[GHC] #1614: Type checker does not use functional dependency to avoid ambiguity

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

[GHC] #1614: Type checker does not use functional dependency to avoid ambiguity

Simon Peyton Jones
Morabbin, what's the semantics of "bump"?   I don't know what you intend by it!

Simon

|  -----Original Message-----
|  From: ghc-tickets-bounces at haskell.org [mailto:ghc-tickets-bounces at haskell.org]
|  On Behalf Of GHC
|  Sent: 26 January 2013 22:12
|  Cc: ghc-tickets at haskell.org
|  Subject: Re: [GHC] #1614: Type checker does not use functional dependency to
|  avoid ambiguity
|  
|  #1614: Type checker does not use functional dependency to avoid ambiguity
|  --------------------------------------+-------------------------------------
|    Reporter:  guest                    |          Owner:
|        Type:  bug                      |         Status:  new
|    Priority:  normal                   |      Milestone:  _|_
|   Component:  Compiler (Type checker)  |        Version:  6.7
|  Resolution:                           |       Keywords:
|          Os:  Unknown/Multiple         |   Architecture:  Unknown/Multiple
|     Failure:  None/Unknown             |     Difficulty:  Unknown
|    Testcase:                           |      Blockedby:
|    Blocking:                           |        Related:
|  --------------------------------------+-------------------------------------
|  Changes (by morabbin):
|  
|    * failure:  => None/Unknown
|  
|  
|  Comment:
|  
|   Bump; chak saya this was fixed in HEAD, and then SPJ closes as wontfix?
|   Was the fix backed out?
|  
|  --
|  Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/1614#comment:11>
|  GHC <http://www.haskell.org/ghc/>
|  The Glasgow Haskell Compiler
|  
|  _______________________________________________
|  ghc-tickets mailing list
|  ghc-tickets at haskell.org
|  http://www.haskell.org/mailman/listinfo/ghc-tickets


Reply | Threaded
Open this post in threaded view
|

[GHC] #1614: Type checker does not use functional dependency to avoid ambiguity

Ian Lynagh-2
On Sat, Jan 26, 2013 at 11:15:56PM +0000, Simon Peyton-Jones wrote:
> Morabbin, what's the semantics of "bump"?   I don't know what you intend by it!

I think the intention is just to draw attention to the ticket.

By the way, thanks for looking at our old tickets! Adding a comment to a
ticket which looks like it it could now be easily closed, or where it is
unclear what the status is (as with this ticket), or where some other
easy action is needed, is very useful.

If there's a particular bug that's affecting you, then adding a comment
to it is also useful, as it helps us prioritise.

However, I think that just adding a comment like "Bump; still relevant."
such as in
    http://hackage.haskell.org/trac/ghc/ticket/1526#comment:7
is counter-productive, as all open tickets are assumed to still be
relevant by default, so it just generates mails that people then spend
time reading and deleting.


Thanks
Ian

> |  -----Original Message-----
> |  From: ghc-tickets-bounces at haskell.org [mailto:ghc-tickets-bounces at haskell.org]
> |  On Behalf Of GHC
> |  Sent: 26 January 2013 22:12
> |  Cc: ghc-tickets at haskell.org
> |  Subject: Re: [GHC] #1614: Type checker does not use functional dependency to
> |  avoid ambiguity
> |  
> |  #1614: Type checker does not use functional dependency to avoid ambiguity
> |  --------------------------------------+-------------------------------------
> |    Reporter:  guest                    |          Owner:
> |        Type:  bug                      |         Status:  new
> |    Priority:  normal                   |      Milestone:  _|_
> |   Component:  Compiler (Type checker)  |        Version:  6.7
> |  Resolution:                           |       Keywords:
> |          Os:  Unknown/Multiple         |   Architecture:  Unknown/Multiple
> |     Failure:  None/Unknown             |     Difficulty:  Unknown
> |    Testcase:                           |      Blockedby:
> |    Blocking:                           |        Related:
> |  --------------------------------------+-------------------------------------
> |  Changes (by morabbin):
> |  
> |    * failure:  => None/Unknown
> |  
> |  
> |  Comment:
> |  
> |   Bump; chak saya this was fixed in HEAD, and then SPJ closes as wontfix?
> |   Was the fix backed out?
> |  
> |  --
> |  Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/1614#comment:11>
> |  GHC <http://www.haskell.org/ghc/>
> |  The Glasgow Haskell Compiler


Reply | Threaded
Open this post in threaded view
|

[GHC] #1614: Type checker does not use functional dependency to avoid ambiguity

Simon Peyton Jones
In reply to this post by Simon Peyton Jones
Ah, so the mysterious and highly-active morabbin is none other than our old friend and drinking companion Andy Adams-Moran!  Welcome Andy! Thanks for looking through these old tickets.

|  "Bump" is a cutesy internet name for this kind of bubbling something up to the
|  top of the dev list. Please have another look at this ticket; here's what
| I see (or not), maybe we can get rid of it."

The trouble is that there aren't enough of these GHC devs to deal with the hundreds of open tickets, bumped or not.  When adding to a long-existing ticket, there are really three modes of operation

* This ticket is mission-critical to me; here's why; please please help.

* Here is some new information. Eg. I tripped over this myself; here is the repro case; or I think I may have an idea what's going on.

* I'm going to have ago at fixing this.  Here's how I propose to do it.  Please yell if I'm going awry here.

But your "bump" seems to mean "I've looked at this and it still seems to be open", and I don't know what to do about that.  I'd love to have them *all* attended to, but we need a lot more people to help if we are to achieve that.

To simply indicate interest, the protocol is to add yourself to the cc list.  Long cc lists get more attention. Well some more attention!  http://hackage.haskell.org/trac/ghc/wiki/WorkingConventions/BugTracker

Simon

|  -----Original Message-----
|  From: Andrew Adams-Moran [mailto:morabbin at gmail.com]
|  Sent: 26 January 2013 23:42
|  To: Simon Peyton-Jones
|  Cc: ghc-devs at haskell.org
|  Subject: Re: [GHC] #1614: Type checker does not use functional dependency to
|  avoid ambiguity
|  
|  
|  Hi Simon;
|  
|  "Bump" is a cutesy internet name for this kind of bubbling something up to the
|  top of the dev list.
|  
|      "Please have another look at this ticket; here's what I see (or not), maybe we
|  can get rid of it."
|  
|  In this case, it's also means that the commit is result of my ongoing attempt to
|  revivify the bug sweep. So folks should feel free to ignore.
|  
|  (Also, do let me know if this is counter to GHC conventions, and I'll adapt
|  accordingly.)
|  
|  Cheers,
|  
|  Andy
|  
|  On Jan 26, 2013, at 3:15 PM, Simon Peyton-Jones <simonpj at microsoft.com>
|  wrote:
|  
|  > Morabbin, what's the semantics of "bump"?   I don't know what you intend by
|  it!
|  >
|  > Simon
|  >
|  > |  -----Original Message-----
|  > |  From: ghc-tickets-bounces at haskell.org [mailto:ghc-tickets-
|  bounces at haskell.org]
|  > |  On Behalf Of GHC
|  > |  Sent: 26 January 2013 22:12
|  > |  Cc: ghc-tickets at haskell.org
|  > |  Subject: Re: [GHC] #1614: Type checker does not use functional dependency
|  to
|  > |  avoid ambiguity
|  > |
|  > |  #1614: Type checker does not use functional dependency to avoid ambiguity
|  > |  --------------------------------------+-------------------------------------
|  > |    Reporter:  guest                    |          Owner:
|  > |        Type:  bug                      |         Status:  new
|  > |    Priority:  normal                   |      Milestone:  _|_
|  > |   Component:  Compiler (Type checker)  |        Version:  6.7
|  > |  Resolution:                           |       Keywords:
|  > |          Os:  Unknown/Multiple         |   Architecture:  Unknown/Multiple
|  > |     Failure:  None/Unknown             |     Difficulty:  Unknown
|  > |    Testcase:                           |      Blockedby:
|  > |    Blocking:                           |        Related:
|  > |  --------------------------------------+-------------------------------------
|  > |  Changes (by morabbin):
|  > |
|  > |    * failure:  => None/Unknown
|  > |
|  > |
|  > |  Comment:
|  > |
|  > |   Bump; chak saya this was fixed in HEAD, and then SPJ closes as wontfix?
|  > |   Was the fix backed out?
|  > |
|  > |  --
|  > |  Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/1614#comment:11>
|  > |  GHC <http://www.haskell.org/ghc/>
|  > |  The Glasgow Haskell Compiler
|  > |
|  > |  _______________________________________________
|  > |  ghc-tickets mailing list
|  > |  ghc-tickets at haskell.org
|  > |  http://www.haskell.org/mailman/listinfo/ghc-tickets
|  >
|  > _______________________________________________
|  > ghc-devs mailing list
|  > ghc-devs at haskell.org
|  > http://www.haskell.org/mailman/listinfo/ghc-devs



Reply | Threaded
Open this post in threaded view
|

[GHC] #1614: Type checker does not use functional dependency to avoid ambiguity

Simon Peyton Jones
Please do endanger the code base!  GHC needs people willing to have a go.  

* We try to review offered patches promptly (yell if this does not happen)

* Worth discussing the general approach before investing a lot of time;
   the more substantial the change/fix the more discussion!

Simon
|  -----Original Message-----
|  From: Andrew Adams-Moran [mailto:morabbin at gmail.com]
|  Sent: 27 January 2013 00:19
|  To: Simon Peyton-Jones
|  Cc: ghc-devs at haskell.org
|  Subject: Re: [GHC] #1614: Type checker does not use functional dependency to
|  avoid ambiguity
|  
|  
|  Hi all!
|  
|  Thanks Simon; I figured this was good way for me to learn the conventions of the
|  team without endangering the code base.
|  
|  Thanks for the feedback, and consider it taken on board.
|  
|  Cheers,
|  
|  Andy
|  
|  On Jan 26, 2013, at 4:11 PM, Simon Peyton-Jones <simonpj at microsoft.com>
|  wrote:
|  
|  > Ah, so the mysterious and highly-active morabbin is none other than our old
|  friend and drinking companion Andy Adams-Moran!  Welcome Andy! Thanks for
|  looking through these old tickets.
|  >
|  > |  "Bump" is a cutesy internet name for this kind of bubbling something up to
|  the
|  > |  top of the dev list. Please have another look at this ticket; here's what
|  > | I see (or not), maybe we can get rid of it."
|  >
|  > The trouble is that there aren't enough of these GHC devs to deal with the
|  hundreds of open tickets, bumped or not.  When adding to a long-existing ticket,
|  there are really three modes of operation
|  >
|  > * This ticket is mission-critical to me; here's why; please please help.
|  >
|  > * Here is some new information. Eg. I tripped over this myself; here is the repro
|  case; or I think I may have an idea what's going on.
|  >
|  > * I'm going to have ago at fixing this.  Here's how I propose to do it.  Please yell
|  if I'm going awry here.
|  >
|  > But your "bump" seems to mean "I've looked at this and it still seems to be
|  open", and I don't know what to do about that.  I'd love to have them *all*
|  attended to, but we need a lot more people to help if we are to achieve that.
|  >
|  > To simply indicate interest, the protocol is to add yourself to the cc list.  Long cc
|  lists get more attention. Well some more attention!
|  http://hackage.haskell.org/trac/ghc/wiki/WorkingConventions/BugTracker
|  >
|  > Simon
|  >
|  > |  -----Original Message-----
|  > |  From: Andrew Adams-Moran [mailto:morabbin at gmail.com]
|  > |  Sent: 26 January 2013 23:42
|  > |  To: Simon Peyton-Jones
|  > |  Cc: ghc-devs at haskell.org
|  > |  Subject: Re: [GHC] #1614: Type checker does not use functional dependency
|  to
|  > |  avoid ambiguity
|  > |
|  > |
|  > |  Hi Simon;
|  > |
|  > |  "Bump" is a cutesy internet name for this kind of bubbling something up to
|  the
|  > |  top of the dev list.
|  > |
|  > |      "Please have another look at this ticket; here's what I see (or not), maybe
|  we
|  > |  can get rid of it."
|  > |
|  > |  In this case, it's also means that the commit is result of my ongoing attempt
|  to
|  > |  revivify the bug sweep. So folks should feel free to ignore.
|  > |
|  > |  (Also, do let me know if this is counter to GHC conventions, and I'll adapt
|  > |  accordingly.)
|  > |
|  > |  Cheers,
|  > |
|  > |  Andy
|  > |
|  > |  On Jan 26, 2013, at 3:15 PM, Simon Peyton-Jones <simonpj at microsoft.com>
|  > |  wrote:
|  > |
|  > |  > Morabbin, what's the semantics of "bump"?   I don't know what you intend
|  by
|  > |  it!
|  > |  >
|  > |  > Simon
|  > |  >
|  > |  > |  -----Original Message-----
|  > |  > |  From: ghc-tickets-bounces at haskell.org [mailto:ghc-tickets-
|  > |  bounces at haskell.org]
|  > |  > |  On Behalf Of GHC
|  > |  > |  Sent: 26 January 2013 22:12
|  > |  > |  Cc: ghc-tickets at haskell.org
|  > |  > |  Subject: Re: [GHC] #1614: Type checker does not use functional
|  dependency
|  > |  to
|  > |  > |  avoid ambiguity
|  > |  > |
|  > |  > |  #1614: Type checker does not use functional dependency to avoid
|  ambiguity
|  > |  > |  --------------------------------------+-------------------------------------
|  > |  > |    Reporter:  guest                    |          Owner:
|  > |  > |        Type:  bug                      |         Status:  new
|  > |  > |    Priority:  normal                   |      Milestone:  _|_
|  > |  > |   Component:  Compiler (Type checker)  |        Version:  6.7
|  > |  > |  Resolution:                           |       Keywords:
|  > |  > |          Os:  Unknown/Multiple         |   Architecture:  Unknown/Multiple
|  > |  > |     Failure:  None/Unknown             |     Difficulty:  Unknown
|  > |  > |    Testcase:                           |      Blockedby:
|  > |  > |    Blocking:                           |        Related:
|  > |  > |  --------------------------------------+-------------------------------------
|  > |  > |  Changes (by morabbin):
|  > |  > |
|  > |  > |    * failure:  => None/Unknown
|  > |  > |
|  > |  > |
|  > |  > |  Comment:
|  > |  > |
|  > |  > |   Bump; chak saya this was fixed in HEAD, and then SPJ closes as
|  wontfix?
|  > |  > |   Was the fix backed out?
|  > |  > |
|  > |  > |  --
|  > |  > |  Ticket URL:
|  <http://hackage.haskell.org/trac/ghc/ticket/1614#comment:11>
|  > |  > |  GHC <http://www.haskell.org/ghc/>
|  > |  > |  The Glasgow Haskell Compiler
|  > |  > |
|  > |  > |  _______________________________________________
|  > |  > |  ghc-tickets mailing list
|  > |  > |  ghc-tickets at haskell.org
|  > |  > |  http://www.haskell.org/mailman/listinfo/ghc-tickets
|  > |  >
|  > |  > _______________________________________________
|  > |  > ghc-devs mailing list
|  > |  > ghc-devs at haskell.org
|  > |  > http://www.haskell.org/mailman/listinfo/ghc-devs
|  >