8886a9940d
The basic idea is that in the main mungedocs we run the entirefile and create an annotated set of lines about that file. All mungers then act on a struct mungeLines instead of on a bytes array. Making use of the metadata where appropriete. Helper functions exist to make updating a 'macro block' extremely easy.
996 B
996 B
Documentation Mungers
Basically this is like lint/gofmt for md docs.
It basically does the following:
- iterate over all files in the given doc root.
- for each file split it into a slice (mungeLines) of lines (mungeLine)
- a mungeline has metadata about each line typically determined by a 'fast' regex.
- metadata contains things like 'is inside a preformmatted block'
- contains a markdown header
- has a link to another file
- etc..
- if you have a really slow regex with a lot of backtracking you might want to write a fast one to limit how often you run the slow one.
- each munger is then called in turn
- they are given the mungeLines
- they create an entirely new set of mungeLines with their modifications
- the new set is returned
- the new set is then fed into the next munger.
- in the end we might commit the end mungeLines to the file or not (--verify)