zlib 0.5.3.2 broken?

Hi all, I've received a number of reports of a broken zlib 0.5.3.2 (and experienced it myself). Is this a generally known issue? As a temporary workaround, I've released a new version of zlib-bindings that has an upper bound on zlib to avoid 0.5.3.2 (and later). Once a new patched version is released, I'll release an updated version of zlib-bindings to remove that overly restrictive upper bound. Michael

I recently ran into this as well. I found this stack overflow question
[1], where Daniel Fischer notes that a proper solution has been found,
and it shouldn't be long until it reaches hackage. That was one and a
half weeks ago.
Erik
[1] http://stackoverflow.com/questions/8961413/zlib-build-error-with-ghc
On Thu, Feb 2, 2012 at 06:59, Michael Snoyman
Hi all,
I've received a number of reports of a broken zlib 0.5.3.2 (and experienced it myself). Is this a generally known issue?
As a temporary workaround, I've released a new version of zlib-bindings that has an upper bound on zlib to avoid 0.5.3.2 (and later). Once a new patched version is released, I'll release an updated version of zlib-bindings to remove that overly restrictive upper bound.
Michael
_______________________________________________ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe

I had an issue as well and the explanation is on StackOverflow, so I
suppose it is a known issue:
http://stackoverflow.com/questions/8961413/zlib-build-error-with-ghc
Putting a double # in front of the offending preprocessor directives
fixed it for me.
JP
On Thu, Feb 2, 2012 at 6:59 AM, Michael Snoyman
Hi all,
I've received a number of reports of a broken zlib 0.5.3.2 (and experienced it myself). Is this a generally known issue?
As a temporary workaround, I've released a new version of zlib-bindings that has an upper bound on zlib to avoid 0.5.3.2 (and later). Once a new patched version is released, I'll release an updated version of zlib-bindings to remove that overly restrictive upper bound.
Michael
_______________________________________________ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
-- JP Moresmau http://jpmoresmau.blogspot.com/

On Thu, 2 Feb 2012 07:59:08 +0200, Michael Snoyman
Hi all,
I've received a number of reports of a broken zlib 0.5.3.2 (and experienced it myself). Is this a generally known issue?
As a temporary workaround, I've released a new version of zlib-bindings that has an upper bound on zlib to avoid 0.5.3.2 (and later). Once a new patched version is released, I'll release an updated version of zlib-bindings to remove that overly restrictive upper bound.
Michael
Hi Michael, zlib 0.5.3.2 builds fine, with ghc 7.0.4 Cabal-1.10.2.0 it just needs some help to find the header file containing the macros generated by cabal. To avoid the error: Stream.hsc:86:21: error: missing binary operator before token "(" in the Gentoo Haskell overlay I pass this extra option to the runghc ./Setup.hs configure: --hsc2hs-options=--include=${S}/dist/build/autogen/cabal_macros.h where ${S} is the source code directory. With ghc 7.2.2 and 7.4.1-rc2 it compiles fine with or without this configure option. Mark

bzlib-0.5.0.2 suffers from the exact same problem. I send a bug report to the author a few days ago, but I can imagine he's very busy. It might help if we can send patches that fix the compile error.
participants (5)
-
Erik Hesselink
-
JP Moresmau
-
Mark Wright
-
Michael Snoyman
-
Roel van Dijk