ngrok exposes local servers behind NATs and firewalls to the public internet over secure tunnels.
For serving static files Amazon S3 is a good solution. In addition, if you have files in github or as a gist then you can not really serve from there because the Content-Type header does not come out the way that the browsers want to see that file. You can get around this issue by serving these files through a service that can recognize the file content type.
Atom Editor: https://atom.io/
- Atom's Github package works with atom editor: https://github.atom.io/
Create Git aliases - list of aliases
co = checkout
ci = commit
st = status
add-commit = !git add -A && git commit
alias = ! git config --get-regexp ^alias\. | sed -e s/^alias\.// -e s/\ /\ =\ /
RawGit is a service that solves this problem. You can serve github and gist files with proper Content-Type header using RawGit. This works for file types such as HTML, JSON, CSS, JavaScript.
Try these two examples:
- Raw github gist html file opens in the browser in raw html format. Try here.
- Same gist html file hosted via rawgit opens in the browser in rendered form. Try here.
FILE CONTENT-TYPE HEADER VIA GIST:
$ curl -s -I https://gist.githubusercontent.com/akaak/840a8c44a9054a35fcb13198a7d6dc9b/raw/ac3423492386541a280892f711b11197c6f83687/simple-page.html | grep "^Content-Type"
Content-Type: text/plain; charset=utf-8`
gives the Content-Type of text/plain. So, if you just pull up this url in the browser the browser does not render the page in HTML.
This is where https://rawgit.com/ comes in useful. By running a github file or gist url through rawgit this service creates a caches version of this file and serves it via CDN and most importantly serves with the proper Content-Type header based on the file type.
FILE CONTENT-TYPE HEADER VIA RAWGIT:
$ curl -s -I https://cdn.rawgit.com/akaak/840a8c44a9054a35fcb13198a7d6dc9b/raw/6f3b746a534e8f456492568f67ca9acade0372c1/simple-page.html | grep "^Content-Type"
Content-Type: text/html;charset=utf-8
curl -s -I https://cdn.rawgit.com/akaak/840a8c44a9054a35fcb13198a7d6dc9b/raw/6f3b746a534e8f456492568f67ca9acade0372c1/simple-page.html | grep "^Content-Type"`
Content-Type: text/html;charset=utf-8
You can use the -s flag to use the curl in silent mode. This does not display the progress meter -
via stackoverflow post
- Basic HTTP Auth for S3; protecting a Static website via basic authentication (service provided by http://s3auth.com)
Cron script to run every 10 minutes. Use http://crontab.guru for examples for cron and use crontab to setup cron.
*/10 * * * * /PATH-TO-SCRIPT/SCRIPT.sh
The SCRIPT.sh would be a shell script with the script to take the website screenshot and output to a file with a timestamp.
#!/bin/bash
/usr/local/bin/webkit2png -TF --delay=5 -o /PATH-TO-OUTPUT/FILENAME-`date +%m%d%Y-%H-%M-%S` --ignore-ssl-check http://www.WEBSITE.com
exit
Related Article: http://stackoverflow.com/questions/927358/how-to-undo-last-commits-in-git
Follow these four (4) steps to roll back or undo the latest commit
The commit message does not even get into the git log
Check the log with git log before and after the undo.
git reset --hard HEAD^
git commit -m 'rolling back'
git log --pretty=oneline
git push origin master --force
Check git log in simple (pretty) format
git log --pretty=oneline is useful to see the summary of log in one line. Git's viewing the commit history has many useful commands.
Generate a text file with 6 random dictionary words per line with a total of 10 Lines. (from http://www.skorks.com/2010/03/how-to-quickly-generate-a-large-file-on-the-command-line-with-linux/)
ruby -e 'a=STDIN.readlines;10.times do;b=[];6.times do; b << a[rand(a.size)].chomp end; puts b.join(" "); end' < /usr/share/dict/words > file.txt
last updated nov 2016 by @akaak