Re: debuggin types in records
"Daniel C. Wang" <[email protected]> Sat, 27 Jan 2007 12:08:09 -0800
| Newsgroups | gmane.comp.lang.sml.smlnj |
|---|---|
| Message-ID | <[email protected]> |
then you must have omitted a field or misspelled a field name.
a sanity check would be to see if replacing the whole record with a hole
type checks.
brian wrote:
> Well I replaced all of the record field assignments using hole and
> still get the same error...
>
> What does that tell me ???
>
> Grinds teeth...
>
> Brian
>
> On Jan 25, 2007, at 10:29 PM, Daniel C. Wang wrote:
>
>> My favorite type-debugging trick is the following
>>
>> fun hole () = raise (Fail "place holder")
>> than I start replacing expressions with hole() until the type error
>> goes away. Once, I find the expression that makes the problem go away
>> I at least know what to blame.
>>
>> brian wrote:
>>> I ran across the rather self-explanatory error
>>>
>>> operator domain: XEventTypes.key_xevtinfo
>>> operand: {child:word, event:word, event_pt:Geometry.point,
>>> keycode:int, mbut_state:XProtTypes.mbutton_state,
>>> mod_state:XProtTypes.modkey_state, root:word,
>>> root_pt:Geometry.point, same_screen:bool,
>>> time:Word32.word}
>>>
>>> The problem is that I _think_ the error is due to the fact that
>>> I've got the type of one of the entries wrong. Naturally I think
>>> I've got it right, and clearly I am wrong :-)
>>>
>>> Is there some way to turn up some debugging flags so I would know
>>> which one ?
>>>
>>> Any good stratgey for figuring this out ?
>>>
>>>
>
>
-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV