Menu Docs
Página inicial do Docs
/ / /
Driver C
/ /

"distinct" e "mapReduce"

Nesta página

  • Configurar
  • Comando "diferenciado"
  • "mapReduce" - Exemplo básico
  • "mapReduce" - Exemplo mais complicado
  • Executando os exemplos

Este documento fornece alguns exemplos práticos e simples para demonstrar os comandos distinct e mapReduce .

Primeiro, escreveremos um código para inserir dados de amostra:

bool
insert_data (mongoc_collection_t *collection)
{
mongoc_bulk_operation_t *bulk;
enum N { ndocs = 4 };
bson_t *docs[ndocs];
bson_error_t error;
int i = 0;
bool ret;
bulk = mongoc_collection_create_bulk_operation_with_opts (collection, NULL);
docs[0] = BCON_NEW ("x", BCON_DOUBLE (1.0), "tags", "[", "dog", "cat", "]");
docs[1] = BCON_NEW ("x", BCON_DOUBLE (2.0), "tags", "[", "cat", "]");
docs[2] = BCON_NEW ("x", BCON_DOUBLE (2.0), "tags", "[", "mouse", "cat", "dog", "]");
docs[3] = BCON_NEW ("x", BCON_DOUBLE (3.0), "tags", "[", "]");
for (i = 0; i < ndocs; i++) {
mongoc_bulk_operation_insert (bulk, docs[i]);
bson_destroy (docs[i]);
docs[i] = NULL;
}
ret = mongoc_bulk_operation_execute (bulk, NULL, &error);
if (!ret) {
fprintf (stderr, "Error inserting data: %s\n", error.message);
}
mongoc_bulk_operation_destroy (bulk);
return ret;
}
/* A helper which we'll use a lot later on */
void
print_res (const bson_t *reply)
{
char *str;
BSON_ASSERT (reply);
str = bson_as_canonical_extended_json (reply, NULL);
printf ("%s\n", str);
bson_free (str);
}

É assim que se usa o comando distinct para obter os valores distintos de x que são maiores que 1:

distinct.c
bool
distinct (mongoc_database_t *database)
{
bson_t *command;
bson_t reply;
bson_error_t error;
bool res;
bson_iter_t iter;
bson_iter_t array_iter;
double val;
command = BCON_NEW ("distinct",
BCON_UTF8 (COLLECTION_NAME),
"key",
BCON_UTF8 ("x"),
"query",
"{",
"x",
"{",
"$gt",
BCON_DOUBLE (1.0),
"}",
"}");
res = mongoc_database_command_simple (database, command, NULL, &reply, &error);
if (!res) {
fprintf (stderr, "Error with distinct: %s\n", error.message);
goto cleanup;
}
/* Do something with reply (in this case iterate through the values) */
if (!(bson_iter_init_find (&iter, &reply, "values") && BSON_ITER_HOLDS_ARRAY (&iter) &&
bson_iter_recurse (&iter, &array_iter))) {
fprintf (stderr, "Couldn't extract \"values\" field from response\n");
goto cleanup;
}
while (bson_iter_next (&array_iter)) {
if (BSON_ITER_HOLDS_DOUBLE (&array_iter)) {
val = bson_iter_double (&array_iter);
printf ("Next double: %f\n", val);
}
}
cleanup:
/* cleanup */
bson_destroy (command);
bson_destroy (&reply);
return res;
}

Um exemplo simples usando a estrutura de redução de mapa. Apenas adiciona o número de ocorrências de cada "etiqueta".

Primeiro defina as funções map e reduce :

constantes.c
const char *const COLLECTION_NAME = "things";
/* Our map function just emits a single (key, 1) pair for each tag
in the array: */
const char *const MAPPER = "function () {"
"this.tags.forEach(function(z) {"
"emit(z, 1);"
"});"
"}";
/* The reduce function sums over all of the emitted values for a
given key: */
const char *const REDUCER = "function (key, values) {"
"var total = 0;"
"for (var i = 0; i < values.length; i++) {"
"total += values[i];"
"}"
"return total;"
"}";
/* Note We can't just return values.length as the reduce function
might be called iteratively on the results of other reduce
steps. */

Execute o comando mapReduce . Use os auxiliares de comando genéricos (por exemplo mongoc_database_command_simple). Os ajudantes do comando de leitura (por exemplo mongoc_database_read_command_with_opts) porque são consideradas operações de leitura repetíveis. Se as leituras repetíveis estiverem ativadas, essas operações tentarão novamente uma vez em um erro repetível, gerando um comportamento indesejável para mapReduce.

map-reduce-basic.c
bool
map_reduce_basic (mongoc_database_t *database)
{
bson_t reply;
bool res = false;
bson_error_t error;
mongoc_cursor_t *cursor = NULL;
bool query_done = false;
const char *out_collection_name = "outCollection";
mongoc_collection_t *out_collection = NULL;
/* Empty find query */
bson_t find_query = BSON_INITIALIZER;
/* Construct the mapReduce command */
/* Other arguments can also be specified here, like "query" or
"limit" and so on */
bson_t *const command = BCON_NEW ("mapReduce",
BCON_UTF8 (COLLECTION_NAME),
"map",
BCON_CODE (MAPPER),
"reduce",
BCON_CODE (REDUCER),
"out",
BCON_UTF8 (out_collection_name));
res = mongoc_database_command_simple (database, command, NULL, &reply, &error);
if (!res) {
fprintf (stderr, "MapReduce failed: %s\n", error.message);
goto cleanup;
}
/* Do something with the reply (it doesn't contain the mapReduce results) */
print_res (&reply);
/* Now we'll query outCollection to see what the results are */
out_collection = mongoc_database_get_collection (database, out_collection_name);
cursor = mongoc_collection_find_with_opts (out_collection, &find_query, NULL, NULL);
query_done = true;
/* Do something with the results */
const bson_t *doc = NULL;
while (mongoc_cursor_next (cursor, &doc)) {
print_res (doc);
}
if (mongoc_cursor_error (cursor, &error)) {
fprintf (stderr, "ERROR: %s\n", error.message);
res = false;
goto cleanup;
}
cleanup:
/* cleanup */
if (query_done) {
mongoc_cursor_destroy (cursor);
mongoc_collection_destroy (out_collection);
}
bson_destroy (&reply);
bson_destroy (command);
return res;
}

Você deve ter o conjunto de réplicas em execução para isso.

Neste exemplo , entramos em contato com um secundário no conjunto de réplicas e fizemos uma redução de mapa "inline", para que os resultados sejam retornados imediatamente:

map-reduce-advanced.c
bool
map_reduce_advanced (mongoc_database_t *database)
{
bson_t *command;
bson_error_t error;
bool res = true;
mongoc_cursor_t *cursor;
mongoc_read_prefs_t *read_pref;
const bson_t *doc;
/* Construct the mapReduce command */
/* Other arguments can also be specified here, like "query" or "limit"
and so on */
/* Read the results inline from a secondary replica */
command = BCON_NEW ("mapReduce",
BCON_UTF8 (COLLECTION_NAME),
"map",
BCON_CODE (MAPPER),
"reduce",
BCON_CODE (REDUCER),
"out",
"{",
"inline",
"1",
"}");
read_pref = mongoc_read_prefs_new (MONGOC_READ_SECONDARY);
cursor = mongoc_database_command (database, MONGOC_QUERY_NONE, 0, 0, 0, command, NULL, read_pref);
/* Do something with the results */
while (mongoc_cursor_next (cursor, &doc)) {
print_res (doc);
}
if (mongoc_cursor_error (cursor, &error)) {
fprintf (stderr, "ERROR: %s\n", error.message);
res = false;
}
mongoc_cursor_destroy (cursor);
mongoc_read_prefs_destroy (read_pref);
bson_destroy (command);
return res;
}

Veja como executar o código de exemplo

Basic-aggregation.c
/*
* Copyright 2009-present MongoDB, Inc.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#include <mongoc/mongoc.h>
#include <stdio.h>
#include "constants.c"
#include "../doc-common-insert.c"
#include "distinct.c"
#include "map-reduce-basic.c"
#include "map-reduce-advanced.c"
int
main (int argc, char *argv[])
{
mongoc_database_t *database = NULL;
mongoc_client_t *client = NULL;
mongoc_collection_t *collection = NULL;
mongoc_uri_t *uri = NULL;
bson_error_t error;
char *host_and_port = NULL;
int exit_code = EXIT_FAILURE;
if (argc != 2) {
fprintf (stderr, "usage: %s CONNECTION-STRING\n", argv[0]);
fprintf (stderr, "the connection string can be of the following forms:\n");
fprintf (stderr, "localhost\t\t\t\tlocal machine\n");
fprintf (stderr, "localhost:27018\t\t\t\tlocal machine on port 27018\n");
fprintf (stderr,
"mongodb://user:pass@localhost:27017\t"
"local machine on port 27017, and authenticate with username "
"user and password pass\n");
return exit_code;
}
mongoc_init ();
if (strncmp (argv[1], "mongodb://", 10) == 0) {
host_and_port = bson_strdup (argv[1]);
} else {
host_and_port = bson_strdup_printf ("mongodb://%s", argv[1]);
}
uri = mongoc_uri_new_with_error (host_and_port, &error);
if (!uri) {
fprintf (stderr,
"failed to parse URI: %s\n"
"error message: %s\n",
host_and_port,
error.message);
goto cleanup;
}
client = mongoc_client_new_from_uri (uri);
if (!client) {
goto cleanup;
}
mongoc_client_set_error_api (client, 2);
database = mongoc_client_get_database (client, "test");
collection = mongoc_database_get_collection (database, COLLECTION_NAME);
printf ("Inserting data\n");
if (!insert_data (collection)) {
goto cleanup;
}
printf ("distinct\n");
if (!distinct (database)) {
goto cleanup;
}
printf ("map reduce\n");
if (!map_reduce_basic (database)) {
goto cleanup;
}
printf ("more complicated map reduce\n");
if (!map_reduce_advanced (database)) {
goto cleanup;
}
exit_code = EXIT_SUCCESS;
cleanup:
if (collection) {
mongoc_collection_destroy (collection);
}
if (database) {
mongoc_database_destroy (database);
}
if (client) {
mongoc_client_destroy (client);
}
if (uri) {
mongoc_uri_destroy (uri);
}
if (host_and_port) {
bson_free (host_and_port);
}
mongoc_cleanup ();
return exit_code;
}

Se você quiser tentar o exemplo de redução de mapa avançada com um secundário, inicie um conjunto de réplicas (instruções de como fazer isso podem ser encontradas aqui).

Caso contrário, basta iniciar uma instância do MongoDB:

$ mongod

Agora compile e execute o programa de exemplo:

$ cd examples/basic_aggregation/
$ gcc -Wall -o agg-example basic-aggregation.c $(pkg-config --cflags --libs libmongoc-1.0)
$ ./agg-example localhost
Inserting data
distinct
Next double: 2.000000
Next double: 3.000000
map reduce
{ "result" : "outCollection", "timeMillis" : 155, "counts" : { "input" : 84, "emit" : 126, "reduce" : 3, "output" : 3 }, "ok" : 1 }
{ "_id" : "cat", "value" : 63 }
{ "_id" : "dog", "value" : 42 }
{ "_id" : "mouse", "value" : 21 }
more complicated map reduce
{ "results" : [ { "_id" : "cat", "value" : 63 }, { "_id" : "dog", "value" : 42 }, { "_id" : "mouse", "value" : 21 } ], "timeMillis" : 14, "counts" : { "input" : 84, "emit" : 126, "reduce" : 3, "output" : 3 }, "ok" : 1 }

Voltar

Exemplos de framework de aggregation