|
From: | Alejandro López-Valencia |
Subject: | Re: [Groff] win32 experiences? |
Date: | Wed, 27 Oct 2004 18:07:15 -0500 |
User-agent: | Mozilla/5.0 (Windows; U; Windows NT 5.0; rv:1.7.3) Gecko/20040913 Thunderbird/0.8 Mnenhy/0.6.0.104 |
On 27/10/2004 04:55 p.m., Jorgen Grahn wrote:
On Wed Oct 27 22:47:59 2004, address@hidden wrote:Alas, I have been forced to delete linux and switch to winbl0ze (if you
[snip]
Plain cygwin (http://.../setup.exe and so on) worked fine last time I tried. Although I suspect it lags behind with respect to recent groff releases.
Yup. I compile my own CVS snapshots and package them to be installed with setup.exe. The best of both worlds: Bleeding edge and standard package management.
My only positive experience with win32 groff so far was with a clean compile with help of cygwin (everything ran normal, just had to dl win32 joe editor because of the pesky CRLF pairs :-))I cannot recall having problems with that (i.e. cygwin's groff not liking CRLF endings). On the other hand, cygwin's CRLF handling is a bit unpredictable...
It is not groff (that is, troff and the post processors) that have problems with CRLF files. Rather, it is some of the preprocessors. E.g., tbl dies unceremoniously if given a CRLF file. Perhaps it should be considered a bug: some parts of groff are able to handle EOL transparently, while other parts don't.
A Cygwin specific solution is to have your working space in a text mount, which will make Cygwin translate EOLs from DOS to UNIX and back transparently (as in behind your back, will kill binary object files, so don't compile anything under such conditions). May or may not work, YMMV.
[Prev in Thread] | Current Thread | [Next in Thread] |