Re: Beta 1 completion items

[Marty K. <mrk...@co...>]

On 11/29/2011 06:20 AM, Greg White wrote:
> Hi Marty,
>
> I think we're talking at cross purposes? I am referring to an objective I've been referring to some time, for splitting off the javadoc documentation of pvDataJava, pvAccessJava and pvDataJava into 2 separate systems of documents. Firstly, the exiting javadoc. Second, a new document called something like "EPICS V4 technical guide". This technical guide would just be started by concatenating the exiting overview docs, *but is not intended to be automatically generated. It's just one of our normative documents, like ay other.
>

OK but I must stay focused.

I will think about my priorities after I have what I have to do for 
masar so that I can leave the rest to Guobao.

I have NTNameValuePair and NTTable ready.
I am working on generic server support for channelRPC in pvIOCCPP.
It is working but I have some clean up to do.

NOTE that this also satisfies one generic goal: Make it easy to 
implement server side of  C++ services.
This is for channelRPC. I can also see a way to make channelPutGet easy 
to use.
When both are ready then pvIOCCPP support for implementing services will 
be ready.

Then I have to make it easy for client side of masar to use channelRPC 
to talk to server.
Later this should be part of ezpva but for now it will just be part of 
masar.

Then I have to implement what is like local gatherGet.

Then I will be done with what is required from me for masar and can 
think about my priorities.

Marty
> On your questions of automating production though, I've seen that maven is producing javadoc. See the "gettingStarted" I pushed yesterday [1] for how to build all core EPICS V4 using maven rather than eclipse. I think the javadoc product of
> these maven builds is published automatically now to our sourceForge web site? *Matej* - is that right? Sorry, don't have time to check myself right now. Eve if it isn't, it clearly can be.
>
> Cheers
> Greg
>
> [1] http://epics-pvdata.sourceforge.net/gettingStarted.txt
>
>
>
> On 29 Nov 2011, at 11:41, Marty Kraimer wrote:
>
>> On 11/29/2011 05:14 AM, Greg White wrote:
>>> Marty, re items that would constitute a beta 1 version of EPICS V4 as a whole [1], how should we proceed with the line item "+ pvData and pvIOC documentation separate to JavaDoc (Marty, Greg)." I suggested we copy out the Overviews and put them in a separate (normative) document, and the edit down the actual overview.htmls so they're summaries.
>>>
>>> Sound good? When? Shall I extract the overviews and paste them together, and you précis the existing overviews?
>>>
>>> Cheers
>>> Greg
>>>
>>> [1] http://sourceforge.net/mailarchive/message.php?msg_id=28436546
>>>
>>>
>>>
>> Here is the current status of the javaDOC.
>>
>> In each of pvDataJava, pvAccessJava, and pvIOCJava:
>>
>> In src there is an overview.html.
>>
>> In some packages there is a package.html.
>>
>> To prepare pure javaDOC I have been using eclipse.
>>
>> I click on project/"Generate JavaDOC"
>>
>> I select the project and then next
>> use defaults and then next
>> check overview AND pick correct overview.html file. It remembers the
>> last so must check this every time.
>> click on finish.
>>
>> So what to do to automate?
>>
>> I will guess that maven can be used to automate this process for each
>> project.
>> It can select a target location. But what?
>> It should be somewhere so that it can automatically copied to the
>> sourceforge WEB site.
>> But where?
>>
>> This does not handle having JUST overview  and package docs.
>>
>> What to do?
>>
>> Can a tool automatically copy the overview.html and the package.html
>> files somewhere and create an index,html file that has links to each
>> copied file?
>> ALSO copy any .gif files.
>>
>> The destination should also be somewhere which can automatically copied
>> to the sourceforge WEB site.
>> But where
>>
>> If this is done what will be produced will be fine except for a few
>> broken links.
>> That would be a good start.
>>
>> The broken links can be fixed by just changing the overview.html and
>> package.html files to get rid of the bad links.
>>
>> Marty
>>
>>
>>
>> ------------------------------------------------------------------------------
>> All the data continuously generated in your IT infrastructure
>> contains a definitive record of customers, application performance,
>> security threats, fraudulent activity, and more. Splunk takes this
>> data and makes sense of it. IT sense. And common sense.
>> http://p.sf.net/sfu/splunk-novd2d
>