Comment author: @gasche

As always, patches here or pull-requests on
https://github.com/ocaml/ocaml-manual
are warmly welcome.

Comment author: @c-cube

Also, the change in attributes syntax is breaking code. Should this be in a separate bug report? I understand a new syntax can be introduced, but forbidding the old syntax in a minor version should not happen.

Comment author: @gasche

As already pointed out on caml-list,
https://sympa.inria.fr/sympa/arc/caml-list/2015-06/msg00115.html
the attributes syntax change was asked for and supported by today ppx users (the one that gave feedback anyway, including ppx_deriving's developer), so I don't think the "should not happen" is fair here -- people also complain when known-to-be-suboptimal choices are kept forever for backward-compatibility reasons.

I'm not aware of any plans to restore the old syntax: the people that participated to the discussion seemed to consider it was worse than the 4.02.2 one. If you want to restart the discussion, feel free to open a new bug report, or (even better I think) participate to the discussion in
#152

Comment author: @diml

I added a small description of nonrec (rev 16186) and updated attributes placement (rev 16185)

Comment author: @Octachron

I have a look at the description of the nonrec keyword, and I find the wording a little confusing

In a set of definitions, references in the type equation,
representation or constraint to one of the type constructor name being
defined is considered as recursive, unless "type" is followed by
"nonrec".

First, shouldn't the "is" be an "are" here?
Another point, I wonder if the "the type equation, representation or constraint" enumeration could not be replaced by "the right-hand side of definitions" since the notions of type equation, representation or constraint are only described later in the text?

Lastly, It may make sense to move the paragraph after the paragraph describing simple type definitions.

Comment author: @diml

octachron, can you submit a pull-request on github?

Comment author: @Octachron

Certainly ocaml/ocaml-manual#11.

Comment author: @alainfrisch

Bumping Severity so that we don't forget to check that the doc has been updated.

Comment author: @alainfrisch

"#-prefixed operators" have been documented in commit 1c396b9.

New placement rules for attributes, in commit 861a0db. Fixed in commit 74ac42c (attributes on record fields can also be put after the semi-colon).