[ Sorry for the long delay. ]

Could you remind me of the purpose of this change?
Was it only to add a `separator' argument to ewoc?

While, I'm here I noticed that your code did not follow our indentation
and layout conventions, so I've fixed that in the attached patch (where
I also replaced the ewoc-provides-variable-separator with a subfeature).

Finally, regarding this comment:

   ;; A start-marker's insertion-type should already be `t', but some
   ;; callers want to be 100% sure it is, so this function exists.

is it the case that it would be a bug if the insertion-type was
changed to nil, or are there legitimate cases where the insertion-type
is changed to nil (and if so, which are these)?  I ask because if
there's no legitimate case, then we should just `assert' it.


        Stefan


>>>>> "Tom" == Tom Breton (Tehom) <tehom@panix.com> writes:

>> Keep only `separator', but with the added twist that a non-string,
>> non-nil argument means the empty string (aka mean "nosep").  That should
>> preserve backward compatibility with a mostly clean API.

> Done.

>> I think you're worrying for no good reason.  If you really want to
>> handle version dependencies right, you don't want it in a Lisp variable
>> but in a header understood and processed by some package manager (ELPA,
>> for instace).  Otherwise, you'll have to deal (one way or another) with
>> runtime checks, and in most cases the calling package will simply not
>> work with an older version.

> Yes, I have to deal with runtime checks.

> Anyways, I have replaced the version variable with a variable called
> `ewoc-provides-variable-separator' that is bound just if ewoc provides
> a variable separator.  I hope that will satisfy.

> Attached is
>  * new ewoc patch against 23.1
>  * test file

> Thank you for your help, Stefan.

>         Tom Breton (Tehom)