[kepler-dev] Re: [seek-dev] Query Builder Code Review

Fri Aug 20 06:07:09 PDT 2004

Bertram brought up a good point, although the QB converts cross table 
references into joins it doesn't do anything with a join in the lower 
panel in the Adv tab. I will create a bug for that.

Rod

Bertram Ludaescher wrote:

>If I'm not mistaken, in QBE joins can be simply indicated by the use
>of the same variable in multiple places.
>
>Projections are indicated by a .P 
>
>Here are some notes/intros:
>
>http://www.mills.edu/ACAD_INFO/MCS/CS/S99MCS187/notes/l10
>http://www.cs.wisc.edu/~dbbook/openAccess/firstEdition/slides/pdfslides/mod3l6.pdf
>
>This "semi-graphical" version (actually only tables are needed -- even 
>HTML could do) might be yet another tab, say "old-fashioned (aka power 
>user ;)" 
>
>just a thought...
>
>Bertram
>
>  
>
>>>>>>"RS" == Rod Spears <rods at ku.edu> writes:
>>>>>>            
>>>>>>
>RS> 
>RS> I have put together an overview document <http://129.237.201.166/QB/> to 
>RS> help with the code review.
>RS> 
>RS> At Ricardo's suggestion, I used the "Maven" tool to generate a source 
>RS> code listing, a cross-reference, metrics and JavaDoc 
>RS> <http://129.237.201.166/QB/index.html#Query_Builder_DesignArchitecture>.
>RS> 
>RS> The links in that section should be enough to do the code review. When 
>RS> looking at the metrics make sure you familiarize yourself with the 
>RS> various terms, some mean the opposite that you might think.
>RS> 
>RS> Also, there seems to be some varying feelings and opinions about the 
>RS> coding style for Kepler. My approach which I am sure differs from Kepler 
>RS> is as follows:
>RS> 
>RS> 1) All method arguments start with a lowercase "a"
>RS> 2) All class data members all start with "m" (unless they are final)
>RS> 3) All local variables have no prefix.
>RS> 
>RS> It is my belief that you should always be able to look at a variable in 
>RS> the code and understand it's scope. I disagree with the approach where 
>RS> both the method arguments and the local variables have no prefix and are 
>RS> often distinguished by a "this.", which is rather verbose at times.
>RS> 
>RS> 
>RS> Rod
>RS> 
>RS> <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
>RS> <html>
>RS> <head>
>RS>   <meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
>RS>   <title></title>
>RS> </head>
>RS> <body bgcolor="#ffffff" text="#000000">
>RS> <meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
>RS> <title></title>
>RS> I have put together an overview <a href="http://129.237.201.166/QB/">document</a>
>RS> to help with the code review.<br>
>RS> <br>
>RS> At Ricardo's suggestion, I used the "Maven" tool to generate a <a
>RS>  href="http://129.237.201.166/QB/index.html#Query_Builder_DesignArchitecture">source
>RS> code listing, a cross-reference, metrics and JavaDoc</a>.<br>
>RS> <br>
>RS> The links in that section should be enough to do the code review. When
>RS> looking at the metrics make sure you familiarize yourself with the
>RS> various terms, some mean the opposite that you might think.<br>
>RS> <br>
>RS> Also, there seems to be some varying feelings and opinions about the
>RS> coding style for Kepler. My approach which I am sure differs from
>RS> Kepler is as follows:<br>
>RS> <br>
>RS> 1) All method arguments start with a lowercase "a" <br>
>RS> 2) All class data members all start with "m" (unless they are final)<br>
>RS> 3) All local variables have no prefix. <br>
>RS> <br>
>RS> It is my belief that you should always be able to look at a variable in
>RS> the code and understand it's scope. I disagree with the approach where
>RS> both the method arguments and the local variables have no prefix and
>RS> are often distinguished by a "this.", which is rather verbose at times.<br>
>RS> <br>
>RS> <br>
>RS> Rod<br>
>RS> <br>
>RS> </body>
>RS> </html>
>  
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mercury.nceas.ucsb.edu/ecoinformatics/pipermail/kepler-dev/attachments/20040820/1083a957/attachment.htm