Add definition of full expression

[andralex]

cc @WalterBright

[dlang-bot]

Thanks for your pull request, @andralex!

Bugzilla references

Your PR doesn't reference any Bugzilla issue.

If your PR contains non-trivial changes, please reference a Bugzilla issue or create a manual changelog.

[andralex]

Urgh. Is there a way to see a diff with wraparound enabled?

[jacob-carlborg]

It's quite difficult to review lines that are this long.

[jacob-carlborg]

Thanks.

[CyberShadow]

Urgh. Is there a way to see a diff with wraparound enabled?

M-x fill-paragraph, amend, force-push

In situations where newlines do not show up in rendered text, I found that adding them at sentence breaks makes sense for version controlled text (diffs are good), and it's not unlike how statements are generally one per line in programs.

[andralex]

@CyberShadow I'm experimenting with Visual Studio Code...

[WalterBright]

see nits

[andralex]

@WalterBright done

Nits fixed.

[thewilsonator]

./spec/expression.dd:27:Note: Although the definition is straightforward, a few subtleties exist related to function literals. In the
Can we please get rid of this already?

[andralex]

"This" being what?

[thewilsonator]

The whitespace checker.

[andralex]

@thewilsonator I guess it could fix the error itself. I introduced that because I was using the online editor.

BTW I see a bunch of errors in magic.d when looking at the doc builter log http://dtest.dlang.io/results/e63eaaa2e744e490663fb29307ffb6ff65b803de/b435eade3ae96f349e71710e47a353608fff6a8b/build.log @CyberShadow could they be fixed?

[thewilsonator]

I guess it could fix the error itself.

Yes it should.

I introduced that because I was using the online editor.

Yes I do that all the time, the whitespace checker is a PITA.

BTW I see a bunch of errors in magic.d when looking at the doc builter log

They look almost like the ones I had in #2453 (comment) (see the force pushes just below).

[CyberShadow]

@CyberShadow could they be fixed?

I think the authors of the relevant code are no longer maintaining it (DDox), so someone else would need to adopt / look into it. In any case, those errors are "benign", and indicate that the tool just failed to grok the syntax of that file.

[wilzbach]

In any case, those errors are "benign", and indicate that the tool just failed to grok the syntax of that file.

Well, actually the errors are coming from libdparse trying to parse newer code from Phobos for Ddoc (<-not the c here) as we have to parse the AST again to be able to rewrite the unittests, so even if we drop Ddox, this would still be there.

See also: https://github.com/dlang/dlang.org/blob/master/ddoc/source/assert_writeln_magic.d

In theory, upgrading libdparse should do the trick, but last time I did I ran into issues: #2551

[CyberShadow]

actually the errors are coming from libdparse trying to parse newer code from Phobos for Ddoc

Thanks for the correction, I did mix that up. To be fair, libdparse is also no longer maintained by its original author, as with DDox.