I need to control the string encoding (utf-8) but i didn't find a transformer for.
Do you have a tip to test the encoding ?
Thanks for your help.
Alexy
Of course,
I need to make sure that the attribute type is "string" and check its encoding to utf-8.
I can not find transformers to perform these actions.
Thanks in advance
Hi,
If you will change the type of attribute(s) to utf-8 unconditionally, the AttributeEncoder might help you.
However, there isn't a transformer to check if the type is "string". If you have to check that, consider using a Python script (PythonCaller). The "FMEFeature.getAttributeType" method returns an integer identifier indicating the internal attribute type of specified attribute.
Takashi
Hi
If you need to know if a particular attribute is unicode or not, you can do the following in a PythonCaller:
David
If you need to know if a particular attribute is unicode or not, you can do the following in a PythonCaller:
import fme
import fmeobjects
def FeatureProcessor(feature):
s = feature.getAttribute('MyString') # Change attribute name as needed
if isinstance(s, str):
print "ordinary string"
elif isinstance(s, unicode):
print "unicode string"
else:
print "not a string"
David
david_r wrote:
Hi
If you need to know if a particular attribute is unicode or not, you can do the following in a PythonCaller:
David
If you need to know if a particular attribute is unicode or not, you can do the following in a PythonCaller:
import fme
import fmeobjects
def FeatureProcessor(feature):
s = feature.getAttribute('MyString') # Change attribute name as needed
if isinstance(s, str):
print "ordinary string"
elif isinstance(s, unicode):
print "unicode string"
else:
print "not a string"
David
@david_r, aside, how did you paste the script into the code block? I tried a few times in other Q&A;, but the script will not be contained in the code block correctly anyway. See here. I gave up and reported it to Safe...
takashi wrote:
@david_r, aside, how did you paste the script into the code block? I tried a few times in other Q&A;, but the script will not be contained in the code block correctly anyway. See here. I gave up and reported it to Safe...
Yeah, that code block thingy isn't very good... Basically, you can't have any blank lines, or it'll mess it up.
david_r wrote:
Yeah, that code block thingy isn't very good... Basically, you can't have any blank lines, or it'll mess it up.
I found a workaround for the blank lines. You can toggle into HTML tag mode when editing your post and replace the blank lines with a <br> tag. Then you get two blank lines for the price of one ;-)
david_r wrote:
I found a workaround for the blank lines. You can toggle into HTML tag mode when editing your post and replace the blank lines with a <br> tag. Then you get two blank lines for the price of one ;-)
Great, thanks. I'll try it at the next chance. Hope the editor will be fixed as soon as possible. @mitahajirakar, @dewetvannieker
takashi wrote:
Hi,
If you will change the type of attribute(s) to utf-8 unconditionally, the AttributeEncoder might help you.
However, there isn't a transformer to check if the type is "string". If you have to check that, consider using a Python script (PythonCaller). The "FMEFeature.getAttributeType" method returns an integer identifier indicating the internal attribute type of specified attribute.
Takashi
Tank you.
Takashi, do you know if a grid of interpretation of "feature.getAttributeType"'s result is available somewhere?
alexy wrote:
Tank you.
Takashi, do you know if a grid of interpretation of "feature.getAttributeType"'s result is available somewhere?
You can find required information in the API reference. Go to:
Knowledge Center home > FME Documentation > Python FME Objects API Reference
takashi wrote:
You can find required information in the API reference. Go to:
Knowledge Center home > FME Documentation > Python FME Objects API Reference
Link for the lazy: getAttributeType
But as you can tell, it won't help telling the difference between a unicode or a regular string. You will have to use the Python code below for that.
david_r wrote:
Link for the lazy: getAttributeType
But as you can tell, it won't help telling the difference between a unicode or a regular string. You will have to use the Python code below for that.
Thanks for the link for lazy :-)
I thought that FME_ATTR_STRING (=11) returned by the getAttributeType indicates "string". Was I wrong?
david_r wrote:
Link for the lazy: getAttributeType
But as you can tell, it won't help telling the difference between a unicode or a regular string. You will have to use the Python code below for that.
@takashi, you're right about FME_ATTR_STRING, of course. My point was that it won't tell you if the attribute is a string encoded to the current locale (e.g. cp1252 here in western Europe) or if it is in unicode (utf-8 or utf-16).
david_r wrote:
Link for the lazy: getAttributeType
But as you can tell, it won't help telling the difference between a unicode or a regular string. You will have to use the Python code below for that.
@david_r, thanks for the clarification. Yes, of course it's necessary to add some codes to check the encoding.
david_r wrote:
I found a workaround for the blank lines. You can toggle into HTML tag mode when editing your post and replace the blank lines with a <br> tag. Then you get two blank lines for the price of one ;-)
@mitahajirakar, thanks for your efforts. Regards.
Reply
Rich Text Editor, editor1
Editor toolbars
Press ALT 0 for help
Helpful Members This Week
Enter your E-mail address. We'll send you an e-mail with instructions to reset your password.