4 ms·
The example of grabbing a page and feeding it into d3 is whizbang! ( https://github.com/pigshell/pigshell#hello-world https://github.com/pigshell/pigshell#hello
by andyidsinga 11y ago
The example of grabbing a page and feeding it into d3 is whizbang! ( https://github.com/pigshell/pigshell#hello-world https://github.com/pigshell/pigshell#hello-world )
I tried to cat this hn page, but also got the CORS error -- trying to route through [h]ttp://localhost:50937/...
need to run psty for proxying certain things and local access ...so its not "pure client-side Javascript app running in the browser"
pretty cool nonetheless -- maybe psty could be baked into a chrome or FF extention (?)
- tubelite 11y agoTry ycat https://news.ycombinator.com https://news.ycombinator.com | hgrep .athing | html ycat uses YQL to bypass CORS. There are a few interesting workflows which can be done completely in the browser without a proxy: e.g. copying files from one Google Drive account to another, backing up Google Drive to Dropbox, and so on.