Ergonomic, hermetic, user-friendly software build system for Python, Go, Java, Kotlin, Protobuf, Scala, Shell. Pants lets you fearlessly scale up your codebase!

Pants

I always like to hear that the right thing was already done!

```&gt;&gt;&gt; Revision.semver('1.1.1-pre0') &lt; Revision.semver('1.1.1-pre6')
True
&gt;&gt;&gt; Revision.semver('1.1.1-pre0') &gt; Revision.semver('1.1.1-pre6')
False```

```&gt;&gt;&gt; Revision.semver('1.1.1') &lt; Revision.semver('1.1.1-pre0')
True```

<@U0607UFLH>: but that means that putting 1.1.1 in there should be...safe

I should maybe open an issue to track though - presumably we might want to have that last case reversed someday

<@U06A03HV1>: missing it on the second fetch, don't see it

I think this has been hammered home, but pep is not semver

opened <https://github.com/pantsbuild/pants/issues/3382>

<@U04S45AHA>: outside of the `if not vt.valid` check, there is another call to `self._resolve`

either way... asked <@U0XUTHRPZ> to try and make that a bit tighter

On the server bug, it's almost not relevant

closer approximation to pypi versioning: ```[illuminati ~]$ pex packaging
Python 2.7.10 (default, Dec 16 2015, 14:09:45) 
[GCC 4.2.1 Compatible Apple LLVM 7.0.2 (clang-700.1.81)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
(InteractiveConsole)
&gt;&gt;&gt; from packaging.version import Version
&gt;&gt;&gt; Version('1.1.1') &lt; Version('1.1.1-pre0')
False```

I feel like folks don't give a damn on this version thing and yet we have a pep

<@U054MQUPL>: added that to <https://github.com/pantsbuild/pants/issues/3382>

Somehow we still use - I gave up after 2  tries

yeah. I agree but am following docs/matt until we publish something to the contrary

<@U06A03HV1>: _resolve is just poorly named perhaps, its ~5 lines and resolves and address to a target by injecting in the graph or fetching if already there.

PyPi does not understand the `pre` syntax, we will have to "hide" those versions in the UI manually.