Tangle

Tangle is an unmaintained hypertext topology analysis tool for static websites, written in Python and hosted on github. You set it loose on a collection of HTML files and other paraphernalia in a directory, and it spits out information about how the documents are all connected. This can be useful, for example, if you want to move an HTML document from one directory to another but don't know what local links will be broken.

There were a couple of problems with Tangle:

• Uses html5lib, so doesn't necessarily parse strange cases correctly
• Enjoy parsing CSS and JavaScript links

Both of these problems are due to Tangle using python rather than being based in a browser. But using a browser for this code would slow it down considerably, whereas a tool for reading files in a directory can be a lot faster. A browser based tool could perhaps cache results into local storage and then only update based on modification times. But even fetching modification times over HTML using HEAD is slower than doing it on the filesystem. Perhaps modification times could be aggregated into a file somehow, annotating the filesystem using actual files.

@@ Behind the name.

Reforms

A successor Tangle architecture could be a mixture of Python and other languages. It could be a federation of small tools doing their jobs properly, in the unix way:

"This is the Unix philosophy: Write programs that do one thing and do it well. Write programs to work together. Write programs to handle text streams, because that is a universal interface." (Doug McIlroy)

Perhaps these days we'd use JSON as a universal interface, as a subset of a text stream. At any rate, here are some possible small tools:

• Script that uses WebKit to turn HTML into a DOM (dumped as JSON?)
  
• Extract links from a (JSON?) DOM
• Get a topology from a set of files
• Spot orphans within a topology
• Create a graph from a topology