# Dynamic amount of field

**URL:** <https://forum.cloverdx.com/t/dynamic-amount-of-field/998>\
**Category:** CloverDX Platform\
**Created:** [January 15, 2012, 12:00am UTC](https://forum.cloverdx.com/t/dynamic-amount-of-field/998 "2012-01-15T00:00:00Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![eyal](https://avatars.discourse-cdn.com/v4/letter/e/76d3ee/32.png) [@eyal](https://forum.cloverdx.com/u/eyal)\
**Post date:** [January 15, 2012, 12:00am UTC](https://forum.cloverdx.com/t/dynamic-amount-of-field/998/1 "2012-01-15T00:00:00Z")

</div>

Hi,  
the file i am reading has few versions.  
each version has different amount of fields, as reasult i cant use the same metadata (actually the not using the same graph).  
is there a way to have dynamic metada or import that will skip missing fields.  
i will add the difference is only in the last fields, there is a basic common amount of fileds and order.

---

<div class="post-metadata">

**Author:** ![jurban](https://avatars.discourse-cdn.com/v4/letter/j/cc9497/32.png) [@jurban](https://forum.cloverdx.com/u/jurban)\
**Post date:** [January 16, 2012, 1:49pm UTC](https://forum.cloverdx.com/t/dynamic-amount-of-field/998/2 "2012-01-16T13:49:25Z")

</div>

Hi,

this is unfortunately currently not possible - you will need to create separate metadata for each version of the file.

Regards,  
Jaro

---

<div class="post-metadata">

**Author:** ![eyal](https://avatars.discourse-cdn.com/v4/letter/e/76d3ee/32.png) [@eyal](https://forum.cloverdx.com/u/eyal)\
**Post date:** [January 16, 2012, 2:52pm UTC](https://forum.cloverdx.com/t/dynamic-amount-of-field/998/3 "2012-01-16T14:52:24Z")

</div>

thank you for the replay.  
how will the graph will know which metadata to use?  
for example i have 3 version of the file ( 10,12 and 14 fields):  
should i make 3 data readers with seperate data reader, while 2 readers will always give error, one will always be ok.

---

<div class="post-metadata">

**Author:** ![dpavlis](https://avatars.discourse-cdn.com/v4/letter/d/90ced4/32.png) [@dpavlis](https://forum.cloverdx.com/u/dpavlis)\
**Post date:** [January 17, 2012, 9:34pm UTC](https://forum.cloverdx.com/t/dynamic-amount-of-field/998/4 "2012-01-17T21:34:22Z")

</div>

One solution to your problem is to use parameters and drive which data file and which metadata will be used. You can create a template/skeleton transformation graph and by defining parameter value modify it to read particular data/format.

See the attached CloverDesigner project - look at “WHICH\_DATA” parameter and how is it used to define “fileURL” for the reader and which metadata definition will be linked in.

Parameters are also described in [Clover User Doc](http://www.cloveretl.com/documentation/UserGuide).

- [Metadata\_Example\_Project.zip](https://cdn.cloverdx.com/forum/107%5Fe3b2158ccaa9791e2b9c8378917236f4/Metadata%5FExample%5FProject.zip)