This is an export script for the Omnivore read-it-later solution written in Python.

Its main purpose was to create a backup of all links saved in Omnivore, as long as the Omnivore app still lacked an export function.

Since Omnivore also lacked functionality to show the number of articles, this repository also provides a script to show these numbers.

Meanwhile, Omnivore provides a way to export the links in the web UI via the menu option "Settings" - "Export Data", so that the export script is not really needed any more.

But unfortunately, the Omnivore team decided to shut down the service in November 2024 and recommends using Readwise (paid service) or Wallabag (open-source) instead.

While Readwise has released an Omnivore importer, Wallabag was initially lacking such a feature.

Therefore, this repository also contains a script to convert the exported links from Omnivore to the import format of Wallabag v2.

Meanwhile, Wallabag also integrated a function to import from Omnivore, which will be available in version 2.6.10 of Wallabag and can be already used in the hosted solution, so that this script is not needed any more either.

To run this script, you need to install gql with httpx support first:

pip install --pre -U gql[httpx]

Next, you must create an API key for Omnivore.

Then, change the global variable API_KEY in the script omnivore-export.py, or set the environment variable OMNIVORE_API_KEY.

If you're not using the free hosted Omnivore, you must also change the global variable API_URL or set the environment variable OMNIVORE_API_URL.

You can change the path for the exported data with the global variable BACKUP_PATH or the environment variable OMNIVORE_BACKUP_PATH. The current date is automatically added to the backup filename, unless you change the global variable WITH_DATE to False or set the environment variable OMNIVORE_WITH_DATE to no.

There are some more global variables that you can change in the script:

• SEARCH = "in:all" - change if you don't want to export everything
• LIMIT = 100 - the batch size when querying the API (max. 100)
• TIMEOUT = 15 - the request timeout in seconds when querying the API
• WITH_CONTENT = False - change if you want to export the content as well

Finally, just run the script via Python:

python omnivore-export.py 

This repository also contains a script omnivore-summary.py that can be used to print a summary of the data store in Omnivore.

After configuring it in the same way as the export script, run:

python omnivore-summary.py 

Instead of setting parameters in the script or via environment variables, you can also pass them as options on the command line. You can show the exact command line syntax by running the script with the --help option.

In order to convert your Omnivore links to Wallabag, you need to export them first using the function "Settings" - "Export Data" in the Omnivore web UI.

Then you need to unpack the downloaded zip file into a sub-directory name "export" that should be in the same directory as the conversion script omnivore-to-wallabag.py. You can also specify a different input directory by changing the global variable OMNIVORE_EXPORT_DIR in the script.

Then simply run the script:

python omnivore-to-wallabag.py

It will produce an output file with the name wallabag.json. If your export from Omnivore contains more than 500 links, then it will produce multiple enumerated output files instead, because Wallabag might run out of memory when the imported files are too huge. You can specify the batch size using the global variable BATCH_SIZE in the script.

Finally, import the JSON files generated by the script (in the correct order) into Wallabag. You can find the import function in the web UI under "My account" - "Import" - "Wallabag v2" - "Import contents" - "Upload file".

If you imported the wrong files and want to start from scratch, you can use the function "Config" - "Reset area" - "Remove all entries" in Wallabag.

Note that the script currently does not convert highlights, just the links and the content.