Skip to content

Latest commit

 

History

History
59 lines (42 loc) · 1.4 KB

README.md

File metadata and controls

59 lines (42 loc) · 1.4 KB

Wordclouds in go.

alt text

How to use

wordCounts := map[string]int{"important":42, "noteworthy":30, "meh":3}

w := wordclouds.NewWordcloud(
	wordCounts,
	wordclouds.FontFile("fonts/myfont.ttf"),
	wordclouds.Height(2048),
	wordclouds.Width(2048),
)

img := w.Draw()

Options

  • Output height and width
  • Font: Must be a valid TTF file.
  • Font max,min size
  • Colors
  • Background color
  • Placement : random or circular
  • Masking

Masking

A list of bounding boxes where the algorithm can not place words can be provided.

The Mask function can be used to create such a mask given a file and a masking color.

boxes := wordclouds.Mask(
	conf.Mask.File,
	conf.Width,
	conf.Height,
	conf.Mask.Color,
)

See the example folder for a fully working implementation.

Speed

Most wordclouds should take a few seconds to be generated. A spatial hashmap is used to find potential collisions.

There are two possible placement algorithm choices:

  1. Random: the algorithms randomly tries to place the word anywhere in the image space.
    • If it can't find a spot after 500000 tries, it gives up and moves on to the next word. It's quite slow.
  2. Spiral: the algorithm starts to place the words on concentric circles starting at the center of the image. It is very fast and is the default algorithm

Contributing

Feel free to create pull requests, I'll gladly review them!