documents the limitation of attributes on overridden functions

[ghost]

Just to prevent any report tagged "reject-valid" after this D front-end change: dlang/dmd#6570

[wilzbach]

please use a macro for such links, e.g RELATIVE_LINK2 would work ;-)

[adamdruppe]

On Mon, Feb 27, 2017 at 09:17:56PM -0800, Sebastian Wilzbach wrote: + $(P It's not allowed to mark an overridden method with the attributes + $(LINK2 attribute.html#disable, $(D @disable)) or please use a macro for such links, e.g `RELATIVE_LINK2` would work ;-)

I'm not really following this (just saw an email), but why? Linking with the ddoc macros is complicated enough as it is, so why use another for a standard url feature like relative linking? I don't think I have ever seen RELATIVE_LINK2 before (though again, I haven't been following closely) and don't really see why you'd need it, even for other bases like library-prerelease. I can understand having a macro for linking to the spec which prepends /whatever-base/spec/ to it, but I'd call that SPEC_LINK not RELATIVE_LINK2...

[wilzbach]

I'm not really following this (just saw an email), but why? Linking with the ddoc macros is complicated enough as it is, so why use another for a standard url feature like relative linking? I don't think I have ever seen RELATIVE_LINK2 before (though again, I haven't been following closely) and don't really see why you'd need it, even for other bases like library-prerelease.

I have just been trying to review this and compared the link style with the prevalent one in the document.
My assumption for the need was that from the spec a PDF, mobi and Kindle ebook is generated?

[adamdruppe]

On Mon, Feb 27, 2017 at 09:28:10PM -0800, Sebastian Wilzbach wrote: I have just been trying to review this and compared the link style with the prevalent one in the document.

huh, yeah, it is used elsewhere so I guess we can just be consistent with that here.

My assumption for the need was that from the spec a PDF, mobi and Kindle ebook is generated?

As far as I can tell, the definition is pretty consistently just an alias for LINK2. It is reasonable for a contributor to be confused by it...

[wilzbach]

huh, yeah, it is used elsewhere so I guess we can just be consistent with that here.

@bbasile would you mind updating your PR, so that we can merge it and move forward?

[adamdruppe]

@wilzbach RELATIVE_LINK2 cannot possibly replace that LINK2... the definition in html.ddoc puts a # on the href meaning it can't cross pages.

[ghost]

@wilzbach, not done because not needed. LINK2 is good here.

[ghost]

@wilzbach, do we agree on this PR finally ? That's already too much discussion for such a detail.

[wilzbach]

@bbasile: Oh I am sorry. I totally forgot about this PR. You are absolutely right: these small PRs shouldn't get stalled for such a long time.