Medup

Sync all content from Medium or Dev.to with local folder via API

Table of Contents

• Usage
• Features
• Getting Started
• Contributing
• Contributors
• License

Usage

Export post

Export a single medium post to local

medup https://medium.com/notes-and-tips-in-full-stack-development/medup-backups-articles-8bf90179b094

Export a single dev.to post to local

medup https://dev.to/jetthoughts/the-trial-period-for-staff-augmentation-in-jetthoughts-1h5c

Author's posts

Export all articles written by single author to special folder posts/miry

medup -u miry -d posts/miry

Alternative:

medup @miry -d posts/miry

Similar for Dev.to

medup --platform=devto @miry
medup --platform=devto -u miry
medup https://dev.to/miry

User's recommendations

Export all articles recommended by miry:

medup -u miry -d posts/recommendations -r

Alternative:

medup @miry -d posts/recommendations -r

Publication's posts

Export all articles written by single author to special folder posts/miry

medup -p jetthoughts -d posts/jetthoughts

Alternative:

medup jetthoughts -d posts/jetthoughts

Similar for Dev.to

medup --platform=devto jetthoughts
medup --platform=devto -p jetthoughts
medup https://dev.to/jetthoughts

Assets

By default all images are embedded in the article. It is possible to save images in separate folder with arguments - --assets-images

$ medup @miry --assets-images -d posts/miry
$ tree posts/miry
posts/miry
├── 2017-03-20-war-with-ads-and-trackers.md
├── 2017-03-26-setup-k8s-v1-6-0-rc-1-cluster.md
├── 2017-03-28-docker-network-performance.md
...
└── assets
    ├── 0*BmauN6EER-PdvCee.jpeg
    ├── 0*FbFs8aNmqNLKw4BM.jpeg
    ├── 0*LZaURw4xtfA74nu9.jpeg
...

Specify assets directory

To make it is even more agile and work with Jekyl, put assets in different directory - --assets-dir=DIR

$ medup @miry --assets-images -d site/posts --assets-dir=site/assets
$ tree site
site
├── assets
│   ├── 0*BmauN6EER-PdvCee.jpeg
│   ├── 0*FbFs8aNmqNLKw4BM.jpeg
│   ├── 0*LZaURw4xtfA74nu9.jpeg
...
└── posts
    ├── 2017-03-20-war-with-ads-and-trackers.md
    ├── 2017-03-26-setup-k8s-v1-6-0-rc-1-cluster.md
    ├── 2017-03-28-docker-network-performance.md
...

Specify assets base path

Also change the path in the document to find assets with --assets-base-path=URI. Default base path is ./assets/ASSET_NAME.XXX. Example image: ![](./assets/1*hBBZVQZHFYJ0WWaj08bHuA.jpeg). With --assets-base-path=/static the src would be /static/ASSET_NAME.XXX. Example image: ![](/static/1*hBBZVQZHFYJ0WWaj08bHuA.jpeg).

$ medup @miry --assets-images -d tmp/posts --assets-dir=tmp/static --assets-base-path=/static

ENVIRONMENT

• MEDUP_GITHUB_API_TOKEN Use this GitHub personal access token when accessing the GitHub Gist.

Mix

The neat feature to mix mutliple rules in single command

medup -u miry @pftg -p jetthoughts notes-and-tips-in-full-stack-development \
      https://medium.com/notes-and-tips-in-full-stack-development/medup-backups-articles-8bf90179b094 \
      https://dev.to/jetthoughts \
      https://dev.to/jetthoughts/the-trial-period-for-staff-augmentation-in-jetthoughts-1h5c

It would download all articles for:

1. user @miry
2. user @pftg
3. publication jetthoughts
4. publication notes-and-tips-in-full-stack-development
5. single post "Medup backups articles"
6. publication jetthoughts from dev.to
7. single post "The Trial Period for Staff Augmentation in JetThoughts" from dev.to

Features

• Discover all articles from user account available in public
• Allow to download all recommended articles by user
• Discover all articles from publisher, that available in public
• Download images used inside article or in assets folder
• Save posts in markdown format
• Convert a single article in markdown format
• Extract gists and convert to code blocks

Getting Started

Homebrew

Install medup tool to the MacOS or Linux via Hombrew:

brew tap miry/medup https://codeberg.org/miry/medup
brew install medup

Run export command for Medium author miry and save articles to local folder:

medup -u miry -d .
medup -d . https://medium.com/notes-and-tips-in-full-stack-development/medup-backups-articles-8bf90179b094

Docker

Docker way to make same job:

docker run -v <path to local articles folder>:/posts -it miry/medup -u <user>

Download all articles that user has recommended:

docker run -v <path to local articles folder>:/posts -it miry/medup -u <user> --recommended -d posts/recommended

Crystal

Run dumping with source code

shards install
crystal run src/cli.cr -- -u <medium user> -d <destination folder>

Example download all posts from author https://medium.com/@miry to local folder posts/miry

crystal run src/cli.cr -- -u miry -d posts/miry

Build

Build a application binary and execute:

shards install
rake build
_output/medup -u <medium user> -d <destination folder>

Result

By default all articles destination folder is posts.

In the directory, you can find 2 format of files: .json and .md.

• JSON format is the raw, what Medium returns.
• Markdown format is simple implementation of block formated text.

Images encoded in the result document. Rich elements content are located in posts/assets and integrated with <IFRAME>.

Markdown

Example of exported document.

1. Original Medium post vs Exported Markdown:
2. Exported Markdown in Github pages vs original Dev.to

Demo

There are possible solution to test export markdown files. Run a test Jekyll server with posts. Example:

rake demo:serve

And after it finishes you can test http://localhost:4000. Update posts for demo:

medup -v7 -d demo/_posts/ --assets-dir=demo/assets --assets-base-path=/assets <url>|<@user>|<publisher>

Test Bridgetown use next command:

rake demo:bridgetown:serve

Bridgetown has a bit different structure for posts. Use to update posts:

medup -v7 -d demo/src/_posts/ --assets-dir=demo/src/assets --assets-base-path=/assets <url>|<@user>|<publisher>

Contributing

1. Fork it ( https://github.com/miry/medup/fork )
2. Create your feature branch (git checkout -b my-new-feature)
3. Commit your changes (git commit -am 'Add some feature')
4. Push to the branch (git push origin my-new-feature)
5. Create a new Pull Request

Contributors

• @miry Michael Nikitochkin - creator, maintainer
• @alexanderadam Alexander - supporter
• @clawfire Thibault Milan - supporter

License

This project is under the LGPL-3.0 license.

Play ownself

FAQ

• Clean unused assets:

for i in $(ls -1 assets/*); do grep -q "$i" *.md || echo $i; done | xargs rm

• Detect potential articles with IFRAME content for gists:

for i in $(ls -1 assets/*.html); do grep "gist.github.com" -q $i && grep -n "$i" *.md; done

• Detect potential articles' url with IFRAME content for gists:

for i in $(ls -1 assets/*.html); do grep "gist.github.com" -q $i && (grep -l "$i" *.md | xargs head | grep url); done

• Detect potential articles' url with IFRAME content for youtube:

for i in $(ls -1 assets/*.html); do grep "schema=youtube" -q $i && (grep -l "$i" *.md | xargs head | grep url); done

Medium API

1. Medium API documentation

2. Generate token on the page https://medium.com/me/settings

3. Create environment variable MEDIUM_TOKEN=<token>

4. Verify token with sample query:

curl -H "Authorization: Bearer <token>" https://api.medium.com/v1/me
{"data":{"id":"number","username":"miry","name":"Michael Nikitochkin","url":"https://medium.com/@miry","imageUrl":"https://cdn-images-1.medium.com/fit/c/400/400/0*KgbjgGnH-csHuB8j."}}

Crawler

1. Check public information not covered by API

curl "https://medium.com/@miry?format=json" | cut -c17- && : Remove in the front from response some strange JS code.

2. Pagination

curl "https://medium.com/@miry/latest?format=json&limit=100" | cut -c17- && : Remove in the front from response some strange JS code.

3. Post info

curl -s -H "Content-Type: application/json" https://medium.com/@miry/c35b40c499e\?format\=json\&limit\=100

4. Stream

curl -s -H "Content-Type: application/json" "https://medium.com/_/api/users/fdf238948af6/profile/stream?source=overview" | cut -c17-
curl -s -H "Content-Type: application/json" "https://medium.com/_/api/users/fdf238948af6/profile/stream?limit=100&page=3" | cut -c17- > stream.json
cat stream.json| jq ".payload.references.Post[].title"
cat stream.json| jq ".payload.paging.next"

5. Recommendations

curl -s -H "Content-Type: application/json" https://medium.com/@miry/has-recommended | cut -c17-
curl -s -H "Content-Type: application/json" "https://medium.com/_/api/users/fdf238948af6/profile/stream?source=has-recommended" | cut -c17-