Fix parseMMCIF and fetchPDB to get CIF files

prody/proteins/cifheader.py

@@ -877,6 +877,7 @@ def _getPolymers(lines, **kwargs):
     polymers = dict()
 
     entities = defaultdict(list)
+    longSeq = kwargs.get('longSeq', False)
 
     # SEQRES block
     items1 = parseSTARSection(lines, '_entity_poly', report=False)

prody/proteins/localpdb.py

@@ -251,7 +251,7 @@ def fetchPDB(*pdb, **kwargs):
                       'specify another folder'.format(folder))
 
     if compressed is not None and not compressed:
-        filedict = findPDBFiles(folder, compressed=True)
+        filedict = findPDBFiles(folder, compressed=True, format=format_)
         not_found, decompress = [], not_found
         for i, pdb in decompress:
             if pdb in filedict:

prody/proteins/wwpdb.py

@@ -341,10 +341,14 @@ def fetchPDBviaHTTP(*pdb, **kwargs):
         try:
             url = getURL(pdb)
             if format != 'pdb':
-                url = url.replace('.pdb', extension)
+                if url.find('.pdb') != -1:
+                    url = url.replace('.pdb', extension)
+                elif url.find('.ent') != -1:
+                    url = url.replace('.ent', extension)
 
                 if url.find('divided/pdb') != -1:
                     url = url.replace('divided/pdb', 'divided/' + long_format)
+
             handle = openURL(url)
         except Exception as err:
             if not _isExtendedPDBID(pdb):

prody/tests/proteins/test_ciffile.py

@@ -121,7 +121,14 @@ def testLongChainArgument(self):
                         self.no_pdb['segment_SX0_atoms'],
                         'parseMMCIF failed to parse correct number of atoms '
                         'when segment SX0 is specified')
-
+
+    def testSkipPDBs(self):
+        """Test outcome of valid and invalid *segment* arguments."""
+
+        pdb = self.altlocs['pdb']
+        _ = parsePDB(pdb)
+        _ = parseMMCIF(pdb)
+
     def testUniteChainsArgument(self):
         """Test outcome of valid and invalid *segment* arguments."""
 
@@ -468,6 +475,11 @@ def testAltlocAllMultiModels(self):
         self.assertEqual(ag.numCoordsets(), self.multi['models'],
             'parsePDB failed to parse correct number of coordsets ({0}) with altloc "all"'.format(self.multi['models']))
 
+    def testLongSeqFix(self):
+        """Test the outcome of parsing 3o21 from id, which failed previously."""
+
+        _ = parseMMCIF('3o21')
+
     def testStopScipion(self):
         """Test number of coordinate sets and atoms for PyMOL CIF file with altloc='all'."""
 

prody/utilities/pathtools.py

@@ -216,6 +216,12 @@ def gunzip(filename, outname=None):
             else:
                 outname = filename
 
+        # Quick gzip magic number check
+        with open(filename, 'rb') as f:
+            magic = f.read(2)
+        if magic != b'\x1f\x8b':
+            raise ValueError(f"File {filename} is not a valid gzip file")
+
         inp = gzip_open(filename, 'rb')
         data = inp.read()
         inp.close()
@@ -224,6 +230,10 @@ def gunzip(filename, outname=None):
         out.close()
         return outname
     else:
+        # filename is a buffer, so check directly without opening
+        if filename[:2] != b'\x1f\x8b':
+            raise ValueError("Buffer is not gzipped or is invalid content")
+
         result = None
         try:
             from StringIO import StringIO