Hello,
I try to convert a CSV files (UTF8) which have accents in several attributes.
When I parameter the CSV in FME, (with UTF8 enconding) the accents
appear to be correctly interpret in the preview window but it isn't the
case in the workspace. Instead of "téléphone" I have "Téléphone" for
example.
Can someone help me?
Thank you in advance
Hi @jen
do you have accented characters in attribute names or in attribute values? Also, what FME version/build do you use.
Hi @LenaAtSafe
The accented characters are in attribute names and I use FME2013
I wonder if the problem can come from the langage use in FME which is english while I try to import french data.
Thank you for your help :)
Hello @jen
Attribute names are a lot harder to handle than values and I think you may have to upgrade your FME to a newer version. For example, I can do this in FME2016.1 easily by setting the encoding parameter to UTF8:
Demo: http://screencast.com/t/ZupRjVm8W
But I just tried this in 2013 and - although the CSV preview recognized the data - it did not appear correctly in Workbench. I think workspaces just didn't work in UTF8 at that point. They do now, so if you can upgrade it would probably work.
Hope this helps
Mark
Hi @jen
I totally agree with @Mark2AtSafe - the best option would be to upgrade your FME. Since 2013, encoding support has been significantly improved.
With some tricks you might be able to read your data even with FME 2013 though (this might be different for different formats, I have tested CSV only). While you see the attribute name as Téléphone under the Reader User Attributes and on the canvas, there is also the original téléphone attribute which is read by the Reader but not exposed. You can simply use AttributeExposer to expose this hidden attribute.
Please take a look at the attached demo.
Hi,
I will keep in mind the AttributeExposer trick and try to update my FME version.
Thank you for all your assistance .