search cancel

Web Viewer 14.0 - Multi Million line reports, Convert AFP File to PDF


Article ID: 212562


Updated On:


Output Management Web Viewer


The below values should be changed in my environment?


X2YY_PAGELIM=20000            Max number of pages (0-999999999)
X2YY_LINELIM=100000           Max number of lines (0-999999999)

We generate reports with millions of lines I should set the above values, increasing them from 20000 to what?

Below are the values in my system
#X2YY_A2PD=A2PDL               AFP-to-PDF transform node unc  
X2YY_A2PD=A2PD               AFP-to-PDF transform node unc

Is it recommended we comment out x2yy_A2PD=A2PD
and change x2yy_a2pd=A2PDL ?


Release : 14.0

Component : CA Output Management Web Viewer

CA View with Stand-Alone Transformers

CA Spool


The reason these files were not being transformed is because they exceeded the X2YY_PAGELIM or X2YY_LINELIM value set in STDENVX.  Given the large size of these AFP files, some guidance on what resources are required by the stand alone transformers to do the transformation was needed.

CPU and memory usage is determined by the complexity of the AFP data. When concerned about how long the wait for the transformation to complete will be, linearized pdf project file, A2PDL.d2eproj should be used so that the first pages are viewable as quickly as possible.

The amount of space available in the filesystem with the transformer directories also needs to be considered.

The files stored in /input and written to /output will be large.  
From the page:
CA View 14.0 CA Spool Java Transformers
If X2YY_SUBID is specified, this parameter specifies the name of the CA Spool printer node that drives the AFP-to-PDF transformation. AFP-to-PDF transformation is not performed if this parameter is not specified.
If X2YY_PROC is specified, this parameter must be set to value of "A2PD" or "A2PDL".
You can optionally specify X2YY_A2PD = A2PDL. This setting generates a linearized PDF, which loads faster in most browsers than a non-linearized PDF.

Additional Information

Unfortunately, there is no way to estimate the space requirement ahead of time. It's a tuning effort. Some guidelines:

-          Queue and reports do not have to be large.
o   The reports files in these directories are around 25k or less and unless you are retaining them for debugging or a transformation failed, they are cleaned at the end of the transformation.

-          Input needs to be large as the report files in this directory depend on the size of the report.
o   It’s not just the number of pages/lines in the report but also the length of each line that determines the size of the input report files.
o   The input report files are cleaned at the end of the transformation.
o   If more than one transformation is being requested (for example - user1 is requesting a transform of report ABC and user2 is requesting a transform of report XYZ) then each will have a file in the input directory so it needs to be large enough to hold both requests.

-          Output needs to be large as this is where the PDF output files are stored.  The size of those files depend not only on the size of the report but the report complexity as well.
o   Whether or not output files are cleaned at the end of the transformation depends on the X2YY_RETAIN setting.  Longer X2YY_RETAIN times allow the output file to be reused instead of having to transform the file again – the cost is the output pdf file will remain in the output directory during that time.
o   If more than one transformation is being requested or if previous transformations are retained then the output directory needs to be large enough to hold them all.