Text extraction and columns

I am trying to extract text in continuous form -- right across lines when in normal paragraphs, successive lines within a column when the text is in columns.

GetPageText( 2 ) does a good job of distinguishing columns and keeping their contents separate when extracting text. But in ordinary paragraphs, GetPageText(2) often breaks lines mid-word, and that's a pain.

GetPageText( 3 ) provides better detail in ordinary paragraphs, but intermittently (not always!) it runs columns together.

Is there any way to GetPageText( 3 ) to return the text in each column separately [without sacrificing its ability to handle ordinary paragraphs]?

StMike38

Ingo,

I agree that option 4 is the best for getting at continuous text. On some PDFs it nicely puts out a word at a time. But in a 2 MB PDF file it suddenly decides to output one or two letters at a time. With variable width fonts, there is no sure way of recognizing / calculating space between words. Attempts so far have two undesirable results -- some words wind up fragmented, while parts of separate words get pushed together.

Is there any way to get control so that option 4 actually gets a word at a time?

StMike38