Macbook M1|2|3 support - Githubissues

SAP / credential-digger

A Github scanning tool that identifies hardcoded credentials while filtering the false positive data through machine learning models :lock:

Apache License 2.0

312 stars 47 forks source link

Macbook M1|2|3 support #249

Open marcorosa opened 2 years ago

marcorosa commented 2 years ago

At the current stage we are not supporting M1 (or M2) out-of-the-box. Some dependencies are not actually compatible (e.g., hyperscan). Moreover, also docker may have some architectural issues.

Initial steps to do:

verify hyperscan package ✅
verify hyperscan wrt #231 ✅
verify the other dependencies (most of all, tensorflow and related)
verify docker (is it supported or do we really need workarounds like export DOCKER_DEFAULT_PLATFORM=linux/amd64?)

After these verifications, we will identify the steps needed to support arm

marcorosa commented 9 months ago

Update as of end of 2023 on this issue:

hyperscan package has received recent contributions that fixed several issues. Support to ARM architecture is in their roadmap now (yet, without any ETD).
Apple released M3. Once hyperscan supports ARM, we will have to test the tool also on M3

marcorosa commented 5 months ago

A way to install Credential Digger on ARM has been successfully tested. See wiki here.

Missing dependencies to have a full integration instead of this workaround:

[x] tensorflow-text ~(as of today, direct download from an external OS project)~ [edit Jul 2024]
[ ] tf-models-official

marcorosa commented 5 months ago

Another option: play with pyproject.toml and provide two different requirements file (i.e., requirements.txt and requirements-arm.txt), but this will burden ourselves with the maintenance of 2 requirements files.

marcorosa commented 1 month ago

tensorflow-text starts to support arm64 architecture with version 2.17.0 binaries on pypi