forrest-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From David Crossley <>
Subject Re: Odd 'diffs' when building incubator site
Date Mon, 23 Jan 2006 02:35:12 GMT
Jean T. Anderson wrote:
> This will demonstrate how sadly far behind I am on emails.
> I'd like to resurrect one issue from the post at 

> . The original subject was "Re: [Fwd: [ Re: [RT] Super 
> Simple Site Generation Tool]]"
> Ross Gardler wrote on Sat, 31 Dec 2005:
> ><vast snippage> 
> > Leo Simons wrote:
> ...
> > > $ svn status
> > > ?      forrest_07_branch
> > > M      site-author/index.html
> > > M      site-publish/index.pdf
> > > M      site-publish/projects/axion.pdf
> > > M      site-publish/incubation/incubation-process.png
> > > M      site-publish/index.html
> > > $ svn diff site-publish/index.html
> > > ...
> > > -<a name="N1002D"></a><a name="ApacheCon+US+2005"></a>
> > > +<a name="N10033"></a><a name="ApacheCon+US+2005"></a>
> > > ...
> > >
> > > Huh? Why such a big diff? I can sort-of understand why the index.* files
> > > have been modified (though this isn't exactly a pretty diff to look at)
> > > but what's with the image or the axion.pdf? Moreoever, why on earth did
> > > it spend 5 minutes generating files that are now exactly the same as
> > > before?? Abort, abort, abort.

I think that i commented on some of that elsewhere.

My guess is that adding a new section, or perhaps
changing the section name, causes the auto-generated
anchors to be different.

> >I can't explain the axion and png differences. Perhaps this is a badly 
> >configured SVN setup since both of these are binary files.
> I can explain the png differences (though I can't fix them). And it's 
> possible David already explained this in a followup; if so, my apologies.
> I mentioned some operating system differences when building forrest on 
> the dev list last August: 

> .

Ah, one of the many things that i intended to
follow up on. Sorry.

> A succinct summary is I use Linux and everyone else on Derby uses 
> Windows. Derby uses the official 0.7 release. I always see diff's on 
> linux with png files generated from aart and a handful of the 
> site/skin/*.css files. Infrequently I also see unexplainable diff's with 
> other files of any type, such as html, zip, pdf. Since everyone else is 
> on Windows, I just don't commit those files that don't make sense.
> I should have followed up and filed a Jira issue, but didn't see it as 
> that big of a deal. But now I'm feeling guilty because I've started 
> helping David publish incubator changes and do see diffs on the 
> incubator site that don't make sense, so understand Leo's confusion.

Interesting, now i am seeing some diffs following
your recent commits at Incubator. I am on Mac OS X.
I saw diffs with axion.pdf and incubation-process.png

> Using the forrest_07_branch I don't see diffs with the css files that I 
> see with the official 0.7 release, but I do see diffs with many png's 
> under site-publish/skin/images. And I occasionally do see diffs with 
> other files.

Would you please provide a list of the images problems.
I have seen them occur elsewhere but never been able
to track down the cause.

> (Initially I also saw huge diffs when my environment 
> defaulted to the formal 0.7 release that I use for building derby, but 
> now that I know what that looks like I won't make that mistake again).
> I can think of these reasons for encountering unexplainable diffs:
> - Operating system differences with generation of the png or other files

There should not be.

We do have one particularly nasty issue for Windows users:
 "Inconsistent Line Endings in generated sites"

> - Somebody modified source but didn't publish the changes, so the next
> person somes along and sees unexpected changes on the publish side.

Yes always a problem. This can usually be seen by following
the svn commits mailing list for the particular project.

> - Differences when using a different release of forrest (in my case this 
> produced such a huge change I knew something was wrong)

We are trying to minimise such changes in output
formatting, but that will still be a problem.

- Another issue that causes line-endings problems
arises when committers do not configure their svn client.

> I don't know that this info helps the incubator, but it might help some 
> other project that has multiple developers on different operating 
> systems working on the same web site.

This is all a huge problem on opensource projects where
committers are on such a variety of operating systems
and have varying quality of SVN client configuration.


View raw message