填写 PDF 表单

这款工具的作用

A fillable PDF carries its own questions. Each box on the page is a named field, and filling one writes a value into the document rather than painting text over it. This tool reads the fields out of the file you drop in, gives you a control for each one typed to what that field actually is - a text box, a tick, a set of radio buttons, a dropdown, a list - and writes your answers back into the same document. The result is still a form afterwards, unless you ask for it to be flattened.

Use it for the forms that arrive by email and are meant to go back the same way: a tax declaration, a bank mandate, a school enrolment sheet, a supplier onboarding pack. It is also the right tool when a reader on your phone will not let you type into a form that Acrobat fills without complaint, which usually means the reader cannot draw the field's appearance rather than that the form is broken.

工作原理

1. Drop the PDF onto this page. Its fields are read and listed with whatever values are already in them.
2. Work down the list. Each field gets the control its type calls for, and the field's tooltip is shown where the document provides one.
3. Decide whether to flatten. Left off, the answers stay editable; turned on, they are drawn into the page and the fields are removed.
4. Leave the font embedded if any answer uses accents or a non-Latin script.
5. Press Fill form and the download starts on its own.

Answers are written as field values, which is what makes them behave like answers. A recipient's reader shows them in the field, a form-data export picks them up, and anyone with the same document can correct a typo without redoing the page. Drawing text on top - what happens when you fill a form in an image editor or with a generic PDF annotator - looks identical on screen and does none of it.

Appearance streams are regenerated field by field rather than in one pass, and then the document's /NeedAppearances flag is cleared. That flag tells a reader to throw away the streams and draw its own, which is exactly how a filled form ends up looking different in Acrobat, Preview and Chrome. Doing the fields one at a time means an odd field that cannot be drawn costs that field's appearance instead of the whole job, and the count of skipped fields is reported rather than hidden.

Some fields refuse a value, and the tool carries on. A read-only field, or a dropdown given a value that is not in its option list, is counted as skipped and the other forty answers are still written. Values naming a field this document does not have are dropped quietly, because that is almost always what is left over after swapping one file for another.

Embedding a font matters more than it sounds. With it on, Noto Sans is embedded - around 2 MB, served from this site's own origin and cached after the first use - so an answer containing an accent, Greek or Cyrillic appears as you typed it. With it off, the reader's built-in Helvetica is used, which adds no bytes but can only encode WinAnsi. Neither covers CJK, Arabic or Indic.

Nothing is uploaded, which for this tool is worth more than usual: the answers on a form are frequently a full name, an address, a bank account and a date of birth. The document and everything you type into it stay in the browser tab, the fill runs in a Web Worker, and the only thing the network tab shows is the allowance check - a tool name and a hash, never a field you filled in.