Convert files

What it does

It takes files of different kinds all at once and converts each one wherever it can go. Images become PNG, JPG, WebP or a one-page PDF; an old Excel spreadsheet becomes .xlsx; an old Word document, an RTF, a saved web page or a plain text file becomes .docx.

It is the same work the single tool pages do, with one difference: there the target format is decided by the page, and you have to know which one to open first. Here the files go in as they are — twenty at a time, mixed — and the choice is per row.

The format is read from the bytes, not the extension

A file named .xls is very often something else entirely: half the web’s «export to Excel» produces an HTML table, and some business software writes a CSV and gives it the wrong extension. On the Word side, a .doc is often an RTF. Opened with the right program they work, and that convinces you they are what they claim to be.

Here the first bytes are read and what was found is said, before converting. If you loaded what you believed was an Excel spreadsheet and the row says «an HTML page or table», that line is the answer to every question that would come after — and the destinations in the menu are the real ones for that file, not the ones the extension promised.

A .doc and an .xls also have the exact same signature: they are the same Microsoft container. For those, the inside is read too, because the difference between a document and a spreadsheet is not a detail when the menu offers both.

What becomes what

Every row’s menu holds only the destinations for which an engine really exists. There is no entry that opens and then fails, and no hidden entry: if a row has no menu, the reason is written beside it.

Images are read in every format your browser can open and written back in three — PNG, JPG and WebP — plus PDF. An .xlsx or a .docx that lands in here has no menu, and the row says it is already in a new format: converting it would remake the same file while looking like it had done something.

A PDF is not read, it is merged

Nothing comes out of a PDF, and that is not an oversight. Reading a PDF — pulling out its text, its tables, its images — is a different class of work from everything else on this page, and doing it halfway would produce files that open and do not hold what they seemed to.

What can be done is done: from two PDFs up, a button appears that merges them into a single file, in the order they sit in the list.

What it does not do

Quality is not adjustable and dimensions are not changed. JPEG comes out at a high, fixed quality, images are not resized, and transparency gets white without asking. Those are the three things the single tools let you choose: one more control per row, across twenty rows, costs more than it gives.

Macros, charts, layout and complex formatting do not come through. The data and the text come out, not the work done around the data and the text.

Password-protected files do not open. There is no partial way to read them, and trying would produce a document of random characters. The protection has to be removed first, with the program that applied it.

Animated images become their first frame, and it is worth knowing before converting a GIF.

Large files are a memory problem. Everything happens in the browser, so the limit is the tab’s. Past a certain size a warning appears — not a refusal, because on a recent machine that file goes through, and saying no would be a lie about the reader’s machine.