Philou's Jekyll and Minimal Mistakes Blogging Goodies

Tips and automation scripts for blogging with Jekyll and Minimal Mistakes.

Installation

Create a sub-module in your blog repo under the _scripts directory.

Usage

Run the scripts from the root of your blog repo, _scripts/_<command>. They will print out an example usage line.

Scripts

_absolutify_links.sh to make link absolute so that they display well emails
_check_fixmes.sh to grep all FIXME comments that might be left in the files
_check_links.sh to grep for all bad links (ex: relative, or local only)
_grep_tweets.sh greps a post for bold sentences that could be tweeted
_new_post.sh creates a new post skeleton
_preview.sh runs a local server to see local changes
_word_count.rb Computes the number of words in a post

Notes and Troubleshooting

Here are common reminders and fixes.

How to add comments in markdown

[comment]: # (this is a comment)

See https://stackoverflow.com/questions/4823468/comments-in-markdown for the explanation

How to setup on windows

Check the ruby version in ../.ruby-version.
Head to https://rubyinstaller.org/downloads/, and download the appropriate Ruby+Devkit installer
Run the installer, when asked for what to install in the command line, choose option '3'
Open a git-bash and head to the blog main dir
Run bundle install
Run _scripts/_preview.sh and check that the website is available

How to update gem dependencies

Gems need to be updated regularly:

brew upgrade rbenv ruby-build

rbenv install --list

rbenv install xxx

Update .ruby-version

bundle update

Test with _scripts/_preview.sh and _scripts/_local_preview.sh

How to include figures with links and credits

Here is the template I am using :

{% assign figure_path="/imgs/...todo.jpg" | absolute_url %}

{% assign figure_alt="TODO" %}

{% capture figure_caption %}
TODO markdown
{% endcapture %}

{% include figure image_path=figure_path alt=figure_alt caption=figure_caption %}

Images best sizes

Teaser images : 600x315, 72dpi, compressed for minimal size (appears on the front page!)
Open Graph images : 1800x945, 250dpi, not compressed to appear well on high precision devices

How to license an image

Visual stamp

Might be easier to resize the image to 1800x945 ratio (same as teaser images)
Stamp the image, use gimp and add the layer from insert-cc.xmp
Resize image as needed
Add credits to the image in the page

{% assign figure_path="/imgs/...todo.jpg" | absolute_url %}

{% assign figure_alt="TODO" %}

{% capture figure_caption %}
By [Philippe Bourgau]({{site.url}}), under [CC BY-SA 4.0](http://creativecommons.org/licenses/by-sa/4.0/), [high resolution image]({{site.url}}{{site.baseurl}}/imgs/...todo.jpg)
{% endcapture %}

{% include figure image_path=figure_path alt=figure_alt caption=figure_caption %}

Add metadata

Generate .xmp metadata from Creative Commons website
Add the metadata to images with exiftool -tagsFromFile CC_Attribution-ShareAlike_4.0_International.xmp xxx1.jpg xxx2.jpg...

Allowing people to retweet an image

From https://www.christiankonline.com/twitter-retweet-link-tutorial/

First send a tweet with the high-res image (cf Images best sizes)
Write down the tweet-id
Add this markdown under the image, with the new tweet-id [Retweet this image](https://twitter.com/intent/retweet?tweet_id=\<TWEET_ID\>)

Extracting bold sections from a post

This can be useful to generate tweets or variations for Recurpost.

_scripts/_grep_tweets.sh POST_FILE_PATH

This dump usually needs further editing.

TODO:

This does not take the quotes into account
It would be nice to uppercase the first letter

Dumping the word count per post

Just run the following command:

find _posts -type f -and -name "*.markdown" -exec bundle exec ruby _word_count.rb {} \; > tmp/word_counts.csv

You'll need the following pre-requisites:

Ruby
Python3
Init submodule markdown-word-count

philou/jekyll-minimal-mistakes-goodies