File: ets.html

package info (click to toggle)
erlang-doc-html 1%3A10.b.1a-1
links: PTS
area: main
in suites: sarge
size: 22,488 kB
ctags: 9,933
sloc: erlang: 505; ansic: 323; perl: 61; sh: 45; makefile: 39
file content (2034 lines) | stat: -rw-r--r-- 65,623 bytes
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<!-- This document was generated using DocBuilder 3.3.2 -->
<HTML>
<HEAD>
  <TITLE>ets</TITLE>
  <SCRIPT type="text/javascript" src="../../../../doc/erlresolvelinks.js">
</SCRIPT>
  <STYLE TYPE="text/css">
<!--
    .REFBODY     { margin-left: 13mm }
    .REFTYPES    { margin-left: 8mm }
-->
  </STYLE>
</HEAD>
<BODY BGCOLOR="#FFFFFF" TEXT="#000000" LINK="#0000FF" VLINK="#FF00FF"
      ALINK="#FF0000">
<!-- refpage -->
<CENTER>
<A HREF="http://www.erlang.se">
  <IMG BORDER=0 ALT="[Ericsson AB]" SRC="min_head.gif">
</A>
<H1>ets</H1>
</CENTER>

<H3>MODULE</H3>
<DIV CLASS=REFBODY>
ets
</DIV>

<H3>MODULE SUMMARY</H3>
<DIV CLASS=REFBODY>
Built-In Term Storage
</DIV>

<H3>DESCRIPTION</H3>
<DIV CLASS=REFBODY>

<P>This module is an interface to the Erlang built-in term storage BIFs.
These provide the ability to store very large quantities of data in
an Erlang runtime system, and to have constant access time to the
data. (In the case of <CODE>ordered_set</CODE>, see below, access time is
proportional to the logarithm of the number of objects stored).
<P>Data is organized as a set of dynamic tables, which can store
tuples. Each table is created by a process. When the process
terminates, the table is automatically destroyed. Every table has
access rights set at creation.
<P>Tables are divided into four different types, <CODE>set</CODE>,
<CODE>ordered_set</CODE>, <CODE>bag</CODE> and <CODE>duplicate_bag</CODE>. 
A <CODE>set</CODE> or <CODE>ordered_set</CODE> table can only have one object
associated with each key. A <CODE>bag</CODE> or <CODE>duplicate_bag</CODE> can
have many objects associated with each key.
<P>The number of tables stored at one Erlang node is limited.
The current default limit is approximately 1400 tables. The upper
limit can be increased by setting the environment variable
<CODE>ERL_MAX_ETS_TABLES</CODE> before starting the Erlang runtime system
(i.e. with the <CODE>-env</CODE> option to <CODE>erl</CODE>/<CODE>werl</CODE>).
The actual limit may be slightly higher than the one specified, but
never lower.
<P>Note that there is no automatic garbage collection for tables.
Even if there are no references to a table from any process, it will
not automatically be destroyed unless the owner process terminates.
It can be destroyed explicitly by using <CODE>delete/1</CODE>.
<P>Some implementation details:
<P>
<UL>

<LI>
In the current implementation, every object insert and look-up 
        operation results in one copy of the object.
</LI>


<LI>
This module provides very limited support for concurrent updates.
        No locking is available, but the <CODE>safe_fixtable/2</CODE> function can
        be used to guarantee that a sequence of <CODE>first/1</CODE> and
        <CODE>next/2</CODE> calls will traverse the table without errors even if
        another process (or the same process) simultaneously deletes or
        inserts objects in the table.
</LI>


<LI>
<CODE>'$end_of_table'</CODE> should not be used as a key since this
        atom is used to mark the end of the table when using
        <CODE>first</CODE>/<CODE>next</CODE>.
</LI>


</UL>

<P>In general, the functions below will exit with reason <CODE>badarg</CODE> if
any argument is of the wrong format, or if the table identifier is
invalid.<BR>

The type <CODE>tid()</CODE> is used to denote a table identifier. Note that
the internal structure of this type is implementation-specific.
</DIV>

<H3>EXPORTS</H3>

<P><A NAME="all/0"><STRONG><CODE>all() -&#62; [Tab]</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Returns a list of all tables at the node. Named tables are
         given by their names, unnamed tables are given by their
         table identifiers.
</DIV>

<P><A NAME="delete/1"><STRONG><CODE>delete(Tab) -&#62; true</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Deletes the entire table <CODE>Tab</CODE>.
</DIV>

<P><A NAME="delete/2"><STRONG><CODE>delete(Tab, Key) -&#62; true</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Key = term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Deletes all objects with the key <CODE>Key</CODE> from the table
         <CODE>Tab</CODE>.
</DIV>

<P><A NAME="delete_all_objects/1"><STRONG><CODE>delete_all_objects(Tab) -&#62; true</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Delete all objects in the ETS table <CODE>Tab</CODE>. The
        deletion is atomic.
</DIV>

<P><A NAME="delete_object/2"><STRONG><CODE>delete_object(Tab,Object) -&#62; true</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Object = tuple()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Delete the exact object <CODE>Object</CODE> from the ETS table,
        leaving objects with the same key but other differences
        (useful for type <CODE>bag</CODE>).
</DIV>

<P><A NAME="file2tab/1"><STRONG><CODE>file2tab(Filename) -&#62; {ok,Tab} | {error,Reason}</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Filename = string() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Reason = term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Reads a file produced by <CODE>tab2file/2</CODE> and creates the
         corresponding table <CODE>Tab</CODE>.
</DIV>

<P><A NAME="first/1"><STRONG><CODE>first(Tab) -&#62; Key | '$end_of_table'</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Key = term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Returns the first key <CODE>Key</CODE> in the table <CODE>Tab</CODE>. 
         If the table is of the <CODE>ordered_set</CODE> type, the first key
         in Erlang term order will be returned. If the table is of any
         other type, the first key according to the table's internal
         order will be returned. If the table is empty,
         <CODE>'$end_of_table'</CODE> will be returned.
<P>Use <CODE>next/2</CODE> to find subsequent keys in the table.
</DIV>

<P><A NAME="fixtable/2"><STRONG><CODE>fixtable(Tab, true|false) -&#62; true | false</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>
<TABLE CELLPADDING=4>
  <TR>
    <TD VALIGN=TOP><IMG ALT="Warning!" SRC="warning.gif"></TD>
    <TD>

<P>The function is retained for backwards compatibility only.
         Use <CODE>safe_fixtable/2</CODE> instead.    </TD>
  </TR>
</TABLE>

<P>Fixes a table for safe traversal. The function is primarily used
         by the Mnesia DBMS to implement functions which allow write
         operations in a table, although the table is in the process of
         being copied to disk or to another node. It does not keep track
         of when and how tables are fixed.
</DIV>

<P><A NAME="foldl/3"><STRONG><CODE>foldl(Function, Acc0, Tab) -&#62; Acc1</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Function = fun(A, AccIn) -&#62; AccOut</CODE></STRONG><BR>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Acc0 = Acc1 = AccIn = AccOut = term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P><CODE>Acc0</CODE> is returned if the table is empty.
        This function is similar to <CODE>lists:foldl/3</CODE>. The order in
        which the elements of the table are traversed is unspecified,
        except for tables of type <CODE>ordered_set</CODE>, for which they
        are traversed first to last. Since <CODE>safe_fixtable/2</CODE> is
        called, the table must be public or owned by the calling process.

</DIV>

<P><A NAME="foldr/3"><STRONG><CODE>foldr(Function, Acc0, Tab) -&#62; Acc1</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Function = fun(A, AccIn) -&#62; AccOut</CODE></STRONG><BR>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Acc0 = Acc1 = AccIn = AccOut = term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P><CODE>Acc0</CODE> is returned if the table is empty.
        This function is similar to <CODE>lists:foldr/3</CODE>. The order in
        which the elements of the table are traversed is unspecified,
        except for tables of type <CODE>ordered_set</CODE>, for which they
        are traversed last to first. Since <CODE>safe_fixtable/2</CODE> is
        called, the table must be public or owned by the calling process.

</DIV>

<P><A NAME="from_dets/2"><STRONG><CODE>from_dets(Tab, DetsTab) -&#62; Tab</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>DetsTab = atom()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>     Fills an already created ETS table with the objects in the
        already opened Dets table named <CODE>DetsTab</CODE>. The ETS table
        is emptied before the objects are inserted.
        
</DIV>

<P><A NAME="fun2ms/1"><STRONG><CODE>fun2ms(LiteralFun) -&#62; MatchSpec</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>LiteralFun = fun() literal</CODE></STRONG><BR>
<STRONG><CODE>MatchSpec = term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>      Pseudo function that by means of a <CODE>parse_transform</CODE>
         translates the <STRONG>literal</STRONG> <CODE>fun()</CODE> typed as parameter in
         the function call to a match specification as described in
         the <CODE>match_spec</CODE> manual of <CODE>ERTS</CODE> users guide.
         (with literal I mean that the <CODE>fun()</CODE> needs to
         textually be written as the parameter of the function, it
         cannot be held in a variable which in turn is passed to the 
         function). 
        
<P>      The parse transform is implemented in the module
         <CODE>ms_transform</CODE> and the source <STRONG>must</STRONG> include the
         file <CODE>ms_transform.hrl</CODE> in <CODE>stdlib</CODE> for this
         pseudo function to work. Failing to include the hrl file in
         the source will result in a runtime error, not a compile
         time ditto. The include file is easiest included by adding
         the line
         <CODE>-include_lib(&#34;stdlib/include/ms_transform.hrl&#34;).</CODE> to
         the source file.
        
<P>      The <CODE>fun()</CODE> is very restricted, it can take only a
         single parameter (the object to match), a sole variable or a
         tuple. It needs to use the <CODE>is_</CODE>XXX guard tests and one
         cannot use language constructs that have no representation
         in a match_spec (like <CODE>if</CODE>, <CODE>case</CODE>,
         <CODE>receive</CODE> etc). The return value from the fun will be
         the return value of the resulting match_spec.
        
<P>      Example:
        
<PRE>
2&#62; ets:fun2ms(fun({M,N}) when N &#62; 3 -&#62; M end).
[{{'$1','$2'},[{'&#62;','$2',3}],['$1']}]
        
</PRE>

<P>      Variables from the environment can be imported, so that this
         works:
        
<PRE>
2&#62; X=3.                                       
3
3&#62; ets:fun2ms(fun({M,N}) when N &#62; X -&#62; M end).
[{{'$1','$2'},[{'&#62;','$2',{const,3}}],['$1']}]
        
</PRE>

<P>      The imported variables will be replaced by match_spec
         <CODE>const</CODE> expressions, which is consistent with the
         static scoping for erlang <CODE>fun()</CODE>'s. Local or global
         function calls can not be in the guard or body of the fun
         however. Calls to builtin match_spec functions of course is
         allowed:
        
<PRE>
4&#62; ets:fun2ms(fun({M,N}) when N &#62; X, is_atomm(M) -&#62; M end). 
Error: fun containing local erlang function calls
('is_atomm' called in guard) cannot be translated into match_spec
{error,transform_error}
5&#62; ets:fun2ms(fun({M,N}) when N &#62; X, is_atom(M) -&#62; M end). 
[{{'$1','$2'},[{'&#62;','$2',{const,3}},{is_atom,'$1'}],['$1']}]
        
</PRE>

<P> As you can see by the example, the function can be called from
the shell too. The <CODE>fun()</CODE> needs to be literally in the
call when used from the shell as well. Other means than the 
         parse_transform are used in the shell case, but more or less
the same restrictions apply (the exception being records,
as they are not handled by the shell).
        
<P>
<TABLE CELLPADDING=4>
  <TR>
    <TD VALIGN=TOP><IMG ALT="Warning!" SRC="warning.gif"></TD>
    <TD>

<P>      If the parse_transform is not applied to a module which calls this
         pseudo function, the call will fail in runtime (with a 
         <CODE>badarg</CODE>). The module <CODE>ets</CODE> actually exports a
         function with this name, but it should never really be called
         except for when using the function in the shell. If the
         <CODE>parse_transform</CODE> is properly applied by including
         the <CODE>ms_transform.hrl</CODE> header file, compiled code
         will never call the function, but the function call is
         replaced by a literal match_spec.
             </TD>
  </TR>
</TABLE>

<P>      More information is provided by the <CODE>ms_transform</CODE>
         manual page in <CODE>stdlib</CODE>.
        
</DIV>

<P><A NAME="i/0"><STRONG><CODE>i() -&#62; void()</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY>

<P>Displays information about all ETS tables on tty.
</DIV>

<P><A NAME="i/1"><STRONG><CODE>i(Tab) -&#62; void()</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Browses the table <CODE>Tab</CODE> on tty.
</DIV>

<P><A NAME="info/1"><STRONG><CODE>info(Tab) -&#62; [{Item,Value}] | undefined</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Item, Value - see below</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Returns information about the table <CODE>Tab</CODE> as a list of
         <CODE>{Item,Value}</CODE> tuples:
<P>
<UL>

<LI>
         <CODE>Item=memory, Value=int()</CODE><BR>

         The number of words allocated to the table.
         
</LI>


<LI>
         <CODE>Item=owner, Value=pid()</CODE><BR>

         The pid of the owner of the table.
         
</LI>


<LI>
         <CODE>Item=name, Value=atom()</CODE><BR>

         The name of the table.
         
</LI>


<LI>
         <CODE>Item=size, Value=int()</CODE><BR>

         The number of objects inserted in the table.
         
</LI>


<LI>
         <CODE>Item=node, Value=atom()</CODE><BR>

         The node where the table is stored. This field is no longer
         meaningful as tables cannot be accessed from other nodes.
         
</LI>


<LI>
         <CODE>Item=named_table, Value=true|false</CODE><BR>

         Indicates if the table is named or not.
         
</LI>


<LI>
         <CODE>Item=type, Value=set|ordered_set|bag|duplicate_bag</CODE><BR>

         The table type.
         
</LI>


<LI>
         <CODE>Item=keypos, Value=int()</CODE><BR>

         The key position.
         
</LI>


<LI>
         <CODE>Item=protection, Value=public|protected|private</CODE><BR>

         The table access rights.
         
</LI>


</UL>

</DIV>

<P><A NAME="info/2"><STRONG><CODE>info(Tab, Item) -&#62; Value | undefined</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Item, Value - see below</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Returns the information associated with <CODE>Item</CODE> for
         the table <CODE>Tab</CODE>. In addition to the <CODE>{Item,Value}</CODE>
         pairs defined for <CODE>info/1</CODE>, the following items are
         allowed:
<P>
<UL>

<LI>
         <CODE>Item=fixed, Value=true|false</CODE><BR>

         Indicates if the table is fixed by any process or not.
         
</LI>


<LI>
         <CODE>Item=safe_fixed, Value={FirstFixed,Info}|false
         </CODE><BR>

         If the table has been fixed using <CODE>safe_fixtable/2</CODE>,
         the call returns a tuple where <CODE>FirstFixed</CODE> is the time
         when the table was first fixed by a process, which may or may
         not be one of the processes it is fixed by right now.<BR>

         <CODE>Info</CODE> is a possibly empty lists of tuples
         <CODE>{Pid,RefCount}</CODE>, one tuple for every process the table is
         fixed by right now. <CODE>RefCount</CODE> is the value of the
         reference counter, keeping track of how many times the table
         has been fixed by the process.<BR>

         If the table never has been fixed, the call returns
         <CODE>false</CODE>.<BR>

         
</LI>


</UL>

</DIV>

<P><A NAME="init_table/2"><STRONG><CODE>init_table(Name, InitFun) -&#62; true</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Name = atom()</CODE></STRONG><BR>
<STRONG><CODE>InitFun = fun(Arg) -&#62; Res</CODE></STRONG><BR>
<STRONG><CODE>Arg = read | close</CODE></STRONG><BR>
<STRONG><CODE>Res = end_of_input | {[object()], InitFun} | term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Replaces the existing objects of the table <CODE>Tab</CODE> with
        objects created by calling the input function <CODE>InitFun</CODE>,
        see below. This function is provided for compatibility with
        the Dets module, it's not more efficient than filling a table
        by using <CODE>ets:insert/2</CODE>.

        
<P>When called with the argument <CODE>read</CODE> the function
<CODE>InitFun</CODE> is assumed to return <CODE>end_of_input</CODE> when
there is no more input, or <CODE>{Objects, Fun}</CODE>, where
<CODE>Objects</CODE> is a list of objects and <CODE>Fun</CODE> is a new
input function. Any other value Value is returned as an error
<CODE>{error, {init_fun, Value}}</CODE>. Each input function will be
called exactly once, and should an error occur, the last
function is called with the argument <CODE>close</CODE>, the reply
of which is ignored.
<P>If the type of the table is <CODE>set</CODE> and there is more
        than one object with a given key, one of the objects is
        chosen. This is not necessarily the last object with the given
        key in the sequence of objects returned by the input
        functions. This holds also for duplicated
        objects stored in tables of type <CODE>duplicate_bag</CODE>.
</DIV>

<P><A NAME="insert/2"><STRONG><CODE>insert(Tab, ObjectOrObjects) -&#62; true</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>ObjectOrObjects = tuple() | [tuple()]</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Inserts the object or all of the objects in the list
         <CODE>ObjectOrObjects</CODE> into the table <CODE>Tab</CODE>. If there
         already exists an object with the same key as one of the
         objects, and the table is a <CODE>set</CODE> or <CODE>ordered_set</CODE>
         table, the old object will be replaced. If the list contains
         more than one object with the same key and the table is a
         <CODE>set/ordered_set</CODE>, one will be inserted, which one is
         not defined.
</DIV>

<P><A NAME="insert_new/2"><STRONG><CODE>insert_new(Tab, ObjectOrObjects) -&#62; bool()</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>ObjectOrObjects = tuple() | [tuple()]</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>This function works exactly like insert/2, with the
        exception that instead of overwriting objects with the same
        key (in the case of sets or ordered_sets) or adding more
        objects with keys already existing in the table (in the case
        of bags and duplicate_bags), it simply returns <CODE>false</CODE>. If
        <CODE>ObjectOrObjects</CODE> is a list, the function checks
        <STRONG>every</STRONG> key prior to inserting anything. Nothing will be
        inserted if not <STRONG>all</STRONG> keys present in the list are
        absent from the table.

</DIV>

<P><A NAME="is_compiled_ms/1"><STRONG><CODE>is_compiled_ms(Term) -&#62; bool()</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Term = term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>This function is used to check if a term is a valid
         compiled match specification. The compiled match_spec is an
         opaque datatype which can <STRONG>not</STRONG> be sent between erlang
         nodes nor be stored on disk. Any attempt to create an external
         representation of a compiled match specification will result
         in an empty binary (<CODE>&#60;&#60;&#62;&#62;</CODE>). As an example,
         the following expression:
<PRE>
          ets:is_compiled_ms(ets:match_spec_compile([{'_',[],[true]}])).
        
</PRE>

<P>will yield <CODE>true</CODE>, while the following expressions:
<PRE>
          MS = ets:match_spec_compile([{'_',[],[true]}]),
          Broken = binary_to_term(term_to_binary(MS)),
          ets:is_compiled_ms(Broken).
        
</PRE>

<P>will yield false, as the variable <CODE>Broken</CODE> will contain a
         compiled match specification that has passed through external
         representation.
<P>
<TABLE CELLPADDING=4>
  <TR>
    <TD VALIGN=TOP><IMG ALT="Note!" SRC="note.gif"></TD>
    <TD>

<P>The fact that compiled match_specifications has no external
         representation is for performance reasons. It may be subject
         to change in future releases, while this interface will
         still remain for backward compatibility reasons.    </TD>
  </TR>
</TABLE>

</DIV>

<P><A NAME="last/1"><STRONG><CODE>last(Tab) -&#62; Key | '$end_of_table'</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Key = term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Returns the last key <CODE>Key</CODE> according to Erlang term order
         in the table <CODE>Tab</CODE> of the <CODE>ordered_set</CODE> type. If
         the table is of any other type, the function is synonymous to
         <CODE>first/2</CODE>. If the table is empty, <CODE>'$end_of_table'</CODE> is
         returned.
<P>Use <CODE>prev/2</CODE> to find preceding keys in the table.
</DIV>

<P><A NAME="lookup/2"><STRONG><CODE>lookup(Tab, Key) -&#62; [Object]</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Key = term()</CODE></STRONG><BR>
<STRONG><CODE>Object = tuple()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Returns a list of all objects with the key <CODE>Key</CODE> in
         the table <CODE>Tab</CODE>.
<P>If the table is of type <CODE>set</CODE> or <CODE>ordered_set</CODE>,
         the function returns either the empty list or a list with one
         element, as there cannot be more than one object with the same
         key. If the table is of type <CODE>bag</CODE> or <CODE>duplicate_bag</CODE>,
         the function returns a list of arbitrary length.
<P>Note that the time order of object insertions is preserved;
         The first object inserted with the given key will be first
         in the resulting list, and so on.
<P>Insert and look-up times in tables of type <CODE>set</CODE>, <CODE>bag</CODE>
         and <CODE>duplicate_bag</CODE> are constant, regardless of the size of
         the table. For the <CODE>ordered_set</CODE> data-type, time is
         proportional to the (binary) logarithm of the number of objects.
        
</DIV>

<P><A NAME="lookup_element/3"><STRONG><CODE>lookup_element(Tab, Key, Pos) -&#62; Elem</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Key = term()</CODE></STRONG><BR>
<STRONG><CODE>Pos = int()</CODE></STRONG><BR>
<STRONG><CODE>Elem = term() | [term()]</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>If the table <CODE>Tab</CODE> is of type <CODE>set</CODE> or
         <CODE>ordered_set</CODE>, the function returns the <CODE>Pos</CODE>:th
         element of the object with the key <CODE>Key</CODE>.
<P>If the table is of type <CODE>bag</CODE> or <CODE>duplicate_bag</CODE>,
         the functions returns a list with the <CODE>Pos</CODE>:th element of
         every object with the key <CODE>Key</CODE>.
<P>If no object with the key <CODE>Key</CODE> exists, the function will
         exit with reason <CODE>badarg</CODE>.
</DIV>

<P><A NAME="match/2"><STRONG><CODE>match(Tab, Pattern) -&#62; [Match]</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Pattern = tuple()</CODE></STRONG><BR>
<STRONG><CODE>Match = [term()]</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Matches the objects in the table <CODE>Tab</CODE> against the pattern
         <CODE>Pattern</CODE>.
<P>A pattern is a term that may contain:
<P>
<UL>

<LI>
bound parts (Erlang terms),
</LI>


<LI>
<CODE>'_'</CODE> which matches any Erlang term, and
</LI>


<LI>
pattern variables: <CODE>'$N'</CODE> where <CODE>N</CODE>=0,1,...
</LI>


</UL>

<P>The function returns a list with one element for each matching
         object, where each element is an ordered list of pattern variable
         bindings. An example:
<PRE>
&#62; <STRONG>ets:match(T, '$1').</STRONG> % Matches every object in the table
[{rufsen,dog,7},{brunte,horse,5},{ludde,dog,5}]
&#62; <STRONG>ets:match(T, {'_',dog,'$1'}).</STRONG>
[[7],[5]]
&#62; <STRONG>ets:match(T, {'_',cow,'$1'}).</STRONG>
[]
        
</PRE>

<P>If the key is specified in the pattern, the match is very
         efficient. If the key is not specified, i.e. if it is a variable
         or an underscore, the entire table must be searched. The search
         time can be substantial if the table is very large.
<P>On tables of the <CODE>ordered_set</CODE> type, the result is in the
         same order as in a <CODE>first/next</CODE> traversal.
</DIV>

<P><A NAME="match/3"><STRONG><CODE>match(Tab, Pattern, Limit) -&#62; {[Match],Continuation} | '$end_of_table'</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Pattern = tuple()</CODE></STRONG><BR>
<STRONG><CODE>Match = [term()]</CODE></STRONG><BR>
<STRONG><CODE>Continuation = term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Works like <CODE>ets:match/2</CODE> but only returns a limited
        (<CODE>Limit</CODE>) number of matching objects. The
        <CODE>Continuation</CODE> term can then be used in subsequent calls
        to <CODE>ets:match/1</CODE> to get the next chunk of matching
        objects. This is a space efficient way to work on objects in a
        table which is still faster than traversing the table object
        by object using <CODE>ets:first/1</CODE> and <CODE>ets:next/1</CODE>.
<P><CODE>'$end_of_table'</CODE> is returned if the table is empty.
</DIV>

<P><A NAME="match/1"><STRONG><CODE>match(Continuation) -&#62; {[Match],Continuation} | '$end_of_table'</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Match = [term()]</CODE></STRONG><BR>
<STRONG><CODE>Continuation = term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Continues a match started with <CODE>ets:match/3</CODE>. The next
         chunk of the size given in the initial <CODE>ets:match/3</CODE>
         call is returned together with a new <CODE>Continuation</CODE>
         that can be used in subsequent calls to this function.
<P>'$end_of_table' is returned when there are no more
         objects in the table.
</DIV>

<P><A NAME="match_delete/2"><STRONG><CODE>match_delete(Tab, Pattern) -&#62; true</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Pattern = tuple()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Deletes all objects which match the pattern <CODE>Pattern</CODE> from
         the table <CODE>Tab</CODE>. See <CODE>match/2</CODE> for a description of
         patterns.
</DIV>

<P><A NAME="match_object/2"><STRONG><CODE>match_object(Tab, Pattern) -&#62; [Object]</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Pattern = Object = tuple()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Matches the objects in the table <CODE>Tab</CODE> against the pattern
         <CODE>Pattern</CODE>. See <CODE>match/2</CODE> for a description of patterns.
         The function returns a list of all objects which match the pattern.
        
<P>If the key is specified in the pattern, the match is very
         efficient. If the key is not specified, i.e. if it is a variable
         or an underscore, the entire table must be searched. The search
         time can be substantial if the table is very large.
<P>On tables of the <CODE>ordered_set</CODE> type, the result is in the
         same order as in a <CODE>first/next</CODE> traversal.
</DIV>

<P><A NAME="match_object/3"><STRONG><CODE>match_object(Tab, Pattern, Limit) -&#62; {[Match],Continuation} | '$end_of_table'</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Pattern = tuple()</CODE></STRONG><BR>
<STRONG><CODE>Match = [term()]</CODE></STRONG><BR>
<STRONG><CODE>Continuation = term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Works like <CODE>ets:match_object/2</CODE> but only returns a limited
        (<CODE>Limit</CODE>) number of matching objects. The
        <CODE>Continuation</CODE> term can then be used in subsequent calls
        to <CODE>ets:match_object/1</CODE> to get the next chunk of matching
        objects. This is a space efficient way to work on objects in a
        table which is still faster than traversing the table object
        by object using <CODE>ets:first/1</CODE> and <CODE>ets:next/1</CODE>.
<P><CODE>'$end_of_table'</CODE> is returned if the table is empty.
</DIV>

<P><A NAME="match_object/1"><STRONG><CODE>match_object(Continuation) -&#62; {[Match],Continuation} | '$end_of_table'</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Match = [term()]</CODE></STRONG><BR>
<STRONG><CODE>Continuation = term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Continues a match started with
         <CODE>ets:match_object/3</CODE>. The next
         chunk of the size given in the initial <CODE>ets:match_object/3</CODE>
         call is returned together with a new <CODE>Continuation</CODE>
         that can be used in subsequent calls to this function.
<P>'$end_of_table' is returned when there are no more
         objects in the table.
</DIV>

<P><A NAME="match_spec_compile/1"><STRONG><CODE>match_spec_compile(MatchSpec) -&#62; CompiledMatchSpec</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>MatchSpec = term()</CODE></STRONG><BR>
<STRONG><CODE>CompiledMatchSpec = term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>This function transforms a match specification into a
        internal representation that can be used in subsequent calls
        to <CODE>ets:match_spec_run/2</CODE>. The internal representation is
        opaque and can not be converted to external term format and
        then back again without losing its properties (meaning it can
        not be sent to a process on another node and still remain a
        valid compiled match specification, nor can it be stored on
        disk). The validity of a compiled match specification can be
        checked using <CODE>ets:is_compiled_ms/1</CODE>.
<P>If the term <CODE>MatchSpec</CODE> can not be compiled (does not
represent a valid match specification), a <CODE>badarg</CODE> fault
is thrown.
<P>
<TABLE CELLPADDING=4>
  <TR>
    <TD VALIGN=TOP><IMG ALT="Note!" SRC="note.gif"></TD>
    <TD>

<P>This function has limited use in normal code, it's used by Dets
        to perform the <CODE>dets:select</CODE> operations.    </TD>
  </TR>
</TABLE>

</DIV>

<P><A NAME="match_spec_run/2"><STRONG><CODE>match_spec_run(List,CompiledMatchSpec) -&#62; list()</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>List = [ tuple() ]</CODE></STRONG><BR>
<STRONG><CODE>CompiledMatchSpec = term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>This function executes the matching specified in a
compiled match specification on a list of tuples. The
<CODE>CompiledMatchSpec</CODE> term should be the result of a
call to <CODE>ets:match_spec_compile/1</CODE> and is hence the
internal representation of the match_spec one wants to use.
<P>The matching will be executed on each element in <CODE>List</CODE>
and the function returns a <CODE>list()</CODE> containing all
results. If an element in <CODE>List</CODE> does not match, nothing
is returned for that element. The length of the result list is
therefore equal or less than the the length of the parameter
<CODE>List</CODE>. The two calls in the following example will give
the same result (but certainly not the same execution
time...):
<PRE>
       Table = ets:new...
       MatchSpec = ....
       % The following call...
       ets:match_spec_run(ets:tab2list(Table),
                          ets:match_spec_compile(MatchSpec)),
       % ...will give the same result as the more common (and more efficient)
       ets:select(Table,MatchSpec),
       
</PRE>

<P>
<TABLE CELLPADDING=4>
  <TR>
    <TD VALIGN=TOP><IMG ALT="Note!" SRC="note.gif"></TD>
    <TD>

<P>This function has limited use in normal code, it's used by Dets
to perform the <CODE>dets:select</CODE> operations and by Mnesia
during transactions.    </TD>
  </TR>
</TABLE>

</DIV>

<P><A NAME="member/2"><STRONG><CODE>member(Tab, Key) -&#62; true | false</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Key = term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Works like lookup/2, but does not return the objects. The
        function returns <CODE>true</CODE> if one or more elements in the
        table has the key <CODE>Key</CODE>, <CODE>false</CODE> otherwise.
</DIV>

<P><A NAME="new/2"><STRONG><CODE>new(Name, Options) -&#62; tid()</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Name = atom()</CODE></STRONG><BR>
<STRONG><CODE>Options = [Option]</CODE></STRONG><BR>
<STRONG><CODE>Option = Type | Access | named_table | {keypos,Pos}</CODE></STRONG><BR>
<STRONG><CODE>Type = set | ordered_set | bag | duplicate_bag</CODE></STRONG><BR>
<STRONG><CODE>Access = public | protected | private</CODE></STRONG><BR>
<STRONG><CODE>Pos = int()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Creates a new table and returns a table identifier which can be
         used in subsequent operations. The table identifier can be sent to
         other processes so that a table can be shared between different
         processes within a node.
        
<P>The parameter <CODE>Options</CODE> is a list of atoms which specifies
         table type, access rights, key position and if the table is named
         or not. If one or more options are left out, the default values
         are used. This means that not specifying any options (<CODE>[]</CODE>) is
         the same as specifying <CODE>[set,protected,{keypos,1}]</CODE>.
<P>
<UL>

<LI>
<CODE>set</CODE>
         The table is a <CODE>set</CODE> table - one key, one object, no
         order among objects. This is the default table type.<BR>

         
</LI>


<LI>
<CODE>ordered_set</CODE>
         The table is a <CODE>ordered_set</CODE> table - one key, one
         object, ordered in Erlang term order, which is the order
         implied by the &#60; and &#62; operators. Tables of this type
         have a somewhat different behavior in some situations
         than tables of the other types.<BR>

         
</LI>


<LI>
<CODE>bag</CODE>
         The table is a <CODE>bag</CODE> table which can have many objects,
         but only one instance of each object, per key.<BR>

         
</LI>


<LI>
<CODE>duplicate_bag</CODE>
         The table is a <CODE>duplicate_bag</CODE> table which can have many
         objects, including multiple copies of the same object, per
         key.<BR>

         
</LI>


<LI>
<CODE>public</CODE>
         Any process may read or write to the table.<BR>

         
</LI>


<LI>
<CODE>protected</CODE>
         The owner process can read and write to the table. Other
         processes can only read the table. This is the default
         setting for the access rights.<BR>

         
</LI>


<LI>
<CODE>private</CODE>
         Only the owner process can read or write to the table.<BR>

         
</LI>


<LI>
<CODE>named_table</CODE>
         If this option is present, the name <CODE>Name</CODE> is associated
         with the table identifier. The name can then be used
         instead of the table identifier in subsequent operations.<BR>

         
</LI>


<LI>
<CODE>{keypos,Pos}</CODE>
         Specfies which element in the stored tuples should be used as
         key. By default, it is the first element, i.e. <CODE>Pos=1</CODE>.
         However, this is not always appropriate. In particular,
         we do not want the first element to be the key if we want to
         store Erlang records in a table.<BR>

         Note that any tuple stored in the table must have at least
         <CODE>Pos</CODE> number of elements.<BR>

         
</LI>


</UL>

</DIV>

<P><A NAME="next/2"><STRONG><CODE>next(Tab, Key1) -&#62; Key2 | '$end_of_table'</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Key1 = Key2 = term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Returns the next key <CODE>Key2</CODE>, following the key <CODE>Key1</CODE>
         in the table <CODE>Tab</CODE>. If the table is of the <CODE>ordered_set</CODE>
         type, the next key in Erlang term order is returned. If the table
         is of any other type, the next key according to the table's
         internal order is returned. If there is no next key,
         <CODE>'$end_of_table'</CODE> is returned.
        
<P>Use <CODE>first/1</CODE> to find the first key in the table.
<P>Unless a table of type <CODE>set</CODE>, <CODE>bag</CODE> or
         <CODE>duplicate_bag</CODE> is protected using <CODE>safe_fixtable/2</CODE>,
         see below, a traversal may fail if concurrent updates are made
         to the table.
         If the table is of type <CODE>ordered_set</CODE>, the function returns
         the next key in order, even if the object does no longer exist.
</DIV>

<P><A NAME="prev/2"><STRONG><CODE>prev(Tab, Key1) -&#62; Key2 | '$end_of_table'</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Key1 = Key2 = term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Returns the previous key <CODE>Key2</CODE>, preceding the key
         <CODE>Key1</CODE> according the Erlang term order in the table
         <CODE>Tab</CODE> of the <CODE>ordered_set</CODE> type. If the table is of
         any other type, the function is synonymous to <CODE>next/2</CODE>.
         If there is no previous key, <CODE>'$end_of_table'</CODE> is returned.
        
<P>Use <CODE>last/1</CODE> to find the last key in the table.
</DIV>

<P><A NAME="rename/2"><STRONG><CODE>rename(Tab, Name) -&#62; Name</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = Name = atom()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Renames the named table <CODE>Tab</CODE> to the new name <CODE>Name</CODE>.
         Afterwards, the old name can not be used to access the table.
         Renaming an unnamed table has no effect.
</DIV>

<P><A NAME="repair_continuation/2"><STRONG><CODE>repair_continuation(Continuation, MatchSpec) -&#62; Continuation</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Continuation = term()</CODE></STRONG><BR>
<STRONG><CODE>MatchSpec = term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>This function can be used to restore an opaque continuation
returned by ets:select/3 or ets:select/1 if the continuation has
passed through external term format (been sent between nodes or
stored on disk).
<P>The reason for this function is that continuation terms
contain compiled match specifications and therefore will be
invalidated if converted to external term format. Given that the
original match specification is kept intact, the continuation
can be restored, meaning it can once again be used in subsequent
ets:select/1 calls even though it has been stored on disk or
on another node.
<P>As an example, the following seqence of calls will fail:
<PRE>
      T=ets:new(x,[]),
      ...
      {_,C} = ets:select(T,ets:fun2ms(fun({N,_}=A) 
                                        when (N rem 10) =:= 0 -&#62; 
                                          A 
                                      end),10),
      Broken = binary_to_term(term_to_binary(C)),
      ets:select(Broken).
      
</PRE>

<P>...while the following sequence will work:
<PRE>
      T=ets:new(x,[]),
      ...
      MS = ets:fun2ms(fun({N,_}=A) 
                        when (N rem 10) =:= 0 -&#62; 
                          A 
                       end),
      {_,C} = ets:select(T,MS,10),
      Broken = binary_to_term(term_to_binary(C)),
      ets:select(ets:repair_continuation(Broken,MS)).
      
</PRE>

<P>...as the call to <CODE>ets:repair_continuation/2</CODE> will
reestablish the (deliberately) invalidated continuation
<CODE>Broken</CODE>.
<P>
<TABLE CELLPADDING=4>
  <TR>
    <TD VALIGN=TOP><IMG ALT="Note!" SRC="note.gif"></TD>
    <TD>

<P>This function is very rarely needed in application code. It
is used by Mnesia to implement distributed <CODE>select/3</CODE> and
<CODE>select/1</CODE> sequences. A normal application would either use
Mnesia or keep the continuation from beeing converted to
external format.
<P>The reason for not having an external representation of compiled 
match specifications is performance. It may be subject to change in 
future releases, while this interface will remain for backward 
compatibility.    </TD>
  </TR>
</TABLE>

</DIV>

<P><A NAME="safe_fixtable/2"><STRONG><CODE>safe_fixtable(Tab, true|false) -&#62; true</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Fixes a table of the <CODE>set</CODE>, <CODE>bag</CODE> or
         <CODE>duplicate_bag</CODE> table type for safe traversal.
<P>A process fixes a table by calling <CODE>safe_fixtable(Tab,true)</CODE>.
         The table remains fixed until the process releases it by calling
         <CODE>safe_fixtable(Tab,false)</CODE>, or until the process terminates.
        
<P>If several processes fix a table, the table will remain fixed
         until all processes have released it (or terminated).
         A reference counter is kept on a per process basis, and N
         consecutive fixes requires N releases to actually release
         the table.
<P>When a table is fixed, a sequence of <CODE>first/1</CODE> and
         <CODE>next/2</CODE> calls are guaranteed to succeed even if objects
         are removed during the traversal. An example:
<PRE>
clean_all_with_value(Tab,X) -&#62;
  safe_fixtable(Tab,true),
  clean_all_with_value(Tab,X,ets:first(Tab)),
  safe_fixtable(Tab,false).
          
clean_all_with_value(Tab,X,'$end_of_table') -&#62;
  true;
clean_all_with_value(Tab,X,Key) -&#62;
  case ets:lookup(Tab,Key) of
    [{Key,X}] -&#62;
       ets:delete(Tab,Key);
    _ -&#62;
       true
  end,
  clean_all_with_value(Tab,X,ets:next(Tab,Key)).
        
</PRE>

<P>Note that no deleted objects are actually removed from a fixed
         table until it has been released. If a process fixes a table but
         never releases it, the memory used by the deleted objects will
         never be freed. The performance of operations on the table will
         also degrade significantly.
<P>Use <CODE>info/2</CODE> to retrieve information about which processes
         have fixed which tables. A system with a lot of processes fixing
         tables may need a monitor which sends alarms when tables have
         been fixed for too long.
<P>Note that for tables of the <CODE>ordered_set</CODE> type,
         <CODE>safe_fixtable/2</CODE> is not necessary as calls to <CODE>first/1</CODE>
         and <CODE>next/2</CODE> will always succeed.
</DIV>

<P><A NAME="select/2"><STRONG><CODE>select(Tab, MatchSpec) -&#62; [Object]</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Object = tuple()</CODE></STRONG><BR>
<STRONG><CODE>MatchSpec = term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>      Matches the objects in the table <CODE>Tab</CODE> using a
         match_spec as described in ERTS users guide. This is a more
         general call than the <CODE>ets:match/2</CODE> and
         <CODE>ets:match_object/2</CODE> calls. In its simplest forms the 
         match_spec's
         look like this:
         
<P>
<UL>

<LI>
MatchSpec = [MatchFunction]
         
</LI>


<LI>
MatchFunction = {MatchHead, [Guard], [Result]}
         
</LI>


<LI>
MatchHead = &#34;Pattern as in ets:match&#34;
         
</LI>


<LI>
Guard = {&#34;Guardtest name&#34;, ...}
         
</LI>


<LI>
Result = &#34;Term construct&#34;
         
</LI>


</UL>

<P>      This means that the match_spec is always a list of one or
         more tuples (of arity 3). The tuples first element should be
         a pattern as described in the documentation of
         <CODE>ets:match/2</CODE>. The second element of the tuple should
         be a list of 0 or more guard tests (described below). The
         third element of the tuple should be a list containing a
         description of the value to actually return. In almost all
         normal cases the list contains exactly one term which fully
         describes the value to return for each object.
         
<P>      The return value is constructed using the &#34;match variables&#34;
         bound in the MatchHead or using the special match variables
         <CODE>'$_'</CODE> (the whole matching object) and <CODE>'$$' (all match
         variables in a list), so that the following
         &#60;c&#62;ets:match/2</CODE> expression:
         
<PRE>
            ets:match(Tab,{'$1','$2','$3'})
          
</PRE>

<P> is exactly equivalent to:
         
<PRE>
            ets:select(Tab,[{{'$1','$2','$3'},[],['$$']}])
          
</PRE>

<P>      - and the following <CODE>ets:match_object/2</CODE> call:
         
<PRE>
            ets:match_object(Tab,{'$1','$2','$1'})
          
</PRE>

<P>      is exactly equivalent to
         
<PRE>
            ets:select(Tab,[{{'$1','$2','$1'},[],['$_']}])
          
</PRE>

<P>      Composite terms can be constructed in the <CODE>Result</CODE> part
         either by simply writing a list, so that this code:
         
<PRE>
            ets:select(Tab,[{{'$1','$2','$3'},[],['$$']}])
          
</PRE>

<P>      gives the same output as:
         
<PRE>
            ets:select(Tab,[{{'$1','$2','$3'},[],[['$1','$2','$3']]}])
          
</PRE>

<P>      i.e. all the bound variables in the match head as a list. If
         tuples are to be constructed, one has to write a tuple of
         arity 1 with the single element in the tuple being the tuple
         one wants to construct (as an ordinary tuple could be mistaken
         for a <CODE>Guard</CODE>). Therefore the following call:
         
<PRE>
            ets:select(Tab,[{{'$1','$2','$1'},[],['$_']}])
          
</PRE>

<P>      gives the same output as:
         
<PRE>
            ets:select(Tab,[{{'$1','$2','$1'},[],[{{'$1','$2','$3'}}]}])
          
</PRE>

<P>      - this syntax is equivalent to the syntax used in the trace
         patterns (see the <CODE>dbg</CODE> module in the
         <CODE>runtime_tools</CODE> application).
         
<P>      The <CODE>Guard</CODE>'s are constructed as tuples where the first
         element is the name of the test (again, see the
         <CODE>match_spec</CODE> documentation in ERTS users guide) and the
         rest of the elements are the parameters of the test. To check
         for a specific type (say a list) of the element bound to the
         match variable <CODE>'$1'</CODE>, one would write the test as
         <CODE>{is_list, '$1'}</CODE>. If the test fails, the object in the
         table won't match and the next <CODE>MatchFunction</CODE> (if any)
         will be tried. Most guard tests present in erlang can be used,
         but only the new versions prefixed <CODE>is_</CODE> are allowed
         (like <CODE>is_float</CODE>, <CODE>is_atom</CODE> etc). An exact list of
         the allowed guard tests is present in the <CODE>match_spec</CODE>
         section of ERTS users guide. 
         
<P>      The <CODE>Guard</CODE> section can also contain logic and arithmetic
         operations, which are written with the same syntax as the
         guard tests (prefix notation), so that a guard test written in
         erlang looking like this:
         
<PRE>
            is_integer(X), is_integer(Y), X + Y &#60; 4711
          
</PRE>

<P>      is expressed like this (X replaced with '$1' and Y with
         '$2'):
         
<PRE>
            [{is_integer, '$1'}, {is_integer, '$2'}, {'&#60;', {'+', '$1',
            '$2'}, 4711}]
          
</PRE>

<P>      A complete list of the operators is present in the match_spec
         section of ERTS users guide.
         
</DIV>

<P><A NAME="select/3"><STRONG><CODE>select(Tab, MatchSpec, Limit) -&#62; {[Match],Continuation} | '$end_of_table'</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Object = tuple()</CODE></STRONG><BR>
<STRONG><CODE>MatchSpec = term()</CODE></STRONG><BR>
<STRONG><CODE>Continuation = term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Works like <CODE>ets:select/2</CODE> but only returns a limited
        (<CODE>Limit</CODE>) number of matching objects. The
        <CODE>Continuation</CODE> term can then be used in subsequent calls
        to <CODE>ets:select/1</CODE> to get the next chunk of matching
        objects. This is a space efficient way to work on objects in a
        table which is still faster than traversing the table object
        by object using <CODE>ets:first/1</CODE> and <CODE>ets:next/1</CODE>.
<P><CODE>'$end_of_table'</CODE> is returned if the table is empty.
</DIV>

<P><A NAME="select/1"><STRONG><CODE>select(Continuation) -&#62; {[Match],Continuation} | '$end_of_table'</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Match = [term()]</CODE></STRONG><BR>
<STRONG><CODE>Continuation = term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Continues a match started with
         <CODE>ets:select/3</CODE>. The next
         chunk of the size given in the initial <CODE>ets:select/3</CODE>
         call is returned together with a new <CODE>Continuation</CODE>
         that can be used in subsequent calls to this function.
<P>'$end_of_table' is returned when there are no more
         objects in the table.
</DIV>

<P><A NAME="select_delete/2"><STRONG><CODE>select_delete(Tab, MatchSpec) -&#62; NumDeleted</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Object = tuple()</CODE></STRONG><BR>
<STRONG><CODE>MatchSpec = term()</CODE></STRONG><BR>
<STRONG><CODE>NumDeleted = integer()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>      Matches the objects in the table <CODE>Tab</CODE> using a
         match_spec as described in ERTS users guide. 
         If the match_spec returns the atom <CODE>true</CODE> for an
         object, that object is removed from the table. For any
         other result from the match_spec the object is retained.
         This is a more
         general call than the <CODE>ets:match_delete/2</CODE> call. 
         
<P>The function returns the number of objects actually
         deleted from the table.
</DIV>

<P><A NAME="select_count/2"><STRONG><CODE>select_count(Tab, MatchSpec) -&#62; NumMatched</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Object = tuple()</CODE></STRONG><BR>
<STRONG><CODE>MatchSpec = term()</CODE></STRONG><BR>
<STRONG><CODE>NumMatched = integer()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>      Matches the objects in the table <CODE>Tab</CODE> using a
         match_spec as described in ERTS users guide. 
         If the match_spec returns the atom <CODE>true</CODE> for an
         object, that object considered a match and is counted. For any
         other result from the match_spec the object is not
         considered a match and is therefore not counted.
         
<P>The function could be described as a match_delete/2 that
         does not actually delete any elements, but only counts them.
<P>The function returns the number of objects matched.
</DIV>

<P><A NAME="slot/2"><STRONG><CODE>slot(Tab, I) -&#62; [Object] | '$end_of_table'</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>I = int()</CODE></STRONG><BR>
<STRONG><CODE>Object = tuple()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>This function is mostly for debugging purposes, Normally
         one should use <CODE>first/next</CODE> or <CODE>last/prev</CODE> instead.
<P>Returns all objects in the <CODE>I</CODE>:th slot of the table
         <CODE>Tab</CODE>. A table can be traversed by repeatedly calling
         the function, starting with the first slot <CODE>I=0</CODE> and
         ending when <CODE>'$end_of_table'</CODE> is returned.
         The function will fail with reason <CODE>badarg</CODE> if the <CODE>I</CODE>
         argument is out of range.
<P>Unless a table of type <CODE>set</CODE>, <CODE>bag</CODE> or
         <CODE>duplicate_bag</CODE> is protected using <CODE>safe_fixtable/2</CODE>,
         see above, a traversal may fail if concurrent updates are made
         to the table.
         If the table is of type <CODE>ordered_set</CODE>, the function returns
         a list containing the <CODE>I</CODE>:th object in Erlang term order.
</DIV>

<P><A NAME="tab2file/2"><STRONG><CODE>tab2file(Tab, Filename) -&#62; ok | {error,Reason}</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Filename = string() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Reason = term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Dumps the table <CODE>Tab</CODE> to the file <CODE>Filename</CODE>.
         The implementation of this function is not efficient.
</DIV>

<P><A NAME="tab2list/1"><STRONG><CODE>tab2list(Tab) -&#62; [Object]</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Object = tuple()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>Returns a list of all objects in the table <CODE>Tab</CODE>.
</DIV>

<P><A NAME="table/2"><STRONG><CODE>table(Tab [, Options]) -&#62; QueryHandle</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>QueryHandle = -a query handle, see qlc(3)-</CODE></STRONG><BR>
<STRONG><CODE>Options = [Option] | Option</CODE></STRONG><BR>
<STRONG><CODE>Option = {n_objects, NObjects}
| {traverse, TraverseMethod}</CODE></STRONG><BR>
<STRONG><CODE>NObjects = default | integer() &#62; 0</CODE></STRONG><BR>
<STRONG><CODE>TraverseMethod = first_next
| last_prev
| select
| {select, MatchSpec}</CODE></STRONG><BR>
<STRONG><CODE>MatchSpec = term()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P><A NAME="qlc_table"><!-- Empty --></A>Returns a QLC (Query List
         Comprehension) query handle. The module <CODE>qlc</CODE>
         implements a query language aimed mainly at Mnesia but ETS
         tables, Dets tables, and lists are also recognized by QLC
         as sources of data. Calling <CODE>ets:table/1,2</CODE> is the
         means to make the ETS table <CODE>Tab</CODE> usable to QLC.
<P>When there are only simple restrictions on the key position
         QLC uses <CODE>ets:lookup/2</CODE> to look up the keys, but when
         that is not possible the whole table is traversed. The
         option <CODE>traverse</CODE> determines how this is done:
<P>
<UL>

<LI>
<CODE>first_next</CODE>. The table is traversed one key at
         a time by calling <CODE>ets:first/1</CODE> and
         <CODE>ets:next/2</CODE>.<BR>


</LI>


<LI>
<CODE>last_prev</CODE>. The table is traversed one key at
         a time by calling <CODE>ets:last/1</CODE> and
         <CODE>ets:prev/2</CODE>.<BR>


</LI>


<LI>
<CODE>select</CODE>. The table is traversed by calling
         <CODE>ets:select/3</CODE> and <CODE>ets:select/1</CODE>. The option
         <CODE>n_objects</CODE> determines the number of objects
         returned (the third argument of <CODE>select/3</CODE>); the
         default is to return <CODE>100</CODE> objects at a time. The
         match specification (the second argument of
         <CODE>select/3</CODE>) is assembled by QLC: simple filters are
         translated into equivalent match specifications while
         more complicated filters have to be applied to all
         objects returned by <CODE>select/3</CODE> given a match
         specification that matches all objects.<BR>


</LI>


<LI>
<CODE>{select, MatchSpec}</CODE>. As for <CODE>select</CODE>
         the table is traversed by calling <CODE>ets:select/3</CODE> and
         <CODE>ets:select/1</CODE>. The difference is that the match
         specification is explicitly given. This is how to state
         match specifications that cannot easily be expressed
         within the syntax provided by QLC.<BR>

         
</LI>


</UL>

<P>The following example uses an explicit match specification
         to traverse the table:

<PRE>
1&#62; <STRONG>ets:insert(Tab = ets:new(t, []), [{1,a},{2,b},{3,c},{4,d}]),</STRONG>
<STRONG>MS = ets:fun2ms(fun({X,Y}) when (X &#62; 1) or (X &#60; 5) -&#62; {Y} end),</STRONG>
<STRONG>QH1 = ets:table(Tab, [{traverse, {select, MS}}]).</STRONG>
</PRE>

<P>An example with implicit match specification:
<PRE>
2&#62; <STRONG>QH2 = qlc:q([{Y} || {X,Y} &#60;- ets:table(Tab), (X &#62; 1) or (X &#60; 5)]).</STRONG>
</PRE>

<P>The latter example is in fact equivalent to the former which 
         can be verified using the function <CODE>qlc:info/1</CODE>:
<PRE>
3&#62; <STRONG>qlc:info(QH1) =:= qlc:info(QH2).</STRONG>
true
</PRE>

<P><CODE>qlc:info/1</CODE> returns information about a query handle,
         and in this case identical information is returned for the
         two query handles.
</DIV>

<P><A NAME="test_ms/2"><STRONG><CODE>test_ms(Tuple, MatchSpec) -&#62; {ok, Result} | {error, Errors}</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tuple = tuple()</CODE></STRONG><BR>
<STRONG><CODE>MatchSpec = term()</CODE></STRONG><BR>
<STRONG><CODE>Result = term()</CODE></STRONG><BR>
<STRONG><CODE>Errors = [{warning|error, string()}]
</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>      This function is a utility to test the <CODE>match_spec</CODE>'s
         used in calls to <CODE>ets:select/2</CODE>. The function both
         tests the <CODE>MatchSpec</CODE> for &#34;syntactic&#34; correctness and
         runs the match_spec against the object <CODE>Tuple</CODE>. If the
         match_spec contains errors, the tuple <CODE>{error, Errors}</CODE>
         is returned where <CODE>Errors</CODE> is a list of natural
         language descriptions of what was wrong with the
         match_spec. If the match_spec is syntactically OK, the
         function returns <CODE>{ok,Term}</CODE> where <CODE>Term</CODE> is what
         would have been the result in a real <CODE>ets:select/2</CODE>
         call or <CODE>false</CODE> if the <CODE>match_spec</CODE> does not match
         the object <CODE>Tuple</CODE>.
         
<P>      This is a useful debugging and test tool, especially when
         writing complicated <CODE>ets:select/2</CODE> calls.
         
</DIV>

<P><A NAME="to_dets/2"><STRONG><CODE>to_dets(Tab, DetsTab) -&#62; Tab</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>DetsTab = atom()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>     Fills an already created/opened Dets table with the objects in the
        already opened ETS table named <CODE>Tab</CODE>. The Dets table
        is emptied before the objects are inserted.
        
</DIV>

<P><A NAME="update_counter/6"><STRONG><CODE>update_counter(Tab, Key, {Pos,Incr,Threshold,SetValue}) -&#62; Result</CODE></STRONG></A><BR>
<A NAME="update_counter/4"><STRONG><CODE>update_counter(Tab, Key, {Pos,Incr}) -&#62; Result</CODE></STRONG></A><BR>
<A NAME="update_counter/3"><STRONG><CODE>update_counter(Tab, Key, Incr) -&#62; Result</CODE></STRONG></A><BR>

<DIV CLASS=REFBODY><P>Types:
  <DIV CLASS=REFTYPES>
<P>
<STRONG><CODE>Tab = tid() | atom()</CODE></STRONG><BR>
<STRONG><CODE>Key = term()</CODE></STRONG><BR>
<STRONG><CODE>Pos = Incr = Threshold = SetValue = Result = int()</CODE></STRONG><BR>

  </DIV>
</DIV>

<DIV CLASS=REFBODY>

<P>This functions provides an efficient way to update a counter,
         without the hassle of having to look up an object, update the
         object by incrementing an element and insert the resulting object
         into the table again.
<P>It will destructively update the object with key <CODE>Key</CODE> in
         the table <CODE>Tab</CODE> by adding <CODE>Incr</CODE> to the element at
         the <CODE>Pos</CODE>:th position. The new counter value is returned.
         If no position is specified, the element directly following
         the key (<CODE>&#60;keypos&#62;+1</CODE>) is updated.
<P>If a <CODE>Threshold</CODE> is specified, the counter will be
         reset to the value <CODE>SetValue</CODE> if the following
         conditions occur:
<P>
<UL>

<LI>
The <CODE>Incr</CODE> is not negative (<CODE>&#62;= 0</CODE>) and the
         result would be greater than (<CODE>&#62;</CODE>) <CODE>Threshold</CODE>
</LI>


<LI>
The <CODE>Incr</CODE> is negative (<CODE>&#60; 0</CODE>) and the
         result would be less than (<CODE>&#60;</CODE>)
         <CODE>Threshold</CODE>
</LI>


</UL>

<P>The function will fail with reason <CODE>badarg</CODE> if:
<P>
<UL>

<LI>
the table is not of type <CODE>set</CODE> or <CODE>ordered_set</CODE>,
         
</LI>


<LI>
no object with the right key exists,
</LI>


<LI>
the object has the wrong arity,
</LI>


<LI>
the element to update is not an integer, or,
</LI>


<LI>
any of <CODE>Pos</CODE>, <CODE>Incr</CODE>, <CODE>Threshold</CODE> or
         <CODE>SetValue</CODE> is not an integer
</LI>


</UL>

</DIV>

<H3>AUTHORS</H3>
<DIV CLASS=REFBODY>
 Claes Wikstrom, Tony Rogvall, Patrik Nyblom - support@erlang.ericsson.se<BR>

</DIV>
<CENTER>
<HR>
<SMALL>stdlib 1.13.2<BR>
Copyright &copy; 1991-2004
<A HREF="http://www.erlang.se">Ericsson AB</A><BR>
</SMALL>
</CENTER>
</BODY>
</HTML>