wrap_log_reader
Module
Module Summary
Description
This module makes it possible to read internally formatted wrap disk logs, see disk_log(3). wrap_log_reader does not interfere with disk_log activities; there is however a bug in this version of the wrap_log_reader, see section Known Limitations.
A wrap disk log file consists of many files, called index files. A log file can be opened and closed. Also, a single index file can be opened separately. If a non-existent or non-internally formatted file is opened, an error message is returned. If the file is corrupt, no attempt is made to repair it, but an error message is returned.
If a log is configured to be distributed, it is possible that all items are not logged on all nodes. wrap_log_reader only reads the log on the called node; it is up to the user to be sure that all items are read.
Data Types
Continuation returned by open/1,2 or chunk/1,2.
Exports
Types
{Continuation2, Terms :: [term()]} |
{Continuation2,
Terms :: [term()],
Badbytes :: integer() >= 0} |
{Continuation2, eof} |
{error, Reason :: term()}
Enables to efficiently read the terms that are appended to a log. Minimises disk I/O by reading 64 kilobyte chunks from the file.
The first time chunk() is called, an initial continuation returned from open/1 or open/2 must be provided.
When chunk/3 is called, N controls the maximum number of terms that are read from the log in each chunk. Defaults to infinity, which means that all the terms contained in the 8K chunk are read. If less than N terms are returned, this does not necessarily mean that end of file is reached.
Returns a tuple {Continuation2, Terms}, where Terms is a list of terms found in the log. Continuation2 is yet another continuation that must be passed on to any subsequent calls to chunk(). With a series of calls to chunk(), it is then possible to extract all terms from a log.
Returns a tuple {Continuation2, Terms, Badbytes} if the log is opened in read only mode and the read chunk is corrupt. Badbytes indicates the number of non-Erlang terms found in the chunk. Notice that the log is not repaired.
Returns {Continuation2, eof} when the end of the log is reached, and {error, Reason} if an error occurs.
The returned continuation either is or is not valid in the next call to this function. This is because the log can wrap and delete the file into which the continuation points. To ensure this does not occur, the log can be blocked during the search.
Closes a log file properly.
Types
{ok, Continuation :: continuation()} |
{error, Reason :: tuple()}
Filename specifies the name of the file to be read.
N specifies the index of the file to be read. If N is omitted, the whole wrap log file is read; if it is specified, only the specified index file is read.
Returns {ok, Continuation} if the log/index file is opened successfully. Continuation is to be used when chunking or closing the file.
Returns {error, Reason} for all errors.
Known Limitations
This version of wrap_log_reader does not detect if disk_log wraps to a new index file between a call to wrap_log_reader:open() and the first call to wrap_log_reader:chunk(). If this occurs, the call to chunk() reads the last logged items in the log file, as the opened index file was truncated by disk_log.