File: incompatible.html

package info (click to toggle)
erlang-doc-html 1%3A11.b.2-1
links: PTS
area: main
in suites: etch, etch-m68k
size: 23,284 kB
ctags: 10,724
sloc: erlang: 505; ansic: 323; makefile: 62; perl: 61; sh: 45
file content (682 lines) | stat: -rw-r--r-- 17,913 bytes
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<!-- This document was generated using DocBuilder 3.3.3 -->
<HTML>
<HEAD>
  <TITLE>Potential Incompatibilities</TITLE>
  <SCRIPT type="text/javascript" src="../doc/erlresolvelinks.js">
</SCRIPT>
</HEAD>
<BODY BGCOLOR="#FFFFFF" TEXT="#000000" LINK="#0000FF" VLINK="#FF00FF"
      ALINK="#FF0000">
<CENTER>
<A HREF="http://www.erlang.se"><IMG BORDER=0 ALT="[Ericsson AB]" SRC="min_head.gif"></A>
</CENTER>

<P>This document contains a list of potential incompatibilities
between Erlang 5.5/OTP R11B (unpatched) and the previous version
Erlang 5.4/OTP R10B (unpatched), and is an extract from the release
notes for the respective applications.
<P>Some of the potential incompatibilities listed below have been
introduced in R10B patches, although as a general rule, this is
avoided.
<P>Note: This document was compiled at the time when R11B was released
and does not list any potential incompatibilities introduced in
R11B patches. If using a patched version of R11B, please refer to
the README for the respective patches for information about
additional potential incompatibilites.<A NAME="1"><!-- Empty --></A>
<H2>1 Compiler</H2>

<P>
<UL>

<LI>
                 The compiler options <CODE>ignore_try</CODE> and
         <CODE>ignore_cond</CODE>, which allowed code that used unquoted
         <CODE>try</CODE> or <CODE>cond</CODE> as atoms or record tags, has
         been removed. Old code that depended on the options need to be
         revised to have occurrences of <CODE>try</CODE> or <CODE>cond</CODE>
         as atom or record tags single-quoted. (Note: Although
         <CODE>cond</CODE> is a reserved keyword, there is no
         <CODE>cond</CODE> statement. It might be introduced in a future
         release.)<BR>


                 Own Id: OTP-6058<BR>


</LI>


</UL>
<A NAME="2"><!-- Empty --></A>
<H2>2 Erts</H2>

<P>
<UL>

<LI>
                 <STRONG>NOTE</STRONG>: The design of SMP support for drivers is
         ongoing. There will probably be incompatible driver
         changes (only affecting drivers run on the runtime system
         with SMP support) released as patches for R11B.<BR>


                 <STRONG>Potential incompatibility</STRONG>: Previously, specific
         driver call-backs were always called from the same
         thread. This is <STRONG>not</STRONG> true in the runtime system
         with SMP support. Calls to call-backs will be made from
         different threads, e.g., two consecutive calls to exactly
         the same call-back can be made from two different
         threads. This will in <STRONG>most</STRONG> cases not be a problem.
         All calls to call-backs are synchronized, i.e., only one
         call-back will be called at a time.<BR>


                 Own Id: OTP-6006 Aux Id: OTP-6095<BR>


</LI>


<LI>
                 Linked-in driver modifications.<BR>

        
<UL>

<LI>
 Linked-in drivers must be recompiled. <BR>

</LI>


<LI>
 The <CODE>refc</CODE> field in the <CODE>ErlDrvBinary</CODE>
         type has been removed. The reference count can be
         accessed via API functions. For more information see the
         <CODE>erl_driver(1)</CODE> man page. <BR>

</LI>


</UL>

                 Own Id: OTP-6095 Aux Id: OTP-6006<BR>


</LI>


</UL>
<A NAME="3"><!-- Empty --></A>
<H2>3 Inets</H2>

<P>
<UL>

<LI>
                 The internal design of using blocking <CODE>gen_tcp:recv</CODE>
         with a timeout and retries resulted in code that was hard to
         get a good overview of, and ultimate led to situations where
         the client got the wrong answer or no answer at all.
         The errors where many times very timing dependent and mainly
         effected the chunk-related functions, so if you where lucky
         you probably would not have noticed. The internal design was
         changed to use <CODE>gen_tcp</CODE> active once semantics. The API
         is not effected except for the function <CODE>ftp:quote/2</CODE>
         which now returns a list of strings (ftp result lines) where
         the line endings &#34;\r\n&#34; has been removed. This was
         the original intention for the return value of
         <CODE>ftp:quote/2</CODE> but it was non trivial to make a good such
         solution with the old design and a compromise was made.<BR>

                 Introduced in Inets 4.5<BR>

                 Own Id: OTP-5623<BR>


</LI>


<LI>
                 When generating dynamic HTTP response bodies the the default
         content-type is now set to &#34;text/html&#34; instead of &#34;text/plain&#34;
         which is more intuitive.<BR>

                 Introduced in Inets 4.5<BR>

                 Own Id: OTP-5626<BR>


</LI>


<LI>
                 The ftp client now supports passive mode. Actually the ftp
         client will always try to use passive mode and if it fails it
         will use active mode instead. It is also possible to force
         the ftp client to use active mode, if that is desired, by
         calling <CODE>ftp:force_active/1</CODE> this way you can get the old
         behavior.<BR>

                 Introduced in Inets 4.4.1<BR>

                 Own Id: OTP-5148<BR>


</LI>


<LI>
                 The pipeline timeout was changed to be zero by default to
         avoid that people by accident would create connection
         processes that never dies and eats up the socket resources.
        <BR>

                 Introduced in Inets 4.3<BR>

                 Own Id: OTP-5442<BR>


</LI>


</UL>
<A NAME="4"><!-- Empty --></A>
<H2>4 Kernel</H2>

<P>
<UL>

<LI>
                 Previously <CODE>unlink/1</CODE> and <CODE>erlang:demonitor/2</CODE>
         behaved completely asynchronous. This had one undesirable
         effect, though. You could never know when you were
         guaranteed <STRONG>not</STRONG> to be affected by a link that you
         had unlinked or a monitor that you had demonitored.<BR>

                 The new behavior of <CODE>unlink/1</CODE> and
         <CODE>erlang:demonitor/2</CODE> can be viewed as two operations
         performed atomically. Asynchronously send an unlink
         signal or a demonitor signal, and ignore any future
         results of the link or monitor.<BR>

                 <STRONG>NOTE</STRONG>: This change can cause some obscure code
         to fail which previously did not. For example, the
         following code might hang:<BR>

        
<PRE>
          Mon = erlang:monitor(process, Pid),
          %% ...
          exit(Pid, bang),
          erlang:demonitor(Mon),
          receive
          {'DOWN', Mon, process, Pid, _} -&#62; ok
          %% We were previously guaranteed to get a down message
          %% (since we exited the process ourself), so we could
          %% in this case leave out:
          %% after 0 -&#62; ok
          end,
        
</PRE>

                 Own Id: OTP-5772<BR>


</LI>


<LI>
                 Global no longer allows the registration of a process
         under more than one name. If the old (buggy) behavior is
         desired the Kernel application variable
         <CODE>global_multi_name_action</CODE> can be given the value
         <CODE>allow</CODE>.<BR>

                 Own Id: OTP-5640 Aux Id: OTP-5603<BR>


</LI>


<LI>
                 The <CODE>auth</CODE> module API is deprecated.<BR>

                 Own Id: OTP-6037<BR>


</LI>


<LI>
                 Several bug fixes and improvements in the global name
         registration facility (see <CODE>global(3)</CODE>):<BR>

        
<UL>

<LI>
the name resolving procedure did not always unlink no
         longer registered processes;
</LI>


<LI>
the global name server could sometimes hang when a
         <CODE>nodedown</CODE> was immediately followed by a
         <CODE>nodeup</CODE>;
</LI>


<LI>
global names were not always unregistered when a node
         went down;
</LI>


<LI>
it is now possible to set and delete locks at
         the same time as the global name server is resolving
         names--the handling of global locks has been separated
         from registration of global names;
</LI>


</UL>

                 As of this version, <CODE>global</CODE> no longer supports nodes
         running Erlang/OTP R7B or earlier.<BR>

                 Introduced in Kernel 2.10.11
        <BR>
         Own Id: OTP-5563<BR>


</LI>


<LI>
                 The previously undocumented and UNSUPPORTED <CODE>zlib</CODE>
         module has been updated in an incompatible way and many
         bugs have been corrected. It is now also documented.<BR>

                 Introduced in Kernel 2.10.11
        <BR>
         Own Id: OTP-5715<BR>


</LI>


</UL>
<A NAME="5"><!-- Empty --></A>
<H2>5 Megaco</H2>

<P>
<UL>

<LI>
        The old version info print function,
         <CODE>format_versions/1</CODE>, have been deprecated.<BR>

                 Introduced in Megaco 3.2.7
        <BR>
Own Id: OTP-5953<BR>


</LI>


<LI>
        The <CODE>conn_info</CODE> function once again returns
         the <STRONG>next</STRONG> transaction id.<BR>

                 Introduced in Megaco 3.2.5
        <BR>
Own Id: OTP-5887<BR>

        Aux Id: Seq 10184
<BR>

</LI>


<LI>
        <CODE>localControlDescriptor</CODE> list of <CODE>localParm</CODE> now
         at least has to be of length 1.<BR>

                 Introduced in Megaco 3.2.4
        <BR>
Own Id: OTP-5882
<BR>

</LI>


<LI>
        The <CODE>ok</CODE> return value from
         <CODE>test_digit_event</CODE> and 
         <CODE>eval_digit_map</CODE> has been changed.<BR>

                 Introduced in Megaco 3.2.3/3.2.2
        <BR>
Own Id: OTP-5826 Aux Id: Seq 10085
<BR>

</LI>


<LI>
        The (successful) return value from the 
         <CODE>test_digit_event</CODE> and 
         <CODE>eval_digit_map</CODE> 
         has been changed.<BR>

        Introduced in Megaco 3.2
        <BR>
Own Id: OTP-5750 Aux Id: Seq 10085
<BR>

</LI>


<LI>
        The <CODE>File</CODE> argument to the function 
         <CODE>enable_trace</CODE> no 
         longer sets up <CODE>dbg</CODE> to write the trace events directly
         to file but instead to be written to a plain text file using 
         <CODE>io:format</CODE>.<BR>

        Also <CODE>enable_trace</CODE>
         no longer accepts the argument 
         <CODE>{io, Verbosity}</CODE>.<BR>

        Own Id: OTP-5447<BR>

                 Introduced in Megaco 3.0.1
<BR>

</LI>


<LI>
        Package number scheme (among other things) was changed
         in version 1 of the standard.
        <BR>
         Introduced in Megaco 3.0
        <BR>
Own Id: OTP-5272<BR>


</LI>


<LI>
        The config property <STRONG>orig_pending_limit</STRONG> has been 
         renamed to <STRONG>sent_pending_limit</STRONG>.
        <BR>
         Introduced in Megaco 3.0
        <BR>
Own Id: OTP-5220<BR>


</LI>


</UL>
<A NAME="6"><!-- Empty --></A>
<H2>6 OS_Mon</H2>

<P>
<UL>

<LI>
        The behaviour of OS_Mon 2.0 is
         backwards compatible under normal operation, but has changed
         somewhat in error situations: The services do not terminate
         and the API functions do not raise exceptions in all cases
         where they did before. Also, in the case where a service does
         terminate, the exit reason may be different.
        <BR>


</LI>


<LI>
                 The memsup part of the OS_Mon application has been made
         more stable. If there are (possibly temporary) problems
         collecting memory data, the interface functions
         (<CODE>get_memory_data/0</CODE>,
         <CODE>get_system_memory_data/0</CODE>) now do not fail, but
         return the previously collected value, if any, or a dummy
         value otherwise. Also, a warning message is printed.<BR>

                 Introduced in OS_Mon 1.8
        <BR>
         Own Id: OTP-5798<BR>


</LI>


</UL>
<A NAME="7"><!-- Empty --></A>
<H2>7 SASL</H2>

<P>
<UL>

<LI>
                 A new option <CODE>{update_paths,Bool}</CODE> has been added
         for <CODE>release_handler:install_release/2</CODE>. It
         indicates if all application code paths should be updated
         (<CODE>Bool==true</CODE>), or if only code paths for modified
         applications should be updated (<CODE>Bool==false</CODE>,
         default).<BR>

                 <CODE>release_handler:set_unpacked/2</CODE> now returns an
         error tuple if a specified application directory does not
         exist.<BR>

                 Introduced in SASL 2.1
        <BR>
         Own Id: OTP-5761<BR>


</LI>


</UL>
<A NAME="8"><!-- Empty --></A>
<H2>8 SNMP</H2>

<P>
<UL>

<LI>
        [manager] The 
         <CODE>snmpm_network_interface</CODE> 
         behaviour has been updated (see 
         <CODE>inform_response</CODE> and 
         definition of the manager net if).<BR>

                 Introduced in SNMP 4.5
        <BR>
Own Id: OTP-5733<BR>


</LI>


<LI>
        [agent] The format of the info returned by the agent 
         info retrieval function has been changed. The info 
         can be converted to the old format by calling the 
         <CODE>old_info_format</CODE> function. 
        <BR>
         Introduced in SNMP 4.4
        <BR>
Own Id: OTP-5666<BR>


</LI>


</UL>
<A NAME="9"><!-- Empty --></A>
<H2>9 SSL</H2>

<P>
<UL>

<LI>
                 The state of a connection in active mode could be in a
         restrictive state, so that an internal <CODE>tcp_closed</CODE>
         message was incorrectly considered illegal, resulting in a
         premature termination of the connection process.
        <BR>
         Introduced in SSL 3.0.11
        <BR>
         Own Id: OTP-5972 Aux Id: seq10188 
<BR>

</LI>


</UL>
<A NAME="10"><!-- Empty --></A>
<H2>10 STDLIB</H2>

<P>
<UL>

<LI>
                 <CODE>ets:info/1</CODE> has been corrected to behave according
         to the documentation and return a list of tuples, not a
         tuple with tuples.<BR>

                 Own Id: OTP-5639<BR>


</LI>


<LI>
                 Corrected <CODE>filelib:ensure_dir/1</CODE> which sometimes
         returned <CODE>true</CODE> and sometimes <CODE>ok</CODE> to always
         return <CODE>ok</CODE> when successful. This goes against the
         documentation which said <CODE>true</CODE>, but <CODE>ok</CODE> was
         judged to be a more logical return value.<BR>

                 Own Id: OTP-5960 Aux Id: seq10240 <BR>


</LI>


<LI>
                 The functions <CODE>lists:ukeysort/2</CODE> and
         <CODE>lists:ukeymerge/3</CODE> have been changed in such a way
         that two tuples are considered equal if their keys
         match.<BR>

                 For the sake of consistency, <CODE>lists:usort/2</CODE> and
         <CODE>lists:umerge/3</CODE> have been modified too: two elements
         are considered equal if they compare equal.<BR>

        The <CODE>file_sorter</CODE> module has been modified in a
         similar way: the <CODE>unique</CODE> option now applies to the
         key (<CODE>keysort()</CODE> and <CODE>keymerge()</CODE>) and the
         ordering function (the option <CODE>{order, Order} </CODE>).<BR>

                 Own Id: OTP-6019<BR>


</LI>


<LI>
                 The <CODE>public</CODE> option has been removed from
         <CODE>digraph:new/1</CODE>. The reason is that several
         functions in the <CODE>digraph</CODE> module are implemented
         using multiple ETS accesses, which is not thread safe.
         (Thanks to Ulf Wiger.)<BR>

                 Own Id: OTP-5985<BR>


</LI>


<LI>
                 Referencing a so far undeclared record from the default
         value of some record declaration is from now on considered
         an error by the linter. It is also an error if the default
         value of a record declaration uses or binds a variable.<BR>

                 Own Id: OTP-5878<BR>


</LI>


<LI>
                 <CODE>filelib:wildcard/2</CODE> was broken (it ignored its
         second argument).<BR>

                 Also, <CODE>filelib:wildcard(&#34;Filename&#34;)</CODE> (where the
         argument does not contain any meta-characters) would
         always return <CODE>[&#34;Filename&#34;]</CODE>. Corrected so that an
         empty list will be returned if <CODE>&#34;Filename&#34;</CODE> does not
         actually exist. (Same correction in
         <CODE>filelib:wildcard/2</CODE>.) (This change is a slight
         incompatibility.)<BR>

                 <CODE>filelib:wildcard/1,2</CODE> will generate a different
         exception when given bad patterns such as <CODE>&#34;{a,&#34;</CODE>. The
         exception used to be caused by
         <CODE>exit(missing_delimiter)</CODE> but is now
         <CODE>erlang:error({badpattern,missing_delimiter})</CODE>.<BR>

                 Introduced in STDLIB 1.13.7
        <BR>
         Own Id: OTP-5523 Aux Id: seq9824 <BR>


</LI>


<LI>
                 <CODE>gb_sets:del_element/2</CODE> was changed to do the
         same as <CODE>gb_sets:delete_any/2</CODE> which was the
         original intention, not as <CODE>gb_sets:delete/2</CODE>. Code
         that relies on <CODE>gb_sets:del_element/2</CODE> causing an
         error if the element does not exist must be changed to
         call <CODE>gb_sets:delete/2</CODE> instead.<BR>

                 The documentation was also updated to explicitly
         document functions that were only referred to as
         &#34;aliases&#34; of a documented function. Also, a list of all
         functions common to the <CODE>gb_sets</CODE>, <CODE>sets</CODE>, and
         <CODE>ordsets</CODE> was added.<BR>

                 Introduced in STDLIB 1.13.3
        <BR>
         Own Id: OTP-5277<BR>


</LI>


</UL>
<CENTER>
<HR>
<SMALL>
Copyright &copy; 1991-2006
<A HREF="http://www.erlang.se">Ericsson AB</A><BR>
</SMALL>
</CENTER>
</BODY>
</HTML>