Erlang logo
Reference Manual
Release Notes
PDF
Top

Kernel
Reference Manual
Version 2.16.1


Expand All
Contract All

Table of Contents

wrap_log_reader

MODULE

wrap_log_reader

MODULE SUMMARY

A function to read internally formatted wrap disk logs

DESCRIPTION

wrap_log_reader is a function to read internally formatted wrap disk logs, refer to disk_log(3). wrap_log_reader does not interfere with disk_log activities; there is however a known bug in this version of the wrap_log_reader, see chapter bugs below.

A wrap disk log file consists of several files, called index files. A log file can be opened and closed. It is also possible to open just one index file separately. If an non-existent or a non-internally formatted file is opened, an error message is returned. If the file is corrupt, no attempt to repair it will be done but an error message is returned.

If a log is configured to be distributed, there is a possibility that all items are not loggen on all nodes. wrap_log_reader does only read the log on the called node, it is entirely up to the user to be sure that all items are read.

DATA TYPES

continuation()

Continuation returned by open/1,2 or chunk/1,2.

EXPORTS

chunk(Continuation) -> chunk_ret()
chunk(Continuation, N) -> chunk_ret()

Types:

Continuation = continuation()
N = infinity | integer() >= 1
chunk_ret() = {Continuation2, Terms :: [term()]}
            | {Continuation2,
               Terms :: [term()],
               Badbytes :: integer() >= 0}
            | {Continuation2, eof}
            | {error, Reason :: term()}

This function makes it possible to efficiently read the terms which have been appended to a log. It minimises disk I/O by reading large 8K chunks from the file.

The first time chunk is called an initial continuation returned from the open/1, 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. Default is 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.

The chunk function returns a tuple {Continuation2, Terms}, where Terms is a list of terms found in the log. Continuation2 is yet another continuation which must be passed on into any subsequent calls to chunk. With a series of calls to chunk it is then possible to extract all terms from a log.

The chunk function 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. Note also that the log is not repaired.

chunk returns {Continuation2, eof} when the end of the log is reached, and {error, Reason} if an error occurs.

The returned continuation may or may not be valid in the next call to chunk. This is because the log may wrap and delete the file into which the continuation points. To make sure this does not happen, the log can be blocked during the search.

close(Continuation) -> ok | {error, Reason}

Types:

Continuation = continuation()
Reason = file:posix()

This function closes a log file properly.

open(Filename) -> open_ret()
open(Filename, N) -> open_ret()

Types:

Filename = string() | atom()
N = integer()
open_ret() = {ok, Continuation :: continuation()}
           | {error, Reason :: tuple()}

Filename specifies the name of the file which is to be read.

N specifies the index of the file which is to be read. If N is omitted the whole wrap log file will be read; if it is specified only the specified index file will be read.

The open function returns {ok, Continuation} if the log/index file was successfully opened. The Continuation is to be used when chunking or closing the file.

The function returns {error, Reason} for all errors.

Bugs

This version of the wrap_log_reader does not detect if the disk_log wraps to a new index file between a wrap_log_reader:open and the first wrap_log_reader:chunk. In this case the chuck will actually read the last logged items in the log file, because the opened index file was truncated by the disk_log.

See Also