<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
<title></title>
</head>
<body bgcolor="#ffffff" text="#000000">
Dave Burris wrote:
<blockquote cite="mid430C9755.4000300@kwburris.com" type="cite">
<meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
<title></title>
Kenneth Brody wrote:
<blockquote
cite="mid20050824101234.cvzgisc88gco88ww@webmail.spamcop.net"
type="cite">
<pre wrap="">Quoting Dave Burris (Tue, 23 Aug 2005 22:39:46 -0400):
[...]
</pre>
<blockquote type="cite">
<pre wrap="">I am running this code from *report. The strange thing is that the
ascii file is generated correctly every time. The pdf file is also
generated correctly when there are no lookup and getnext commands that
run. I walk through the code and watch the system command run, but no
pdf file is created. If I run the same command from the command line,
the pdf is created ok, and if I re-run the report for an invoice with no
invoice items (just time entries so the above lookup returns no invoice
items), the pdf is created successfully. It is like filePro fails to
execute the command completely, but I see no error. The system command
is:
::system "/appl/fp/text2pdf"<if
</pre>
</blockquote>
<pre wrap=""><!---->
What if you replace this with:
SYSTEM "/appl/fp/text2pdf.debug" < if
and create a "text2pdf.debug" script with:
echo "Running text2pdf.debug on file:"
ls -l $1
read junk
/appl/fp/text2pdf $1
echo "text2pdf returned with an exit of " $?
read junk
Is there any difference in the display with LOOKUP/GETNEXT and without?
</pre>
</blockquote>
On OpenServer 5.0.6, I tried the above code and the text2pdf.debug code
never seems to run. The screen flickers a little and that is it. So I
changed the command to:<br>
<br>
SYSTEM NOREDRAW "/appl/fp/text2pdf.debug" < if<br>
<br>
Now the error stays on the screen. It says: /appl/fp/text2pdf.debug:
cannot open any more files<br>
<br>
On OpenServer 5.0.7, the text2pdf.debug command runs, and I see the
following:<br>
Running text2pdf.debug on file:<br>
-rw------ 1 fielpro group 3727 Aug 24 11:42
/tmp/AE053181<br>
<br>
/appl/fp/text2pdf: cannot make pipe<br>
/tmp/AE053181.txt: No such file or directory<br>
text2pdf returned with an exit of 0<br>
<br>
Here is the text2pdf script:<br>
<pre>#!/bin/sh
# Print a pdf file by first converting it to postscript and then sending it to
# the gslp print script
textfile=$1".txt"
pdffile=$1".pdf"
fpdir=/appl/fp
dtox $1 | $fpdir/unpcl > $textfile
$fpdir/text2ps $textfile > /tmp/1.ps
$fpdir/ps2pdf /tmp/1.ps $pdffile > /dev/null 2>&1
rm -f /tmp/1.ps
rm -f /tmp/$textfile
</pre>
</blockquote>
Thanks for the help with debugging. Based on the errors we saw there,
I increased the NOFILES kernel parameter from 110 to 220. This seems
to fix the issue. Does this indicate a problem with the way I am
generating the report? <br>
<br>
<br>
<br>
<br>
</body>
</html>