<?xml version="1.0" encoding="latin1" ?>
<!DOCTYPE erlref SYSTEM "erlref.dtd">
<erlref>
<header>
<copyright>
<year>2001</year><year>2010</year>
<holder>Ericsson AB. All Rights Reserved.</holder>
</copyright>
<legalnotice>
The contents of this file are subject to the Erlang Public License,
Version 1.1, (the "License"); you may not use this file except in
compliance with the License. You should have received a copy of the
Erlang Public License along with this software. If not, it can be
retrieved online at http://www.erlang.org/.
Software distributed under the License is distributed on an "AS IS"
basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See
the License for the specific language governing rights and limitations
under the License.
</legalnotice>
<title>file_sorter</title>
<prepared>Hans Bolinder</prepared>
<responsible>nobody</responsible>
<docno></docno>
<approved>nobody</approved>
<checked>no</checked>
<date>2001-03-13</date>
<rev>PA1</rev>
<file>file_sorter.sgml</file>
</header>
<module>file_sorter</module>
<modulesummary>File Sorter</modulesummary>
<description>
<p>The functions of this module sort terms on files, merge already
sorted files, and check files for sortedness. Chunks containing
binary terms are read from a sequence of files, sorted
internally in memory and written on temporary files, which are
merged producing one sorted file as output. Merging is provided
as an optimization; it is faster when the files are already
sorted, but it always works to sort instead of merge.
</p>
<p>On a file, a term is represented by a header and a binary. Two
options define the format of terms on files:
</p>
<list type="bulleted">
<item><c>{header, HeaderLength}</c>. HeaderLength determines the
number of bytes preceding each binary and containing the
length of the binary in bytes. Default is 4. The order of the
header bytes is defined as follows: if <c>B</c> is a binary
containing a header only, the size <c>Size</c> of the binary
is calculated as
<c><![CDATA[<<Size:HeaderLength/unit:8>> = B]]></c>.
</item>
<item><c>{format, Format}</c>. The format determines the
function that is applied to binaries in order to create the
terms that will be sorted. The default value is
<c>binary_term</c>, which is equivalent to
<c>fun binary_to_term/1</c>. The value <c>binary</c> is
equivalent to <c>fun(X) -> X end</c>, which means that the
binaries will be sorted as they are. This is the fastest
format. If <c>Format</c> is <c>term</c>, <c>io:read/2</c> is
called to read terms. In that case only the default value of
the <c>header</c> option is allowed. The <c>format</c> option
also determines what is written to the sorted output file: if
<c>Format</c> is <c>term</c> then <c>io:format/3</c> is called
to write each term, otherwise the binary prefixed by a header
is written. Note that the binary written is the same binary
that was read; the results of applying the <c>Format</c>
function are thrown away as soon as the terms have been
sorted. Reading and writing terms using the <c>io</c> module
is very much slower than reading and writing binaries.
</item>
</list>
<p>Other options are:
</p>
<list type="bulleted">
<item><c>{order, Order}</c>. The default is to sort terms in
ascending order, but that can be changed by the value
<c>descending</c> or by giving an ordering function <c>Fun</c>.
An ordering function is antisymmetric, transitive and total.
<c>Fun(A, B)</c> should return <c>true</c> if <c>A</c>
comes before <c>B</c> in the ordering, <c>false</c> otherwise.
An example of a typical ordering function is less than or equal
to, <c>=</2</c>.
Using an ordering function will slow down the sort
considerably. The <c>keysort</c>, <c>keymerge</c> and
<c>keycheck</c> functions do not accept ordering functions.
</item>
<item><c>{unique, bool()}</c>. When sorting or merging files,
only the first of a sequence of terms that compare equal (<c>==</c>)
is output if this option is set to <c>true</c>. The default
value is <c>false</c> which implies that all terms that
compare equal are output. When checking files for
sortedness, a check that no pair of consecutive terms
compares equal is done if this option is set to <c>true</c>.
</item>
<item><c>{tmpdir, TempDirectory}</c>. The directory where
temporary files are put can be chosen explicitly. The
default, implied by the value <c>""</c>, is to put temporary
files on the same directory as the sorted output file. If
output is a function (see below), the directory returned by
<c>file:get_cwd()</c> is used instead. The names of
temporary files are derived from the Erlang nodename
(<c>node()</c>), the process identifier of the current Erlang
emulator (<c>os:getpid()</c>), and a timestamp
(<c>erlang:now()</c>); a typical name would be
<c>fs_mynode@myhost_1763_1043_337000_266005.17</c>, where
<c>17</c> is a sequence number. Existing files will be
overwritten. Temporary files are deleted unless some
uncaught EXIT signal occurs.
</item>
<item><c>{compressed, bool()}</c>. Temporary files and the
output file may be compressed. The default value
<c>false</c> implies that written files are not
compressed. Regardless of the value of the <c>compressed</c>
option, compressed files can always be read. Note that
reading and writing compressed files is significantly slower
than reading and writing uncompressed files.
</item>
<item><c>{size, Size}</c>. By default approximately 512*1024
bytes read from files are sorted internally. This option
should rarely be needed.
</item>
<item><c>{no_files, NoFiles}</c>. By default 16 files are
merged at a time. This option should rarely be needed.
</item>
</list>
<p>To summarize, here is the syntax of the options:</p>
<list type="bulleted">
<item>
<p><c>Options = [Option] | Option</c></p>
</item>
<item>
<p><c>Option = {header, HeaderLength} | {format, Format} | {order, Order} | {unique, bool()} | {tmpdir, TempDirectory} | {compressed, bool()} | {size, Size} | {no_files, NoFiles}</c></p>
</item>
<item>
<p><c>HeaderLength = int() > 0</c></p>
</item>
<item>
<p><c>Format = binary_term | term | binary | FormatFun</c></p>
</item>
<item>
<p><c>FormatFun = fun(Binary) -> Term</c></p>
</item>
<item>
<p><c>Order = ascending | descending | OrderFun</c></p>
</item>
<item>
<p><c>OrderFun = fun(Term, Term) -> bool()</c></p>
</item>
<item>
<p><c>TempDirectory = "" | file_name()</c></p>
</item>
<item>
<p><c>Size = int() >= 0</c></p>
</item>
<item>
<p><c>NoFiles = int() > 1</c></p>
</item>
</list>
<p>As an alternative to sorting files, a function of one argument
can be given as input. When called with the argument <c>read</c>
the function is assumed to return <c>end_of_input</c> or
<c>{end_of_input, Value}}</c> when there is no more input
(<c>Value</c> is explained below), or <c>{Objects, Fun}</c>,
where <c>Objects</c> is a list of binaries or terms depending on
the format and <c>Fun</c> is a new input function. Any other
value is immediately returned as value of the current call to
<c>sort</c> or <c>keysort</c>. Each input function will be
called exactly once, and should an error occur, the last
function is called with the argument <c>close</c>, the reply of
which is ignored.
</p>
<p>A function of one argument can be given as output. The results
of sorting or merging the input is collected in a non-empty
sequence of variable length lists of binaries or terms depending
on the format. The output function is called with one list at a
time, and is assumed to return a new output function. Any other
return value is immediately returned as value of the current
call to the sort or merge function. Each output function is
called exactly once. When some output function has been applied
to all of the results or an error occurs, the last function is
called with the argument <c>close</c>, and the reply is returned
as value of the current call to the sort or merge function. If a
function is given as input and the last input function returns
<c>{end_of_input, Value}</c>, the function given as output will
be called with the argument <c>{value, Value}</c>. This makes it
easy to initiate the sequence of output functions with a value
calculated by the input functions.
</p>
<p>As an example, consider sorting the terms on a disk log file.
A function that reads chunks from the disk log and returns a
list of binaries is used as input. The results are collected in
a list of terms.</p>
<pre>
sort(Log) ->
{ok, _} = disk_log:open([{name,Log}, {mode,read_only}]),
Input = input(Log, start),
Output = output([]),
Reply = file_sorter:sort(Input, Output, {format,term}),
ok = disk_log:close(Log),
Reply.
input(Log, Cont) ->
fun(close) ->
ok;
(read) ->
case disk_log:chunk(Log, Cont) of
{error, Reason} ->
{error, Reason};
{Cont2, Terms} ->
{Terms, input(Log, Cont2)};
{Cont2, Terms, _Badbytes} ->
{Terms, input(Log, Cont2)};
eof ->
end_of_input
end
end.
output(L) ->
fun(close) ->
lists:append(lists:reverse(L));
(Terms) ->
output([Terms | L])
end. </pre>
<p>Further examples of functions as input and output can be found
at the end of the <c>file_sorter</c> module; the <c>term</c>
format is implemented with functions.
</p>
<p>The possible values of <c>Reason</c> returned when an error
occurs are:</p>
<list type="bulleted">
<item>
<p><c>bad_object</c>, <c>{bad_object, FileName}</c>.
Applying the format function failed for some binary,
or the key(s) could not be extracted from some term.</p>
</item>
<item>
<p><c>{bad_term, FileName}</c>. <c>io:read/2</c> failed
to read some term.</p>
</item>
<item>
<p><c>{file_error, FileName, Reason2}</c>. See
<c>file(3)</c> for an explanation of <c>Reason2</c>.</p>
</item>
<item>
<p><c>{premature_eof, FileName}</c>. End-of-file was
encountered inside some binary term.</p>
</item>
</list>
<p><em>Types</em></p>
<pre>
Binary = binary()
FileName = file_name()
FileNames = [FileName]
ICommand = read | close
IReply = end_of_input | {end_of_input, Value} | {[Object], Infun} | InputReply
Infun = fun(ICommand) -> IReply
Input = FileNames | Infun
InputReply = Term
KeyPos = int() > 0 | [int() > 0]
OCommand = {value, Value} | [Object] | close
OReply = Outfun | OutputReply
Object = Term | Binary
Outfun = fun(OCommand) -> OReply
Output = FileName | Outfun
OutputReply = Term
Term = term()
Value = Term</pre>
</description>
<funcs>
<func>
<name>sort(FileName) -> Reply</name>
<name>sort(Input, Output) -> Reply</name>
<name>sort(Input, Output, Options) -> Reply</name>
<fsummary>Sort terms on files.</fsummary>
<type>
<v>Reply = ok | {error, Reason} | InputReply | OutputReply</v>
</type>
<desc>
<p>Sorts terms on files.
</p>
<p><c>sort(FileName)</c> is equivalent to
<c>sort([FileName], FileName)</c>.
</p>
<p><c>sort(Input, Output)</c> is equivalent to
<c>sort(Input, Output, [])</c>.
</p>
<p></p>
</desc>
</func>
<func>
<name>keysort(KeyPos, FileName) -> Reply</name>
<name>keysort(KeyPos, Input, Output) -> Reply</name>
<name>keysort(KeyPos, Input, Output, Options) -> Reply</name>
<fsummary>Sort terms on files by key.</fsummary>
<type>
<v>Reply = ok | {error, Reason} | InputReply | OutputReply</v>
</type>
<desc>
<p>Sorts tuples on files. The sort is performed on the
element(s) mentioned in <c>KeyPos</c>. If two tuples
compare equal (<c>==</c>) on one element, next element according to
<c>KeyPos</c> is compared. The sort is stable.
</p>
<p><c>keysort(N, FileName)</c> is equivalent to
<c>keysort(N, [FileName], FileName)</c>.
</p>
<p><c>keysort(N, Input, Output)</c> is equivalent to
<c>keysort(N, Input, Output, [])</c>.
</p>
<p></p>
</desc>
</func>
<func>
<name>merge(FileNames, Output) -> Reply</name>
<name>merge(FileNames, Output, Options) -> Reply</name>
<fsummary>Merge terms on files.</fsummary>
<type>
<v>Reply = ok | {error, Reason} | OutputReply</v>
</type>
<desc>
<p>Merges terms on files. Each input file is assumed to be
sorted.
</p>
<p><c>merge(FileNames, Output)</c> is equivalent to
<c>merge(FileNames, Output, [])</c>.
</p>
</desc>
</func>
<func>
<name>keymerge(KeyPos, FileNames, Output) -> Reply</name>
<name>keymerge(KeyPos, FileNames, Output, Options) -> Reply</name>
<fsummary>Merge terms on files by key.</fsummary>
<type>
<v>Reply = ok | {error, Reason} | OutputReply</v>
</type>
<desc>
<p>Merges tuples on files. Each input file is assumed to be
sorted on key(s).
</p>
<p><c>keymerge(KeyPos, FileNames, Output)</c> is equivalent
to <c>keymerge(KeyPos, FileNames, Output, [])</c>.
</p>
<p></p>
</desc>
</func>
<func>
<name>check(FileName) -> Reply</name>
<name>check(FileNames, Options) -> Reply</name>
<fsummary>Check whether terms on files are sorted.</fsummary>
<type>
<v>Reply = {ok, [Result]} | {error, Reason}</v>
<v>Result = {FileName, TermPosition, Term}</v>
<v>TermPosition = int() > 1</v>
</type>
<desc>
<p>Checks files for sortedness. If a file is not sorted, the
first out-of-order element is returned. The first term on a
file has position 1.
</p>
<p><c>check(FileName)</c> is equivalent to
<c>check([FileName], [])</c>.
</p>
</desc>
</func>
<func>
<name>keycheck(KeyPos, FileName) -> CheckReply</name>
<name>keycheck(KeyPos, FileNames, Options) -> Reply</name>
<fsummary>Check whether terms on files are sorted by key.</fsummary>
<type>
<v>Reply = {ok, [Result]} | {error, Reason}</v>
<v>Result = {FileName, TermPosition, Term}</v>
<v>TermPosition = int() > 1</v>
</type>
<desc>
<p>Checks files for sortedness. If a file is not sorted, the
first out-of-order element is returned. The first term on a
file has position 1.
</p>
<p><c>keycheck(KeyPos, FileName)</c> is equivalent
to <c>keycheck(KeyPos, [FileName], [])</c>.
</p>
<p></p>
</desc>
</func>
</funcs>
</erlref>
