Check whether page cropping hides part of a table

Where the extra work starts

Check the full source before blaming the extractor so you do not spend time repairing output from an incomplete view.

ParseMill processes native invoice PDFs and clear printed English PDF scans. Try a typical supplier invoice and a difficult one. Check the extracted fields, correct exceptions and open the workbook before committing to a monthly plan.

Cropping is a presentation or preparation step, not proof that excluded content is irrelevant. It also should not be treated as redaction or as evidence that the file became smaller.

A practical way to get it right

Open the original and prepared file side by side using approved tools. Compare page bounds, first and last columns, header rows, units, footnotes and continuation indicators. Check whether a value at the right or bottom edge is clipped rather than blank.

Count expected rows in the original region and compare with the crop. Record source-page mapping for any excerpt and identify context retained elsewhere. If the crop omits a heading or qualification needed to understand values, revise the prepared scope or use the original source for review.

After any authorized preparation, reopen the saved file and repeat text selection and table-fit checks. Do not assume the parser sees exactly the same visible region as the viewer; inspect the actual candidate output against the full source relationship.

Adobe: Crop PDF pages states that cropped content is hidden rather than discarded. Compare page bounds and do not treat a crop as secure removal or a guaranteed size reduction.

What this looks like on a real task

Synthetic example. The people, references and figures below are invented to demonstrate the method.

This crop is not ready for a complete table task. Copy the boundary, finding and action columns and add the original page reference. If only a subset of rows is intentionally needed, define that subset explicitly; do not call an excerpt a complete extraction of the original table.

Catch the cases that create rework

Resetting page bounds may reveal content hidden by cropping, but use only an authorized copy and preserve the original. Hidden content is not automatically relevant or trustworthy; it needs the same source and scope review. A screenshot of the crop cannot establish whether underlying text remains in the PDF.

If preparation creates a new image-only PDF, the native-text fit may be lost even though the crop looks clear. Test the saved derivative rather than judging its appearance alone.

Adobe: Copy content from PDFs explains that Snapshot copies a region as an image. Use it for visual comparison only; verify the actual saved PDF’s text and content separately.

Decide whether the change is worth keeping

Accept the prepared file only when the intended rows, columns and qualifications are accounted for and the source mapping is clear. Keep the preparation note with the output and verify edge rows after export. If crop behavior remains uncertain, use the complete source in a reviewed method rather than building a register from an unexplained partial view.