XML tag filter (loop) problem

[chrisnta]

I have the following problem. I use xml to get data from a ISP with TroiURL. Everything works fine, but I try to replace the tags to more simple tags. Like this one:

5

to

5

I have got this working with the calculation (see end) However, it takes only the first available line and not the whole text or other lines.

<?xml version=1.0?>

5

OH3 9YH

Z

G

2097152

G

HSHSH

City

E

Y

G

4096000

1235

Home 512k Classic

adsl

Y

524288

extract ( "

[comment]

May I ask why are you bothering with this in Filemaker? There are much better tools for this task. Even if you HAVE to do this in Filemaker, you could import the data as is (using a custom sylesheet), then export it to the desired format (using another stylesheet).

[chrisnta]

The data will be saved in filemaker. It is not meant to convert the xml to something else to export. I just need it readable for others. Also I can only send it per https post request, thats the reason why I'm using troiurl otherwise I could use filemaker tools. I already using a simelar function to get data within the tags in a field. However, something like checking ADSL availabilty returning more option or set of data not just one set. Here I just would like to present a readable list.

[comment]

Yes, I can see why you would want to use a plugin to download the file. What I don't see is why, once you have done that, you don't proceed to import it. If you need a more readable format, you could output it from within Filemaker - same as with any other data.

[chrisnta]

Do you mean when I receive the xml response in a field than export the field content has a xml file and import the xml file back to filemaker with a style sheet?

I just need to think about the procedure, I want to keep it simple for the user.

[comment]

Yes - except it would probably be better to save the response directly to a local file (if your plugin allows this), instead of going through a Filemaker field. A file exported as field contents is invariable encoded in UTF-16, and this might cause problems with importing. In any case, why don't you try it and see - the entire process can be scripted, so it would be completely transparent to the user.