|
From: | JonY |
Subject: | Re: [Lzip-bug] [PATCH] lzip on Windows |
Date: | Tue, 13 Oct 2009 09:05:05 +0800 |
User-agent: | Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.1.12) Gecko/20080213 Thunderbird/2.0.0.12 Mnenhy/0.7.5.0 |
On 10/13/2009 03:33, Antonio Diaz Diaz wrote:
Dear Jon, Thank you for the patch, but I'm afraid I can't accept it by a number of reasons. It is too intrusive. I am pretty sure lines like this snprintf( buf, sizeof buf, "%"LZIPLL"d %s", num, p ); will break compilation on some other systems. It is for a system too different from posix systems, and there is at least a way of compiling lzip in windows that does not need to obscure the code with conditional compilation constructs; Cygwin.
Hi,It is also possible to use "%ll" on MinGW via MinGW printf wrappers, but it requires a fairly modern MinGW support CRT, possibly GCC too. The Windows specific _setmode() however, is still needed for piping to tar.
Do you recommend that I try to remove as much ifdefs as possible?
It is for a concrete compiler, MSVC. But you say that "it doesn't work with MSVC yet"!
No, I have not done work for a MSVC port yet, it will come later if needed, implementing a minimalist stdint.h for MSVC is not hard.
Right now it works with MinGW GCC.
Have you considered the possibility of maintaining a port? Ports for systems very different from the original help to maintain the upstream source clean and simple.
Do you mean maintaining local patches? If so, that is a possibility, but I would like to push changes upstream if possible.
I am relying on the test suite to tell me if lzip is actually working, any chance of adding a testsuite for lziprecover?
[Prev in Thread] | Current Thread | [Next in Thread] |