Removing regEx

[Former Member]

Hi,

I have a string in the below format.

<?xml version="1.0" encoding="UTF-8"?>#<Data>##><Row>#####<Entry>Test1</Entry>#####<Entry>Dtest2</Entry>#####<Entry>Test3</Entry>#####<Entry>-</Entry>#####<Entry>-</Entry>#####<.........so on

i wanted this string without '#' as below:

<?xml version="1.0" encoding="UTF-8"?><Data>><Row><Entry>Test1</Entry><Entry>Dtest2</Entry><Entry>Test3</Entry><Entry>-</Entry><Entry>-</Entry><.........so on

I tried using REPLACE ALL OCCURENCES REGEX '[[:punct:]]' for string with ''.

This is removing even the '<' , '>', '/' , etc...[Its resulting only the alphabets and numbers.

Can some one please provide a feasible solution.

Regards,

Rajkamal

[Former Member]

>

> Hi,Thanks for the response.
> 
> The above code did not work 
> 
> Nothing changes.
> 
> Regards,
> Rajkamal

Already had a feeling this would be your reply.

Do as Matt requested.

[Former Member]

REPLACE ALL OCCURENCES OF REGEX '\Q#\E' IN gv_string WITH ''.

[Former Member]

Hi,Thanks for the response.

The above code did not work

Nothing changes.

Regards,

Rajkamal

[matt]

Go into debug and tell me the hex value of one of these "#"

[Former Member]

>

> Hi,Thanks for the response.
> 
> The above code did not work 
> 
> Nothing changes.
> 
> Regards,
> Rajkamal

Already had a feeling this would be your reply.

Do as Matt requested.

[Former Member]

Hi Matt,

Here is the hex value for '#' 0023.

Regards,

Rajkamal

[matt]

That's good. So it really is a '#'.

You could use the non-regex REPLACE ALL OCCURENCES OF '#' IN my_string WITH space. And then use CONDENSE NO-GAPS.

Regex for me has always been trial and error, using the online tutorial and online test tools, which you can find with google.

[Former Member]

Funny enough it does work for me then.

[Former Member]

Oh no! Even this is not working for me!

I would like to explain my scenario a bit brief .

I am calling a URL using web service. The methods called for this are : CALL METHOD http_client->send , CALL METHOD http_client->receive

Then storing the received data as: result = http_client->response->get_cdata( ). [This is resulting in as XML with '#' as mentioned above].

Regards,

Rajkamal

[Former Member]

Is the result an internal table? A single variable? What's the type definition?

[Former Member]

The result is a variable.

result TYPE string

[Former Member]

data: gv_string type string.

gv_string = '<?xml version="1.0" encoding="UTF-8"?>#<Data>##><Row>#####<Entry>Test1</Entry>#####<Entry>Dtest2</Entry>#####<Entry>Test3</Entry>#####<Entry>-</Entry>#####<Entry>-</Entry>#####<.........>'.

write: / gv_string.
REPLACE ALL OCCURENCES OF REGEX '\Q#\E' IN gv_string WITH ''.
write: / gv_string.

My example that worked.

[Former Member]

Yes, the above code worked for me as well. Is it like the hard-coded values accept the statement REPLACE ALL OCCURENCES OF REGEX '\Q#\E' IN gv_string WITH '', but not the string which we get dynamically?

[Former Member]

Hi Matt/Maen,

The issue is resolved. The string with #### is not exactly'#' value, but they are actually representing horizontal tab and new line.

To overcome this, i used :

REPLACE ALL OCCURRENCES OF SUBSTRING cl_abap_char_utilities=>horizontal_tab IN result WITH ''.

REPLACE ALL OCCURRENCES OF SUBSTRING cl_abap_char_utilities=>newline IN result WITH ''.

Thanks for your time.

Regards,

Rajkamal

[matt]

So the hex value of the # in your input string wasn't x0023. Which is what we suspected in the first place!