1.0 This work incorporated my personal preference and removed the need to serve markdown indirectly. Also relatively convenient and portable.

1.1 Now the repo is serving as a subfolder, which leaves your work directory (markdown and html) cleaner.

1.2 Centered around a new html-rich "slide.md" file, with +js,css customizable scripts. Also have md_strip_html.py to strip down to plain markdown file.

Thanks to theno's selection of tools and jgm's pandoc for converting markdown to html.

plain text

• 1
  • 1.1
    • 1.1.1
      • 1.1.1.1
        • 1.1.1.1.1
• formula
  • $e^{ \pm i\theta } = \cos \theta \pm i\sin \theta$
  • $\displaystyle F1=\frac{2PrecisionRecall}{Precision+Recall} = \frac{2TP}{2TP+FP+FN}$

1. d1
   1. d2
      1. d3
         1. d4
            1. d5

Contact Rui for any questions 😊

apt update && apt intall curl wget pandoc # sudo if needed

# download the latest release to subfolder [reveal_slides]
curl -s https://api.github.com/repos/ruixiao85/reveal_slides/releases/latest \
| grep "tarball_url" \
| cut -d \" -f 4 \
| xargs wget -qO - | tar zxvf - && \
mv `ls -td *reveal_slides*/ | head -1` reveal_slides

If you don't have any, you can copy the included example to your work directory:

cp reveal_slides/slide.md . # template needed to create slides
cp reveal_slides/slide*.html . # pre-rendered html slides

bash reveal_slides/md_revealjs.sh # [input markdown] [output html]

Optional positional arguments allow you to pass other markdown file and output html. If not provided, default to [slide.md] and output to [slide.html]. The script content is shown below:

#!/usr/bin/env bash
echo "self, scpt=${0}"; scpt=${0}
FI=${1:-slide.md}; echo "arg1, fille_input=$FI"
FO=${2:-${FI%.*}.html}; echo "arg2, file_output=$FO"

SCRIPT_DIR=$( cd -- "$( dirname -- "${BASH_SOURCE[0]}" )" &> /dev/null && pwd )

cat ${SCRIPT_DIR}/revealjs_before.html > ${FO}
pandoc -f markdown -t revealjs --no-highlight --number-sections --slide-level 3 ${FI} >> ${FO}
cat ${SCRIPT_DIR}/revealjs_after.html >> ${FO}

bash reveal_slides/md_revealjs.py # [input markdown] [output html] [heading split level, 0-nosplit, 1-split@heading1 '#']

This allows you to further split at high heading levels, for example, "#". As a result, the input markdown file will be split into multiple markdown files and lower-level headings will move up accordingly.

• main.md

# topic 1
## section 1.1
## section 1.2
# topic 2
## section 2.1

• main_topic1.md

# section 1.1
# section 1.2

• main_topic2.md

# section 2.1

python3 reveal_slides/md_strip_html.py # [input markdown with html] [output markdown without html]

This removes all style blocks, any line starts or ends with '<' or '>', and convert img tag src="pic.jpg" to markdown styled .

• You should be able to directly open the output html in any browser. The subfolders [css, js, lib, plugin] are required for the html file to function properly.

• "#" heading 1 may require the entire slide without any other content for the verticle movement to work (2-level hierachy).