cayenne-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Aristedes Maniatis <>
Subject Re: docbook - continued
Date Mon, 29 Aug 2011 00:30:32 GMT
On Sun Aug 28 23:40:28 2011, Andrus Adamchik wrote:
> On Aug 28, 2011, at 12:30 PM, Andrus Adamchik wrote:
>> I am less concerned about syntax coloring though, more about code blocks standing
out on the page via a different background color and/or border. Especially since the PDF version
is likely be black and white (?)
> Same thing for the tables. I just started on Appendix A (a list of Cayenne properties)
and the big font used for the<code>  blocks and the lack of borders and backgrounds
make them a bit unreadable in either HTML or PDF formats.
> Also even though we only have like 5% of the new docs ready, I think this maybe the 5%
that matters. I almost finished all the DI and ServerRuntime chapters. So we should consider
publishing this on the site and with the upcoming release.
> BTW, working on the docs now allows me to take a user view of the framework, and I am
noticing things that we can polish a bit. Two most obvious points:
> 1. DI property names per Appendix A (make them shorter and more symmetrical)
> 2. Default modeler object naming (UntitledDomain, UntiltledDomainMap) trickles to the
XML file naming, so we'd rather come up with something that makes sense. E.g. "project" for
domains (there can be only 1 per project in 3.1), "datamap", "datamap1", etc. for DataMaps.
These will result in cleaner file names like cayenne-project.xml, etc.).
> Andrus

Let's get what we have published and I can continue to work on the 
formatting to make it cleaner and more readable. There is no reason why 
the PDF cannot be coloured.

Please don't worry about the XML file naming. That will not be visible 
in the html or pdf output, so use whatever naming makes sense.

I'll try and make some time in the next week to review what you've 
already done and what I can do to make it prettier.

As for the publishing, I am not sure what the current Apache thinking 
is for best practice:

* Jenkins: publish -> p.a.o
* Build script on p.a.o
* Jenkins build, script on p.a.o to pull results



Aristedes Maniatis
GPG fingerprint CBFB 84B4 738D 4E87 5E5C  5EFA EF6A 7D2E 3E49 102A

View raw message