One should be careful with the structures to be submitted to a structure2name conversion.
The one you showed for chlorobenzene are formally correct
however the first one is the preferred one because it describes a structure lower in energy than the one with separated formal charges and charge separation costs energy. In addition to this, specific to this example, recall chlorine is more electronegative than carbon; in Pauling's scale (reference), EN(Cl) 3.16, EN(C) 2.55 - thus, it is not that likely chlorine would donate valence electrons to the more electropositive carbon.
Grossman condensed the (empiric) rules about resonance formulae to a set of four of decreasing order of importance. To quote (emphasis in the original):
No first-row atom (B, C, N, O) can have more than eight electrons in its valence shell. (The octet rule is less sacred for heavier main group elements such as P and S, and it does not hold at all for transition metals.)
Common error alert: Resonance structures in which all atoms are surrounded by an octet of electrons are almost always lower in energy than resonance structures in which one or more atoms are electron-deficient. However, if there are electron-deficient atoms, they should be electropositive (C, B), not electronegative (N, O, halogen).
Resonance structures with charge separation are usually higher in energy than those in which charges can be neutralized.
If charge is separated, then electronegative atoms should gain the formal negative charge and electropositive ones should gain the formal positive charge.
Now you may set the description of e.g., DMSO
into context, too.
Robert B. Grossman. The Art of Writing Resonable Organic Reaction Mechanisms. Springer 2003 (second edition), p. 5. He moderates a companion web-site pointing to all (currently) three editions, supplementary material, and errata, too.