| andi303 |
| Newbie |
|
|
|
|
| None Specified |
|
| Friday, March 6, 2009 |
| Wednesday, April 10, 2013 5:39:30 AM |
9
[0.05% of all post / 0.00 posts per day] |
|
I'm trying to map XSLX to MySQL tables.
However, the ODBC-driver complains about the encoding:
Quote:Transformation-Error in Database SQL-execution:
Error in: INSERT INTO `products` (`product_sku`) VALUES (?)
Reason: [MySQL][ODBC 5.2(w) Driver][mysqld-5.1.49-3~bpo50+1-log]Server does not support 4-byte encoded UTF8 characters.
I know that 4-byte utf8 is only supported by MySQL 5.5, but why do I need 4-byte for an XSLX?
|
The schema for .mff allows a variable amount (growable) of input parameters, e.g.:
<component name="concat" growable="true" growablebasename="value">
However, according to the MapForce help, no arrays are allowed for own functions written in Java or .NET.
How does that fit together?
Same for output...
I'm currently serializing the input/output for all my functions, however, this is getting boring...
Thanks in advance!
|
I think it is rather pre-processing. Your (post-processing) solution will only work with a relation table as a source.
You can map to a temporary table, and then map to the final table with build-in functions. However, IMHO this is only a somewhat inefficient workaround.
|
You have to distinguish between empty cells and undefined cells. Normally a cell is undefined as long as you don't enter any value or apply any formatting.
Therefore, you have to check whether the cell exists in your mapping. And if not, just set an empty string instead.
This can be quite cumbersome if you have to merge a lot of columns, maybe you want to build a user-defined funcion. MapForce does not have an option to choose how to handle undefined cells...
|
Because I am not a fan of "reinventing the wheel"-programming.
And yes, supporting build-in db functions (not only digest) is definitively an essential function of db mapping.
|
Bump.
Five years later, MapForce ist still missing some of the most essential DB functions...
|
Thanks!
Yes, I'm using constants since version 2008...
|
Hi all,
is it possible to put the columns annotations in the first row of the resulting excel document?
Thanks in advance!
|
Hello all,
I'm using Diffdog with the MySQL ODBC-driver for merging two databases. There is a problem in Diffdogs behavior when merging the rows.
An example: you have one database row which only exists in the source table and you want to insert this row in the target table. Diffdog then uses the auto-increment number (set in MySQL) for the new primary key.
This behaviour ensures that the primary key is unique, however, it doesn't necessarily keeps the consistency between tables. If you insert multiple rows from different tables of the source database, you may want to keep the foreign key constraints. However, I can't see any option in Diffdog to ensure the consistency of foreign key constraints in the target database.
There should be at least an option "keep the primary key". Diffdog then just has to check if there would be any double primary keys in the target database when inserting rows from the source.
I've tried to insert table rows via ODBC using Access. In this case, it is possible to ignore the auto-increment value from MySQL. So there is no restriction in the ODBC driver.
Anyone else who thinks this might be a useful option?
Andreas
|
|
