Chris Friedline

You've probably figured this out by now, but I deal with this by creating two different `cventries`, the first contains the org, job title, dates, location and one set of...

> I will review the Python port and link to it here in a subsequent PR. Hi there. Wondering if there's been any movement on the review of the python...

I did some experimenting. My current versions of pandas is 1.1.4 and pyranges 0.0.87. Backing down pandas to 1.0.3 has this same problem above. When I take pyranges back to...

Iterating versions after 0.0.78, things are fine with pandas 1.0.3 until 0.0.82, where it breaks. 

Also works with 0.0.78 and latest pandas. So something happened between 0.0.81 and 0.0.82, I guess. It's hard to see b/c the repo is not tagged. Would be great to...

With pandas 1.1.4, I incremented versions of pyranges again. Same as above - starts failing at 0.0.82.

Perhaps it's related to strandedness. Try to replicate this? ``` import numpy as np np.random.seed(0) gr3, gr4, gr5 = pr.random(strand=True), pr.random(strand=False), pr.random(strand=True) pr.concat([gr3, gr4]), pr.concat([gr3, gr5]), pr.__version__, pd.__version__ ``` yields...

Might be it. Here's my data again: `pr.concat([gr1.unstrand(), gr2])` - works ``` +--------------+-----------+-----------+-------+ | Chromosome | Start | End | +2 | | (object) | (int64) | (int64) | ......

I can confirm that on my system, after installing (`poetry add git+https://github.com/biocore-ntnu/pyranges.git#fix_160`) and restarting my kernel, that the output count is as expected. However, the resulting ranges object still shows...

Looks good on my end!