Trouble rebasing RMG-Cat on latest master

[rwest]

The catmerge branch had RMG-cat ready for merging with RMG master.
#1537
I tried rebasing it onto the current master, and pushed the result to
https://github.com/rwest/rmg-py/tree/catmerge2
I had to resolve only a few minor conflicts, which I think I did carefully.
The unit tests all pass.

However, running
python rmg.py examples/rmg/catalysis/ch4_o2/input.py
results in

Initialization complete. Starting model generation.

Generating initial reactions...
Error: Problem family: Surface_Adsorption_vdW
Error: Problem reactants: (Molecule(SMILES="C"), Molecule(SMILES="C"))
Traceback (most recent call last):
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/scoop_framework/util.py", line 112, in __call__
    return self.myfn(*args, **kwargs)
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/rmg/react.py", line 75, in reactSpecies
    reactions = getDB('kinetics').generate_reactions_from_families(speciesTuple)
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/data/kinetics/database.py", line 523, in generate_reactions_from_families
    reaction_list.extend(self.react_molecules(combo, products=products, only_families=only_families, prod_resonance=resonance))
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/data/kinetics/database.py", line 549, in react_molecules
    reaction_list.extend(family.generateReactions(molecules, products=products, prod_resonance=prod_resonance))
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/data/kinetics/family.py", line 1649, in generateReactions
    self.__generateReactions(reactants, products=products, forward=False, prod_resonance=prod_resonance))
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/data/kinetics/family.py", line 1900, in __generateReactions
    productStructures = self.__generateProductStructures(reactantStructures, [mapB, mapA], forward)
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/data/kinetics/family.py", line 1525, in __generateProductStructures
    productStructures = self.applyRecipe(reactantStructures, forward=forward)
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/data/kinetics/family.py", line 1314, in applyRecipe
    self.reverseRecipe.applyForward(reactantStructure, unique)
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/data/kinetics/family.py", line 364, in applyForward
    return self.__apply(struct, True, unique)
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/data/kinetics/family.py", line 287, in __apply
    atom2 = struct.getLabeledAtom(label2)
  File "rmgpy/molecule/molecule.py", line 1264, in rmgpy.molecule.molecule.Molecule.getLabeledAtom
    def getLabeledAtom(self, label):
  File "rmgpy/molecule/molecule.py", line 1270, in rmgpy.molecule.molecule.Molecule.getLabeledAtom
    raise ValueError('No atom in the molecule has the label "{0}".'.format(label))
ValueError: No atom in the molecule has the label "*2".

Traceback (most recent call last):
  File "rmg.py", line 173, in <module>
    main()
  File "rmg.py", line 167, in main
    rmg.execute(**kwargs)
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/rmg/main.py", line 667, in execute
    trimolecularReact=self.trimolecularReact)
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/rmg/model.py", line 679, in enlarge
    unimolecularReact, bimolecularReact, trimolecularReact=trimolecularReact)
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/rmg/react.py", line 143, in reactAll
    rxns = list(react(*spcTuples))
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/rmg/react.py", line 59, in react
    spcTuples)
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/scoop_framework/util.py", line 154, in map_
    return map(WorkerWrapper(args[0]), *args[1:], **kwargs)
  File "/Users/rwest/anaconda/envs/rmg230/lib/python2.7/site-packages/scoop-0.7.2.0-py2.7.egg/scoop/fallbacks.py", line 49, in wrapper
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/scoop_framework/util.py", line 112, in __call__
    return self.myfn(*args, **kwargs)
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/rmg/react.py", line 75, in reactSpecies
    reactions = getDB('kinetics').generate_reactions_from_families(speciesTuple)
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/data/kinetics/database.py", line 523, in generate_reactions_from_families
    reaction_list.extend(self.react_molecules(combo, products=products, only_families=only_families, prod_resonance=resonance))
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/data/kinetics/database.py", line 549, in react_molecules
    reaction_list.extend(family.generateReactions(molecules, products=products, prod_resonance=prod_resonance))
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/data/kinetics/family.py", line 1649, in generateReactions
    self.__generateReactions(reactants, products=products, forward=False, prod_resonance=prod_resonance))
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/data/kinetics/family.py", line 1900, in __generateReactions
    productStructures = self.__generateProductStructures(reactantStructures, [mapB, mapA], forward)
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/data/kinetics/family.py", line 1525, in __generateProductStructures
    productStructures = self.applyRecipe(reactantStructures, forward=forward)
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/data/kinetics/family.py", line 1314, in applyRecipe
    self.reverseRecipe.applyForward(reactantStructure, unique)
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/data/kinetics/family.py", line 364, in applyForward
    return self.__apply(struct, True, unique)
  File "/Users/rwest/Code/Cat/RMG-Py/rmgpy/data/kinetics/family.py", line 287, in __apply
    atom2 = struct.getLabeledAtom(label2)
  File "rmgpy/molecule/molecule.py", line 1264, in rmgpy.molecule.molecule.Molecule.getLabeledAtom
    def getLabeledAtom(self, label):
  File "rmgpy/molecule/molecule.py", line 1270, in rmgpy.molecule.molecule.Molecule.getLabeledAtom
    raise ValueError('No atom in the molecule has the label "{0}".'.format(label))
ValueError: No atom in the molecule has the label "*2".

[mjohnson541]

This is most likely related to the merging of automatic tree generation part 3, which added a set of switches that RMG uses to determine if a tree is an old tree or one of the new trees, however, I don't see any obvious reason why Surface_Adsorption_vdW would run into problems since it's bimolecular and has two trees.

[rwest]

The product is an adsorbate but physisorption (vdW) is represented as a disconnected graph (no bond). Could be the reason? Something splitting the structure into two molecules?

[mjohnson541]

I wouldn't think the product would matter that much, but that certainly seems to make more sense than anything else. I can take a look at this later.

[mjohnson541]

The issue was unrelated to the products, it was actually an issue related to improperly overwritten references in template handling. During the review of atg3 @mliu49 and I removed a deepcopy that was slowing RMG down because we didn't think it would be problematic for ordinary trees. Fortunately I had already fixed this issue (both the references and the speed issue) in atg4 and I've ported forward a fix to catmerge2.

[rwest]

Thanks Matt. I got it onto catmerge #1537 and tested a couple of models