<?xml version="1.0" encoding="utf-8" ?>
<!DOCTYPE cref SYSTEM "cref.dtd">
<cref>
<header>
<copyright>
<year>1996</year><year>2016</year>
<holder>Ericsson AB. All Rights Reserved.</holder>
</copyright>
<legalnotice>
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
</legalnotice>
<title>erl_error</title>
<prepared>Torbjörn Törnkvist</prepared>
<responsible>Torbjörn Törnkvist</responsible>
<docno></docno>
<approved>Bjarne Däcker</approved>
<checked>Torbjörn Törnkvist</checked>
<date>1996-10-14</date>
<rev>A</rev>
<file>erl_error.xml</file>
</header>
<lib>erl_error</lib>
<libsummary>Error print routines.</libsummary>
<description>
<p>This module contains some error printing routines taken
from "Advanced Programming in the UNIX Environment"
by W. Richard Stevens.</p>
<p>These functions are all called in the same manner as
<c>printf()</c>, that is, with a string containing format
specifiers followed by a list of corresponding arguments. All output from
these functions is to <c>stderr</c>.</p>
</description>
<funcs>
<func>
<name><ret>void</ret><nametext>erl_err_msg(FormatStr, ... )</nametext></name>
<fsummary>Non-fatal error, and not system call error.</fsummary>
<type>
<v>const char *FormatStr;</v>
</type>
<desc>
<p>The message provided by the caller is printed. This
function is simply a wrapper for <c>fprintf()</c>.</p>
</desc>
</func>
<func>
<name><ret>void</ret><nametext>erl_err_quit(FormatStr, ... )</nametext></name>
<fsummary>Fatal error, but not system call error.</fsummary>
<type>
<v>const char *FormatStr;</v>
</type>
<desc>
<p>Use this function when a fatal error has occurred that
is not because of a system call. The message provided by the
caller is printed and the process terminates with exit
value <c>1</c>. This function does not return.</p>
</desc>
</func>
<func>
<name><ret>void</ret><nametext>erl_err_ret(FormatStr, ... )</nametext></name>
<fsummary>Non-fatal system call error.</fsummary>
<type>
<v>const char *FormatStr;</v>
</type>
<desc>
<p>Use this function after a failed system call. The message
provided by the caller is printed followed by a string
describing the reason for failure.</p>
</desc>
</func>
<func>
<name><ret>void</ret><nametext>erl_err_sys(FormatStr, ... )</nametext></name>
<fsummary>Fatal system call error.</fsummary>
<type>
<v>const char *FormatStr;</v>
</type>
<desc>
<p>Use this function after a failed system call. The message
provided by the caller is printed followed by a string
describing the reason for failure, and the process
terminates with exit value <c>1</c>. This function does not
return.</p>
</desc>
</func>
</funcs>
<section>
<title>Error Reporting</title>
<p>Most functions in <c>Erl_Interface</c> report failures to the caller by
returning some otherwise meaningless value (typically
<c>NULL</c>
or a negative number). As this only tells you that things did not
go well, examine the error code in <c>erl_errno</c> if you
want to find out more about the failure.</p>
</section>
<funcs>
<func>
<name><ret>volatile int</ret><nametext>erl_errno</nametext></name>
<fsummary>Variable <c>erl_errno</c> contains the
Erl_Interface error number. You can change the value if you wish.
</fsummary>
<desc>
<p><c>erl_errno</c> is initially (at program startup) zero
and is then set by many <c>Erl_Interface</c> functions on failure to
a non-zero error code to indicate what kind of error it
encountered. A successful function call can change
<c>erl_errno</c> (by calling some other function that
fails), but no function does never set it to zero. This means
that you cannot use <c>erl_errno</c> to see <em>if</em> a
function call failed. Instead, each function reports failure
in its own way (usually by returning a negative number or
<c>NULL</c>), in which case you can examine
<c>erl_errno</c> for details.</p>
<p><c>erl_errno</c> uses the error codes defined in your
system's <c><errno.h></c>.</p>
<note>
<p><c>erl_errno</c> is a "modifiable lvalue" (just
like ISO C defines <c>errno</c> to be) rather than a
variable. This means it can be implemented as a macro
(expanding to, for example, <c>*_erl_errno()</c>).
For reasons of thread safety (or task safety), this is exactly what
we do on most platforms.</p>
</note>
</desc>
</func>
</funcs>
</cref>