.in>
* Fixed inconsistency about Intel representation
* Fixed inconsistency in Hitron vs Hitron Technologies
* Added Oren, who I missed the first time around (sorry!)
Signed-off-by: Frederik Van Bogaert <frederik.vanbogaert@mind.be>
We are adding unit tests based on Google Test and Google Mock. Add
support for them to CMake.
GTest support is conditional on the BUILD_TESTS option. GTest-based
tests can use the BUILD_TESTS variable to decide if they should be
built or not.
Since tests will be created in other components than framework, move the
BUILD_TESTS option to the top-level CMakeLists.txt. Same for
enable_testing, and make that conditional on BUILD_TESTS as well.
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Mario Maz <mmazzapater@gmail.com>
REUSE requires the full license texts to exist in the LICENSES
directory. We have three licenses at the moment:
- BSD-2-Clause-Patent for most of the code;
- ISC for the parts taken from iw;
- MIT for easylogging++.
Replace LICENSE with a symlink to the text in LICENSES. The text in
LICENSES is taken as is from the SPDX definition. It differs only in
whitespace, and in the copyright line itself. Since it's better to use
the literal text from the SPDX definition, move the copyright years to
the AUTHORS.md file. While we're at it, correct those years: the start
year is 2016 for some of the code, and the end year is 2020 now.
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Since the file is no longer similar to the code from meshComms, replace
the plain copyright statement of the Broadband Forum to an explicit
reference like we do for nl80211_client.
Also update AUTHORS.md with this information.
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Add additional copyright holders found by grepping for copyright
statements in the source code. Some of these are copyright holders that
are not explicitly mentioned in the logs (Intel, MaxLinear), others are
from code that we copied from somewhere else.
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Add a list of authors. This allows us to avoid specifying a name in the
copyright line in each file. Over time, other people will start
modifying that file so the name specified in the copyright line will
probably go out of date very quickly.
The initial list is populated based on git shortlog -s and removing
duplicates.
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>