<div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr">Right, I see, thanks.<div><br></div><div>This is what I was attempting so far:</div><div><br></div><div><div>data DsMessage =</div><div> DsUnknownMessage !DiagnosticMessage</div><div> | DsLiftedTcRnMessage !TcRnMessage</div><div> -- ^ A diagnostic coming straight from the Typecheck-renamer.</div></div><div><br></div><div>and later:</div><div><br></div><div><div>liftTcRnMessages :: MonadIO m => IORef (Messages TcRnMessage) -> m (IORef (Messages DsMessage))</div><div>liftTcRnMessages ref = liftIO $ do</div><div> oldContent <- readIORef ref</div><div> newIORef (DsLiftedTcRnMessage <$> oldContent)</div></div><div><br></div><div>...</div><div><br></div><div><div>mkDsEnvsFromTcGbl :: MonadIO m</div><div> => HscEnv -> IORef (Messages TcRnMessage) -> TcGblEnv</div><div> -> m (DsGblEnv, DsLclEnv)</div><div>mkDsEnvsFromTcGbl hsc_env msg_var tcg_env</div><div> = do { cc_st_var <- liftIO $ newIORef newCostCentreState</div><div> ...</div><div> ; msg_var' <- liftTcRnMessages msg_var</div><div> ; return $ mkDsEnvs unit_env this_mod rdr_env type_env fam_inst_env</div><div> msg_var' cc_st_var complete_matches</div><div> }</div></div><div><br></div><div><br></div><div>While this typechecks, I wonder if that's the right way to think about it -- from your reply, it seems like the dependency is in the opposite direction -- we need to store desugaring diagnostics in the TcM due to TH splicing, not the other way around.</div><div><br></div><div>I'll explore the idea of adding a second IORef.</div><div><br></div><div>Thanks!</div><div><br></div></div></div></div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Tue, 30 Mar 2021 at 10:51, Simon Peyton Jones <<a href="mailto:simonpj@microsoft.com">simonpj@microsoft.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-style:solid;border-left-color:rgb(204,204,204);padding-left:1ex">
<div lang="EN-GB" style="word-wrap:break-word">
<div class="gmail-m_5021421256253852227WordSection1">
<p class="MsoNormal"><span>I think the main reason is that for Template Haskell the renamer/type-checker need to run the desugarer. See the call to initDsTc in GHC.Tc.Gen.Splice.<u></u><u></u></span></p>
<p class="MsoNormal"><span><u></u> <u></u></span></p>
<p class="MsoNormal"><span>I suppose an alternative is that the TcGblEnv could have a second IORef to use for error messages that come from desugaring during TH splices.<u></u><u></u></span></p>
<p class="MsoNormal"><span><u></u> <u></u></span></p>
<p class="MsoNormal"><span>Nothing deeper than that I think.<u></u><u></u></span></p>
<p class="MsoNormal"><span><u></u> <u></u></span></p>
<p class="MsoNormal"><span>Simon<u></u><u></u></span></p>
<p class="MsoNormal"><span><u></u> <u></u></span></p>
<div style="border-style:none none none solid;border-left-width:1.5pt;border-left-color:blue;padding:0cm 0cm 0cm 4pt">
<div>
<div style="border-style:solid none none;border-top-width:1pt;border-top-color:rgb(225,225,225);padding:3pt 0cm 0cm">
<p class="MsoNormal"><b><span lang="EN-US">From:</span></b><span lang="EN-US"> ghc-devs <<a href="mailto:ghc-devs-bounces@haskell.org" target="_blank">ghc-devs-bounces@haskell.org</a>>
<b>On Behalf Of </b>Alfredo Di Napoli<br>
<b>Sent:</b> 30 March 2021 08:42<br>
<b>To:</b> Simon Peyton Jones via ghc-devs <<a href="mailto:ghc-devs@haskell.org" target="_blank">ghc-devs@haskell.org</a>><br>
<b>Subject:</b> Why TcLclEnv and DsGblEnv need to store the same IORef for errors?<u></u><u></u></span></p>
</div>
</div>
<p class="MsoNormal"><u></u> <u></u></p>
<div>
<div>
<div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
Hello folks,<u></u><u></u></p>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
<u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
as some of you might know me and Richard are reworking how GHC constructs, emits and deals with errors and warnings (See <a href="https://nam06.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgitlab.haskell.org%2Fghc%2Fghc%2F-%2Fwikis%2FErrors-as-(structured)-values&data=04%7C01%7Csimonpj%40microsoft.com%7C49c033aa2865495eb07c08d8f34f70cd%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C637526870280012102%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=ui4JWOp1gl5Yh%2FOYDqcBLXxTm%2FGnQRi0cDshUVEjwmk%3D&reserved=0" target="_blank">https://gitlab.haskell.org/ghc/ghc/-/wikis/Errors-as-(structured)-values</a>
and #18516).<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
<u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
To summarise very briefly the spirit, we will have (eventually) proper domain-specific types instead of SDocs. The idea is to have very precise and "focused" types for the different phases of the compilation pipeline, and a "catch-all" monomorphic `GhcMessage`
type used for the final pretty-printing and exception-throwing:<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
<u></u> <u></u></p>
</div>
<div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
data GhcMessage where<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
GhcPsMessage :: PsMessage -> GhcMessage<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
GhcTcRnMessage :: TcRnMessage -> GhcMessage<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
GhcDsMessage :: DsMessage -> GhcMessage<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
GhcDriverMessage :: DriverMessage -> GhcMessage<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
GhcUnknownMessage :: forall a. (Diagnostic a, Typeable a) => a -> GhcMessage<u></u><u></u></p>
</div>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
<u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
While starting to refactor GHC to use these types, I have stepped into something bizarre: the `DsGblEnv` and `TcLclEnv` envs both share the same `IORef` to store the diagnostics (i.e. warnings and errors) accumulated during compilation. More specifically, a
function like `GHC.HsToCore.Monad.mkDsEnvsFromTcGbl` simply receives as input the `IORef` coming straight from the `TcLclEnv`, and stores it into the `DsGblEnv`.<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
<u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
This is unfortunate, because it would force me to change the type of this `IORef` to be<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
`IORef (Messages GhcMessage)` to accommodate both diagnostic types, but this would bubble up into top-level functions like `initTc`, which would now return a `Messages GhcMessage`. This is once again unfortunate, because is "premature": ideally it might still
be nice to return `Messages TcRnMessage`, so that GHC API users could get a very precise diagnostic type rather than the bag `GhcMessage` is. It also violates an implicit contract: we are saying that `initTc` might return (potentially) *any* GHC diagnostic
message (including, for example, driver errors/warnings), which I think is misleading.<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
<u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
Having said all of that, it's also possible that returning `Messages GhcMessage` is totally fine here and we don't need to be able to do this fine-grained distinction for the GHC API functions. Regardless, I would like to ask the audience:<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
<u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
* Why `TcLclEnv` and `DsGblEnv` need to share the same IORef?<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
* Is this for efficiency reasons?<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
* Is this because we need the two monads to independently accumulate errors into the<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
same IORef?<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
<u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
Thanks!<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
<u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
Alfredo<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
<u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
<u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
<u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
<u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
<u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
<u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
<u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
<u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
<u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
<u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
<u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal" style="margin-right:0cm;margin-bottom:6pt;margin-left:0cm">
<u></u> <u></u></p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</blockquote></div>