Bulk download: many formats at once
Select as many formats as you need, choose a size and a file condition, and download the whole set as one ZIP archive. Useful for building a fixture directory or testing an importer against many formats in a single pass.
Everything is generated in your browser, so nothing is uploaded and there is no limit on how many files you create.
What comes out is one archive with one file per format, named after the format and its condition, so an invalid CSV and a valid one sit side by side and are told apart by the filename. That is the shape a fixture directory usually wants: a folder you can point a test suite at and iterate over, rather than files fetched one at a time.
The size you choose applies to every file in the set, which is what makes the archive useful for size-boundary work. Formats have different minimum sizes, though — structure costs bytes before your content exists, so an ISO or an XLS cannot be as small as a text file. Where a format cannot reach the size you asked for, it is generated at its own floor rather than being skipped or silently truncated.
The same applies to file conditions. Every format can be generated valid; most can also be generated invalid, corrupted or incomplete, and the damage is applied to the bytes rather than the filename, so a parser rejects the file for the reason you would expect. Selecting several conditions at once produces one file per format per condition, which is how a rejection-path suite gets built in a single download.
Two practical limits, both measured rather than estimated. Generation happens in the tab, so a large selection at a large size is bounded by the memory your browser will give the page; the download hint under the button reports the formats in your selection that take noticeable time. And combining several content options raises each format's minimum size, because every option adds a section of its own.
Verifying the archive is worth doing before you build tests on it. Any ZIP tool will list the entries and their sizes, which confirms the count and the byte sizes you asked for; opening two or three of the files in their real applications confirms the rest. That is the same check used here before a format is enabled — a file that a parser accepts, not a file that merely has the right extension.