| BSON_REFERENCE(3) | libbson | BSON_REFERENCE(3) |
NAME¶
bson_reference - Index
LIBBSON¶
A Cross Platform BSON Library for C
Introduction¶
libbson builds, parses, and iterates BSON documents, the native data format of MongoDB. It also converts BSON to and from JSON, and provides a platform compatibility layer for the MongoDB C Driver.
Tutorial¶
Using libbson In Your C Program¶
Include bson.h¶
All libbson's functions and types are available in one header file. Simply include bson.h:
hello_bson.c
#include <stdio.h>
#include <bson/bson.h>
int
main (int argc, const char **argv)
{
bson_t *b;
char *j;
b = BCON_NEW ("hello", BCON_UTF8 ("bson!"));
j = bson_as_canonical_extended_json (b, NULL);
printf ("%s\n", j);
bson_free (j);
bson_destroy (b);
return 0;
}
CMake¶
The libbson installation includes a CMake config-file package, so you can use CMake's find_package command to import libbson's CMake target and link to libbson (as a shared library):
CMakeLists.txt
# Specify the minimum version you require. find_package (bson-1.0 1.7 REQUIRED) # The "hello_bson.c" sample program is shared among four tests. add_executable (hello_bson ../../hello_bson.c) target_link_libraries (hello_bson PRIVATE mongo::bson_shared)
You can also use libbson as a static library instead: Use the mongo::bson_static CMake target:
# Specify the minimum version you require. find_package (bson-1.0 1.7 REQUIRED) # The "hello_bson.c" sample program is shared among four tests. add_executable (hello_bson ../../hello_bson.c) target_link_libraries (hello_bson PRIVATE mongo::bson_static)
pkg-config¶
If you're not using CMake, use pkg-config on the command line to set header and library paths:
gcc -o hello_bson hello_bson.c $(pkg-config --libs --cflags libbson-1.0)
Or to statically link to libbson:
gcc -o hello_bson hello_bson.c $(pkg-config --libs --cflags libbson-static-1.0)
Creating a BSON Document¶
The bson_t structure¶
BSON documents are created using the bson_t structure. This structure encapsulates the necessary logic for encoding using the BSON Specification. At the core, bson_t is a buffer manager and set of encoding routines.
TIP:
Let's start by creating a new BSON document on the stack. Whenever using libbson, make sure you #include <bson/bson.h>.
bson_t b; bson_init (&b);
This creates an empty document. In JSON, this would be the same as {}.
We can now proceed to adding items to the BSON document. A variety of functions prefixed with bson_append_ can be used based on the type of field you want to append. Let's append a UTF-8 encoded string.
bson_append_utf8 (&b, "key", -1, "value", -1);
Notice the two -1 parameters. The first indicates that the length of key in bytes should be determined with strlen(). Alternatively, we could have passed the number 3. The same goes for the second -1, but for value.
Libbson provides macros to make this less tedious when using string literals. The following two appends are identical.
bson_append_utf8 (&b, "key", -1, "value", -1); BSON_APPEND_UTF8 (&b, "key", "value");
Now let's take a look at an example that adds a few different field types to a BSON document.
bson_t b = BSON_INITIALIZER; BSON_APPEND_INT32 (&b, "a", 1); BSON_APPEND_UTF8 (&b, "hello", "world"); BSON_APPEND_BOOL (&b, "bool", true);
Notice that we omitted the call to bson_init(). By specifying BSON_INITIALIZER we can remove the need to initialize the structure to a base state.
Sub-Documents and Sub-Arrays¶
To simplify the creation of sub-documents and arrays, bson_append_document_begin() and bson_append_array_begin() exist. These can be used to build a sub-document using the parent documents memory region as the destination buffer.
bson_t parent;
bson_t child;
char *str;
bson_init (&parent);
bson_append_document_begin (&parent, "foo", 3, &child);
bson_append_int32 (&child, "baz", 3, 1);
bson_append_document_end (&parent, &child);
str = bson_as_canonical_extended_json (&parent, NULL);
printf ("%s\n", str);
bson_free (str);
bson_destroy (&parent);
{ "foo" : { "baz" : 1 } }
Simplified BSON C Object Notation¶
Creating BSON documents by hand can be tedious and time consuming. BCON, or BSON C Object Notation, was added to allow for the creation of BSON documents in a format that looks closer to the destination format.
The following example shows the use of BCON. Notice that values for fields are wrapped in the BCON_* macros. These are required for the variadic processor to determine the parameter type.
bson_t *doc;
doc = BCON_NEW ("foo",
"{",
"int",
BCON_INT32 (1),
"array",
"[",
BCON_INT32 (100),
"{",
"sub",
BCON_UTF8 ("value"),
"}",
"]",
"}");
Creates the following document
{ "foo" : { "int" : 1, "array" : [ 100, { "sub" : "value" } ] } }
Handling Errors¶
Description¶
Many libbson functions report errors by returning NULL or -1 and filling out a bson_error_t structure with an error domain, error code, and message.
- error.domain names the subsystem that generated the error.
- error.code is a domain-specific error type.
- error.message describes the error.
Some error codes overlap with others; always check both the domain and code to determine the type of error.
| BSON_ERROR_JSON | BSON_JSON_ERROR_READ_CORRUPT_JS BSON_JSON_ERROR_READ_INVALID_PARAM BSON_JSON_ERROR_READ_CB_FAILURE | bson_json_reader_t tried to parse invalid MongoDB Extended JSON. Tried to parse a valid JSON document that is invalid as MongoDBExtended JSON. An internal callback failure during JSON parsing. |
| BSON_ERROR_READER | BSON_ERROR_READER_BADFD | bson_json_reader_new_from_file() could not open the file. |
ObjectIDs¶
Libbson provides a simple way to generate ObjectIDs. It can be used in a single-threaded or multi-threaded manner depending on your requirements.
The bson_oid_t structure represents an ObjectID in MongoDB. It is a 96-bit identifier.
Composition¶
- 4 bytes : The UNIX timestamp in big-endian format.
- 5 bytes : A random number.
- 3 bytes : A 24-bit monotonic counter incrementing from rand() in big-endian.
Sorting ObjectIDs¶
The typical way to sort in C is using qsort(). Therefore, Libbson provides a qsort() compatible callback function named bson_oid_compare(). It returns less than 1, greater than 1, or 0 depending on the equality of two bson_oid_t structures.
Comparing Object IDs¶
If you simply want to compare two bson_oid_t structures for equality, use bson_oid_equal().
Generating¶
To generate a bson_oid_t, you may use the following.
bson_oid_t oid; bson_oid_init (&oid, NULL);
Parsing ObjectID Strings¶
You can also parse a string containing a bson_oid_t. The input string MUST be 24 characters or more in length.
bson_oid_t oid; bson_oid_init_from_string (&oid, "123456789012345678901234");
bson_oid_t oid; bson_oid_init_from_string_unsafe (&oid, "123456789012345678901234");
Hashing ObjectIDs¶
If you need to store items in a hashtable, you may want to use the bson_oid_t as the key. Libbson provides a hash function for just this purpose. It is based on DJB hash.
unsigned hash; hash = bson_oid_hash (oid);
Fetching ObjectID Creation Time¶
You can easily fetch the time that a bson_oid_t was generated using bson_oid_get_time_t().
time_t t;
t = bson_oid_get_time_t (oid);
printf ("The OID was generated at %u\n", (unsigned) t);
Parsing and Iterating BSON Documents¶
Parsing¶
BSON documents are lazily parsed as necessary. To begin parsing a BSON document, use one of the provided Libbson functions to create a new bson_t from existing data such as bson_new_from_data(). This will make a copy of the data so that additional mutations may occur to the BSON document.
TIP:
bson_t *b;
b = bson_new_from_data (my_data, my_data_len);
if (!b) {
fprintf (stderr, "The specified length embedded in <my_data> did not match "
"<my_data_len>\n");
return;
}
bson_destroy (b);
Only two checks are performed when creating a new bson_t from an existing buffer. First, the document must begin with the buffer length, matching what was expected by the caller. Second, the document must end with the expected trailing \0 byte.
To parse the document further we use a bson_iter_t to iterate the elements within the document. Let's print all of the field names in the document.
bson_t *b;
bson_iter_t iter;
if ((b = bson_new_from_data (my_data, my_data_len))) {
if (bson_iter_init (&iter, b)) {
while (bson_iter_next (&iter)) {
printf ("Found element key: \"%s\"\n", bson_iter_key (&iter));
}
}
bson_destroy (b);
}
Converting a document to JSON uses a bson_iter_t and bson_visitor_t to iterate all fields of a BSON document recursively and generate a UTF-8 encoded JSON string.
bson_t *b;
char *json;
if ((b = bson_new_from_data (my_data, my_data_len))) {
if ((json = bson_as_canonical_extended_json (b, NULL))) {
printf ("%s\n", json);
bson_free (json);
}
bson_destroy (b);
}
Recursing into Sub-Documents¶
Libbson provides convenient sub-iterators to dive down into a sub-document or sub-array. Below is an example that will dive into a sub-document named "foo" and print it's field names.
bson_iter_t iter; bson_iter_t child; char *json; if (bson_iter_init_find (&iter, doc, "foo") &&
BSON_ITER_HOLDS_DOCUMENT (&iter) && bson_iter_recurse (&iter, &child)) {
while (bson_iter_next (&child)) {
printf ("Found sub-key of \"foo\" named \"%s\"\n",
bson_iter_key (&child));
} }
Finding Fields using Dot Notation¶
Using the bson_iter_recurse() function exemplified above, bson_iter_find_descendant() can find a field for you using the MongoDB style path notation such as "foo.bar.0.baz".
Let's create a document like {"foo": {"bar": [{"baz: 1}]}} and locate the "baz" field.
bson_t *b; bson_iter_t iter; bson_iter_t baz; b =
BCON_NEW ("foo", "{", "bar", "[", "{", "baz", BCON_INT32 (1), "}", "]", "}"); if (bson_iter_init (&iter, b) &&
bson_iter_find_descendant (&iter, "foo.bar.0.baz", &baz) &&
BSON_ITER_HOLDS_INT32 (&baz)) {
printf ("baz = %d\n", bson_iter_int32 (&baz)); } bson_destroy (b);
Validating a BSON Document¶
If all you want to do is validate that a BSON document is valid, you can use bson_validate().
size_t err_offset;
if (!bson_validate (doc, BSON_VALIDATE_NONE, &err_offset)) {
fprintf (stderr,
"The document failed to validate at offset: %u\n",
(unsigned) err_offset);
}
See the bson_validate() documentation for more information and examples.
UTF-8¶
Encoding¶
Libbson expects that you are always working with UTF-8 encoded text. Anything else is invalid API use.
If you should need to walk through UTF-8 sequences, you can use the various UTF-8 helper functions distributed with Libbson.
Validating a UTF-8 Sequence¶
To validate the string contained in my_string, use the following. You may pass -1 for the string length if you know the string is NULL-terminated.
if (!bson_utf8_validate (my_string, -1, false)) {
printf ("Validation failed.\n");
}
If my_string has NULL bytes within the string, you must provide the string length. Use the following format. Notice the true at the end indicating \0 is allowed.
if (!bson_utf8_validate (my_string, my_string_len, true)) {
printf ("Validation failed.\n");
}
For more information see the API reference for bson_utf8_validate().
Guides¶
Streaming BSON¶
bson_reader_t provides a streaming reader which can be initialized with a filedescriptor or memory region. bson_writer_t provides a streaming writer which can be initialized with a memory region. (Streaming BSON to a file descriptor is not yet supported.)
Reading from a BSON Stream¶
bson_reader_t provides a convenient API to read sequential BSON documents from a file-descriptor or memory buffer. The bson_reader_read() function will read forward in the underlying stream and return a bson_t that can be inspected and iterated upon.
#include <stdio.h>
#include <bson/bson.h>
int
main (int argc, char *argv[])
{
bson_reader_t *reader;
const bson_t *doc;
bson_error_t error;
bool eof;
reader = bson_reader_new_from_file ("mycollection.bson", &error);
if (!reader) {
fprintf (stderr, "Failed to open file.\n");
return 1;
}
while ((doc = bson_reader_read (reader, &eof))) {
char *str = bson_as_canonical_extended_json (doc, NULL);
printf ("%s\n", str);
bson_free (str);
}
if (!eof) {
fprintf (stderr,
"corrupted bson document found at %u\n",
(unsigned) bson_reader_tell (reader));
}
bson_reader_destroy (reader);
return 0;
}
See bson_reader_new_from_fd(), bson_reader_new_from_file(), and bson_reader_new_from_data() for more information.
Writing a sequence of BSON Documents¶
bson_writer_t provides a convenient API to write a sequence of BSON documents to a memory buffer that can grow with realloc(). The bson_writer_begin() and bson_writer_end() functions will manage the underlying buffer while building the sequence of documents.
This could also be useful if you want to write to a network packet while serializing the documents from a higher level language, (but do so just after the packets header).
#include <stdio.h>
#include <bson/bson.h>
#include <assert.h>
int
main (int argc, char *argv[])
{
bson_writer_t *writer;
bson_t *doc;
uint8_t *buf = NULL;
size_t buflen = 0;
bool r;
int i;
writer = bson_writer_new (&buf, &buflen, 0, bson_realloc_ctx, NULL);
for (i = 0; i < 10000; i++) {
r = bson_writer_begin (writer, &doc);
assert (r);
r = BSON_APPEND_INT32 (doc, "i", i);
assert (r);
bson_writer_end (writer);
}
bson_free (buf);
return 0;
}
See bson_writer_new() for more information.
JSON¶
Libbson provides routines for converting to and from the JSON format. In particular, it supports the MongoDB extended JSON format.
Converting BSON to JSON¶
There are often times where you might want to convert a BSON document to JSON. It is convenient for debugging as well as an interchange format. To help with this, Libbson contains the functions bson_as_canonical_extended_json() and bson_as_relaxed_extended_json(). The canonical format preserves BSON type information for values that may have ambiguous representations in JSON (e.g. numeric types).
bson_t *b;
size_t len;
char *str;
b = BCON_NEW ("a", BCON_INT32 (1));
str = bson_as_canonical_extended_json (b, &len);
printf ("%s\n", str);
bson_free (str);
bson_destroy (b);
{ "a" : { "$numberInt": "1" } }
The relaxed format prefers JSON primitives for numeric values and may be used if type fidelity is not required.
bson_t *b;
size_t len;
char *str;
b = BCON_NEW ("a", BCON_INT32 (1));
str = bson_as_relaxed_extended_json (b, &len);
printf ("%s\n", str);
bson_free (str);
bson_destroy (b);
{ "a" : 1 }
Converting JSON to BSON¶
Converting back from JSON is also useful and common enough that we added bson_init_from_json() and bson_new_from_json().
The following example creates a new bson_t from the JSON string {"a":1}.
bson_t *b;
bson_error_t error;
b = bson_new_from_json ("{\"a\":1}", -1, &error);
if (!b) {
printf ("Error: %s\n", error.message);
} else {
bson_destroy (b);
}
Streaming JSON Parsing¶
Libbson provides bson_json_reader_t to allow for parsing a sequence of JSON documents into BSON. The interface is similar to bson_reader_t but expects the input to be in the MongoDB extended JSON format.
/*
* Copyright 2013 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.
*/ /*
* This program will print each JSON document contained in the provided files
* as a BSON string to STDOUT.
*/ #include <bson/bson.h> #include <stdlib.h> #include <stdio.h> int main (int argc, char *argv[]) {
bson_json_reader_t *reader;
bson_error_t error;
const char *filename;
bson_t doc = BSON_INITIALIZER;
int i;
int b;
/*
* Print program usage if no arguments are provided.
*/
if (argc == 1) {
fprintf (stderr, "usage: %s FILE...\n", argv[0]);
return 1;
}
/*
* Process command line arguments expecting each to be a filename.
*/
for (i = 1; i < argc; i++) {
filename = argv[i];
/*
* Open the filename provided in command line arguments.
*/
if (0 == strcmp (filename, "-")) {
reader = bson_json_reader_new_from_fd (STDIN_FILENO, false);
} else {
if (!(reader = bson_json_reader_new_from_file (filename, &error))) {
fprintf (
stderr, "Failed to open \"%s\": %s\n", filename, error.message);
continue;
}
}
/*
* Convert each incoming document to BSON and print to stdout.
*/
while ((b = bson_json_reader_read (reader, &doc, &error))) {
if (b < 0) {
fprintf (stderr, "Error in json parsing:\n%s\n", error.message);
abort ();
}
if (fwrite (bson_get_data (&doc), 1, doc.len, stdout) != doc.len) {
fprintf (stderr, "Failed to write to stdout, exiting.\n");
exit (1);
}
bson_reinit (&doc);
}
bson_json_reader_destroy (reader);
bson_destroy (&doc);
}
return 0; }
Examples¶
The following example reads BSON documents from stdin and prints them to stdout as JSON.
/*
* Copyright 2013 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.
*/ /*
* This program will print each BSON document contained in the provided files
* as a JSON string to STDOUT.
*/ #include <bson/bson.h> #include <stdio.h> int main (int argc, char *argv[]) {
bson_reader_t *reader;
const bson_t *b;
bson_error_t error;
const char *filename;
char *str;
int i;
/*
* Print program usage if no arguments are provided.
*/
if (argc == 1) {
fprintf (stderr, "usage: %s [FILE | -]...\nUse - for STDIN.\n", argv[0]);
return 1;
}
/*
* Process command line arguments expecting each to be a filename.
*/
for (i = 1; i < argc; i++) {
filename = argv[i];
if (strcmp (filename, "-") == 0) {
reader = bson_reader_new_from_fd (STDIN_FILENO, false);
} else {
if (!(reader = bson_reader_new_from_file (filename, &error))) {
fprintf (
stderr, "Failed to open \"%s\": %s\n", filename, error.message);
continue;
}
}
/*
* Convert each incoming document to JSON and print to stdout.
*/
while ((b = bson_reader_read (reader, NULL))) {
str = bson_as_canonical_extended_json (b, NULL);
fprintf (stdout, "%s\n", str);
bson_free (str);
}
/*
* Cleanup after our reader, which closes the file descriptor.
*/
bson_reader_destroy (reader);
}
return 0; }
Use Valgrind to Check For BSON Data Leaks¶
A stack-allocated bson_t contains a small internal buffer; it only heap-allocates additional storage if necessary, depending on its data size. Therefore if you forget to call bson_destroy() on a stack-allocated bson_t, it might or might not cause a leak that can be detected by valgrind during testing.
To catch all potential BSON data leaks in your code, configure the BSON_MEMCHECK flag:
cmake -DCMAKE_C_FLAGS="-DBSON_MEMCHECK -g" .
With this flag set, every bson_t mallocs at least one byte. Run your program's unittests with valgrind to verify all bson_t structs are destroyed.
Set the environment variable MONGOC_TEST_VALGRIND to on to skip timing-dependent tests known to fail with valgrind.
Cross Platform Notes¶
Endianness¶
The BSON specification dictates that the encoding format is in little-endian. Many implementations simply ignore endianness altogether and expect that they are to be run on little-endian. Libbson supports both Big and Little Endian systems. This means we use memcpy() when appropriate instead of dereferencing and properly convert to and from the host endian format. We expect the compiler intrinsics to optimize it to a dereference when possible.
Threading¶
Libbson's data structures are NOT thread-safe. You are responsible for accessing and mutating these structures from one thread at a time.
Libbson requires POSIX threads (pthreads) on all UNIX-like platforms. On Windows, the native threading interface is used. Libbson uses your system's threading library to safely generate unique ObjectIds, and to provide a fallback implementation for atomic operations on platforms without built-in atomics.
API Reference¶
bson_t¶
BSON Document Abstraction
Synopsis¶
#include <bson/bson.h> /**
* bson_empty:
* @b: a bson_t.
*
* Checks to see if @b is an empty BSON document. An empty BSON document is
* a 5 byte document which contains the length (4 bytes) and a single NUL
* byte indicating end of fields.
*/ #define bson_empty(b) /* ... */ /**
* bson_empty0:
*
* Like bson_empty() but treats NULL the same as an empty bson_t document.
*/ #define bson_empty0(b) /* ... */ /**
* bson_clear:
*
* Easily free a bson document and set it to NULL. Use like:
*
* bson_t *doc = bson_new();
* bson_clear (&doc);
* BSON_ASSERT (doc == NULL);
*/ #define bson_clear(bptr) /* ... */ /**
* BSON_MAX_SIZE:
*
* The maximum size in bytes of a BSON document.
*/ #define BSON_MAX_SIZE /* ... */ #define BSON_APPEND_ARRAY(b, key, val) \
bson_append_array (b, key, (int) strlen (key), val) #define BSON_APPEND_ARRAY_BEGIN(b, key, child) \
bson_append_array_begin (b, key, (int) strlen (key), child) #define BSON_APPEND_BINARY(b, key, subtype, val, len) \
bson_append_binary (b, key, (int) strlen (key), subtype, val, len) #define BSON_APPEND_BOOL(b, key, val) \
bson_append_bool (b, key, (int) strlen (key), val) #define BSON_APPEND_CODE(b, key, val) \
bson_append_code (b, key, (int) strlen (key), val) #define BSON_APPEND_CODE_WITH_SCOPE(b, key, val, scope) \
bson_append_code_with_scope (b, key, (int) strlen (key), val, scope) #define BSON_APPEND_DBPOINTER(b, key, coll, oid) \
bson_append_dbpointer (b, key, (int) strlen (key), coll, oid) #define BSON_APPEND_DOCUMENT_BEGIN(b, key, child) \
bson_append_document_begin (b, key, (int) strlen (key), child) #define BSON_APPEND_DOUBLE(b, key, val) \
bson_append_double (b, key, (int) strlen (key), val) #define BSON_APPEND_DOCUMENT(b, key, val) \
bson_append_document (b, key, (int) strlen (key), val) #define BSON_APPEND_INT32(b, key, val) \
bson_append_int32 (b, key, (int) strlen (key), val) #define BSON_APPEND_INT64(b, key, val) \
bson_append_int64 (b, key, (int) strlen (key), val) #define BSON_APPEND_MINKEY(b, key) \
bson_append_minkey (b, key, (int) strlen (key)) #define BSON_APPEND_DECIMAL128(b, key, val) \
bson_append_decimal128 (b, key, (int) strlen (key), val) #define BSON_APPEND_MAXKEY(b, key) \
bson_append_maxkey (b, key, (int) strlen (key)) #define BSON_APPEND_NULL(b, key) bson_append_null (b, key, (int) strlen (key)) #define BSON_APPEND_OID(b, key, val) \
bson_append_oid (b, key, (int) strlen (key), val) #define BSON_APPEND_REGEX(b, key, val, opt) \
bson_append_regex (b, key, (int) strlen (key), val, opt) #define BSON_APPEND_UTF8(b, key, val) \
bson_append_utf8 (b, key, (int) strlen (key), val, (int) strlen (val)) #define BSON_APPEND_SYMBOL(b, key, val) \
bson_append_symbol (b, key, (int) strlen (key), val, (int) strlen (val)) #define BSON_APPEND_TIME_T(b, key, val) \
bson_append_time_t (b, key, (int) strlen (key), val) #define BSON_APPEND_TIMEVAL(b, key, val) \
bson_append_timeval (b, key, (int) strlen (key), val) #define BSON_APPEND_DATE_TIME(b, key, val) \
bson_append_date_time (b, key, (int) strlen (key), val) #define BSON_APPEND_TIMESTAMP(b, key, val, inc) \
bson_append_timestamp (b, key, (int) strlen (key), val, inc) #define BSON_APPEND_UNDEFINED(b, key) \
bson_append_undefined (b, key, (int) strlen (key)) #define BSON_APPEND_VALUE(b, key, val) \
bson_append_value (b, key, (int) strlen (key), (val)) BSON_ALIGNED_BEGIN (128) typedef struct {
uint32_t flags; /* Internal flags for the bson_t. */
uint32_t len; /* Length of BSON data. */
uint8_t padding[120]; /* Padding for stack allocation. */ } bson_t BSON_ALIGNED_END (128);
Description¶
The bson_t structure represents a BSON document. This structure manages the underlying BSON encoded buffer. For mutable documents, it can append new data to the document.
Performance Notes¶
The bson_t structure attempts to use an inline allocation within the structure to speed up performance of small documents. When this internal buffer has been exhausted, a heap allocated buffer will be dynamically allocated. Therefore, it is essential to call bson_destroy() on allocated documents.
Example¶
static void
create_on_heap (void)
{
bson_t *b = bson_new ();
BSON_APPEND_INT32 (b, "foo", 123);
BSON_APPEND_UTF8 (b, "bar", "foo");
BSON_APPEND_DOUBLE (b, "baz", 1.23f);
bson_destroy (b);
}
bson_context_t¶
BSON OID Generation Context
Synopsis¶
#include <bson/bson.h>
typedef enum {
BSON_CONTEXT_NONE = 0,
BSON_CONTEXT_DISABLE_PID_CACHE = (1 << 2),
} bson_context_flags_t;
typedef struct _bson_context_t bson_context_t;
bson_context_t *
bson_context_get_default (void) BSON_GNUC_CONST;
bson_context_t *
bson_context_new (bson_context_flags_t flags);
void
bson_context_destroy (bson_context_t *context);
Description¶
The bson_context_t structure is context for generation of BSON Object IDs. This context allows overriding behavior of generating ObjectIDs. The flags BSON_CONTEXT_NONE, BSON_CONTEXT_THREAD_SAFE, and BSON_CONTEXT_DISABLE_PID_CACHE are the only ones used. The others have no effect.
Example¶
#include <bson/bson.h>
int
main (int argc, char *argv[])
{
bson_context_t *ctx = NULL;
bson_oid_t oid;
/* use default context, via bson_context_get_default() */
bson_oid_init (&oid, NULL);
/* specify a local context for additional control */
ctx = bson_context_new (BSON_CONTEXT_NONE);
bson_oid_init (&oid, ctx);
bson_context_destroy (ctx);
return 0;
}
bson_decimal128_t¶
BSON Decimal128 Abstraction
Synopsis¶
#include <bson/bson.h>
#define BSON_DECIMAL128_STRING 43
#define BSON_DECIMAL128_INF "Infinity"
#define BSON_DECIMAL128_NAN "NaN"
typedef struct {
#if BSON_BYTE_ORDER == BSON_LITTLE_ENDIAN
uint64_t low;
uint64_t high;
#elif BSON_BYTE_ORDER == BSON_BIG_ENDIAN
uint64_t high;
uint64_t low;
#endif
} bson_decimal128_t;
Description¶
The bson_decimal128_t structure represents the IEEE-754 Decimal128 data type.
Example¶
#include <bson/bson.h>
#include <stdio.h>
int
main (int argc, char *argv[])
{
char string[BSON_DECIMAL128_STRING];
bson_decimal128_t decimal128;
bson_decimal128_from_string ("100.00", &decimal128);
bson_decimal128_to_string (&decimal128, string);
printf ("Decimal128 value: %s\n", string);
return 0;
}
bson_error_t¶
BSON Error Encapsulation
Synopsis¶
#include <bson/bson.h>
typedef struct {
uint32_t domain;
uint32_t code;
char message[504];
} bson_error_t;
Description¶
The bson_error_t structure is used as an out-parameter to pass error information to the caller. It should be stack-allocated and does not requiring freeing.
See Handling Errors.
Example¶
bson_reader_t *reader;
bson_error_t error;
reader = bson_reader_new_from_file ("dump.bson", &error);
if (!reader) {
fprintf (
stderr, "ERROR: %d.%d: %s\n", error.domain, error.code, error.message);
}
bson_iter_t¶
BSON Document Iterator
Synopsis¶
#include <bson/bson.h> #define BSON_ITER_HOLDS_DOUBLE(iter) /* ... */ #define BSON_ITER_HOLDS_UTF8(iter) /* ... */ #define BSON_ITER_HOLDS_DOCUMENT(iter) /* ... */ #define BSON_ITER_HOLDS_ARRAY(iter) /* ... */ #define BSON_ITER_HOLDS_BINARY(iter) /* ... */ #define BSON_ITER_HOLDS_UNDEFINED(iter) /* ... */ #define BSON_ITER_HOLDS_OID(iter) /* ... */ #define BSON_ITER_HOLDS_BOOL(iter) /* ... */ #define BSON_ITER_HOLDS_DATE_TIME(iter) /* ... */ #define BSON_ITER_HOLDS_NULL(iter) /* ... */ #define BSON_ITER_HOLDS_REGEX(iter) /* ... */ #define BSON_ITER_HOLDS_DBPOINTER(iter) /* ... */ #define BSON_ITER_HOLDS_CODE(iter) /* ... */ #define BSON_ITER_HOLDS_SYMBOL(iter) /* ... */ #define BSON_ITER_HOLDS_CODEWSCOPE(iter) /* ... */ #define BSON_ITER_HOLDS_INT32(iter) /* ... */ #define BSON_ITER_HOLDS_TIMESTAMP(iter) /* ... */ #define BSON_ITER_HOLDS_INT64(iter) /* ... */ #define BSON_ITER_HOLDS_DECIMAL128(iter) /* ... */ #define BSON_ITER_HOLDS_MAXKEY(iter) /* ... */ #define BSON_ITER_HOLDS_MINKEY(iter) /* ... */ #define BSON_ITER_HOLDS_INT(iter) \
(BSON_ITER_HOLDS_INT32 (iter) || BSON_ITER_HOLDS_INT64 (iter)) #define BSON_ITER_HOLDS_NUMBER(iter) \
(BSON_ITER_HOLDS_INT (iter) || BSON_ITER_HOLDS_DOUBLE (iter)) #define BSON_ITER_IS_KEY(iter, key) \
(0 == strcmp ((key), bson_iter_key ((iter)))) typedef struct {
/*< private >*/ } bson_iter_t;
Description¶
bson_iter_t is a structure used to iterate through the elements of a bson_t. It is meant to be used on the stack and can be discarded at any time as it contains no external allocation. The contents of the structure should be considered private and may change between releases, however the structure size will not change.
The bson_t MUST be valid for the lifetime of the iter and it is an error to modify the bson_t while using the iter.
Examples¶
bson_iter_t iter;
if (bson_iter_init (&iter, my_bson_doc)) {
while (bson_iter_next (&iter)) {
printf ("Found a field named: %s\n", bson_iter_key (&iter));
}
}
bson_iter_t iter;
if (bson_iter_init (&iter, my_bson_doc) && bson_iter_find (&iter, "my_field")) {
printf ("Found the field named: %s\n", bson_iter_key (&iter));
}
bson_iter_t iter; bson_iter_t sub_iter; if (bson_iter_init_find (&iter, my_bson_doc, "mysubdoc") &&
(BSON_ITER_HOLDS_DOCUMENT (&iter) || BSON_ITER_HOLDS_ARRAY (&iter)) &&
bson_iter_recurse (&iter, &sub_iter)) {
while (bson_iter_next (&sub_iter)) {
printf ("Found key \"%s\" in sub document.\n", bson_iter_key (&sub_iter));
} }
bson_iter_t iter; if (bson_iter_init (&iter, my_doc) &&
bson_iter_find_descendant (&iter, "a.b.c.d", &sub_iter)) {
printf ("The type of a.b.c.d is: %d\n", (int) bson_iter_type (&sub_iter)); }
bson_json_reader_t¶
Bulk JSON to BSON conversion
Synopsis¶
#include <bson/bson.h>
typedef struct _bson_json_reader_t bson_json_reader_t;
typedef enum {
BSON_JSON_ERROR_READ_CORRUPT_JS = 1,
BSON_JSON_ERROR_READ_INVALID_PARAM,
BSON_JSON_ERROR_READ_CB_FAILURE,
} bson_json_error_code_t;
Description¶
The bson_json_reader_t structure is used for reading a sequence of JSON documents and transforming them to bson_t documents.
This can often be useful if you want to perform bulk operations that are defined in a file containing JSON documents.
TIP:
Example¶
/*
* Copyright 2013 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.
*/ /*
* This program will print each JSON document contained in the provided files
* as a BSON string to STDOUT.
*/ #include <bson/bson.h> #include <stdlib.h> #include <stdio.h> int main (int argc, char *argv[]) {
bson_json_reader_t *reader;
bson_error_t error;
const char *filename;
bson_t doc = BSON_INITIALIZER;
int i;
int b;
/*
* Print program usage if no arguments are provided.
*/
if (argc == 1) {
fprintf (stderr, "usage: %s FILE...\n", argv[0]);
return 1;
}
/*
* Process command line arguments expecting each to be a filename.
*/
for (i = 1; i < argc; i++) {
filename = argv[i];
/*
* Open the filename provided in command line arguments.
*/
if (0 == strcmp (filename, "-")) {
reader = bson_json_reader_new_from_fd (STDIN_FILENO, false);
} else {
if (!(reader = bson_json_reader_new_from_file (filename, &error))) {
fprintf (
stderr, "Failed to open \"%s\": %s\n", filename, error.message);
continue;
}
}
/*
* Convert each incoming document to BSON and print to stdout.
*/
while ((b = bson_json_reader_read (reader, &doc, &error))) {
if (b < 0) {
fprintf (stderr, "Error in json parsing:\n%s\n", error.message);
abort ();
}
if (fwrite (bson_get_data (&doc), 1, doc.len, stdout) != doc.len) {
fprintf (stderr, "Failed to write to stdout, exiting.\n");
exit (1);
}
bson_reinit (&doc);
}
bson_json_reader_destroy (reader);
bson_destroy (&doc);
}
return 0; }
bson_md5_t¶
BSON MD5 Abstraction
Deprecated¶
All MD5 APIs are deprecated in libbson.
Synopsis¶
typedef struct {
uint32_t count[2]; /* message length in bits, lsw first */
uint32_t abcd[4]; /* digest buffer */
uint8_t buf[64]; /* accumulate block */
} bson_md5_t;
Description¶
bson_md5_t encapsulates an implementation of the MD5 algorithm.
bson_oid_t¶
BSON ObjectID Abstraction
Synopsis¶
#include <bson/bson.h>
typedef struct {
uint8_t bytes[12];
} bson_oid_t;
Description¶
The bson_oid_t structure contains the 12-byte ObjectId notation defined by the BSON ObjectID specification.
ObjectId is a 12-byte BSON type, constructed using:
- a 4-byte value representing the seconds since the Unix epoch (in Big Endian).
- a 5-byte random value.
- a 3-byte counter (Big Endian), starting with a random value.
String Conversion¶
You can convert an Object ID to a string using bson_oid_to_string() and back with bson_oid_init_from_string().
Hashing¶
A bson_oid_t can be used in hashtables using the function bson_oid_hash() and bson_oid_equal().
Comparing¶
A bson_oid_t can be compared to another using bson_oid_compare() for qsort() style comparing and bson_oid_equal() for direct equality.
Validating¶
You can validate that a string containing a hex-encoded ObjectID is valid using the function bson_oid_is_valid().
Example¶
#include <bson/bson.h>
#include <stdio.h>
int
main (int argc, char *argv[])
{
bson_oid_t oid;
char str[25];
bson_oid_init (&oid, NULL);
bson_oid_to_string (&oid, str);
printf ("%s\n", str);
if (bson_oid_is_valid (str, sizeof str)) {
bson_oid_init_from_string (&oid, str);
}
printf ("The UNIX time was: %u\n", (unsigned) bson_oid_get_time_t (&oid));
return 0;
}
bson_reader_t¶
Streaming BSON Document Reader
Synopsis¶
#include <bson/bson.h> typedef struct _bson_reader_t bson_reader_t; bson_reader_t * bson_reader_new_from_handle (void *handle,
bson_reader_read_func_t rf,
bson_reader_destroy_func_t df); bson_reader_t * bson_reader_new_from_fd (int fd, bool close_on_destroy); bson_reader_t * bson_reader_new_from_file (const char *path, bson_error_t *error); bson_reader_t * bson_reader_new_from_data (const uint8_t *data, size_t length); void bson_reader_destroy (bson_reader_t *reader);
Description¶
bson_reader_t is a structure used for reading a sequence of BSON documents. The sequence can come from a file-descriptor, memory region, or custom callbacks.
Example¶
/*
* Copyright 2013 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.
*/ /*
* This program will print each BSON document contained in the provided files
* as a JSON string to STDOUT.
*/ #include <bson/bson.h> #include <stdio.h> int main (int argc, char *argv[]) {
bson_reader_t *reader;
const bson_t *b;
bson_error_t error;
const char *filename;
char *str;
int i;
/*
* Print program usage if no arguments are provided.
*/
if (argc == 1) {
fprintf (stderr, "usage: %s [FILE | -]...\nUse - for STDIN.\n", argv[0]);
return 1;
}
/*
* Process command line arguments expecting each to be a filename.
*/
for (i = 1; i < argc; i++) {
filename = argv[i];
if (strcmp (filename, "-") == 0) {
reader = bson_reader_new_from_fd (STDIN_FILENO, false);
} else {
if (!(reader = bson_reader_new_from_file (filename, &error))) {
fprintf (
stderr, "Failed to open \"%s\": %s\n", filename, error.message);
continue;
}
}
/*
* Convert each incoming document to JSON and print to stdout.
*/
while ((b = bson_reader_read (reader, NULL))) {
str = bson_as_canonical_extended_json (b, NULL);
fprintf (stdout, "%s\n", str);
bson_free (str);
}
/*
* Cleanup after our reader, which closes the file descriptor.
*/
bson_reader_destroy (reader);
}
return 0; }
Character and String Routines¶
We provide a small number of character and string routines to substitute for those that are not available on all platforms, and routines to make UTF-8 character manipulation convenient.
bson_string_t¶
String Building Abstraction
Synopsis¶
#include <bson/bson.h>
typedef struct {
char *str;
uint32_t len;
uint32_t alloc;
} bson_string_t;
Description¶
bson_string_t is an abstraction for building strings. As chunks are added to the string, allocations are performed in powers of two.
This API is useful if you need to build UTF-8 encoded strings.
Example¶
bson_string_t *str;
str = bson_string_new (NULL);
bson_string_append_printf (str, "%d %s %f\n", 0, "some string", 0.123);
printf ("%s\n", str->str);
bson_string_free (str, true);
TIP:
bson_subtype_t¶
Binary Field Subtype
Synopsis¶
#include <bson/bson.h>
typedef enum {
BSON_SUBTYPE_BINARY = 0x00,
BSON_SUBTYPE_FUNCTION = 0x01,
BSON_SUBTYPE_BINARY_DEPRECATED = 0x02,
BSON_SUBTYPE_UUID_DEPRECATED = 0x03,
BSON_SUBTYPE_UUID = 0x04,
BSON_SUBTYPE_MD5 = 0x05,
BSON_SUBTYPE_COLUMN = 0x07,
BSON_SUBTYPE_USER = 0x80,
} bson_subtype_t;
Description¶
This enumeration contains the various subtypes that may be used in a binary field. See http://bsonspec.org for more information.
Example¶
bson_t doc = BSON_INITIALIZER; BSON_APPEND_BINARY (&doc, "binary", BSON_SUBTYPE_BINARY, data, data_len);
bson_type_t¶
BSON Type Enumeration
Synopsis¶
#include <bson/bson.h>
typedef enum {
BSON_TYPE_EOD = 0x00,
BSON_TYPE_DOUBLE = 0x01,
BSON_TYPE_UTF8 = 0x02,
BSON_TYPE_DOCUMENT = 0x03,
BSON_TYPE_ARRAY = 0x04,
BSON_TYPE_BINARY = 0x05,
BSON_TYPE_UNDEFINED = 0x06,
BSON_TYPE_OID = 0x07,
BSON_TYPE_BOOL = 0x08,
BSON_TYPE_DATE_TIME = 0x09,
BSON_TYPE_NULL = 0x0A,
BSON_TYPE_REGEX = 0x0B,
BSON_TYPE_DBPOINTER = 0x0C,
BSON_TYPE_CODE = 0x0D,
BSON_TYPE_SYMBOL = 0x0E,
BSON_TYPE_CODEWSCOPE = 0x0F,
BSON_TYPE_INT32 = 0x10,
BSON_TYPE_TIMESTAMP = 0x11,
BSON_TYPE_INT64 = 0x12,
BSON_TYPE_DECIMAL128 = 0x13,
BSON_TYPE_MAXKEY = 0x7F,
BSON_TYPE_MINKEY = 0xFF,
} bson_type_t;
Description¶
The bson_type_t enumeration contains all of the types from the BSON Specification. It can be used to determine the type of a field at runtime.
Example¶
bson_iter_t iter; if (bson_iter_init_find (&iter, doc, "foo") &&
(BSON_TYPE_INT32 == bson_iter_type (&iter))) {
printf ("'foo' is an int32.\n"); }
bson_unichar_t¶
Unicode Character Abstraction
Synopsis¶
typedef uint32_t bson_unichar_t;
Description¶
bson_unichar_t provides an abstraction on a single unicode character. It is the 32-bit representation of a character. As UTF-8 can contain multi-byte characters, this should be used when iterating through UTF-8 text.
Example¶
static void
print_each_char (const char *str)
{
bson_unichar_t c;
for (; *str; str = bson_utf8_next_char (str)) {
c = bson_utf8_get_char (str);
printf ("The numberic value is %u.\n", (unsigned) c);
}
}
bson_value_t¶
BSON Boxed Container Type
Synopsis¶
#include <bson/bson.h>
typedef struct _bson_value_t {
bson_type_t value_type;
union {
bson_oid_t v_oid;
int64_t v_int64;
int32_t v_int32;
int8_t v_int8;
double v_double;
bool v_bool;
int64_t v_datetime;
struct {
uint32_t timestamp;
uint32_t increment;
} v_timestamp;
struct {
uint32_t len;
char *str;
} v_utf8;
struct {
uint32_t data_len;
uint8_t *data;
} v_doc;
struct {
uint32_t data_len;
uint8_t *data;
bson_subtype_t subtype;
} v_binary;
struct {
char *regex;
char *options;
} v_regex;
struct {
char *collection;
uint32_t collection_len;
bson_oid_t oid;
} v_dbpointer;
struct {
uint32_t code_len;
char *code;
} v_code;
struct {
uint32_t code_len;
char *code;
uint32_t scope_len;
uint8_t *scope_data;
} v_codewscope;
struct {
uint32_t len;
char *symbol;
} v_symbol;
} value;
} bson_value_t;
Description¶
The bson_value_t structure is a boxed type for encapsulating a runtime determined type.
Example¶
const bson_value_t *value;
value = bson_iter_value (&iter);
if (value->value_type == BSON_TYPE_INT32) {
printf ("%d\n", value->value.v_int32);
}
bson_visitor_t¶
Synopsis¶
#include <bson/bson.h>
typedef struct {
/* run before / after descending into a document */
bool (*visit_before) (const bson_iter_t *iter, const char *key, void *data);
bool (*visit_after) (const bson_iter_t *iter, const char *key, void *data);
/* corrupt BSON, or unsupported type and visit_unsupported_type not set */
void (*visit_corrupt) (const bson_iter_t *iter, void *data);
/* normal bson field callbacks */
bool (*visit_double) (const bson_iter_t *iter,
const char *key,
double v_double,
void *data);
bool (*visit_utf8) (const bson_iter_t *iter,
const char *key,
size_t v_utf8_len,
const char *v_utf8,
void *data);
bool (*visit_document) (const bson_iter_t *iter,
const char *key,
const bson_t *v_document,
void *data);
bool (*visit_array) (const bson_iter_t *iter,
const char *key,
const bson_t *v_array,
void *data);
bool (*visit_binary) (const bson_iter_t *iter,
const char *key,
bson_subtype_t v_subtype,
size_t v_binary_len,
const uint8_t *v_binary,
void *data);
/* normal field with deprecated "Undefined" BSON type */
bool (*visit_undefined) (const bson_iter_t *iter,
const char *key,
void *data);
bool (*visit_oid) (const bson_iter_t *iter,
const char *key,
const bson_oid_t *v_oid,
void *data);
bool (*visit_bool) (const bson_iter_t *iter,
const char *key,
bool v_bool,
void *data);
bool (*visit_date_time) (const bson_iter_t *iter,
const char *key,
int64_t msec_since_epoch,
void *data);
bool (*visit_null) (const bson_iter_t *iter, const char *key, void *data);
bool (*visit_regex) (const bson_iter_t *iter,
const char *key,
const char *v_regex,
const char *v_options,
void *data);
bool (*visit_dbpointer) (const bson_iter_t *iter,
const char *key,
size_t v_collection_len,
const char *v_collection,
const bson_oid_t *v_oid,
void *data);
bool (*visit_code) (const bson_iter_t *iter,
const char *key,
size_t v_code_len,
const char *v_code,
void *data);
bool (*visit_symbol) (const bson_iter_t *iter,
const char *key,
size_t v_symbol_len,
const char *v_symbol,
void *data);
bool (*visit_codewscope) (const bson_iter_t *iter,
const char *key,
size_t v_code_len,
const char *v_code,
const bson_t *v_scope,
void *data);
bool (*visit_int32) (const bson_iter_t *iter,
const char *key,
int32_t v_int32,
void *data);
bool (*visit_timestamp) (const bson_iter_t *iter,
const char *key,
uint32_t v_timestamp,
uint32_t v_increment,
void *data);
bool (*visit_int64) (const bson_iter_t *iter,
const char *key,
int64_t v_int64,
void *data);
bool (*visit_maxkey) (const bson_iter_t *iter, const char *key, void *data);
bool (*visit_minkey) (const bson_iter_t *iter, const char *key, void *data);
/* if set, called instead of visit_corrupt when an apparently valid BSON
* includes an unrecognized field type (reading future version of BSON) */
void (*visit_unsupported_type) (const bson_iter_t *iter,
const char *key,
uint32_t type_code,
void *data);
bool (*visit_decimal128) (const bson_iter_t *iter,
const char *key,
const bson_decimal128_t *v_decimal128,
void *data);
void *padding[7];
} bson_visitor_t bson_visitor_t;
Description¶
The bson_visitor_t structure provides a series of callbacks that can be called while iterating a BSON document. This may simplify the conversion of a bson_t to a higher level language structure.
If the optional callback visit_unsupported_type is set, it is called instead of visit_corrupt in the specific case of an unrecognized field type. (Parsing is aborted in either case.) Use this callback to report an error like "unrecognized type" instead of simply "corrupt BSON". This future-proofs code that may use an older version of libbson to parse future BSON formats.
Basic Example¶
#include <bson/bson.h>
#include <stdio.h>
static bool
my_visit_before (const bson_iter_t *iter, const char *key, void *data)
{
int *count = (int *) data;
(*count)++;
/* returning true stops further iteration of the document */
return false;
}
static void
count_fields (bson_t *doc)
{
bson_visitor_t visitor = {0};
bson_iter_t iter;
int count = 0;
visitor.visit_before = my_visit_before;
if (bson_iter_init (&iter, doc)) {
bson_iter_visit_all (&iter, &visitor, &count);
}
printf ("Found %d fields.\n", count);
}
The example below demonstrates how to set your own callbacks to provide information about the location of corrupt or unsupported BSON document entries.
Example Corruption Check¶
#include <bson/bson.h>
#include <stdio.h>
typedef struct {
ssize_t *err_offset;
} my_state_t;
static void
my_visit_corrupt (const bson_iter_t *iter, void *data)
{
*(((my_state_t *) data)->err_offset) = iter->off;
}
static void
my_visit_unsupported_type (const bson_iter_t *iter,
const char *key,
uint32_t type_code,
void *data)
{
*(((my_state_t *) data)->err_offset) = iter->off;
}
static void
find_error_location (bson_t *doc)
{
bson_visitor_t visitors = {0};
bson_iter_t iter;
my_state_t state;
ssize_t err_offset = -1;
visitors.visit_corrupt = my_visit_corrupt;
visitors.visit_unsupported_type = my_visit_unsupported_type;
/* provide additional visitors as needed based on your requirements */
state.err_offset = &err_offset;
if (!bson_iter_init (&iter, doc)) {
printf ("Could not initialize iterator!");
exit (1);
}
if (bson_iter_visit_all (&iter, &visitors, &state) ||
err_offset != -1) {
printf ("Found error at offset %d.\n", err_offset);
} else {
printf ("BSON document had no errors.\n");
}
}
The example below demonstrates how to use a visitor to validate a BSON document's maximum depth.
Example Custom Validation¶
bson-check-depth.c
/* Reports the maximum nested depth of a BSON document. */
#include <bson/bson.h>
#include <assert.h>
#include <stdio.h>
#include <stdlib.h>
typedef struct {
uint32_t depth;
int max_depth;
bool valid;
} check_depth_t;
bool
_check_depth_document (const bson_iter_t *iter,
const char *key,
const bson_t *v_document,
void *data);
static const bson_visitor_t check_depth_funcs = {
NULL,
NULL,
NULL,
NULL,
NULL,
_check_depth_document,
_check_depth_document,
NULL,
};
bool
_check_depth_document (const bson_iter_t *iter,
const char *key,
const bson_t *v_document,
void *data)
{
check_depth_t *state = (check_depth_t *) data;
bson_iter_t child;
BSON_UNUSED (iter);
BSON_UNUSED (key);
if (!bson_iter_init (&child, v_document)) {
fprintf (stderr, "corrupt\n");
return true; /* cancel */
}
state->depth++;
if (state->depth > state->max_depth) {
state->valid = false;
return true; /* cancel */
}
bson_iter_visit_all (&child, &check_depth_funcs, state);
state->depth--;
return false; /* continue */
}
void
check_depth (const bson_t *bson, int max_depth)
{
bson_iter_t iter;
check_depth_t state = {0};
if (!bson_iter_init (&iter, bson)) {
fprintf (stderr, "corrupt\n");
}
state.valid = true;
state.max_depth = max_depth;
_check_depth_document (&iter, NULL, bson, &state);
if (!state.valid) {
printf ("document exceeds maximum depth of %d\n", state.max_depth);
} else {
char *as_json = bson_as_canonical_extended_json (bson, NULL);
printf ("document %s ", as_json);
printf ("is valid\n");
bson_free (as_json);
}
}
int
main (int argc, char **argv)
{
bson_reader_t *bson_reader;
const bson_t *bson;
bool reached_eof;
char *filename;
bson_error_t error;
int max_depth;
if (argc != 3) {
fprintf (stderr, "usage: %s FILE MAX_DEPTH\n", argv[0]);
fprintf (stderr, "Checks that the depth of the BSON contained in FILE\n");
fprintf (stderr, "does not exceed MAX_DEPTH\n");
}
filename = argv[1];
max_depth = atoi (argv[2]);
bson_reader = bson_reader_new_from_file (filename, &error);
if (!bson_reader) {
printf ("could not read %s: %s\n", filename, error.message);
return 1;
}
while ((bson = bson_reader_read (bson_reader, &reached_eof))) {
check_depth (bson, max_depth);
}
if (!reached_eof) {
printf ("error reading BSON\n");
}
bson_reader_destroy (bson_reader);
return 0;
}
bson_writer_t¶
Bulk BSON serialization Abstraction
Synopsis¶
#include <bson/bson.h> typedef struct _bson_writer_t bson_writer_t; bson_writer_t * bson_writer_new (uint8_t **buf,
size_t *buflen,
size_t offset,
bson_realloc_func realloc_func,
void *realloc_func_ctx); void bson_writer_destroy (bson_writer_t *writer);
Description¶
The bson_writer_t API provides an abstraction for serializing many BSON documents to a single memory region. The memory region may be dynamically allocated and re-allocated as more memory is demanded. This can be useful when building network packets from a high-level language. For example, you can serialize a Python Dictionary directly to a single buffer destined for a TCP packet.
Example¶
#include <bson/bson.h>
int
main (int argc, char *argv[])
{
bson_writer_t *writer;
uint8_t *buf = NULL;
size_t buflen = 0;
bson_t *doc;
writer = bson_writer_new (&buf, &buflen, 0, bson_realloc_ctx, NULL);
for (i = 0; i < 1000; i++) {
bson_writer_begin (writer, &doc);
BSON_APPEND_INT32 (&doc, "i", i);
bson_writer_end (writer);
}
bson_writer_destroy (writer);
bson_free (buf);
return 0;
}
System Clock¶
BSON Clock Abstraction
Synopsis¶
int64_t bson_get_monotonic_time (void); int bson_gettimeofday (struct timeval *tv);
Description¶
The clock abstraction in Libbson provides a cross-platform way to handle timeouts within the BSON library. It abstracts the differences in implementations of gettimeofday() as well as providing a monotonic (incrementing only) clock in microseconds.
Memory Management¶
BSON Memory Abstraction.
Description¶
Libbson contains a lightweight memory abstraction to make portability to new platforms easier. Additionally, it helps us integrate with interesting higher-level languages. One caveat, however, is that Libbson is not designed to deal with Out of Memory (OOM) situations. Doing so requires extreme diligence throughout the application stack that has rarely been implemented correctly. This may change in the future. As it stands now, Libbson will abort() under OOM situations.
To aid in language binding integration, Libbson allows for setting a custom memory allocator via bson_mem_set_vtable(). This allocation may be reversed via bson_mem_restore_vtable().
Libbson Versioning¶
Versioning Macros and Functions
Macros¶
The following preprocessor macros can be used to perform various checks based on the version of the library you are compiling against. This may be useful if you only want to enable a feature on a certain version of the library.
Synopsis¶
#define BSON_CHECK_VERSION(major, minor, micro) #define BSON_MAJOR_VERSION (1) #define BSON_MINOR_VERSION (4) #define BSON_MICRO_VERSION (1) #define BSON_VERSION_S "1.4.1" #define BSON_VERSION_HEX \
(BSON_MAJOR_VERSION << 24 | BSON_MINOR_VERSION << 16 | \
BSON_MICRO_VERSION << 8)
Only compile a block on Libbson 1.1.0 and newer.
#if BSON_CHECK_VERSION(1, 1, 0)
static void
do_something (void)
{
}
#endif
AUTHOR¶
MongoDB, Inc
COPYRIGHT¶
2017-present, MongoDB, Inc
| October 20, 2022 | 1.23.1 |