How to upgrade quarto cli using conda ?

[sunvitD]

Description

For some reason Quarto cli was never installed on my windows machine through pip may be because of Anaconda.

Error:

      copying build\lib\quarto_cli\share\deno_std\cache\deps\https\cdn.skypack.dev\010ab5a2e2b8bd9419eceeb3654e9d2f1412afd9065e89a8d3d03c58f29970dd.metadata.json -> build\bdist.win-amd64\wheel\.\quarto_cli\share\deno_std\cache\deps\https\cdn.skypack.dev
      error: could not create 'build\bdist.win-amd64\wheel\.\quarto_cli\share\deno_std\cache\deps\https\cdn.skypack.dev\010ab5a2e2b8bd9419eceeb3654e9d2f1412afd9065e89a8d3d03c58f29970dd.metadata.json': No such file or directory   
      [end of output]

  note: This error originates from a subprocess, and is likely not a problem with pip.
  ERROR: Failed building wheel for quarto-cli
Failed to build quarto-cli
ERROR: ERROR: Failed to build installable wheels for some pyproject.toml based projects (quarto-cli)

So i finally installed quarto using conda with command conda install conda-forge::quarto but the installed version I have is 1.5.57 and even on upgrade it stays the same.

Can it be upgraded using conda command or do I need to find a solution by uninstalling anaconda and quarto and then reinstalling quarto cli ?

[mcanouil]

The Conda version is not officially packaged by the Quarto Team.
I believe the recipe should automatically update to the latest stable or might require a manual intervention on their repository.

https://github.com/conda-forge/quarto-feedstock

[sunvitD]

Thanks for the reply.

[cderv]

For some reason Quarto cli was never installed on my windows machine through pip may be because of Anaconda.

We have some report in previous discussion about pip install quarto-cli not working with conda.

For conda the conda-forge is indeed the one to use

The Conda version is not officially packaged by the Quarto Team. I believe the recipe should automatically update to the latest stable or might require a manual intervention on their repository.

We help on this bundle, but the release still needs to be done :

• quarto v1.6.39 conda-forge/quarto-feedstock#65

It happens there is bundling problem that needs to be solved. So there is a delay on the release unfortunately.

[sunvitD]

Thank you for the reply. I am eagerly waiting for its release as the 1.5 series versions doesn't support branding/logo and many other features. It will be great to have the 1.6+ versions.

[sunvitD]

I have tried again to install quarto-cli using uv pip this time uv pip install 'quarto-cli==1.6.37' and there is some new information in the logs that I have not seen before when I used just pip to install.

Sharing the logs Below:
quarto_install_issue_uv_log.txt

If there is an easy fix to this then please let me know.

Thanks !!!

[cderv]

Thanks a lot. This is quite interesting - thanks for sharing !

To be clear, you tried to do uv pip install quarto-cli==1.6.37 inside a conda environment and uv produced this log ?

[sunvitD]

Glad to share, Yes this is exactly the case.

[sunvitD]

I am not sure if this can be relevant but I have also noticed that even when within a virtual environment Quarto sometimes uses another Python outside the virtual env. This was happening when I was trying to run Quarto using a subprocess within a streamlit app in streamlit cloud.

This is the discussion link with the streamlit dev team.

And as per them "quarto is running the wrong python" "problem is that quarto runs python code using the system python environment instead of the virtual environment."

And eventually he made it work by "by creating a bash script that activates the venv and then runs quarto.".

Not sure if this is helpful as I am not a software dev and do not have that technical skills but thought this could also be related so sharing it here.

[cderv]

We discuss using virtual environment at https://quarto.org/docs/projects/virtual-environments.html

About venv we say

To use the environment you need to activate it.

quarto will know that the environment is activated check environment variable and use this python version in priority. When the virtual environmnent is not activated, quarto python detection will follow the usual logic and find first python version having the desired kernel. So it could check in PATH and use the system one.

I have also noticed that even when within a virtual environment Quarto sometimes uses another Python outside the virtual env.

So did this happened in an activated environment ? Or otherwise what are you calling "within a virtual env" ?

Do you mean installing quarto-cli inside a virtual env and then you would expect that specific version to use the same virtualenv it is installed in without having to activae the venv first ?

I am trying to clarify how you think it should work. You can also look at python detection algorithm in our code base.

Thanks !