| reports portable to multiple RDBMS [message #205360] |
Thu, 07 December 2006 10:52  |
Eclipse User |
|
|
|
Hi,
does anyone have experience in writing reports
that can be used with multiple database engines
including oracle?
The issue is that most RDBMS are case sensitive
and they return column names as they were defined
in CREATE TABLE or SELECT column as label clauses.
But oracle is oracle. It returns column names
uppercased. Always.
BIRT does not hide these differences and it is
case sensitive too.
select id,url from table;
table with two columns:
row["id"] row["url"] - for case sensitive databases
row["ID"] row["URL"] - for oracle
If you use first syntax in oracle, the exception is thrown,
that there is no "id" in row. If you use the second syntax
in mssql, the exception is thrown, that there is no "ID" in
row.
Does any have solution for this serious trouble?
Leos
|
|
|
| Re: reports portable to multiple RDBMS [message #205557 is a reply to message #205360] |
Fri, 08 December 2006 04:08  |
| Eclipse User |
|
|
|
RTFM Leos :-)
Well, I found one solution. Define alias in Output columns.
Data will be available both under column name and alias.
datasetRow["END_TIME"]
datasetRow["end_time"]
If your report is supposed to be run under various RDBMS,
you shall use alias feature.
Jason, could you add this to Birt's FAQ?
Leos
Leos Literak napsal(a):
> Hi,
>
> does anyone have experience in writing reports
> that can be used with multiple database engines
> including oracle?
>
> The issue is that most RDBMS are case sensitive
> and they return column names as they were defined
> in CREATE TABLE or SELECT column as label clauses.
> But oracle is oracle. It returns column names
> uppercased. Always.
>
> BIRT does not hide these differences and it is
> case sensitive too.
>
> select id,url from table;
>
> table with two columns:
> row["id"] row["url"] - for case sensitive databases
> row["ID"] row["URL"] - for oracle
>
> If you use first syntax in oracle, the exception is thrown,
> that there is no "id" in row. If you use the second syntax
> in mssql, the exception is thrown, that there is no "ID" in
> row.
>
> Does any have solution for this serious trouble?
>
> Leos
|
|
|
| Re: reports portable to multiple RDBMS [message #205772 is a reply to message #205557] |
Mon, 11 December 2006 11:12 |
| Eclipse User |
|
|
|
Leos,
I will add ths to the FAQ. Great suggestion.
Jason
"Leos Literak" <literakl@centrum.cz> wrote in message
news:elba2c$bln$1@utils.eclipse.org...
> RTFM Leos :-)
>
> Well, I found one solution. Define alias in Output columns.
> Data will be available both under column name and alias.
>
> datasetRow["END_TIME"]
> datasetRow["end_time"]
>
> If your report is supposed to be run under various RDBMS,
> you shall use alias feature.
>
> Jason, could you add this to Birt's FAQ?
>
> Leos
>
> Leos Literak napsal(a):
>> Hi,
>>
>> does anyone have experience in writing reports
>> that can be used with multiple database engines
>> including oracle?
>>
>> The issue is that most RDBMS are case sensitive
>> and they return column names as they were defined
>> in CREATE TABLE or SELECT column as label clauses.
>> But oracle is oracle. It returns column names
>> uppercased. Always.
>>
>> BIRT does not hide these differences and it is
>> case sensitive too.
>>
>> select id,url from table;
>>
>> table with two columns:
>> row["id"] row["url"] - for case sensitive databases
>> row["ID"] row["URL"] - for oracle
>>
>> If you use first syntax in oracle, the exception is thrown,
>> that there is no "id" in row. If you use the second syntax
>> in mssql, the exception is thrown, that there is no "ID" in
>> row.
>>
>> Does any have solution for this serious trouble?
>>
>> Leos
|
|
|
| Re: reports portable to multiple RDBMS [message #214205 is a reply to message #205557] |
Thu, 25 January 2007 16:28 |
| Eclipse User |
|
|
|
Leos,
This should be in the FAQ now under data access. Sorry this took so long.
Jason
"Leos Literak" <literakl@centrum.cz> wrote in message
news:elba2c$bln$1@utils.eclipse.org...
> RTFM Leos :-)
>
> Well, I found one solution. Define alias in Output columns.
> Data will be available both under column name and alias.
>
> datasetRow["END_TIME"]
> datasetRow["end_time"]
>
> If your report is supposed to be run under various RDBMS,
> you shall use alias feature.
>
> Jason, could you add this to Birt's FAQ?
>
> Leos
>
> Leos Literak napsal(a):
>> Hi,
>>
>> does anyone have experience in writing reports
>> that can be used with multiple database engines
>> including oracle?
>>
>> The issue is that most RDBMS are case sensitive
>> and they return column names as they were defined
>> in CREATE TABLE or SELECT column as label clauses.
>> But oracle is oracle. It returns column names
>> uppercased. Always.
>>
>> BIRT does not hide these differences and it is
>> case sensitive too.
>>
>> select id,url from table;
>>
>> table with two columns:
>> row["id"] row["url"] - for case sensitive databases
>> row["ID"] row["URL"] - for oracle
>>
>> If you use first syntax in oracle, the exception is thrown,
>> that there is no "id" in row. If you use the second syntax
>> in mssql, the exception is thrown, that there is no "ID" in
>> row.
>>
>> Does any have solution for this serious trouble?
>>
>> Leos
|
|
|
Powered by
FUDForum. Page generated in 0.09718 seconds
Home