[BioC] Using htmlpage

[john seers (IFR)]

Hi Jim


>OK, now I understand. You want to build a link using one ID, but 
>supplying a different ID to the end user for them to click. This 
>functionality still doesn't exist for htmlpage(), as there is no way to 
>pass in two sets of IDs, one to build the query and one for the HTML.

Yup, I think you've got it. Close enough anyway.

>And the way htmlpage() is designed, I don't see a simple way to add such 
>functionality without making it more of a kludge-fest than it already is.

Fair enough. 


>I disagree with this statement. The htmlpage() function isn't designed 
>to be completely flexible, nor IMO should it be. 

Well, I won't argue with you over this. It is your software and your design. 

How about htmlpagemoregeneric()?  :)


>If you want complete flexibility, then really what you should do is just 
>use the xtable package directly. You can create a data.frame containing 
>URIs in some of the columns, of whatever form you desire, and then use 
>xtable() to build the table and then print it using print(thextable, 
>type="html"), which is what is done in htmlpage() (see the last two 
>lines of the function).

Yes, probably the advice I was looking for when I first raised the question. I did not know about xtable until I started looking at how htmlpage worked. In hindsight perhaps I was expecting htmlpage to do more and was trying to avoid reinventing the wheel. But I think I can see my way forward now. I am just waiting for some feedback on Mark 1 from my user and then I will knock it into shape if what I have done is useful.

Thanks. 

Regards


John