CSV output is always blank

Question

Im having trouble with one of the reports im working on. Im allowing the user to select from 3 output formats (PDF, CSV, and Data). If i select the PDF it works perfectly (using an RTF template), and if i select DATA it also works fine and returns XML output. But if i select CSV then it comes back empty. No errors... No messages... Nothing...
I have another report that seems to work fine, including the CSV output. This is why it seems baffling to me. The difference between those 2 reports is that the working one uses a SQL query for its data while the one that doesnt work is using a template. Since the template only contains a single query, i did take it out and tried it as a SQL Query but had the same results (blank page).
Im at a loss for things to try. Without any feedback from Publisher, i have no idea where to go. I know that there is data available because several rows are returned with the PDF and Data outputs.
Any suggestions and/or ideas are definitely welcome.
Thanks
--Brett
PS: in case anyone is interested, here's my data template:
<dataTemplate name="CurrCompany" description="Current Company List" dataSourceRef="iub_datasource">
<!-- CVS Version $Revision: 1.4 $ -->
<parameters>
     <parameter name="P_CODE" dataType="character"/>
     <parameter name="P_IND_TYPE" dataType="character"/>
     <parameter name="P_STATUS" dataType="character"/>
     <parameter name="P_ORDER_BY" dataType="character"/>
</parameters>
<dataQuery>
<sqlStatement name="Q_COMPANY">
<![CDATA[
select distinct TRIM(cinfo.fname || ' ' || cinfo.lname) contact_name,
co.companyname company_name,
co2.companyname dba_name,
ci.address1,
ci.address2,
ci.address3,
ci.city,
sc.name state_full_name,
sc.code state_abbreviated,
ci.zipcode,
ty.name company_type
from company co,
company co2,
company_relationshiptype crt,
relationshiptype rst,
company_IndustryType cit,
industryType it,
companyStatus cs,
companyInfo ci,
Company_Companycodetype cct,
Companycodetype ty,
company_contact cc,
Contacttype ct,
Contact con,
contactInfo cinfo,
statecode sc
where co.companystatusid = cs.companystatusid
and cs.name = NVL(:P_STATUS, cs.name)
and co.companyid = ci.companyid
and ci.statecodeid = sc.statecodeid (+)
and co.companyid = cit.companyid (+)
and cit.industrytypeid = it.industrytypeid (+)
and instr( NVL(:P_IND_TYPE, it.description ), it.description ) > 0
and co.companyid = cct.companyid
and cct.companycodetypeid = ty.companycodetypeid
and instr( NVL(:P_CODE,ty.name), ty.name ) > 0
and co.Companyid = cc.companyid
and cc.contacttypeid = ct.contacttypeid
and UPPER(ct.name) = 'REGULATORY'
and cc.contactid = con.contactid
and con.contactid = cinfo.contactid
and co.companyid = crt.companyid (+)
and crt.relationshiptypeid = rst.relationshiptypeid (+)
and rst.name (+)= 'Doing Business As'
and crt.relatedcompanyid = co2.companyid (+)
order by UPPER(DECODE( :P_ORDER_BY, 'Name', company_name || zipcode, 'Type', company_type || company_name, 'Zipcode', zipcode || City || company_name ))
]]>
</sqlStatement>
</dataQuery>
     <dataStructure>
          <group name="G_COMPANY" source="Q_COMPANY">
          <element name="contact_name" value="contact_name"/>
          <element name="company_name" value="company_name"/>
          <element name="dba_name" value="dba_name"/>
          <element name="address1" value="address1"/>
          <element name="address2" value="address2"/>
          <element name="address3" value="address3"/>
          <element name="city" value="city"/>
          <element name="state_full_name" value="state_full_name"/>
          <element name="state_abbreviated" value="state_abbreviated"/>
          <element name="zipcode" value="zipcode"/>
          <element name="company_type" value="company_type"/>
          </group>
     </dataStructure>
</dataTemplate>

Answer 1

Yes, you wont get the csv,
CSV is supported only if the xml is having rowset/row tags.
Since you have your own tags G_COMPANY, as the groupings, you wont get the csv from BIP.
So, use the query directly in BIP , it will automatically add the rowset/row tag, and csv generated is supported for that.
even in data template it is supported if you put rowset and row tags as the group names.
or
you have to write the e-text template with each field separated by the comma , or desired separator.