/dash-docset-tensorflow

dash/zeal docset for tensorflow

Primary LanguageCSSApache License 2.0Apache-2.0

dash-docset-tensorflow

screenshot

View TensorFlow docs in the dash/zeal offline docset browser.

Dash/Zeal Feeds

To use, you can add the following feeds in Dash/Zeal:

  • v1.15: https://raw.githubusercontent.com/ilyakava/dash-docset-tensorflow/master/TensorFlow-v1.15.xml
  • v2.2: https://raw.githubusercontent.com/ilyakava/dash-docset-tensorflow/master/TensorFlow-v2.2.xml

Or download archived docsets as releases here.

Steps to generate the docset

  • Install dashing: go get -u github.com/technosophos/dashing
  • pip install --user python-magic beautifulsoup4 lxml
  • cd THIS_REPO
  • wget -nc -np --domains=www.tensorflow.org -e robots=off --adjust-extension -r 'https://www.tensorflow.org/api_docs/'
  • cp dashing.json icon*.png main.css www.tensorflow.org
  • ./preprocess.sh www.tensorflow.org
  • cd www.tensorflow.org/
  • dashing build will give you a TensorFlow.docset folder.

Right now this dashing.json only roughly parses function names (which is enough for most use cases). Feel free to add more features and contribute!

Installing a local docset into Zeal

  • Make an archive: tar --exclude='.DS_Store' -cvzf <docset name>.tgz <docset name>.docset
  • Make an xml file whose name will be the docset display name in Zeal. Its contents should be:
<entry>
    <version>1.15</version>
    <url>file:///path/to/archive/TensorFlow-vX.X.docset.tgz</url>
</entry>
  • In Zeal go to Tools > Docsets > Add feed and for url type 'file:///path/to/docset.xml'