Add the option to write the results to a JSON file #53
Conversation
|
I closed the previous pull request because it was causing a huge diff, due to changes in line endings. |
|
Hi @cgkantidis ! Thank you for your contribution. I've looked at your PR briefly, and it looks nice. I didn't know cmake has package support. I'll look into it a bit more over the weekend. Generally I think we can include libraries now that the build is done using GitHub actions, so no issue. It would be great if you can make the Windows build green too (probably lacks a header include only). |
|
Thanks @dlidstrom ! Indeed I had to include the There seems to be an issue with CI now, because it can't create a new tag. It seems like some kind of permission issue? Not sure. Please let me know if there's something I can do from my side. |
cgkantidis
force-pushed
the
main
branch
2 times, most recently
from
dd62a51 to
16c2c83
Compare
JSON has wider support than XML across programming languages, so I believe having the option to write the findings in a JSON format make Duplo more versatile, since its output can be consumed by many more tools. I chose to support JSON by adding a dependancy to the nlohmann/json header-only library, which handles escaping of the lines correctly. This has the drawback that we keep all the findings in memory (in a `json` object) instead of writing them directly to the output file. If this is considered a deal-breaker, we can re-implement all the escaping of the lines ourselves, as is done for XML, and write directly to the output file.
cgkantidis
force-pushed
the
main
branch
2 times, most recently
from
6cfdefe to
b8d8c43
Compare
|
Hi @dlidstrom, I resolved the conflict with the other pull requests, and I also added a test for the json output. |
|
This is looking great to me. A big thanks for taking the time to add a test too! |
2 participants
JSON has wider support than XML across programming languages, so I believe having the option to write the findings in a JSON format make Duplo more versatile, since its output can be consumed by many more tools.
I chose to support JSON by adding a dependancy to the nlohmann/json header-only library, which handles escaping of the lines correctly.
This has the drawback that we keep all the findings in memory (in a
jsonobject) instead of writing them directly to the output file.If this is considered a deal-breaker, we can re-implement all the escaping of the lines ourselves, as is done for XML, and write directly to the output file.