[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Groff] Pdfmark
From: |
Keith MARSHALL |
Subject: |
Re: [Groff] Pdfmark |
Date: |
Tue, 28 Sep 2004 16:42:54 +0100 |
Earlier, I wrote:
> After creating pdfmark-parts.mk, your make goes into a recursive
> loop. This would occur if PARTNAMES is not defined in the secondary
> make, and this definition is obtained by reading pdfmark-parts.mk,
> so the first thing to check is that its content is correct. You can
> persuade pdfmake to create just pdfmark-parts.mk, by executing:
>
> ./pdfmake TMACTYPE=ms pdfmark-parts.mk
>
> Now, inspect the content of pdfmark-parts.mk -- it should contain
> exactly one line:
>
> PARTNAMES = %-fp.ps %-toc.ps %-body.ps
>
> Once we've determined whether this is correct, or not, then we can
> start thinking about how to proceed further. (To speed things up,
> if pdfmark-parts.mk doesn't look correct, please post *both* it,
> and the output of "od -b pdfmark-parts.mk", so I can have a look).
I looked into this a bit deeper myself. The sed command I used to
create pdfmark-parts.mk assumes all lines end with LF, as they would
on *nix, but CR isn't handled at all. Indeed, even on *nix, the use
of \n in the input pattern expression isn't guaranteed to work with
all versions of sed -- not even all GNU versions, IIRC!
Please test the attached patch -- the *diff* version is against my
original unpatched tarball, the *patch* version should be used, if
you have already applied the update from yesterday, (which I posted
earlier today); do *not* apply both.
Best regards,
Keith.
pdfmark.tmac.diff-20040928.gz
Description: Binary data
pdfmark.tmac.patch-20040928.gz
Description: Binary data