For example:

{{TRIM:X (disambiguation)}} -> X
{{TRIM:Y, Jr.}} -> Y

Created attachment 6915
patch to add new core parser function {{linkpst:...}}

I'm not sure how this would be useful, but i made a patch which basically takes the link related stuff from Parser::pstPass2 and puts it in a core parser function named linkpst (since it essentially does the link related stuff of the pre-save-transform).

{{linkpst:foo (bar)}} -> foo
{{linkpst:ns:page}} -> page
{{linkpst:ns:bar (baz), fred|}} -> bar
etc.

The only thing i'm unsure about is the regex in $p4. It doesn't appear to match stuff during pst.

Attached:

Maybe it's useful in some templates, to get the name of what an article describes.

Some thoughts for the patch in comment #2:

• do not use SFH_NO_HASH for new parser functions
• do not duplicate code, reuse the one in Parser.php if possible
• it would be very nice to have some test cases for that function in parser tests

Perhaps another parameter indicating which steps are used?

{{linkpst:foo (bar)|B}} -> foo
{{linkpst:foo (bar)|C}} -> foo (bar)
{{linkpst:ns:page|N}} -> page
{{linkpst:ns:page}} -> page
{{linkpst:ns:page|N}} -> page
{{linkpst:ns:page|B}} -> ns:page
{{linkpst:ns:page|B}} -> ns:page
{{linkpst:ns:bar (baz), fred}} -> bar
{{linkpst:ns:bar (baz), fred|BC}} -> ns:bar
{{linkpst:ns:bar (baz), fred|NB}} -> bar, fred
{{linkpst:ns:bar (baz), fred|NC}} -> bar (baz)
{{linkpst:ns:bar (baz), fred|C}} -> ns:bar (baz)

(In reply to comment #2)

I'm not sure how this would be useful...

Neither am I.

(In reply to comment #3)

Maybe it's useful in some templates, to get the name of what an article
describes.

I'm not convinced. Can you explain yourself further?

For example a template {{disambig page}} with the content:

'''{{linkpst:{{PAGENAME}}}}''' may refer to:
{{{1}}}
{{disambig}}

If there's no linkpst, some pages may be displayed like:

'''ABC (disambiguation)''' may be refer to:
*something
*something else
{{disambig}}

Ignore that "be" in "'''ABC (disambiguation)''' may be refer to:" please. :)

• This bug has been marked as a duplicate of bug 7264 ***