[GHC] #1770: GHCi should know about itself,
so that .ghci can be portable
GHC
trac at galois.com
Tue Nov 6 11:09:14 EST 2007
#1770: GHCi should know about itself, so that .ghci can be portable
-----------------------------+----------------------------------------------
Reporter: guest | Owner:
Type: feature request | Status: new
Priority: normal | Milestone: 6.10 branch
Component: GHCi | Version: 6.6.1
Severity: normal | Resolution:
Keywords: | Difficulty: Unknown
Testcase: | Architecture: Unknown
Os: Unknown |
-----------------------------+----------------------------------------------
Comment (by guest):
nice one, thanks!-) i knew about `System.Info`, but only for os-related
information, not for compiler-related stuff. since none of the 100 or so
downloaders of my `.ghci` file mentioned this to me, i guess i'm not the
only one in the dark?
ok, that changes the picture slightly: versioned command definitions for
ghci are still a bit tricky, but possible. if we leave advanced things
like `has` and `exists` for later, that leaves us with `docdir`, `libdir`,
and `info`.
the `System.Info` has the problem that it tells me what, but not where,
about the compiler. one might either extend it with a compiler-specific
`compilerinfo`, from which one could then extract the details directly, or
one might extend it with a `compilerPath`, so that one could ask the
compiler for more details.
unfortunately, either option suffers from issues with relocated ghc
installs. is there any reliable way for a ghc-compiled program to get the
path of the ghc that compiles it?
this is no longer really ghci-specific, eg, users of the ghc api need to
find a suitable value for topdir, which could be extracted from
`{compilerPath} --print-libdir`.
--
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/1770#comment:2>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler
More information about the Glasgow-haskell-bugs
mailing list