Lightweight, extensible, zero-dependency CLI written in Python to help us publish Obsidian notes with Hugo.
My Obsidian vault and my Hugo site are two completely different systems with wildly different directory structures. I do all of my writing in Obsidian and simply use Hugo as a place to publish that writing. As such, I wanted a tool that updated my Hugo content that is driven entirely by changes made in my Obsidian vault.
This tool directly copies specified directories into Hugo content
folder, maintaining the directory structure in the process. It flattens the rest of the vault and dumps into a content/writing
directory all while maintaining leaf bundles.
- Clears Hugo content directory
- Copies Obsidian vault contents into Hugo content directory
- Maintains folder structure for directories you specify
- Flattens and dumps all other notes to
content/writing
- Maintains all notes as Hugo leaf bundles
- Deletes all file types except for
md
files and images that are within a leaf bundle - Replaces Obsidian wiki links (
[[wikilink]]
) with Hugo shortcode links ([wikilink]({{< ref "wikilink" >}})
) - Replaces Obsidian image wiki links (
![[some_image.jpg]]
) with Hugo shortcode links ({{< image some_image.jpg \>}}
) - Replaces obsidian marks (
==important==
) with HTML marks (<mark>important</mark>
) - Removes imported Kindle highlights under any
## Highlights
header - Want to do more? You can write and register custom filters to dynamically include/exclude content from processing and processors to do whatever you want with the file contents.
Obsidian | Hugo |
---|---|
[[/some/wiki/link]] |
[/some/wiki/link]({{< ref "/some/wiki/link" >}}) |
[[/some/wiki/link|Some text]] |
[Some text]({{< ref "/some/wiki/link" >}}) |
[[/some/wiki/link/_index]] |
[/some/wiki/link/]({{< ref "/some/wiki/link/" >}}) |
[[/some/wiki/link#Some Heading|Some Heading Link]] |
[Some Heading Link]({{< ref "/some/wiki/link#some-heading" >}}) |
==foo bar=== |
<mark>foo bar</mark> |
![[some_image.jpg]] |
 |
![[some_image.jpg|692]] |
{{< image some_image.jpg Resize "692x" \>}} |
Note For now, there is no way to escape obsidian wiki links. Every link will be replaced with a hugo link. The only way to get around this is changing the wiki link to don't match the exact sytax, for example by adding an invisible space (Obsidian will highlight the invisible character as a red dot).
However, this still is really really not best practice, so if anyone wants to implement real escaping, please do so.
Obsidian structure must follow Hugo leaf bundle structure. If a note contains images, put the note and all images in the same folder with the note being named index.md to maintain leaf bundles and associated images.
For now, clone the project and install locally.
git clone https://github.com/Westley-Winks/obsidian-to-hugo.git
cd obsidian-to-hugo
make install_locally
usage: __main__.py [-h] [--version] [--hugo-content-dir HUGO_CONTENT_DIR]
[--obsidian-vault-dir OBSIDIAN_VAULT_DIR]
options:
-h, --help show this help message and exit
--version, -v Show the version and exit.
--hugo-content-dir HUGO_CONTENT_DIR
Directory of your Hugo content directory, the obsidian notes
should be processed into.
--obsidian-vault-dir OBSIDIAN_VAULT_DIR
Directory of the Obsidian vault, the notes should be processed
from.
This file goes in your Hugo site in scripts
.
from obsidian_to_hugo import ObsidianToHugo
obsidian_to_hugo = ObsidianToHugo(
obsidian_vault_dir="path/to/obsidian/vault",
hugo_content_dir="path/to/hugo/content",
)
obsidian_to_hugo.run()
You can pass an optional filters
argument to the ObsidianToHugo
constructor. This argument should be a list of functions.
The function will be invoked for each file from the obsidian vault that is copied into the hugo content directory.
Inside the function, you have access to the file path and the file contents.
When the function returns False
, the file will be skipped and not copied
into the hugo content directory.
from obsidian_to_hugo import ObsidianToHugo
def filter_file(file_contents: str, file_path: str) -> bool:
# do something with the file path and contents
if your_condition:
return True # copy file
else:
return False # skip file
obsidian_to_hugo = ObsidianToHugo(
obsidian_vault_dir="path/to/obsidian/vault",
hugo_content_dir="path/to/hugo/content",
filters=[filter_file],
)
obsidian_to_hugo.run()
You can pass an optional processors
argument to the ObsidianToHugo
constructor. This argument should be a list of functions.
The function will be invoked for each file from the obsidian vault that is copied into the hugo content directory. It will be passed the file contents as string, and should return the processed version of the file contents.
Custom processors are invoked after the default processing of the file contents.
from obsidian_to_hugo import ObsidianToHugo
def process_file(file_contents: str) -> str:
# do something with the file contents
return file_contents
obsidian_to_hugo = ObsidianToHugo(
obsidian_vault_dir="path/to/obsidian/vault",
hugo_content_dir="path/to/hugo/content",
processors=[process_file],
)
obsidian_to_hugo.run()
For the directories that you want to copy the exact structure of and put into content
as is, add the direct_copies
argument to the ObsidianToHugo
constructor. This should be a list of paths to the directories in your obsidian vault.
This will put the same structure into a directory of the same name in content
in your Hugo site.
from obsidian_to_hugo import ObsidianToHugo
obsidian_to_hugo = ObsidianToHugo(
obsidian_vault_dir="path/to/obsidian/vault",
hugo_content_dir="path/to/hugo/content",
direct_copies = ["path/to/direct/copy/folder"]
)
obsidian_to_hugo.run()
For the sensitive directories that you don't want obsidian-to-hugo
to look at, add the copy_exclusions
argument to the ObsidianToHugo
constructor. This should be a list of directory names. obsidian-to-hugo
will skip these entirely during the copy step.
from obsidian_to_hugo import ObsidianToHugo
obsidian_to_hugo = ObsidianToHugo(
obsidian_vault_dir="path/to/obsidian/vault",
hugo_content_dir="path/to/hugo/content",
copy_exclusions = ["folders", "to", "skip"]
)
obsidian_to_hugo.run()
This project is geared to my own specific use case and I kind of built it just for me. However, if you have some ideas to make it more robust and generally useful please submit a pull request.
This is a direct fork from David Wolf's obsidian-to-hugo project. Huge thank you to them for doing most of the work. Check out their really cool website while you are at it.