Changes between Version 1 and Version 2 of ExplicitCallStack


Ignore:
Timestamp:
Nov 15, 2006 10:50:44 AM (8 years ago)
Author:
simonpj
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • ExplicitCallStack

    v1 v2  
    33There has been a vigorous thread on error attribution ("I get a `head []` error; but who called `head`?")  
    44 
    5 This page summarises some half baked ideas that Simon and I have been discussing. 
     5This page summarises some half baked ideas that Simon and I have been discussing. See also 
     6  * [http://www.haskell.org/pipermail/haskell-cafe/2006-November/019549.html The Haskell cafe thread] 
     7  * [http://www.cse.unsw.edu.au/~dons/loch.html] 
     8 
    69 
    710== The basic idea ==  
     
    1316to pass a string describing the current location to f. 
    1417 
    15 2.  But that doesn't help with 'head'.  We want to pass head's site to head. That's what jhc does when you give 'head' the a magic SRCLOC_ANNOTATE pragma: 
     182.  But that doesn't help with 'head'.  We want to pass head's site to head. That's what jhc does when you give 'head' the a magic [http://repetae.net/john/computer/jhc/jhc.html SRCLOC_ANNOTATE pragma]: 
    1619        * every call to `head` gets replaced with `head_check $currentLocation` 
    1720        * you get to write `head_check` yourself, with type