Generic escape #625
Generic escape #625
Conversation
|
@huddlej this could certainly be made more pretty, but otherwise pretty straightforward. |
Codecov Report
@@ Coverage Diff @@
## master #625 +/- ##
==========================================
- Coverage 28.59% 28.53% -0.06%
==========================================
Files 39 39
Lines 5355 5365 +10
Branches 1314 1319 +5
==========================================
Hits 1531 1531
- Misses 3768 3778 +10
Partials 56 56
Continue to review full report at Codecov.
|
| @@ -71,7 +71,7 @@ | |||
| "pytest-cov >=2.8.1, ==2.8.*", | |||
| "pytest-mock >= 2.0.0, ==2.0.*", | |||
| "recommonmark >=0.5.0, ==0.*", | |||
| "snakemake >=5.4.0, ==5.*", | |||
| "snakemake >=5.4.0, <5.27", | |||
There was a problem hiding this comment.
We shouldn't need this additional constraint anymore. They changed the minimum required version of Snakemake back to 3.5 in the release they made today (5.27.4). This should fix the error Travis CI originally encountered.
There was a problem hiding this comment.
Darn, it looks like there are still bugs with version <3.7 and the newest Snakemake. There is an issue open for the specific error we see in CI, but the easiest fix for us is the original constraint you had here. Sorry about that!
Snakemake's recent releases (v5.27.*) broke compatibility with Python <3.7. We have to use an earlier version of Snakemake since we still use Python 3.6 in CI.
|
@rneher I reverted back to the Snakemake version constraint you originally had and the CI builds pass. This is interesting though, because our CI builds for Python 3.7 and 3.8 should have passed. I see now that we use Python 3.6 in the conda environment no matter what the CI's Python version is. I'll make an issue for this today, since that is a bug on our side. |
Description of proposed changes
Make the way we delimit special characters for iq tree more generic.
Thank you for contributing to Nextstrain!