update docs

Author: meksor

doc/source/core/run.rst

@@ -11,3 +11,5 @@ Runs
 .. autoclass:: ixmp4.core.run.Run
    :members:
 
+.. autoclass:: ixmp4.core.run.RunCheckpoints
+   :members:

doc/source/usage/locks-and-versioning.rst

@@ -56,8 +56,13 @@ these instances may acquire a lock.
     a **method of coordination** for clients that rely on versioning facilities
     and thus **not a reliable security mechanism** in any manner.
 
-Checkpoints
------------
+For advanced use of the locking mechanism, read about:
+
+ - :meth:`ixmp4.core.run.Run.lock`
+ - :meth:`ixmp4.core.run.Run.unlock`
+
+Checkpoints and Reversion
+-------------------------
 
 Checkpoints are used to label important transactions in the transaction history
 as to provide "anchors" to revert to. They are created automatically at the exit 
@@ -74,14 +79,20 @@ created at any other point.
     #> Checkpoint "Add data" created
 
 If an exception occurs within a :meth:`~ixmp4.core.run.Run.transact` block,
-data in the run will be rolled back to the latest checkpoint on platforms 
+data in the run will be reverted to the latest checkpoint on platforms 
 that support versioning. 
 
+To manually revert to a past state, use the ``revert()`` methods on
+:class:`~ixmp4.core.run.Run` and :class:`~ixmp4.core.checkpoint.Checkpoint`.
+
+ - :meth:`ixmp4.core.run.Run.revert`
+ - :meth:`ixmp4.core.checkpoint.Checkpoint.revert`
+
 Checkpoint Views
 ----------------
 
 Checkpoint views provide read-only access to run data at a specific checkpoint.
-Access them through ``run.checkpoints[checkpoint_id]``.
+Access them through :class:`~ixmp4.core.run.RunCheckpoints` / :attr:`~ixmp4.core.run.Run.checkpoints``.
 
 .. autoclass:: ixmp4.core.checkpoint.Checkpoint
     :members:

ixmp4/core/checkpoint.py

@@ -31,12 +31,11 @@ class Checkpoint(BaseFacadeObject[CheckpointService, CheckpointDto]):
 
         # Create a checkpoint and access it later
         run.checkpoints.create("after scenario setup")
-        checkpoint = run.checkpoints.tabulate().iloc[0]
+        checkpoint = run.checkpoints.list().pop()
 
-        view = run.checkpoints[checkpoint.id]
-        df = view.iamc.tabulate()
-        meta = view.meta
-        scalars = view.optimization.scalars.tabulate()
+        df = checkpoint.iamc.tabulate()
+        meta = checkpoint.meta
+        scalars_df = checkpoint.optimization.scalars.tabulate()
 
         # Revert run to this checkpoint state
         view.revert()