Is there a place except for github and here, like a newsgroup server, where one can ask questions and discuss solutions? Big fan of threaded discussions and easy access from CLI...

boa, such threaded discussions are typically done ... via PRs.

That's why I myself don't over invest time in them, because there are usually many different opinions.

Ack, but PRs are for code changes, right? What I had in mind were developer-to-developer questions and discussions. For example, I see that parts of the code could need some love, like lmdb and other externals. Also, clang-tidy produces lots of output (via the health script you wrote and I like). Another example, the code allocates memory to report "out of memory" situations, which may be less than ideal

;-) Where can we discuss questions like this?

Thanks boa . Yes, this is a good point you're raising. But still for what I know, PRs are kinds of best ways here to prove that a concept works, and that the tests pass. Github offers DRAFT PR, which suits this purpose perfectly.

I'll give you an example, why you should never over invest time, especially when it comes to Linting:

You may also open an Issue on Github, if you only have an idea, or request.

Maybe this would be a better starting point.

I can do that instead. Thanks, I appreciate you helping out a noob here :)

n00b helps n00b

selsta, could Adaptive Mining be next? 7618

So that we finally don't have to ignore the test results.

and thanks luigi1111 for the quick merge.

boa, once the Issue is created and discussed, and you think it's time for a PR, it's best to start the description with "Fixes [link-to-issue]". This will link the Issue with the PR.

And if you like my health script, I'm just generating a range of UT coverage reports. It will take a day I think.

mj-xmr I've added a few comments on 7480

Thanks.

Why does the Saviour of NASA take a group achievement award and present it as a proof of individual glory? twitter.com/hyc_symas/status/1203709575226183683

mj-xmr: The "Fixes [link-to-issue]" is nice, but what if there is no issue except for one I've found in the source code? Should I add the issue to github first, and then add the PR? I'm asking because I fear it may be a lot of "placeholder" issues, e.g. if one starts to clean up the output from the clang-tidy script.

Only actual problems, not "change code because clang whines" please.

I mean, it whines on preposterous shit at times.

True, like gcc's -Wconversion. Nlks

(meh, pressed ENTER too soon.)

What I tried to write, was that -Wconversion is a noisy little fellow, but that occationally it warns about something important. It's easier to see the important stuff if it isn't hidden in lots of noise, IME

(figured out the rebase thing, btw. )

What is the rebase thing ?

Just a question I asked earlier. Should've googled it first. My bad.

boa, regarding what moneromooo said: Monero has a serious bottleneck in reviewing power, because there are much worse problems to solve than the potential ones. So Clang-Tidy warnings have a low priority.

That's why I will refrain from doing random checks from now.

ack. /me traces uses of malloc(), strdup(), getenv() and friends, just to get to know the code base and see if there are issues. There are issues... Not sure if it's just me being a code nazi or if the issues are real. It's fun, though, been missing nit-picking :)

:)

Leaks are definitely problems. As are buffer overflows.

Though some leaks may be ignored if they're one shot.

It's more like failing to test for null returns, and failure to sanitize external data (getenv()). Also quite a few instances of failure to check for error returns from functions, even invoking UB. Mostly in the external directory.

These could be problems. malloc will usually not return NULL for OOM, but blow up later when trying to map.

Checking error returns is usually a good idea, so feel free.

External stuff is probably best sent upstream (though easylogging got abandoned).

boa, enjo.hopto.org/pub/monero/data/5b8d…a4/5b8d714a4-unit-tests-lcov.tar.xz

UTest coverage for master.

Full report with some history will be here about tomorrow: enjo.hopto.org/pub/monero

I hope it helps.

boa, to answer your "issue" question. They are only needed if you needed a "threaded" discussion/.

ok, boomer :)

:)

Hmm, I see that the submodules aren't forked automatically. Oh, dear me, what have I done?

git submodule init/sync/update

yes, did that, but the remote is still the original monero-project repo instead of my forked repo. No biggie though, just wanted others to know that this may be a problem

no need to fork the submodules usually

ack. I'm just messing around in the unbound project. The code needs some love...

the unbound submodule would ideally get bumped to the latest upstream release but that would require some cmake changes

08:18 <mj-xmr> selsta, could Adaptive Mining be next? 7618 <-- yes if no one else leaves any blocking comments

Howard, you know why all the 'Titanic intelligence Saviour of NASA' posts stopped last year? They know you are an embarrassment. Your arse didn't learn anything new in 25 years, and that includes C++. They just can't say it to your face.

boa, since we're talking about memory, what I'd prefer to see, is make_unique() instead of malloc. Firstly it will deallocate the memory on exception and will self throw on misallocation. Also it will hold nullptr initially, which is better than some random location.

if this is not C code, that is.

does anyone know how to get the size of a specific transaction in bytes/kbytes from Daemon RPC? I don't see an output for that anywhere. I'm trying the get_transactions method web.getmonero.org/resources/develop…es/daemon-rpc.html#get_transactions

Doesn't seem to be in the RPC response. Easy to add if needed.

Oh. You can get the dump of it. Divide length by 2.

I see, thx

HI. havovesu.page.link/PocUxdgkJRdpz2cB6

why even attempt with such suspicious looking links lol

To check if your Noscript still works :)

enjo.hopto.org/pub/monero <--- finally ready with LCov reports.

^ please also open against release branch if relevant

ok

done

I think everyone here knows that most Monero exchange withdrawals and desposits are traceable (Breaking Monero - poisoned outputs). How does it feel to go out and lie to people that they are private and then earn less than holding BTC on your bag?