-
Notifications
You must be signed in to change notification settings - Fork 24
Description
Snap2HTML, being a Windows-only program, gets to make certain assumptions about the files it's expected to handle. This also means it can use a bespoke data format to interpret in its html, delimited by *.
The problem for us Linux users is that a file can basically have anything but / and \0 as a character. And while I have no intention to handle \n in filenames, * is problematic since it's the delimiter. This can cause issues like files apparently missing in the manifest. I don't think there's any working around this without resorting to a well-specified data format, and json is the best, most readily available option.
edit feb 2025: I have for a long time been waiting on a new template system to rely on. The original template used in Snap2HTML (and by extension LinuxDir2HTML) is called DynaTree, which itself has been replaced by a software called FancyTree, which once again itself has been replaced by a software called Wunderbaum. Wunderbaum, when I last looked a couple of years ago, was in heavy development and at the time was probably not suitable to use. That situation has changed, and it looks acceptable for use now.
The expected benefits of Wunderbaum are a more robust code base (better export options), a much smaller template (no jQuery dependency), and hopefully faster performance.
So what's that got to do with JSON? Not too much, directly. Snap2HTML's bespoke file formatting, however, is intertwined with the template, and I've been wanting to switch the template for a long time, so these two items have been a sort of atomic unit of work that I wanted to handle all at once. Now that I'm more confident about Wunderbaum's state, I'll put LinuxDir2HTML in my dev queue soon enough.