spotbugs-exclude.xml

<?xml version="1.0" encoding="UTF-8"?>
<!--

    Copyright (c) 2010, 2021 Oracle and/or its affiliates. All rights reserved.

    This program and the accompanying materials are made available under the
    terms of the Eclipse Public License v. 2.0, which is available at
    http://www.eclipse.org/legal/epl-2.0.

    This Source Code may also be made available under the following Secondary
    Licenses when the conditions for such availability set forth in the
    Eclipse Public License v. 2.0 are satisfied: GNU General Public License,
    version 2 with the GNU Classpath Exception, which is available at
    https://www.gnu.org/software/classpath/license.html.

    SPDX-License-Identifier: EPL-2.0 OR GPL-2.0 WITH Classpath-exception-2.0

-->

<!-- FindBugs exclude list for Jakarta Mail -->

<FindBugsFilter>
    <!--
	FindBugs *really* wants us to use zero length arrays instead
	of null.  Unfortunately, there's a bunch of places we can't
	do this for various reasons:
	  - The API specifies null and we can't make an incompatible change.
	  - We use null to indicate no header and an empty array to indicate
	    the header exists but has no entries.
	  - We use null to indicate NIL in the protocol and an empty array
	    to indicate an empty list "()" in the protocol.
	This error occurs often enough that we just ignore it everywhere.
    -->
    <Match>
        <Bug pattern="PZLA_PREFER_ZERO_LENGTH_ARRAYS"/>
    </Match>
    <!--
	There are a bunch of places where FindBugs complains about
	exposing internal representations.  We exclude cases where
	this only happens in internal classes that are never visible
	through the public Jakarta Mail API, or cases where the user
	passes in an object (usually an array) and can only hurt
	themselves by modifying the array while a method is in progress,
	or where the implementation is passing the data back to the user
	(e.g., in an Exception) and never uses it again.
    -->
    <Match>
        <Or>
            <Class name="jakarta.mail.SendFailedException"/>
            <Class name="jakarta.mail.event.MessageCountEvent"/>
            <Class name="jakarta.mail.event.TransportEvent"/>
        </Or>
        <Or>
            <Bug pattern="EI_EXPOSE_REP"/>
            <Bug pattern="EI_EXPOSE_REP2"/>
        </Or>
    </Match>

    <Match>
        <Class name="jakarta.mail.internet.MimeBodyPart"/>
        <Method name="&lt;init&gt;"/> <!-- match constructor -->
        <!-- passed in byte array -->
        <Bug pattern="EI_EXPOSE_REP2"/>
    </Match>
    <Match>
        <Class name="jakarta.mail.util.ByteArrayDataSource"/>
        <Method name="&lt;init&gt;"/> <!-- match constructor -->
        <!-- passed in byte array -->
        <Bug pattern="EI_EXPOSE_REP2"/>
    </Match>

    <!--
	A few places where we catch Exception even though it's not
	explicitly thrown.  We need to make sure that if anything
	goes wrong we clean things up.  Perhaps these should be
	converted to a finally block and a boolean "success" flag?
	Most of these are related to reflection or ClassLoader
	operations, which can fail for all sorts of unexpected reasons.
	Some of these could be converted to use multi-catch with JDK 1.7.
    -->
    <Match>
        <Class name="jakarta.mail.Session"/>
        <Or>
            <Method name="getService"/>
            <Method name="loadAllResources"/>
        </Or>
        <Bug pattern="REC_CATCH_EXCEPTION"/>
    </Match>
    <Match>
        <Class name="jakarta.mail.internet.MimeUtility"/>
        <Or>
            <Method name="&lt;clinit&gt;"/> <!-- match static initializer -->
            <Method name="getEncoding"/>
        </Or>
        <Bug pattern="REC_CATCH_EXCEPTION"/>
    </Match>

    <!--
	Yes, the "next" element in my linked list isn't ever actually
	used, but it feels weird to only have a "prev" element.
    -->
    <Match>
        <Class name="jakarta.mail.EventQueue$QueueElement"/>
        <Field name="next"/>
        <Bug pattern="URF_UNREAD_FIELD"/>
    </Match>

    <!--
	Stupid Serializable EventObject class causes FindBugs to complain
	about transient fields in subclasses.  I don't know why it's
	complaining about these fields but not others, but since I don't
	really expect anyone to serialize these events I'm just ignoring
	this complaint.  Ditto Exception fields.
    -->
    <Match>
        <Class name="jakarta.mail.event.TransportEvent"/>
        <Or>
            <Field name="invalid"/>
            <Field name="validSent"/>
            <Field name="validUnsent"/>
        </Or>
        <Bug pattern="SE_TRANSIENT_FIELD_NOT_RESTORED"/>
    </Match>
    <Match>
        <Class name="jakarta.mail.event.FolderEvent"/>
        <Or>
            <Field name="folder"/>
            <Field name="newFolder"/>
        </Or>
        <Bug pattern="SE_TRANSIENT_FIELD_NOT_RESTORED"/>
    </Match>
    <Match>
        <Class name="jakarta.mail.event.MessageChangedEvent"/>
        <Field name="msg"/>
        <Bug pattern="SE_TRANSIENT_FIELD_NOT_RESTORED"/>
    </Match>
    <Match>
        <Class name="jakarta.mail.event.MessageCountEvent"/>
        <Field name="msgs"/>
        <Bug pattern="SE_TRANSIENT_FIELD_NOT_RESTORED"/>
    </Match>
    <Match>
        <Class name="jakarta.mail.event.TransportEvent"/>
        <Field name="msg"/>
        <Bug pattern="SE_TRANSIENT_FIELD_NOT_RESTORED"/>
    </Match>
    <Match>
        <Class name="jakarta.mail.SendFailedException"/>
        <Or>
            <Field name="invalid"/>
            <Field name="validSent"/>
            <Field name="validUnsent"/>
        </Or>
        <Bug pattern="SE_TRANSIENT_FIELD_NOT_RESTORED"/>
    </Match>

    <!--
	These string comparisons are just optimizations.
    -->
    <Match>
        <Or>
            <Class name="jakarta.mail.URLName"/>
            <Class name="jakarta.mail.internet.InternetAddress"/>
        </Or>
        <Method name="equals"/>
        <Bug pattern="ES_COMPARING_STRINGS_WITH_EQ"/>
    </Match>

    <!--
	This string comparison using == is to determine whether the
	String object is a different String object.
    -->
    <Match>
        <Class name="jakarta.mail.internet.MimeUtility"/>
        <Method name="decodeText"/>
        <Bug pattern="ES_COMPARING_STRINGS_WITH_EQ"/>
    </Match>

    <!--
	We extract the "lang" field of an encoded string but we don't
	currently do anything with it.  Ignore this complaint.
    -->
    <Match>
        <Class name="jakarta.mail.internet.ParameterList"/>
        <Method name="decodeValue"/>
        <Bug pattern="DLS_DEAD_LOCAL_STORE"/>
    </Match>

    <!--
	The call ParameterList.set(null, "DONE") is a kludge used by the
	IMAP provider to indicate that it's done setting parameters.
	In other cases we *want* a null name to cause a NullPointerException.
    -->
    <Match>
        <Class name="jakarta.mail.internet.ParameterList"/>
        <Method name="set"/>
        <Bug pattern="NP_NULL_ON_SOME_PATH"/>
    </Match>

    <!--
	We purposely don't close these streams, which are just wrappers
	around the original stream that needs to remain open.
    -->
    <Match>
        <Class name="jakarta.mail.internet.MimeMultipart"/>
        <Or>
            <Method name="parse"/>
            <Method name="parsebm"/>
        </Or>
        <Bug pattern="OS_OPEN_STREAM"/>
    </Match>

    <!--
	Can't fix these errors in toString until JavaMail 1.5 because
	it's part of the spec.  Sigh.
    -->
    <Match>
        <Or>
            <Class name="jakarta.mail.internet.ContentDisposition"/>
            <Class name="jakarta.mail.internet.ContentType"/>
        </Or>
        <Method name="toString"/>
        <Bug pattern="NP_TOSTRING_COULD_RETURN_NULL"/>
    </Match>

    <!--
	These errors are in code imported from the JDK, where it seems
	to be purposely using the platform default encoding.  It's safer
	to just leave this alone and ignore the errors.
    -->
    <Match>
        <Class name="jakarta.mail.URLName"/>
        <Bug pattern="DM_DEFAULT_ENCODING"/>
    </Match>

    <!--
	We're purposely using the platform default encoding because
	we're trying to discover what the platform default encoding *is*!
    -->
    <Match>
        <Class name="jakarta.mail.internet.MimeUtility"/>
        <Method name="getDefaultJavaCharset"/>
        <Bug pattern="DM_DEFAULT_ENCODING"/>
    </Match>

    <!--
	We're purposely using the platform default encoding when a
	charset hasn't been specified.
    -->
    <Match>
        <Class name="jakarta.mail.internet.ParameterList"/>
        <Method name="combineMultisegmentNames"/>
        <Bug pattern="DM_DEFAULT_ENCODING"/>
    </Match>

    <!--
	MimeBodyPart.headers is set in the constructor.  IMAPBodyPart has
	perhaps paranoid defensive code to set it if it's not set.  That
	code is protected by the object lock, but all of the reads of the
	field are not, thus FindBugs complains.  For now I think it's
	better to leave the defensive code in and ignore this error.
    -->
    <Match>
        <Class name="jakarta.mail.internet.MimeBodyPart"/>
        <Field name="headers"/>
        <Bug pattern="IS2_INCONSISTENT_SYNC"/>
    </Match>

    <!--
	These unread fields are part of the public API.
    -->
    <Match>
        <Class name="jakarta.mail.Service"/>
        <Field name="debug"/>
        <Bug pattern="URF_UNREAD_PUBLIC_OR_PROTECTED_FIELD"/>
    </Match>
    <Match>
        <Class name="jakarta.mail.Quota$Resource"/>
        <Field name="usage"/>
        <Bug pattern="URF_UNREAD_PUBLIC_OR_PROTECTED_FIELD"/>
    </Match>

    <!--
	I know Exception.toString is never supposed to return null,
	but this extra check is cheap insurance.
    -->
    <Match>
        <Class name="jakarta.mail.MessagingException"/>
        <Method name="toString"/>
        <Bug pattern="RCN_REDUNDANT_NULLCHECK_OF_NONNULL_VALUE"/>
    </Match>
    <!--
	I know ClassLoader.getResources is never supposed to return null,
	but this extra check is cheap insurance.
    -->
    <Match>
        <Class name="~jakarta\.mail\.Session\$.*"/>
        <Method name="run"/>
        <Bug pattern="RCN_REDUNDANT_NULLCHECK_OF_NONNULL_VALUE"/>
    </Match>

    <!--
	These names may be confusing, but we're stuck with them.
    -->
    <Match>
        <Class name="jakarta.mail.PasswordAuthentication"/>
        <Method name="getUserName"/>
        <Bug pattern="NM_CONFUSING"/>
    </Match>

    <!--
	This trivial inner class extends ArrayList, but is never serialized,
	so it doesn't really need a serialVersionUID.
    -->
    <Match>
        <Class name="jakarta.mail.internet.ParameterList$MultiValue"/>
        <Bug pattern="SE_NO_SERIALVERSIONID"/>
    </Match>

    <!--
	The Authenticator fields are only accessible by subclasses,
	which are only called through the one synchronized public
	method, assuming the subclass doesn't do something crazy and
	expose them itself.
    -->
    <Match>
        <Class name="jakarta.mail.Authenticator"/>
        <Bug pattern="IS2_INCONSISTENT_SYNC"/>
    </Match>
</FindBugsFilter>