Franco Masotti
cea717a494
|
||
|---|---|---|
| _assets | ||
| _comments | ||
| _includes | ||
| _layouts | ||
| _media/2017-02-11-another-post | ||
| _pages | ||
| _posts | ||
| _sass | ||
| .gitattributes | ||
| .gitignore | ||
| .pre-commit-config.yaml | ||
| LICENSE.md | ||
| Makefile | ||
| Pipfile | ||
| README.md | ||
| _config.yml | ||
| robots.txt | ||
README.md
The Flux Of Thought
A full redesign of the-plain-libre Jekyll theme.
Table of contents
- The Flux Of Thought
- Table of contents
- Demo
- Philosophy
- Versions
- Dependencies
- Installation
- Upgrading
- Building and serving
- Tutorial
- Tests
- This version compared to the The Plain Libre
- Useful resources
- Copyright and License
Demo
https://frnmst.github.io/the-flux-of-thought/
Philosophy
- No Javascript is used. See The JavaScript Trap.
- No local fonts are served.
- No remote fonts are included (called).
- A very minimal CSS is served.
Versions
See all the-flux-of-thought releases.
Dependencies
The theme is known to work with the the packages listed here and they must be installed on your system.
| Package | Executable | Version command | Package Version |
|---|---|---|---|
| Ruby | /bin/ruby |
$ ruby --version |
ruby 2.6.5p114 (2019-10-01 revision 67812) |
| RDoc | /bin/rdoc |
$ rdoc --version |
6.2.0 |
| Jekyll | ~/.gem/ruby/2.5.0/bin/jekyll |
$ jekyll --version |
jekyll 4.0.0 |
| Jekyll Sitemap Generator Plugin | - | - | 1.4.0 |
| GNU Make | /bin/make |
$ make --version |
GNU Make 4.2.1 |
| krmadown | ~/.gem/ruby/2.5.0/bin/kramdown |
$ kramdown --version |
2.1.0 |
| Rouge | ~/.gem/ruby/2.5.0/bin/rougify |
$ rougify --version |
3.13.0 |
Installation
Arch Linux based distros
# pacman -S ruby ruby-rdoc
$ gem update
$ gem install jekyll jekyll-sitemap kramdown rouge
Upgrading
Upgrading from Jekyll 3.x to Jekyll 4.x
In case you get something like 'private': nil is not a symbol nor a string (TypeError),
run:
$ gem uninstall forwardable --version 1.3.0
$ gem install forwardable --version 1.2.0
See also this issue and this document
Upgrading from the flux-of-though 0.x to 1.x
Yaml keys in the configuration file have changed. Be sure to to fix your configuration file appropriately.
Upgrading from the flux-of-though 1.x to 2.x
The original CSS has been replaced with a minimal CSS. See this
Building and serving
Locally
Local serving, on 127.0.0.1
$ make
Global serving, on 0.0.0.0
$ make serve-global
Connect to http://127.0.0.1:5555/the-flux-of-thought/.
GitLab Pages
If you use GitLab Pages put the
following in ./.gitlab-ci.yml. See also https://gitlab.com/pages/jekyll
image: ruby:2.6
test:
stage: test
script:
- gem install 'jekyll:4.0.0' jekyll-sitemap 'forwardable:1.2.0'
- jekyll build -d test
artifacts:
paths:
- test
except:
- master
pages:
stage: deploy
script:
- gem install 'jekyll:4.0.0' jekyll-sitemap 'forwardable:1.2.0'
- jekyll build -d public
artifacts:
paths:
- public
only:
- master
Tutorial
Commenting system
Overview
This theme handles comments in the most static way possible:
a mailto unique link is generated for each post, using the address specified
in the _config.yml file. This idea came out while
reading this
page.
Moreover, for each post a comment section gets exposed. To get comments inside
this section, you must create a new file in the _comments directory using the
corresponding posts path name as file name. Let's say a user sends a comment
for the another-post post. The mailto link will have this structure:
<your_email_username>+2017-02-11-another-post@gmail.com
This works for any mailer system supporting these kind of mail aliases, for example GMAIL.
You may notice the substring 2017-02-11-another-post after the email
username which is part of the post's file name:
./_posts/2017-02-11-another-post.md
Let's ingnore the ./_posts/ and .md substrings for a moment and save
2017-02-11-another-post somewhere. Since a post may have more than a
comment, we must distinguish between one and another. To do this we create a
new directory using the comment path, like the following:
$ mkdir _/comments/2017-02-11-another-post
and we will save the comment files inside to keep things tidy.
We then create a new markdown file corresponding to the comment, and its file name will be used as an id, for example:
0.md
The filename must not have whitespace characters and the substring .md except
at the end of it.
Out final comment path is:
./comments/2017-02-11-another-post/0.md
As a final remark you may have noticed that each comment is referenced as a
paragraph (using the HTML id attribute) and that markdown is enabled by
default.
Comment attributes
Comment files have 5 attributes:
-
layout- Always use
commentas variable.
- Always use
-
date-
Use the format reported by the following date command:
$ date "+%F %T %z"
-
-
from- string representing the name or the id or the mail of the person that commented the post.
-
subject- String representing the mail subject.
For example:
---
layout: comment
title: This is another post
date: 2017-02-11 23:37:05 +0100
from: x.y@y.x, A Mailer
subject: test subject
---
This is a test comment for the `Hello again` post.
```python
def hi
```
Media files
Extending these ideas I came up with the following:
Overview
Just like comments, each media file is contained in one directory corresponding to a post. However, unlike comments, media content may be referenced from more than one post. This is how it works.
File locations
If out example post is ./_posts/2017-02-11-another-post.md and we want to add
media files to it, we must create a corresponding directory in _media:
./_media/2017-02-11-another-post
We can now place our file inside, for example:
./_media/2017-02-11-another-post/terminal.png
Possible combinations
Now, let's go back to ./_posts/2017-02-11-another-post.md. To be able to
display that picture we need to use the include liquid tag. In the simplest
form the only required parameter is the file name:
{% include image.html file="terminal.png" %}
You can also use the alt and/or capiton tags:
{% include image.html file="terminal.png" alt="ter" caption="A terminal example caption" %}
You can also use plain markdown, which is not advisable in this case, since you need to input the whole path:
Calling files from another post
In some cases you may want to recall media files from another post. You can do
that using the otherpost tag. Let's say we are in the
./_posts/2015-09-09-download-this-theme.md post.
{% include image.html file="2017-02-11-another-post/terminal.png" alt="Terminal" caption="A terminal image from the other post" otherpost=true %}
As you can see you don't have to specify the full path but only the post name slash the file name. The rest is filled in automatically.
Finally, notice the otherpost=true tag at the end.
Notice and warning
If you do not include the alt tag a default one will be provided instead.
This include feature is currently available only for images (img HTML tag).
Static pages
Reading this
gave me the idea to move the static pages like index, 404, etc.. in the
./_pages directory. If you want to add new pages remember to add a sensible
permalink for each page and remember to keep include: [ "_pages" ]
and:
collections:
pages:
output: false
in the _config.yml file.
Tags and categories
Overview
In this theme tags and categories are considered the same thing. I will use the word tag to speak about both tags and categories.
I gathered some ideas from here and using the official Jekyll documentation.
A list of tags is present on the home page, i.e: ./_pages/index.html.
This list is just a set of links pointing to the appropriate entry in the
./_pages/tags.md page.
The tags score system
A list of tags is present on the home page, i.e: ./_pages/index.html.
This list is just a set of links pointing to an appropriate entry in the
./_pages/tags.md page, which are presented using a score criteria. All
occurrencies of a tag are counted globally. If that frequency is greater or
equal than the tags.score.min variable in the _config.yml file then the tag
list element is showed along with its score. On the other hand if the frequency
is less than tags.score.min the tag will not be shown. If you want all the
tags to be shown, simply set tags.score.min to 1.
You can limit the number of tags from home and tags page by setting the
tags.score.link_limit variable to a positive integer number. If you set it false
the tags will not be limited.
You may have noticed that the tag list is presented in a descreasing number of occurrencies per tags. This method was chosen so that the relevant content is shown first.
Until version 0.0.5
Until version 0.0.5, the logic for all this was adapted from
here
and
here.
Since version 0.1.0
Since version 0.1.0 sorting is done first by tag score and then
alphabetically.
Some elements of logic were taken from here, here, here, here and the official Liquid documentation.
Tags page
The purpose of this page is to present links to posts for each tag in the website. This page reports all tags in the website without the tags score system except that the list will be sorted with descreasing frequency. The tags reported on the top of the page follow the tags score system.
Below the tags list there are links to each posts, organized by tags. If a user clicks on one of the links in the tags list, the viewport will be moved to the list of posts containing that tag.
Tags in posts
If a post contains at least one tag, this will be reported at the top of the
page with a link pointing to the appropriate bookmark of ./_pages/tags.md
(just like in the home page). A user looking at a post can see all related
posts just by clicking at those tag links.
The tag entry is optional which means that you can specify from 0 to n tags. If you want to add some, simply write your comma separated list of tags in the front matter of your post:
tags: [tag 0, tag 1, ... , tag n]
Let's make an example. What follows is the front matter of the
./_posts/2015-09-09-download-this-theme.md post.
---
title: Download this Theme
updated: 2017-04-09 00:29
categories: Download
tags: [download, terminal, image from another post]
---
As you may have noticed, tags can contain spaces.
Once you run make, you should see the following links
just below the Download this Theme title:
*download* *terminal* *image from another post*
Excerpts
An excerpt is a portion of a post reported in the home page.
Any content between <!--excerpt_start--> and <!--excerpt_end--> tags
in a post will be used as excerpt.
If one or both of these tags are omitted, the first 20 words of the post will be reported
instead. You can specify the number of words in the excerpt by editing the
excerpts.words variable in the ./_config.yml file.
If you are not interested in excerpts set the excerpts.enabled variable to
false in ./_config.yml.
The liquid code that handles the excerpts was inspired by this article. Here is its license and copyright:
The MIT License (MIT)
Copyright (c) 2015 Joshua Beam
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.
Anchor headings
Anchor headings are clickable links beside a <h[1-6]> tag. The purpose of
these elements is to get a quick link to a specific part of a page.
You can modify the appearence of these links from the ./_config.yml file
through the anchor_headings variables.
All this is possible thanks to the Jekyll Pure Liquid Heading Anchors library.
RSS feeds
This theme contains an Atom file useful for RSS feeds in ./_pages/feed.xml.
The link to feed.xml is available in the navigation bar of every page.
The purpose is that your readers open this link with a feed reader so that
they can remain up to date with your posts.
Search Engine Optimizaition (SEO)
I'm a noob in this field so I followed this tutorial and this SEO checker.
- meta
- sitemaps
- images
- always use alt tags. For images in
./_assetsthese are already coded in.
- always use alt tags. For images in
- responsiveness
- the theme is reported responsive by Google
- robots.txt
meta and other SEO tags
meta tags are available in the ./_includes/head.html file. These include:
title, description and canonical links.
The description meta tag uses the first available, in this order, of the following:
- Front matter
descriptiontag - Page excerpt
- Site description
It is advised to use the description tag since excerpts or the site
description may not relate to the (whole) page content. Here's an example:
---
title: This is another post
updated: 2017-04-09 23:00
tags: [other, liquid, terminal]
description: How to use markdown and liquid to render images belonging to a
post.
---
You can also edit the meta variable in ./_config.ymlto limit the number
of words in the description meta tag. Avoid values greater than 160 as
reported by previously cited articles:
Sitemap
A sitemap used for web crawlers is available thanks to the jekyll-sitemap plugin.
There is also a (very) basic sitemap for users under the /sitemap/ permalink
which has a link for it at the top of each page.
robots.txt
This very important file must be put under the root of your domain. It is used
by web crawlers (a.k.a bots, spiders, etc) to get access information about the
website. In our case it is also important to tell these programs that our site
has a sitemap. Read this
and
this
for more information.
For example, since this project is called the-plain-libre and is published
under https://frnmst.github.io/the-plain-libre, we need to put the robots.txt
file just under https://frnmst.github.io/, i.e
https://frnmst.github.io/robots.txt
Since I didn't have neither a "user page" nor a domain I created a dummy user
page repository and I copied the
robots.txt file from the generated ./_site directory from Jekyll. This
works for me. This means that in my case the robots.txt in this repository is
ignored by the bots.
I guess that if you have multiple repository pages you can edit the robots.txt file appropriately and set rules (and sitemap locations) for each one of them.
Mobile support
Mobile support is available for this theme through the CSS file through the
@media rules.
Compressed HTML output
By default Jekyll leaves lots of whitespace in the html source. You can verify this yourself.
Since useless white space means useless data -> we transmit useless bytes -> loading the website requires more time.
For this reason I decided to use a compressed html output thanks to
jekyll-compress-html,
which is a pure liquid layout to be used as a wrapper on the
./_layouts/default.html layout.
jekyll-compress-html is released under the MIT license.
Tests
This theme is guaranteed to work correctly with the following web browsers:
| OS | Browser | Executable | Version command | Browser Version |
|---|---|---|---|---|
| Parabola GNU/Linux-libre | Iceweasel | /bin/iceweasel |
$ iceweasel --version |
Mozilla Iceweasel 61.0.2 |
| Parabola GNU/Linux-libre | GNU Icecat | /bin/icecat |
$ icecat --version |
GNU icecat 60.3.0 |
| Parabola GNU/Linux-libre | w3m | /bin/w3m |
$ w3m -version |
w3m version w3m/0.5.3+git20180125 |
| Android 8.1.0 | Firefox Klar | - | - | 6.1.1 |
HTML validation
This Jekyll theme has been checked with the HTML5 W3C validator
This version compared to the The Plain Libre
There are too many things to compare, but in short terms:
- the CSS file was written from scratch.
- some Liquid templating language parts have been changed also because of the new CSS.
- new tags score system for the index page.
Useful resources
- https://www.w3.org/
- Especially useful for CSS examples: https://www.w3schools.com/default.asp
Copyright and License
The MIT License (MIT)
Copyright (c) 2015 Heiswayi Nrird
Copyright (c) 2017-2020 Franco Masotti franco.masotti@live.com
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.