Proofreader takes a website, or a local file, and tries to proofread it using write-good and nodehun.
npm install proofreader -g
proofreader -u https://raw.githubusercontent.com/GoogleChrome/devtools-docs/master/docs/memory-analysis-101.html
proofreader -f ../devtools-docs/docs/commandline-api.md
proofreader -l list-of-files.txt
proofreader -c custom-config.json -f file.html
- blue suggestions come from write-good
- magenta suggestions come from nodehun
Proofreader can handle both HTML and Markdown files. It distinguishes between these two using MIME types.
Downloads and processes single remote file from given URL.
Processes single local file from given path.
Processes all sources listed in the provided file. Sample list file:
../docs/file.html
/home/developer/otherfile.md
http://localhost/remote-file.md
Path to a custom configuration file (default one is in settings.json
). This file has to be a valid JSON. Sample configuration:
{
"dictionaries": {
"build-in": ["en_US", "en_GB"],
"custom": ["devtools-docs.dic"]
},
"selectors": {
"whitelist": "p, li, h1, h2, h3, h4, th, td, dl, figcaption",
"blacklist": "pre, code"
},
"write-good": {
"weasel": false
}
}
- dictionaries
- build-in - one or two of build in dictionaries (
eng_GB
,eng_US
). E.g. when both American English and British English are allowed,["en_US", "en_GB"]
should be specified. - custom - list of custom dictionaries
- build-in - one or two of build in dictionaries (
- selectors
- whitelist - CSS selector that specifies all elements that should be processed. This also applies to Markdown which is compiled to HTML before processing.
- blacklist - All elements that match this CSS selector will be removed before proofreading.
- write-good - Additional settings for write-good (more details here).
Please note that this project was:
- optimized for Chrome DevTools docs
- optimized for HTML and Markdown
- optimized for English
- by default does not process all the tags, only whitelisted ones (e.g. P, LI, H1, H2, H3)