Actually, this patch as-is is broken, since it depends on takeAllTreesContent working as described in the docs, which it doesn't, as mentioned in #98

It makes sense, I don't think there's a way to do that with the existing parsers/combinators. I will just try and find a different name for the new functions, as lookahead wrongly suggests that nothing is consumed from the stream. I'll probably also rename takeAllTreesContent accordingly, since it clearly falls into the same category of parsers.

After thinking about it, I intend to implement the following:

-- Consume-and-yield events of a single tag tree, as long as it matches given name and attribute parsers
takeTreesContent :: MonadThrow m => NameMatcher a -> AttrParser b -> ConduitM Event Event m (Maybe ())

takeAllTreesContent = takeTreesContent anyName ignoreAttrs

(I've introduced NameMatcher in 48e68c2)

This should cover your use case, as follows:

debugParser :: MonadThrow m => Name -> ConduitM Event o m ()
debugParser name = do
  t <- takeTreesContent (matching (/= name)) ignoreAttrs .| renderBytes def .| foldC
  if null t
  then return Nothing
  else putStrLn t >> return (Just ())

parseAll = do
    manyYield $ choose [fooParser, barParser, debugParser "quux"]
    tagName "quux"  {- ... -}

(I didn't typecheck it, but you get the idea)

If you're okay with that approach, I'll implement it and make a release out of everything you've contributed.

Yeah, that looks fine for my usecase(s).

Implemented in release 1.5.0 (not yet on Hackage). Thank you !

@merijn Release 1.5.0 is now available on Hackage, you can start working with it by setting it explicitly as a dependence of your project, otherwise a version <1.5 will be selected by default by cabal.
Please let me know if you experience any issue.