Parsing basis set for Quantum Chemistry(QC) from JSON files?

[rdguerrerom]

Hi,

Many basis-sets for QC can be downloaded in JSON format from EMSL. The task that is left, is loading the information to an array of structures in C++ parsing the information for the coefficients and the exponents to double-precision floating-point numbers properly. I am wondering what could be a good starting point for using json to accomplish this task efficiently?

Any ideas would be appreciated.

[nlohmann]

Once you have defined the C++ struct you want to put the data in, this section could be interesting: https://github.com/nlohmann/json#arbitrary-types-conversions

[rdguerrerom]

@nlohmann Thanks a lot for your advice,

So far I have the following piece of code:

#include <fstream>
#include <iostream>
#include <nlohmann/json.hpp>
#include <vector>

// for convenience
using json = nlohmann::json;

struct contracted_shell {
  std::vector<int> angular_momentum;
  std::vector<double> coefficients;
  std::vector<double> exponents;
  std::string function_type;
  std::string region;
};
void to_json(json& j, const contracted_shell& s) {
  j = json{{"angular_momentum", s.angular_momentum},
           {"coefficients", s.coefficients},
           {"exponents", s.exponents},
           {"function_type", s.function_type},
           {"region", s.region}};
}
void from_json(const json& j, contracted_shell& s) {
        j.at("angular_momentum").get_to(s.angular_momentum);
        j.at("coefficients").get_to(s.coefficients);
        j.at("exponents").get_to(s.exponents);
        j.at("function_type").get_to(s.function_type);
        j.at("region").get_to(s.region);
    }

struct chemical_element {
  std::vector<contracted_shell> electron_shells;
  std::string references;
};
void to_json(json& j, const chemical_element& ce) {
  j = json{{"electron_shells", ce.electron_shells},
           {"references", ce.references}};
}
void from_json(const json& j, chemical_element& ce) {
        j.at("electron_shells").get_to(ce.electron_shells);
        j.at("references").get_to(ce.references);
    }

struct schema {
  std::string schema_type;
  std::string schema_version;
};
void to_json(json& j, const schema& s) {
  j = json{{"chema_type", s.schema_type}, {"schema_version", s.schema_version}};
}
void from_json(const json& j, schema& s) {
        j.at("schema_type").get_to(s.schema_type);
        j.at("schema_version").get_to(s.schema_version);
    }

struct basis_set {
  std::vector<chemical_element> auxiliaries;
  std::string description;
  std::vector<chemical_element> elements;
  std::string family;
  std::vector<std::string> function_types;
  schema molssi_bse_schema;
  std::string name;
  std::vector<std::string> names;
  std::string revision_date;
  std::string revision_description;
  std::string role;
  std::vector<std::string> tags;
  std::string version;
};
void to_json(json& j, const basis_set& bs) {
  j = json {
    {"auxiliaries", bs.auxiliaries}, {"description", bs.description},
        {"elements", bs.elements}, {"family", bs.family},
        {"function_types", bs.function_types},
        {"molssi_bse_schema", bs.molssi_bse_schema}, {"name", bs.name},
        {"names", bs.names}, {"revision_date", bs.revision_date},
        {"revision_description", bs.revision_description}, {"role", bs.role},
        {"tags", bs.tags}, {"version", bs.version}};
}
void from_json(const json& j, basis_set& bs) {
        j.at("auxiliaries").get_to(bs.auxiliaries);
        j.at("description").get_to(bs.description);
        j.at("elements").get_to(bs.elements);
        j.at("family").get_to(bs.family);
        j.at("function_types").get_to(bs.function_types);
        j.at("molssi_bse_schema").get_to(bs.molssi_bse_schema);
        j.at("name").get_to(bs.name);
        j.at("names").get_to(bs.names);
        j.at("revision_date").get_to(bs.revision_date);
        j.at("revision_description").get_to(bs.revision_description);
        j.at("role").get_to(bs.role);
        j.at("tags").get_to(bs.tags);
        j.at("version").get_to(bs.version);
    }

int main() {
  // read a JSON file
  std::ifstream i("6-31G.json");
  json j;
  i >> j;

  // write prettified JSON to another file
  std::ofstream o("pretty.json");
  o << std::setw(4) << j << std::endl;
  auto basis = j.get<basis_set>(); //nlohmann::detail::type_error: [json.exception.type_error.302]
  std::cout << "Complete" << std::endl;
  return 0;
}

that I want to use to read the content of JSON files with internal structure as the following:

{
   "__comment2__": " content of the file 6-31G.json",
    "molssi_bse_schema": {
        "schema_type": "complete",
        "schema_version": "0.1"
    },
    "revision_description": "Data from Gaussian 09/GAMESS",
    "revision_date": "2018-06-19",
    "elements": {
        "1": {
            "electron_shells": [
                {
                    "function_type": "gto",
                    "region": "valence",
                    "angular_momentum": [
                        0
                    ],
                    "exponents": [
                        "0.1873113696E+02",
                        "0.2825394365E+01",
                        "0.6401216923E+00"
                    ],
                    "coefficients": [
                        [
                            "0.3349460434E-01",
                            "0.2347269535E+00",
                            "0.8137573261E+00"
                        ]
                    ]
                },
                {
                    "function_type": "gto",
                    "region": "valence",
                    "angular_momentum": [
                        0
                    ],
                    "exponents": [
                        "0.1612777588E+00"
                    ],
                    "coefficients": [
                        [
                            "1.0000000"
                        ]
                    ]
                }
            ],
            "references": [
                {
                    "reference_description": "31G Split-valence basis set for H,He",
                    "reference_keys": [
                        "ditchfield1971a"
                    ]
                }
            ]
        },
        "2": {
            "electron_shells": [
                {
                    "function_type": "gto",
                    "region": "valence",
                    "angular_momentum": [
                        0
                    ],
                    "exponents": [
                        "0.3842163400E+02",
                        "0.5778030000E+01",
                        "0.1241774000E+01"
                    ],
                    "coefficients": [
                        [
                            "0.4013973935E-01",
                            "0.2612460970E+00",
                            "0.7931846246E+00"
                        ]
                    ]
                },
                {
                    "function_type": "gto",
                    "region": "valence",
                    "angular_momentum": [
                        0
                    ],
                    "exponents": [
                        "0.2979640000E+00"
                    ],
                    "coefficients": [
                        [
                            "1.0000000"
                        ]
                    ]
                }
            ],
            "references": [
                {
                    "reference_description": "31G Split-valence basis set for H,He",
                    "reference_keys": [
                        "gaussian09e01"
                    ]
                }
            ]
        },
        "6": {
            "electron_shells": [
                {
                    "function_type": "gto",
                    "region": "valence",
                    "angular_momentum": [
                        0
                    ],
                    "exponents": [
                        "0.3047524880E+04",
                        "0.4573695180E+03",
                        "0.1039486850E+03",
                        "0.2921015530E+02",
                        "0.9286662960E+01",
                        "0.3163926960E+01"
                    ],
                    "coefficients": [
                        [
                            "0.1834737132E-02",
                            "0.1403732281E-01",
                            "0.6884262226E-01",
                            "0.2321844432E+00",
                            "0.4679413484E+00",
                            "0.3623119853E+00"
                        ]
                    ]
                },
                {
                    "function_type": "gto",
                    "region": "valence",
                    "angular_momentum": [
                        0,
                        1
                    ],
                    "exponents": [
                        "0.7868272350E+01",
                        "0.1881288540E+01",
                        "0.5442492580E+00"
                    ],
                    "coefficients": [
                        [
                            "-0.1193324198E+00",
                            "-0.1608541517E+00",
                            "0.1143456438E+01"
                        ],
                        [
                            "0.6899906659E-01",
                            "0.3164239610E+00",
                            "0.7443082909E+00"
                        ]
                    ]
                },
                {
                    "function_type": "gto",
                    "region": "valence",
                    "angular_momentum": [
                        0,
                        1
                    ],
                    "exponents": [
                        "0.1687144782E+00"
                    ],
                    "coefficients": [
                        [
                            "0.1000000000E+01"
                        ],
                        [
                            "0.1000000000E+01"
                        ]
                    ]
                }
            ],
            "references": [
                {
                    "reference_description": "6-31G Split-valence basis set",
                    "reference_keys": [
                        "hehre1972a"
                    ]
                }
            ]
        },
        "7": {
            "electron_shells": [
                {
                    "function_type": "gto",
                    "region": "valence",
                    "angular_momentum": [
                        0
                    ],
                    "exponents": [
                        "0.4173511460E+04",
                        "0.6274579110E+03",
                        "0.1429020930E+03",
                        "0.4023432930E+02",
                        "0.1282021290E+02",
                        "0.4390437010E+01"
                    ],
                    "coefficients": [
                        [
                            "0.1834772160E-02",
                            "0.1399462700E-01",
                            "0.6858655181E-01",
                            "0.2322408730E+00",
                            "0.4690699481E+00",
                            "0.3604551991E+00"
                        ]
                    ]
                },
                {
                    "function_type": "gto",
                    "region": "valence",
                    "angular_momentum": [
                        0,
                        1
                    ],
                    "exponents": [
                        "0.1162636186E+02",
                        "0.2716279807E+01",
                        "0.7722183966E+00"
                    ],
                    "coefficients": [
                        [
                            "-0.1149611817E+00",
                            "-0.1691174786E+00",
                            "0.1145851947E+01"
                        ],
                        [
                            "0.6757974388E-01",
                            "0.3239072959E+00",
                            "0.7408951398E+00"
                        ]
                    ]
                },
                {
                    "function_type": "gto",
                    "region": "valence",
                    "angular_momentum": [
                        0,
                        1
                    ],
                    "exponents": [
                        "0.2120314975E+00"
                    ],
                    "coefficients": [
                        [
                            "0.1000000000E+01"
                        ],
                        [
                            "0.1000000000E+01"
                        ]
                    ]
                }
            ],
            "references": [
                {
                    "reference_description": "6-31G Split-valence basis set",
                    "reference_keys": [
                        "hehre1972a"
                    ]
                }
            ]
        }
    },
    "version": "1",
    "function_types": [
        "gto",
        "gto_cartesian"
    ],
    "names": [
        "6-31G"
    ],
    "tags": [],
    "family": "pople",
    "description": "6-31G valence double-zeta",
    "role": "orbital",
    "auxiliaries": {},
    "name": "6-31G"
}

How can I adapt the example you are referring to read the content of the file in one go?

[nlohmann]

Do you need further assistance with this issue?

[rdguerrerom]

No, I managed to load the information I need. Perhaps, will be interesting to have here the code snippet I am using?

[nlohmann]

Yes, why not?