Unhardcode LammpsDumpParser

package/CHANGELOG

@@ -20,6 +20,8 @@ mm/dd/yy richardjgowers, kain88-de, lilyminium, p-j-smith, bdice, joaomcteixeira
   * 0.21.0
 
 Fixes
+  * fixed LammpsDumpParser failing on files with >5 fields in atom lines 
+    (PR #2333)
   * Handle exception when PDBWriter is trying to remove an invalid StringIO
     (Issue #2512)
   * Clarifies density_from_Universe docs and adds user warning (Issue #2372)

package/MDAnalysis/coordinates/LAMMPS.py

@@ -459,6 +459,8 @@ class DumpReader(base.ReaderBase):
     """Reads the default `LAMMPS dump format`_
 
     Expects trajectories produced by the default 'atom' style dump.
+    The id of the atom must be the first field, and the coordinates 
+    must be the last 3 fields.
 
     Will automatically convert positions from their scaled/fractional
     representation to their real values.
@@ -563,9 +565,10 @@ def _read_next_timestep(self):
 
         f.readline()  # ITEM ATOMS etc
         for i in range(self.n_atoms):
-            idx, _, xs, ys, zs = f.readline().split()
+            fields = f.readline().split()
 
-            indices[i] = idx
+            indices[i] = fields[0]
+            xs, ys, zs = fields[-3:]
             ts.positions[i] = xs, ys, zs
 
         order = np.argsort(indices)

package/MDAnalysis/topology/LAMMPSParser.py

@@ -593,9 +593,12 @@ def _parse_box(self, header):
 
 
 class LammpsDumpParser(TopologyReaderBase):
-    """Parses Lammps ascii dump files in 'atom' format
+    """Parses Lammps ascii dump files in 'atom' format.
+
+    The id of the atom must be the first field, and the atom types 
+    must be the second field.
 
-    Only reads atom ids.  Sets all masses to 1.0.
+    Sets all masses to 1.0.
 
     .. versionadded:: 0.19.0
     """
@@ -619,10 +622,10 @@ def parse(self, **kwargs):
 
             fin.readline()  # ITEM ATOMS
             for i in range(natoms):
-                idx, atype, _, _, _ = fin.readline().split()
+                fields = fin.readline().split()
 
-                indices[i] = idx
-                types[i] = atype
+                indices[i] = fields[0]
+                types[i] = fields[1]
 
         order = np.argsort(indices)
         indices = indices[order]

testsuite/MDAnalysisTests/datafiles.py

@@ -129,6 +129,7 @@
     "LAMMPShyd", "LAMMPShyd2",
     "LAMMPSdata_deletedatoms",  # with deleted atoms
     "LAMMPSDUMP",
+    "LAMMPSDUMP_long",  # lammpsdump file with extra columns
     "unordered_res",  # pdb file with resids non sequential
     "GMS_ASYMOPT",  # GAMESS C1  optimization
     "GMS_SYMOPT",   # GAMESS D4h optimization
@@ -440,6 +441,7 @@
 LAMMPShyd2 = resource_filename(__name__, "data/lammps/hydrogen-class1.data2")
 LAMMPSdata_deletedatoms = resource_filename(__name__, 'data/lammps/deletedatoms.data')
 LAMMPSDUMP = resource_filename(__name__, "data/lammps/wat.lammpstrj.bz2")
+LAMMPSDUMP_long = resource_filename(__name__, "data/lammps/wat.lammpstrj_long.bz2")
 
 unordered_res = resource_filename(__name__, "data/unordered_res.pdb")
 

testsuite/MDAnalysisTests/topology/test_lammpsdata.py

@@ -37,6 +37,7 @@
     LAMMPShyd2,
     LAMMPSdata_deletedatoms,
     LAMMPSDUMP,
+    LAMMPSDUMP_long,
 )
 
 
@@ -270,3 +271,7 @@ def test_id_ordering(self):
         u = mda.Universe(self.ref_filename, format='LAMMPSDUMP')
         # the 4th in file has id==13, but should have been sorted
         assert u.atoms[3].id == 4
+
+class TestDumpParserLong(TestDumpParser):
+
+    ref_filename = LAMMPSDUMP_long