Unknown character in content stream

[dreesdm]

I have a workflow set up for some letters to resample for print and optimize. On some of the simplex print letters i been getting this error.

Failure (Critical parser failure: Unknown character in content stream) occurred during Select all

Only seem to be happening on two pages out of 1600. I think it might be a font issue, but that is all the info I get out of the report.

The PDF file goes on to print alright, but like to figure out the source of the problem. Any ideas?

Thanks!

[rhd_ole]

I have a workflow set up for some letters to resample for print and optimize. On some of the simplex print letters i been getting this error.

Failure (Critical parser failure: Unknown character in content stream) occurred during Select all

Only seem to be happening on two pages out of 1600. I think it might be a font issue, but that is all the info I get out of the report.

The PDF file goes on to print alright, but like to figure out the source of the problem. Any ideas?

Thanks!

The times I've seen this it was a non-embedded font, missing x-object which turned out to be bogus, or transparencies. how are you processing these files, PS sever, Fastlane etc?

[dreesdm]

I suspect it is a font or an embedded path around a binary bitmap right now.

They come out of a server based PDF composer called iTextSharp.

[freddyp]

A content stream refers to the content of a page in PDF syntax. An unknown character in the stream probably refers to a syntax error. It is impossible to tell without having the file. Can you share the file, or better still, one of the problem pages? After extracting the problem page check it to see if the problem is still there. Things happen when saving PDF files.

[dreesdm]

I cant share the PDF due to PII data they contain. I think I narrowed down the problem to the way these are made using iTextSharp. The pages are all made independently out of the server system, then placed together into one PDF. So we have hundreds of the same embedded fonts in the PDF that the RIP had to process, and it just gives up.

Would any of you know a way to consolidate embedded font subsets?

[freddyp]

There is an Action "Consolidate fonts". That sounds like it could help.

Not all rips like lots of line-art so it may not be feasible, but outlining all text is also a possible approach worth investigating.