David Crossley wrote:
Jean T. Anderson wrote:

Jean T. Anderson wrote:

... I promise to post an 'svn diff' later today for site-publish\skin\screen.css -- has been too busy to get to yet. If anyone has opinions about which file(s) I should do 'svn diff' on I'll be happy to do that(those).

site-publish\skin\basic.css was smaller so I did that diff instead. It looks like the entire file is duplicated (157 lines in the file, 319 lines in the diff).


Hmm, that sounds like a line-endings problem.

This file is copied from the forrest core during
the build.

I just verified that it has "svn:eol-style : native"
and no dos line-endings. main/webapp/skins/pelt/css/basic.css

Ok, I've been able to verify with mks od that the problem files (basic.css, print.css, screen.css) are getting an extra \r in the end-of-line, but profile.css is ok.

here's a dump of the first part of the source file:

C:\Apache\forrest_07_branch\main\webapp\skins\pelt\css>od -c
basic.css
0000000000     /   *  \r  \n   *       C   o   p   y   r   i   g   h   t
0000000020     2   0   0   2   -   2   0   0   5       T   h   e       A p
0000000040     a   c   h   e       S   o   f   t   w   a   r   e       F o
0000000060     u   n   d   a   t   i   o   n       o   r       i   t   s
0000000100     l   i   c   e   n   s   o   r   s   ,  \r  \n   *       a s
0000000120         a   p   p   l   i   c   a   b   l   e   .  \r  \n   *
\r
0000000140    \n   *       L   i   c   e   n   s   e   d       u   n   d e
0000000160     r       t   h   e       A   p   a   c   h   e       L   i c
0000000200     e   n   s   e   ,       V   e   r   s   i   o   n       2
... .

And here's the file after it has been copied to site-publish:

C:\Apache\incubator\site-publish\skin>od -c basic.css
0000000000     /   *  \r  \r  \n   *       C   o   p   y   r   i   g   h t
0000000020         2   0   0   2   -   2   0   0   5       T   h   e A
0000000040     p   a   c   h   e       S   o   f   t   w   a   r   e F
0000000060     o   u   n   d   a   t   i   o   n       o   r       i   t s
0000000100         l   i   c   e   n   s   o   r   s   ,  \r  \r  \n   *
0000000120     a   s       a   p   p   l   i   c   a   b   l   e   .  \r
\r
0000000140    \n   *  \r  \r  \n   *       L   i   c   e   n   s   e   d
0000000160     u   n   d   e   r       t   h   e       A   p   a   c   h e
0000000200         L   i   c   e   n   s   e   ,       V   e   r   s   i o
0000000220     n       2   .   0       (   t   h   e       "   L   i   c e
...


Is this FOR-492 ?

It's odd that profile.css is ok, but I notice that the source is profile.css.xslt, so it gets processing that the others don't.


Ah, i wonder. How did you obtain this forrest
installation on each platform? No, that will not
be it. You must have an SVN checkout, because we
use forrest_07_branch. I was wondering if it was
the packaged release and you got the *.tar.gz instead
of the *.zip or vice-versa. But no.


I also did a diff on projects\derby.html because to my eye it exhibits something different (595 lines in the file, 54 lines in the diff). There are entries such as these:

  -<a href="#Produce+a+release">Produce a release
  +<a href="#Produce+a+release">Produce a release&#13;

  -<h3 class="underlined_5">Project Setup
  +<h3 class="underlined_5">Project Setup&#13;


Ah, that is a good clue. I note that the source for
that doc is derby.cwiki ... see if it makes a difference
to put a blank line after the heading marker: ! Produce a...

Now the diff shows this:

  -<h4>Produce a release
  -</h4>
  +<h4>Produce a release</h4>

Windows just wants to be different.  :-)

 -jean