FileCarpenter guide
Why some PDF text is editable and some is not
A PDF is a page-description format, not a Word document. Two pages that look identical can store their text in completely different ways.
A PDF does not contain “paragraphs” by default
PDF pages usually describe where individual glyphs are painted. Some producers keep enough font and text information for safe replacement; others split words into many tiny text operators or reuse glyph codes in unusual ways. Editing therefore starts with understanding the page objects, not with assuming a paragraph model.
When ordinary text replacement is possible
Selectable text with a usable font mapping can often be edited structurally. FileCarpenter analyzes the text run, font mapping, spacing and surrounding objects before choosing a replacement path. When the original structure is too ambiguous, the safer outcome is to refuse the structural edit or use a compatibility workflow instead of silently damaging the page.
Scans and outlined lettering are different
A scanned page may contain only pixels; text selection can come from an invisible OCR layer that does not match the visible image. Some design PDFs convert letters to vector outlines, which are shapes rather than text. Those cases require different techniques and should not be presented as ordinary text editing.
What to check after editing
Compare line breaks, fonts, spacing, clipping, tags and nearby graphics. For important documents, reopen the exported PDF in another viewer and keep the original separately.