"zip file is corrupt" when loading a zipped project into SQMtools

[fpusan]

This happens for certain projects when running SQMtools on certain R versions.
I suspect it's related to R's internal unzip engine not being able to deal with files greater than 2Gb.
A workaround is to

1. Move the project.zip file into an empty folder also named project.
2. Uncompress project.zip there (so that the SqueezeMeta_conf.pl file, the results directory, etc... are directly inside the project folder.
3. Load the project normally with SQM=loadSQM("/path/to/project")
   I will try to have this fixed for the next release.

[fpusan]

71d5597 switches from using utils::unzip to using zip::unzip for reading zip files during loadSQM.
This fixes the problem for some projects, but eventually I run into r-lib/zip#100 . The error happened with the example file mentioned by OP as well as with an example zipped SQM project.

[fpusan]

This has mostly stopped ocurring after switching to R4.0 in version 1.7.0 (to be published in the upcoming weeks). loadSQM will now point to this workaround in case people still have issues when loading their zipped projects

[carleton-envbiotech]

I am still running into this issue in R v 4.3.2, even after unzipping the directories manually. Now, it won't interact with the object properly, and it is throwing the following error:

Error in SQM$orfs$tax16S[intersect(presentORFs, rownames(subSQM$orfs$tax16S)),  : 
  incorrect number of dimensions
Error during wrapup: not that many frames on the stack
Error: no more error handlers available (recursive errors?); invoking 'abort' restart

[fpusan]

Does it still happen if you use the SQMtools from our dev branch?

remotes::install_github("jtamames/SqueezeMeta", ref = "develop", subdir="lib/SQMtools")