| Reading delimited data from web [message #650647] | 
Tue, 25 January 2011 14:52   | 
 
Eclipse User  | 
 | 
 | 
   | 
 
Is there a way how to read delimited data from http protocol using BIRT? The sample dataset I would like to use is here ( http://www.cnb.cz/cs/financni_trhy/devizovy_trh/kurzy_devizo veho_trhu/rok.txt?rok=2011 - FX history).  
 
I think the best way would be to use scripted data source, but from examples I have found I was not able to create anything usable. 
 
There are other ways possible, like: 
- use some kind of csv/plain-text jdbc driver 
- use ETL solution to extract data from web and than store data in the database (this works - using Clover does the job well, but I think there must be some simple solution using BIRT only) 
 
I have the following questions: 
- do you know any example, how scripted data source is used to read delimited plain text data from web? 
- is there any plain text jdbc driver you can use in BIRT to extract data from web? 
- any other simple solution? 
 
Thank you for any ideas. 
 
Tomas
 |  
 |  
  | 
 | 
 | 
 | 
 | 
| Re: Reading delimited data from web [message #651275 is a reply to message #651176] | 
Fri, 28 January 2011 11:36   | 
 
Eclipse User  | 
 | 
 | 
   | 
 
Thanks.  I posted it this morning on Birt Exchange. 
 
Jason 
 
On 1/28/2011 3:02 AM, TomasGreif wrote: 
> Jason, it is completely ok if you use it on birt-exchange (anyway, it is 
> your solution). 
> 
> Maybe you can consider also the following typical issues (at least for 
> me) when reading such kind of data: 
> - solving different locale (like different decimal separator - "," 
> instead of "." in the example above, or different date format) 
> - skipping more rows at the beginning (another stable dataset - 
> interbank interest rates, 
>  http://www.cnb.cz/cs/financni_trhy/penezni_trh/pribor/rok.tx t?year=2011)
 |  
 |  
  | 
Powered by 
FUDForum. Page generated in 0.03619 seconds