Ensure that arbitrary compound names can be read by chemdict_tools

PDB announced they're running out of three letter codes for compounds and
switch to longer names soon. This commit ensures that OpenStructure is ready
for that switch.

modules/conop/src/compound_lib.cc

@@ -36,13 +36,23 @@ namespace ost { namespace conop {
 
 namespace {
 
+/*
+COMMENT ON CREATE_CMD
+
+CREATE_CMD specifies so called affinities. e.g. VARCHAR(64) where common sense
+interprets the number 64 as max length of entries in that column.
+However, sqlite 3 totally ignores this an interprets it as TEXT without any
+limits. Long story short, don't worry about formulas longer than 64 characters
+or longer names etc.
+*/
+
 const char* CREATE_CMD[]={
 "CREATE TABLE IF NOT EXISTS chemlib_info (                                      "
 "  creation_date     TIMESTAMP,                                                 "
 "  ost_version_used  VARCHAR(64) NOT NULL);",
 "CREATE TABLE IF NOT EXISTS chem_compounds (                                    "
 "  id                INTEGER PRIMARY KEY AUTOINCREMENT,                         "
-"  tlc               VARCHAR(3) NOT NULL,                                       "
+"  tlc               VARCHAR(5) NOT NULL,                                       "
 "  olc               VARCHAR(1) NOT NULL,                                       "
 "  dialect           VARCHAR(1) NOT NULL,                                       "
 "  chem_class        VARCHAR(1),                                                "

modules/conop/tests/CMakeLists.txt

@@ -9,7 +9,8 @@ set(OST_CONOP_UNIT_TESTS
 
 if (COMPOUND_LIB)
   list(APPEND OST_CONOP_UNIT_TESTS test_compound.py
-                                   test_cleanup.py)
+                                   test_cleanup.py
+                                   test_complib.py)
 endif()
 
 ost_unittest(MODULE conop

modules/conop/tests/test_complib.py

+import unittest, os, sys
+import ost
+from ost import conop
+import subprocess
+import tempfile
+
+
+class TestCompLib(unittest.TestCase):
+
+    def test_three_vs_five_letter_code(self):
+
+        prefix_path = ost.GetPrefixPath()
+        chemdict_tool_path = os.path.join(prefix_path, "bin", "chemdict_tool")
+        if not os.path.exists(chemdict_tool_path):
+            raise RuntimeError("Expect chemdict_tool:", chemdict_tool_path)
+        tmp_dir = tempfile.TemporaryDirectory()
+        compounds_path = os.path.join("testfiles", "test_compounds.cif")
+        complib_path = os.path.join(tmp_dir.name, "test_complib.dat")
+        cmd = [chemdict_tool_path, "create", compounds_path, complib_path]
+        subprocess.run(cmd)
+
+        complib = conop.CompoundLib.Load(complib_path)
+
+        comp_001 = complib.FindCompound("001")
+        comp_hello = complib.FindCompound("hello")
+
+        self.assertFalse(comp_001 is None)
+        self.assertFalse(comp_hello is None)
+
+if __name__ == "__main__":
+    from ost import testutils
+    if testutils.SetDefaultCompoundLib():
+        testutils.RunTests()
+    else:
+        print('No compound lib available. Ignoring test_complib tests.')
\ No newline at end of file