Fix: allow specifying more than one input trace path
[babeltrace.git] / doc / API.txt
CommitLineData
827b30c1
DS
1Babeltrace API documentation
2
3Babeltrace provides trace read and write libraries, as well as a trace
4converter. A plugin can be created for any trace format to allow its
5conversion to/from another trace format.
6
7The main format expected to be converted to/from is the Common Trace
8Format (CTF). The latest version of the CTF specification can be found at:
9 git tree: git://git.efficios.com/ctf.git
10 gitweb: http://git.efficios.com/?p=ctf.git
11
12This document describes the main concepts to use the libbabeltrace,
13which exposes the Babeltrace trace reading capability.
14
15
16TERMINOLOGY
17¯¯¯¯¯¯¯¯¯¯¯
18
19* A "callback" is a reference to a piece of executable code (such as a
20 function) that is passed as an argument to another piece of code
21 (like another function).
22
23* A "context" is a structure that represents an object in which a trace
24 collection is opened.
25
26* An "iterator" is a structure that enables the user to traverse a trace.
27
28* A "trace handle" is a unique identifier representing a trace file.
29 It allows the user to manipulate a trace directly.
30
31
32
33USAGE
34¯¯¯¯¯¯
35
36Context:
37
38In order to use libbabeltrace to read a trace, the first step is to create a
39context structure and to add a trace to it. This is done using the
40bt_context_create() and bt_context_add_trace() functions. As long as this
41context structure is allocated and the trace is valid, the trace can be
42manipulated by the library.
43
44The context can be destroyed by calling one more bt_context_put() than
45bt_context_get(), functions which respectively decrement and increment the
46refcount of the context. These functions ensures that the context won't be
47destroyed when it is in use.
48
49Once a trace is added to the context, it can be read and seeked using iterators
50and callbacks.
51
52
53Iterator:
54
55An iterator can be created using the bt_iter_create() function. As of now, only
56ctf iterator are supported. These are used to traverse a ctf-formatted trace.
57Such iterators can be created with bt_ctf_iter_create().
58
59While creating an iterator, a begin and an end position may be specified. To do
60so, one or two struct bt_iter_pos must be passed. Such struct have two
61attributes: type and u. "type" is the seek type, can be either:
62 BT_SEEK_TIME
63 BT_SEEK_RESTORE
64 BT_SEEK_CUR
65 BT_SEEK_BEGIN
66 BT_SEEK_END
67and "u" is a union of the seek time (if using BT_SEEK_TIME) and the restore
68position (if using BT_SEEK_RESTORE).
69
70Once the iterator is created, various functions become available. We have
71bt_ctf_iter_read_event() which returns the ctf event of the trace where the
72iterator is set. There is also bt_ctf_iter_destroy() which frees the iterator.
73Note that only one iterator can be created in a context at the same time. If
74more than one iterator is being created for the same context, the second
75creation will return NULL. The previous iterator must be destroyed before
76creation of the new iterator. In the future, creation of multiples iterators
77will be allowed.
78
79Finally, we have the bt_ctf_get_iter() function which returns a struct bt_iter
80with which the iterator can be moved using one of these functions:
81 bt_iter_next(), moves the iterator to the next event
82 bt_iter_set_pos(), moves the iterator to the specified position
83
84To get the current position (struct bt_iter_pos) of the iterator, the function
85bt_iter_get_pos() must be used. To create an arbitrary position based on a
86specific time, bt_iter_create_time_pos() is the function to use. The
87bt_iter_pos structure returned by these two functions must be freed with
88bt_iter_free_pos() after use.
89
90
91CTF Event:
92
93A CTF event is obtained from an iterator via the bt_ctf_iter_read_event()
94function or via the call_data parameter of a callback. To read the data of a
95CTF event :
96 * bt_ctf_event_name() returns the name of the event;
97 * bt_ctf_get_timestamp() returns the timestamp of the event
98 offsetted with the system clock
99 source (in ns);
100 * bt_ctf_get_cycles() returns the timestamp of the event as
101 written in the packet (in cycles).
102
103The payload of an event is divided in various scopes depending on the type of
104information. There are six top-level scopes (defined in the bt_ctf_scope enum)
105which can be accessed by the bt_ctf_get_top_level_scope() function :
106 BT_TRACE_PACKET_HEADER = 0,
107 BT_STREAM_PACKET_CONTEXT = 1,
108 BT_STREAM_EVENT_HEADER = 2,
109 BT_STREAM_EVENT_CONTEXT = 3,
110 BT_EVENT_CONTEXT = 4,
111 BT_EVENT_FIELDS = 5.
112
113In order to access a field or a field list, the user needs to pass a scope as
114argument, this scope can be a top-level scope or a scope relative to an
115arbitrary field in the case of compound types (array, sequence, structure or
116variant)
117
118For more information on each scope, see the CTF specifications.
119
120The function to get a field list is the bt_ctf_get_field_list(). The function
121to get the definition of a specific field is bt_ctf_get_field().
122
123Once the field is obtained, we can obtain its name and type using the
124bt_ctf_field_name() and bt_ctf_field_type() functions respectively. The
125possible types are defined in the ctf_type_id enum:
126 CTF_TYPE_UNKNOWN = 0,
127 CTF_TYPE_INTEGER,
128 CTF_TYPE_FLOAT,
129 CTF_TYPE_ENUM,
130 CTF_TYPE_STRING,
131 CTF_TYPE_STRUCT,
132 CTF_TYPE_UNTAGGED_VARIANT,
133 CTF_TYPE_VARIANT,
134 CTF_TYPE_ARRAY,
135 CTF_TYPE_SEQUENCE,
136 NR_CTF_TYPES.
137
138Depending on the field type, we can get informations about the field with these
139functions:
140 * bt_ctf_get_index() return the element at the index
141 position of an array of a sequence;
142
143 * bt_ctf_get_array_len() return the length of an array;
144
145 * bt_ctf_get_int_signedness() return the signedness of an integer;
146
147 * bt_ctf_get_int_base() return the base of an integer;
148
149 * bt_ctf_get_int_byte_order() return the byte order of an integer;
150
151 * bt_ctf_get_int_len() return the size in bits of an integer;
152
153 * bt_ctf_get_encoding() return the encoding of an int or a
154 string defined in the
155 ctf_string_encoding enum:
156 CTF_STRING_NONE = 0,
157 CTF_STRING_UTF8,
158 CTF_STRING_ASCII,
159 CTF_STRING_UNKNOWN.
160
161These functions give access to the value associated with a field :
162 * bt_ctf_get_uint64();
163 * bt_ctf_get_int64();
164 * bt_ctf_get_char_array();
6ab22097
JD
165 * bt_ctf_get_string();
166 * bt_ctf_get_enum_int();
167 * bt_ctf_get_enum_str().
827b30c1
DS
168
169If the field does not exist or is not of the type requested, the value returned
170with these four functions is undefined. To check if an error occured, use the
171bt_ctf_field_get_error() function after accessing a field. If no error
172occured, the function will return 0.
173
174It is also possible to access the declaration fields, the same way as the
175definition ones. bt_ctf_get_event_decl_list() sets a list to an array of
176bt_ctf_event_decl pointers and bt_ctf_get_event_decl_fields() sets a list to an
177array of bt_ctf_field_decl pointers. From the first type, the name of the
178event can be obtained with bt_ctf_get_decl_event_name(). For the second type,
179the field decl name is obtained with bt_ctf_get_decl_field_name().
180
181The declaration functions allow the user to list the events, fields and
182contexts fields enabled in the trace once it is opened, whereas the definition
183functions apply on the current event being read.
184
185
186Callback:
187
188The iterator allow the user to read the trace, in order to access the events
189and fields, the user can either call the functions listed previously on each
190event, or register callbacks functions that are called when specific (or all)
191events are read.
192
193This is done with the bt_ctf_iter_add_callback() function. It requires a valid
194ctf iterator as the first argument. Here are all arguments:
195 iter: trace collection iterator (input)
196 event: event to target. 0 for all events.
197 private_data: private data pointer to pass to the callback
198 flags: specific flags controlling the behavior of this callback
199 (or'd).
200 callback: function pointer to call
201 depends: struct bt_dependency detailing the required computation
202 results. Ends with 0.
203 weak_depends: struct bt_dependency detailing the optional computation
204 results that can be optionally consumed by this
205 callback.
206 provides: struct bt_dependency detailing the computation results
207 provided by this callback.
208 Ends with 0.
209
210"depends", "weak_depends" and "provides" memory is handled by the babeltrace
211library after this call succeeds or fails. These objects can still be used by
212the caller until the babeltrace iterator is destroyed, but they belong to the
213babeltrace library.
214
215As of now the flags and dependencies are not used, the callbacks are
216processed in FIFO order.
217
218Note: once implemented, the dependency graph will be calculated when
219bt_ctf_iter_read_event() is executed after a bt_ctf_iter_add_callback(). It is
220valid to create/add callbacks/read/add more callbacks/read some more.
221
222The callback function passed to bt_ctf_iter_add_callback() must return a
223bt_cb_ret value:
224 BT_CB_OK = 0,
225 BT_CB_OK_STOP = 1,
226 BT_CB_ERROR_STOP = 2,
227 BT_CB_ERROR_CONTINUE = 3.
228
229
230Trace handle:
231
232When a trace is added to a context, bt_context_add_trace() returns a trace
233handle id. This id is associated with its corresponding trace handle. With
234that id, it is possible to manipulate directly the trace.
235
236 * bt_trace_handle_get_path()
237 -> returns the path of the trace handle (path to the trace).
238
239 * bt_trace_handle_get_timestamp_begin()
240 * bt_trace_handle_get_timestamp_end()
241 -> return the creation/destruction timestamps (in ns or cycles
242 depending on the type specified) of the buffers of a
243 trace.
244
245 * bt_ctf_event_get_handle_id()
246 -> returns the handle id associated with an event.
247
248
249For more information on CTF, see the CTF documentation.
This page took 0.03019 seconds and 4 git commands to generate.