#1296 Add "customQuery" option support to ket format (#1297)

api/tests/integration/ref/formats/ket_with_query.py.out

@@ -0,0 +1,6 @@
+*** KET with query components ***
+ket_with_query_components.ket:SUCCEED
+*** KET with query properties ***
+ket_with_query_properties.ket:SUCCEED
+*** KET with custom query ***
+ket_with_custom_query.ket:SUCCEED

api/tests/integration/tests/formats/ket_with_query.py

@@ -0,0 +1,45 @@
+import difflib
+import os
+import sys
+
+
+def find_diff(a, b):
+    return "\n".join(difflib.unified_diff(a.splitlines(), b.splitlines()))
+
+
+sys.path.append(
+    os.path.normpath(
+        os.path.join(os.path.abspath(__file__), "..", "..", "..", "common")
+    )
+)
+from env_indigo import *  # noqa
+
+indigo = Indigo()
+indigo.setOption("json-saving-pretty", True)
+
+ref_path = joinPathPy("ref/", __file__)
+
+
+def check_ket_file(name):
+    filename = os.path.join(ref_path, name)
+
+    mol = indigo.loadQueryMoleculeFromFile(filename)
+    with open(filename, "r") as file:
+        ket_ref = file.read()
+    ket = mol.json()
+    diff = find_diff(ket_ref, ket)
+    if not diff:
+        print(name + ":SUCCEED")
+    else:
+        print(name + ":FAILED")
+        print(diff)
+
+
+print("*** KET with query components ***")
+check_ket_file("ket_with_query_components.ket")
+
+print("*** KET with query properties ***")
+check_ket_file("ket_with_query_properties.ket")
+
+print("*** KET with custom query ***")
+check_ket_file("ket_with_custom_query.ket")

api/tests/integration/tests/formats/ref/ket_with_custom_query.ket

@@ -0,0 +1,123 @@
+{
+    "root": {
+        "nodes": [
+            {
+                "$ref": "mol0"
+            }
+        ]
+    },
+    "mol0": {
+        "type": "molecule",
+        "atoms": [
+            {
+                "label": "C",
+                "location": [
+                    6.3348493576049809,
+                    -5.550074577331543,
+                    0.0
+                ]
+            },
+            {
+                "label": "C",
+                "location": [
+                    8.06515121459961,
+                    -5.549589157104492,
+                    0.0
+                ]
+            },
+            {
+                "label": "C",
+                "location": [
+                    7.2016377449035648,
+                    -5.049966812133789,
+                    0.0
+                ]
+            },
+            {
+                "label": "C",
+                "location": [
+                    8.06515121459961,
+                    -6.55053186416626,
+                    0.0
+                ]
+            },
+            {
+                "label": "C",
+                "location": [
+                    6.3348493576049809,
+                    -6.555019855499268,
+                    0.0
+                ]
+            },
+            {
+                "label": "C",
+                "location": [
+                    7.203820705413818,
+                    -7.050033092498779,
+                    0.0
+                ],
+                "queryProperties": {
+                    "customQuery": "C&X3,N&X2"
+                }
+            }
+        ],
+        "bonds": [
+            {
+                "type": 2,
+                "atoms": [
+                    2,
+                    0
+                ]
+            },
+            {
+                "type": 2,
+                "atoms": [
+                    3,
+                    1
+                ]
+            },
+            {
+                "type": 1,
+                "atoms": [
+                    0,
+                    4
+                ]
+            },
+            {
+                "type": 1,
+                "atoms": [
+                    1,
+                    2
+                ]
+            },
+            {
+                "type": 2,
+                "atoms": [
+                    4,
+                    5
+                ]
+            },
+            {
+                "type": 1,
+                "atoms": [
+                    5,
+                    3
+                ]
+            }
+        ],
+        "sgroups": [
+            {
+                "type": "MUL",
+                "atoms": [
+                    0,
+                    1,
+                    2,
+                    3,
+                    4,
+                    5
+                ],
+                "mul": 1
+            }
+        ]
+    }
+}

core/indigo-core/molecule/molecule_json_saver.h

@@ -95,12 +95,12 @@ namespace indigo
 
         bool Int64(int64_t i64)
         {
-            return pretty_json ? _pretty_writer.Uint(i64) : _writer.Uint(i64);
+            return pretty_json ? _pretty_writer.Uint64(i64) : _writer.Uint64(i64);
         }
 
         bool Uint64(uint64_t u64)
         {
-            return pretty_json ? _pretty_writer.Uint(u64) : _writer.Uint(u64);
+            return pretty_json ? _pretty_writer.Uint64(u64) : _writer.Uint64(u64);
         }
 
         bool Double(double d)

core/indigo-core/molecule/smiles_loader.h

@@ -67,6 +67,9 @@ namespace indigo
         bool ignore_bad_valence;
         bool ignore_no_chiral_flag{false};
 
+        static void readSmartsAtomStr(const std::string& atom_str, std::unique_ptr<QueryMolecule::Atom>& qatom);
+        static void readSmartsBondStr(const std::string& bond_str, std::unique_ptr<QueryMolecule::Bond>& qbond);
+
     protected:
         enum
         {
@@ -178,14 +181,16 @@ namespace indigo
         void _handleCurlyBrace(_AtomDesc& atom, bool& inside_polymer);
         void _handlePolymerRepetition(int i);
 
-        void _readAtom(Array<char>& atom_str, bool first_in_brackets, _AtomDesc& atom, std::unique_ptr<QueryMolecule::Atom>& qatom);
+        static void _readAtom(Array<char>& atom_str, bool first_in_brackets, _AtomDesc& atom, std::unique_ptr<QueryMolecule::Atom>& qatom,
+                              bool smarts_mode = false, bool inside_rsmiles = false);
 
-        bool _readAtomLogic(Array<char>& atom_str, bool first_in_brackets, _AtomDesc& atom, std::unique_ptr<QueryMolecule::Atom>& qatom);
+        static bool _readAtomLogic(Array<char>& atom_str, bool first_in_brackets, _AtomDesc& atom, std::unique_ptr<QueryMolecule::Atom>& qatom,
+                                   bool smarts_mode = false, bool inside_rsmiles = false);
 
         int _parseCurly(Array<char>& curly, int& repetitions);
 
-        void _readBond(Array<char>& bond_str, _BondDesc& bond, std::unique_ptr<QueryMolecule::Bond>& qbond);
-        void _readBondSub(Array<char>& bond_str, _BondDesc& bond, std::unique_ptr<QueryMolecule::Bond>& qbond);
+        static void _readBond(Array<char>& bond_str, _BondDesc& bond, std::unique_ptr<QueryMolecule::Bond>& qbond, bool smarts_mode);
+        static void _readBondSub(Array<char>& bond_str, _BondDesc& bond, std::unique_ptr<QueryMolecule::Bond>& qbond, bool smarts_mode);
         void _readRGroupOccurrenceRanges(const char* str, Array<int>& ranges);
 
     private:

core/indigo-core/molecule/smiles_saver.h

@@ -80,6 +80,9 @@ namespace indigo
 
         const Array<int>& getSavedCisTransParities();
 
+        static std::string writeSmartsAtomStr(QueryMolecule::Atom* atom);
+        static std::string writeSmartsBondStr(QueryMolecule::Bond* bond);
+
     protected:
         void _saveMolecule();
 
@@ -110,8 +113,8 @@ namespace indigo
         void _writeAtom(int idx, bool aromatic, bool lowercase, int chirality) const;
         void _writeChirality(int chirality) const;
         void _writeCharge(int charge) const;
-        void _writeSmartsAtom(int idx, QueryMolecule::Atom* atom, int chirality, int depth, bool has_or_parent, bool has_not_parent) const;
-        void _writeSmartsBond(int idx, QueryMolecule::Bond* bond, bool has_or_parent) const;
+        static void _writeSmartsAtom(Output& output, QueryMolecule::Atom* atom, int aam, int chirality, int depth, bool has_or_parent, bool has_not_parent);
+        static void _writeSmartsBond(Output& output, QueryMolecule::Bond* bond, bool has_or_parent);
         void _markCisTrans();
         void _banSlashes();
         int _calcBondDirection(int idx, int vprev);

core/indigo-core/molecule/src/molecule_json_loader.cpp

@@ -12,6 +12,7 @@
 #include "molecule/molecule.h"
 #include "molecule/molecule_sgroups.h"
 #include "molecule/query_molecule.h"
+#include "molecule/smiles_loader.h"
 
 using namespace rapidjson;
 using namespace indigo;
@@ -524,7 +525,10 @@ void MoleculeJsonLoader::parseAtoms(const rapidjson::Value& atoms, BaseMolecule&
                 auto qProps = a["queryProperties"].GetObject();
                 if (qProps.HasMember("customQuery"))
                 {
-                    // Read custom query
+                    std::string customQuery = qProps["customQuery"].GetString();
+                    std::unique_ptr<QueryMolecule::Atom> atom = make_unique<QueryMolecule::Atom>();
+                    SmilesLoader::readSmartsAtomStr(customQuery, atom);
+                    _pqmol->resetAtom(atom_idx, atom.release());
                 }
                 else
                 {
@@ -676,7 +680,9 @@ void MoleculeJsonLoader::parseBonds(const rapidjson::Value& bonds, BaseMolecule&
             if (b.HasMember("customQuery"))
             {
                 std::string customQuery = b["customQuery"].GetString();
-                // 2do process custom query
+                std::unique_ptr<QueryMolecule::Bond> bond = make_unique<QueryMolecule::Bond>();
+                SmilesLoader::readSmartsBondStr(customQuery, bond);
+                _pqmol->resetBond(bond_idx, bond.release());
             }
 
             if (b.HasMember("cip"))
@@ -865,6 +871,8 @@ void MoleculeJsonLoader::parseSGroups(const rapidjson::Value& sgroups, BaseMolec
                     _pqmol->components[atom_idx] = components_count;
                 }
             }
+            else
+                throw Error("queryProperties is allowed only for queries");
             continue;
         }
         int sg_type = SGroup::getType(sg_type_str.c_str());