tensorflow / tensorflow/tensorboard

warn if both `tensorboard` and `tb-nightly` pip packages are installed

Open
#1,907 7 comments 4 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

theme:usability type:build/install
Dominant language
TypeScript
Stars
7.2k
Forks
1.7k
Avg merge
4d 22h
Merged PRs (30d)
1

Description

This is a common source of inscrutable errors, because pip will happily install both pip packages even though they both ship the tensorboard python package [1], and in doing so the second one will overwrite any files from the first one: https://github.com/pypa/pip/issues/4625. This means you can easily end up with a garbled frankentensorboard package (and it gets even worse if you uninstall one of them but not both of them, since then you get a frankentensorboard with a lot of vital organs removed).

Until pip fixes this mess, it's a fact of life, but we can at least try to warn users about this pitfall ourselves.

[1] It's colloquial to call both of these packages, so I'm prefixing to avoid confusion. See also https://github.com/pypa/packaging.python.org/issues/106

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by locating TensorBoard's startup or environment-detection entry point and inspect how installed Python packages can be identified. Done means users receive a clear warning when both tensorboard and tb-nightly are installed, without changing behavior otherwise; the issue names no files or tests, so add or locate appropriate coverage during investigation.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
tooling
Issue type
Feature
Difficulty
3/5
Estimated time
1-2 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.