I have some data that must be displayed in a PDF report. The data contains HTML tags because it is also displayed in HTML pages. I need to strip the HTML tags before displaying it in the PDF report. The issue is compounded by the fact this is a compound PDF report - I had originally thought using HTMLFORMTYPE=PDF would work, but it doesn't, and even if it did, how would I combine that HTMLFORMTYPE code with the rest of the compound PDF components.

Any ideas?

Thanks,

Can you give a sample of the HTML and the complexity of the size, etc ?

An example:

<FONT color=#000000><P><FONT color=#000000>Results are closed as of January 31, 2011. </FONT></P><P><FONT color=#000000>Zzzzz xxx dddd rrrr tttt. QQQQQ ffff. Tgggggggg ddddd lllll dddddddd rrrrrrr.</FONT></P><P>test21</P></FONT>

I'm looking into using an Oracle function, perhaps this might be easier on the database side...

Is it a single line of HTML or multiple, and how will the be needed ?

I have an idea.

The basics is to TABLE, and use STRREP the < and > characters to {crlf}< and >{crlf}.
Save to a file. Reread with a new master, then toss all lines that have < or > or are blank.

How about adding a -IF [FORMAT] EQ 'PDF' THEN 'one thing' ELSE 'another thing'?

I had the same need so I wrote an SQL procedure to handle it. Here is the code I used:

Create PROCEDURE striptags
 (IN Dirty varchar (4000),
  OUT Clean varchar (4000))
language sql
Begin
 Declare @Clean VARCHAR(4000);
 Declare @Start int;
 Declare @End int;
 set @Clean=Dirty;
 While locate('<', @Clean) > 0 And
      locate('>', @Clean)>0 DO
  Set @Start = locate('<', @Clean);
  set @End = locate('>', @Clean);
    if @start=1 THEN
      set @Clean= substr(@Clean,(@END+1),(length(@Clean)-@end));
    ELSEIF @start>1 and @start<length(@Clean) THEN
      set @Clean=substr(@Clean,1,(@start-1))||
        substr(@Clean,(@end+1),(length(@Clean)-@end));
       elseif @end=length(@Clean) then
        set @Clean=substr(@Clean,1,(@start-1));
      END IF;
     END while;
 set Clean=@Clean;
END

I tried to keep it somewhat simple by using the Oracle PL/SQL regexp_replace function:

SQL SQLORA
select
regexp_replace ( regexp_replace ( comment_text, '&|nbsp;', ' ', 1, 0, 'i' ), '<[^>]*>', ' ' ) as comment_text
from report_comments
where report_cde = 'report1'
and section = 'status'
and line_num = 1
;
TABLE ON TABLE HOLD AS HCOMMENT1
END

But, of course, it's not so simple as I have three nested function calls! And I just realized this does not take care of &nbsp; and other like HTML character entities...

(Updated with simplified code that takes care of HTML tags and &nbsp; - which is good enough for me...).

This message has been edited. Last edited by: Francis Mariani,

Why didn't I think of this before.

How about MARKUP.

DEFINE FILE CAR
TAGGEDLINE/A300 = '<FONT color=#000000><P><FONT color=#000000>Results are closed as of January 31, 2011. </FONT></P><P><FONT color=#000000>Zzzzz xxx dddd rrrr tttt. QQQQQ ffff. Tgggggggg ddddd lllll dddddddd rrrrrrr.</FONT></P><P>test21</P></FONT>';
END
 
TABLE FILE CAR
HEADING
"<TAGGEDLINE"
PRINT COUNTRY
ON TABLE PCHOLD FORMAT PDF
ON TABLE SET STYLE *
 TYPE=REPORT, MARKUP=ON, SIZE=9, ORIENTATION=LANDSCAPE, $
ENDSTYLE
END

I found an old post where you revealed MARKUP to the world! Unfortunately it doesn't take care of hyperlinks and there's a mailto hyperlink in the text

Well there was always that chance that all the tags were not going to be removed/evaluated.

Back to the oracle function I guess.

Does the data have to contain the html tags?

Usually in the reports, I do what alex says.

-IF [FORMAT] EQ 'PDF' THEN 'one thing' ELSE 'html tag'?

"Does the data have to contain the html tags?"

The data is designed to be displayed in HTML and contains HTML. I now need to show it in PDF.