File: gsoc-ideas.html

package info (click to toggle)
checker-framework-java 3.0.1%2Bds2-3
links: PTS, VCS
area: main
in suites: bullseye
size: 22,736 kB
sloc: java: 145,286; xml: 785; sh: 456; makefile: 401; perl: 26
file content (2084 lines) | stat: -rw-r--r-- 84,752 bytes
<!DOCTYPE html>
<html>
<head>
  <title>Checker Framework organization: GSoC ideas 2019</title>
  <meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
  <meta name="viewport" content="width=device-width, initial-scale=1.0" />
  <link rel="icon" type="image/png" href=
  "../logo/Checkmark/CFCheckmark_favicon.png">
</head>
<body>

<img src="../logo/Logo/CFLogo.png" alt="Checker Framework logo" />

<h1>GSoC ideas 2019</h1> <!-- omit from toc -->

<p>Contents:</p>
<!-- start toc.  do not edit; run html-update-toc instead -->
<ul>
  <li><a href="#introduction">Introduction</a>
    <ul>
      <li><a href="#get-started">How to get started: do a case study</a></li>
      <li><a href="#ask-questions">How to get help and ask questions</a></li>
      <li><a href="#apply">How to apply</a></li>
    </ul></li>
  <li><a href="#case-studies">Case studies</a>
    <ul>
      <li><a href="#library-annotations">Library annotations</a>
        <ul>
          <li><a href="#choose-a-library">Choosing a library to annotate</a></li>
        </ul></li>
      <li><a href="#case-study-signature">Signature strings</a></li>
      <li><a href="#case-study-android-support">Android support annotations</a></li>
      <li><a href="#case-study-signedness">Signed and unsigned numbers</a></li>
      <li><a href="#optional-case-study">Java's Optional class</a></li>
      <li><a href="#case-study-nullness">Null pointer exceptions</a>
        <ul>
          <li><a href="#case-study-nullness-guava">Guava library</a></li>
          <li><a href="#case-study-nullness-bazel">Bazel tool</a></li>
          <li><a href="#case-study-nullness-bcel">BCEL library</a></li>
        </ul></li>
      <li><a href="#index-errors">Improving error messages</a></li>
      <li><a href="#sound-by-default">Sound checking by default</a></li>
      <li><a href="#compare-other-tools">Comparison to other tools</a></li>
    </ul></li>
  <li><a href="#new-type-systems">New type systems</a>
    <ul>
      <li><a href="#non-empty-checker">Non-Empty Checker for precise handling of Queue.peek() and poll()</a></li>
      <li><a href="#Make_programs_deterministic">Make programs deterministic</a></li>
      <li><a href="#custom-tainting-checking">Custom tainting checking</a></li>
      <li><a href="#Bounded-size_strings">Bounded-size strings</a></li>
      <li><a href="#overflow">Overflow checking</a></li>
      <li><a href="#lock-ordering">Lock ordering</a></li>
      <li><a href="#index-checker-mutable-length">Index checking for mutable length data structures</a></li>
      <li><a href="#nullness-bug-detector">Nullness bug detector</a></li>
      <li><a href="#typestate">Stateful type systems</a></li>
      <li><a href="#your-own-new-type-system">Invent your own new type system</a></li>
    </ul></li>
  <li><a href="#cf-other">Type system enhancements and tools</a>
    <ul>
      <li><a href="#annotated-jdk">JDK annotations</a></li>
      <li><a href="#flow-expression-parser">Flow expression parser</a></li>
      <li><a href="#asm">Upgrade to a newer version of ASM</a></li>
      <li><a href="#analysis_diffs">Tool for analysis diffs</a></li>
      <li><a href="#Whole-program_type_inference">Whole-program type inference</a></li>
      <li><a href="#dataflow">Dataflow enhancements</a></li>
      <li><a href="#Purity_analysis">Purity (side effect) analysis</a></li>
      <li><a href="#javadoc">Javadoc support</a></li>
      <li><a href="#performance">Performance improvements</a></li>
      <li><a href="#run-time-checking">Run-time checking</a></li>
      <li><a href="#ide-support">IDE and build system support</a></li>
      <li><a href="#exhaustive-testing">Model checking of a type system</a></li>
    </ul></li>
</ul>
<!-- end toc -->

<h1 id="introduction">Introduction</h1>

<p>
  The <a href="https://checkerframework.org/">Checker Framework</a> is an
  innovative programming tool that helps you prevent bugs at development
  time, before they escape to production.
</p>

<p>
  Java's type system prevents some bugs, such as <code>int count =
  "hello";</code>.  However, it does not prevent other bugs, such as null
  pointer dereferences, concurrency errors, disclosure of private
  information, incorrect internationalization, out-of-bounds indices, and
  so forth.  <em>Pluggable type-checking</em> replaces a
  programming language's built-in type system with a more powerful,
  expressive one.
</p>

<p>
  We have created around 20
  <a href="https://checkerframework.org/manual/#introduction">new type
  systems</a>, and other people have created
  <a href="https://checkerframework.org/manual/#third-party-checkers">many
  more</a>.
  The more powerful type system is not just a
  bug-finding tool:  it is a verification tool that gives a guarantee that
  no errors (of certain types) exist in your program.  Even though it is
  powerful, it is easy to use.  It follows the standard typing rules
  that programmers already know, and it fits into their workflow.
</p>

<p>
  The Checker Framework is popular:  it is used daily at Google, Amazon,
  Uber, on Wall Street, and in other companies from big to small.  It is
  attractive to programmers who care about their craft and the quality of
  their code.  The Checker Framework is the motivation for Java's type
  annotations feature.  It has received multiple awards.
  <!-- at conferences such as JavaOne. -->
  With this widespread use, there is a need for people to help with the
  project:  everything from bug fixes, to new features, to case studies, to
  integration with other tools.  We welcome your contribution!
</p>

<p>
  Why should you join this project?  It's popular, so you will have an
  impact.  It makes code more robust and secure, which is a socially
  important purpose.  Past GSOC students have had great success.
  (David Lazar became a graduate student at MIT; multiple students
  have published papers in scientific conferences.)  You will
  get to scratch your own itch by creating tools that solve problems that
  frustrate you.  And, we have a lot of fun on this project!
</p>

<p>
  <b>Prerequisites:</b> You should be very comfortable with the Java
  programming language and its type system.  You should know how a type
  system helps you and where it can hinder you.  You should be willing to
  dive into and understand a moderately-sized codebase.
  You should understand fundamental object-oriented programming concepts,
  such as
  <a href="https://en.wikipedia.org/wiki/Liskov_substitution_principle">behavioral
  subtyping</a>:  subtyping theory
  permits argument types to change contravariantly (even though Java forbids it
  for reasons related to overloading), whereas return types may change
  <a href="https://en.wikipedia.org/wiki/Covariance_and_contravariance_%28computer_science%29">covariantly</a>
  both in theory and in Java.
</p>


<p>
<b>Potential projects:</b>
Most of this document lists potential summer projects.  The projects are
grouped roughly from easiest to most challenging.  Many of the projects are
applicable beyond Google Summer of Code.
<!--
You can find more potential projects in the
<a href="https://github.com/typetools/checker-framework/issues">issue
  tracker</a>.
-->
</p>



<h2 id="get-started">How to get started: do a case study</h2>

<p>
  To <b>get started</b>, first do a case study of using the Checker
  Framework.  Do this before submitting your proposal.
</p>
<ol>
  <li>
    <a href="https://checkerframework.org/manual/#installation">Install</a>
    the Checker Framework.
  <li>
    Choose a library or program to type-check.
    <br/>
    This should be an already-existing
    program; it should not be newly-written code, nor a set of tiny
    benchmarks that aim to expose corner cases.
    <!-- (Corner cases are great to
    discover, but they are only worth addressing if they come up in real code.)
    -->
    The library or program should be under active maintenance; don't choose one
    that has not had a commit in several years.
    You will find the case study easier if you are already familiar with
    the program, or if it is written in good style.
    <br/>
    If you need ideas of what program to annotate, you could
    see <a href="#library-annotations">library annotations</a> below.
    Don't choose part of the JDK as your first case study.
  <li>
    Choose one type system, from
    among <a href="https://checkerframework.org/manual/#introduction">those
    distributed with the Checker Framework</a>, that is appropriate for the
    program.
  <li>
    Annotate the program, based on its documentation.
    <br/>
    Please do <em>not</em> make changes unrelated to annotating the
    program, such as inserting/removing whitespace or sorting
    the <code>import</code> statements.  Doing so bloats the size of the
    diffs and makes it hard to understand the essential changes.
  <li>
    Run the type-checker.  If it issues
    warnings, <a href="https://checkerframework.org/manual/#handling-warnings">correct them</a>.
    This might require adding more annotations,
    fixing bugs in the program, or suppressing warnings.
    Be sure that the program's test suite continues to pass.
    Repeat until
    the type-checker passes on the program.
    <ul>
      <li>Don't add an <code>if</code> statement that always succeeds, just
        to suppress a warning.  Convince yourself that both branches can
        execute, or else don't add the <code>if</code> statement.
      <li>If you add a <code>@SuppressWarnings</code> annotation,
        <a href="https://checkerframework.org/manual/#suppresswarnings-best-practices-justification">explain
        why</a> the checker warning is a false positive and you are certain
        the code is safe.
    </ul>
  <li>
    Share it with us so that
    we can evaluate it and give you feedback.  (If the code compiles
    without type-checking warnings, and you didn't
    use <code>@SuppressWarnings</code>, then the annotations are correct,
    your program is correct, and you don't need feedback.  Congratulations!
    You can try a more significant case study.)

    <p>
    Share the case study as soon
    as you finish it or as soon as you have a question that is not answered
    in the <a href="https://checkerframework.org/manual/">manual</a>;
    don't wait until you submit your proposal.
    The subject line should be descriptive (not just "Case study", but
    "Nullness case study of Apache Commons Exec library").
    You should give us access to
    <ul>
      <li>the original (unannotated) version of the program,
      <li>the annotated version of the program, and
      <li>the exact command that runs the type-checker from the command
        line.
    </ul>
    The best way to give all this information is a pointer to your GitHub
    fork of the library.  In your fork, the master branch will be unchanged
    from upstream.  You will create a new branch that contains the
    annotated version.  The only differences between the two branches
    should be related to the annotations you wrote.  Building the annotated
    branch should run the type-checker.
    It is helpful to create a pull request from the annotated to the
    unannotated branch in your fork; you won't merge this pull request, but
    it enables you and others to view and comment on your changes.
</ol>

<p>
Once you have done this work on a small program such as from your
coursework, you can repeat the process with an open-source program or library.
</p>

<p>
The primary result of your case study is that you will discover bugs in the
subject program, or you will verify that it has no bugs (of some particular
type).  If you found bugs in open-source code, report them to the program's
maintainer, and let us know when they are resolved.  If you verified
open-source code to be correct, that is great too; let us know and point us
at the fully-annotated, verified program.
</p>

<p>
Another outcome of your case study is that you may discover bugs, limitations,
or usability problems in the Checker Framework.  Please
<a href="https://checkerframework.org/manual/#reporting-bugs">report them</a>.
We'll try to fix them, or they might give you inspiration for
improvements you would like to make to the Checker Framework this summer.
You can also try to fix them yourself and submit a
<a href="https://github.com/typetools/checker-framework/pulls">pull
  request</a>, but that is <em>not</em> a requirement.
You may discuss your ideas with us by sending mail
to <a href="https://groups.google.com/forum/#!forum/checker-framework-gsoc">checker-framework-gsoc@googlegroups.com</a>.
</p>

<p>
  Note that we do <em>not</em> recommend that you run many different checkers on
  small, artificial programs.  Instead, run one checker on a more
  substantial program.
  </p>

<p>
Why should you start with a case study, instead of diving right into fixing
bugs, designing a new type system, or making other changes to the Checker
Framework?
Before you can contribute to any project, you must
understand the tool from a user point of view, including its strengths,
weaknesses, and how to use it.  Therefore, you need to complete a
substantive case study first.
</p>


<h2 id="ask-questions">How to get help and ask questions</h2>

<p>
We are very happy to answer your questions, and we are eager to interact
with you.
Before you ask a question, read these &ldquo;getting started&rdquo;
instructions (that is, this file) and search in the
<a href="https://checkerframework.org/manual/">Checker Framework manual</a>
for the answer.
Don't send us a message
that says nothing but &ldquo;please guide me&rdquo;
or &ldquo;tell me how to fix this bug&rdquo;.  Such a message shows
that you haven't thought about the problem and haven't tried to solve it
yourself.  It also shows that you have not read this document, and we don't
want to work with people who cannot read instructions!
</p>

<p>
Your questions should show that you will be a productive colleague over the
summer:  tell us what you have tried, tell us what went wrong or
where you got stuck, and ask a concrete technical question that will
help you get past your problem.  If you can do that, then definitely ask
your question, because we don't want you to be stuck or frustrated.
</p>

<p>
Whenever you send email (related to GSoC or not),
please use standard email etiquette, such as:  avoid all-caps; use a
descriptive subject line; don't put multiple different topics in a single
email message; start a new thread with a new subject line
when you change the topic; don't clutter discussions with irrelevant
remarks; don't use screenshots (unless there is a problem with a GUI), but
instead cut-and-paste the code into your message; if you are making a guess, clearly indicate that it is a guess and
your grounds for it.  If you violate these basic rules, you will
look unprofessional, and we don't want you to give a bad impression.  Bug
reports should be
<a href="https://checkerframework.org/manual/#reporting-bugs">complete</a>
and should usually be
<a href="https://checkerframework.org/manual/#reporting-bugs">reported</a>
to the issue tracker.
</p>

<p>
Some GSOC projects have a requirement to fix an issue in the issue tracker.
We do not, because it is unproductive.
Don't try to start fixing issues before you
understand the Checker Framework from the user point of view, which will
not happen until you have completed a case study on an open-source program.
</p>



<h2 id="apply">How to apply</h2>

<p>
  To <b>apply</b>, you will submit a single PDF through the Google Summer
  of Code website.  This PDF should contain two main parts.  We suggest
  that you number the parts and subparts to ensure that you don't forget anything, and
  that we don't overlook anything in your application.  You might find it
  easiest to create multiple PDFs for the different parts, then concatenate
  them before uploading to the website, but how you create your proposal is
  entirely up to you.
</p>

<ol>
  <li>The proposal itself:  what project you want to work on during the
    summer.  You might propose to do a project listed on this webpage, or
    you might propose a different project.

      <p>The proposal should have a descriptive title, both in the PDF and
      in the GSoC submission system.  Don't use a title like "Checker
      Proposal" or "Proposal for GSoC".
      Don't distract from content with gratuitous graphics.

    <p>If you want to create a new type system (whether one proposed on
    this webpage or one of your own devising), then your proposal should be
    the type system's user manual.  You don't have to integrate it in the Checker
    Framework repository (in other words, use any word processor or text
    editor you want to create a PDF file you will submit), but you should describe
    your proposed checker's <a href="https://checkerframework.org/manual/#creating-parts-of-a-checker">parts</a>
    in precise English or simple formalisms and you should follow the
    suggested <a href="https://checkerframework.org/manual/#creating-documenting-a-checker">structure</a>.

    <p>List the tasks or subparts that are required to complete your
    project.  This will help you discover a part that you had forgotten.
    We do not require a detailed timeline, because at this point, you don't
    know enough to create one.
    </p>

    <p>Never literally cut-and-paste text that was not written by you, because
    that would be plagiarism.  If you quote from text written by someone
    else, give proper credit.</p>

    <p>
    If you want to do exactly what is already listed on this page, then
    just say that (but be specific about which one!), and it will not hurt
    your chances of being selected.  However, you might have specific ideas
    about extensions, about details that are not mentioned on this webpage,
    about implementation strategies, and so forth.  If you want to do a
    case study, say what program you will do your case study on.  Don't
    submit a proposal that is just a rearrangement of text that already
    appears on this page or in the Checker Framework manual, because it does
    not help us to assess your likelihood of being successful.  (You can
    propose an idea that's here, but show what progress you have made.)
    </p>
  </li>

  <li>Your qualifications.  Please convince us that you
    are likely to be successful in your proposed summer project.

    <ol>
      <li>A URL that points to a code sample.
        Don't write any new code, but provide code you wrote in the
        past, such as for a class assignment
        or a project you have worked on outside class.
        It does not need to have anything to do with
        the Checker Framework project.  It should be your own personal work.
        The purpose is to assess your programming skills so we can assign you
        to an appropriate project.
        A common problem is to submit undocumented code; we expect every
        programmer to write documentation when working on the Checker
        Framework.
        Don't put a lot of different files in Google Drive and share that
        URL; it's better to upload a single <code>.zip</code> file or
        provide a GitHub URL.
      </li>
      <li>
        What you have done to prepare yourself
        for working with the Checker Framework during the summer.
        You may wish to structure this as a list.
    Examples of items in the list include:
    <ul>
      <li>A URL for code you have annotated as a case study.  Please indicate the
    original unannotated code, the annotated code, and the exact command to
    run the type-checker from the command line.  Ensure that the GSoC
    mentors can compile your code.
    (It is acceptable to use the same code, or different code, for this
      item and the code sample above.)
      </li>
      <li>URLs for bugs or pull requests that you have filed.</li>
      <li>Information about other projects you have done, or classes you
        have taken, that prepare you for your proposed summer task.  This
        is optional, because it might already appear in your resume.</li>
    </ul>
  </li>
  <li>A resume.
    A <a href="https://en.wikipedia.org/wiki/R%C3%A9sum%C3%A9">resume</a>
    contains a brief description of your skills and your job or project
    experience.  It will often list classes you have taken so far and your
    GPA.  It should not be longer than one page.</li>
  <li>An unofficial transcript or grade report (don't spend
    money for an official one).</li>
  </ol>
  </li>
</ol>

<p>
The <b>best way</b> to impress us is by doing a thoughtful job in the case
study.  The case study is even more important than the proposal text,
because it shows us your abilities.
The case study may result in you submitting issues against the issue tracker of the
program you are annotating or of the Checker Framework.
Pull requests against our GitHub project are a plus but are not required:
good submitted bugs are just as valuable as bug fixes!
You can also make a good impression by correctly answering questions from
other students on the GSOC mailing list.
</p>

<p>
Get feedback!  Feel free to <a href="#ask-questions">ask questions</a>
to make your application more
competitive.  We want you to succeed.  Historically, students who start
early and get feedback are most successful.  You can submit a draft
proposal via the Google Summer of Code website, and we will review it.  We
do <em>not</em> receive any notification when you submit a draft
proposal, so if you want feedback, please tell us that.
Also, we can only see draft proposals; we cannot see final proposals until
after the application deadline has passed.
</p>


<h1 id="case-studies">Case studies</h1>

<p>
These projects take an existing type-checker, apply it to a
codebase (you can choose your favorite one, or you can ask for
suggestions), and determine whether the type system is easy to use and
whether it is effective in revealing or preventing defects.
Case studies are our most important source of new ideas and improvements:
our most useful features have arisen as a result of an observation made
during a case study.  Many people have started out &ldquo;just&rdquo; doing a case
study but have ended up making deep, fundamental contributions and even
publishing scientific papers about their discoveries.
</p>

<p>
You should do a small case study during the application process (or maybe a
large one, depending on your ambition).  A case study is the best way to
learn about the Checker Framework, determine whether you would enjoy
joining the project during the summer, and show your aptitude so that you
will be chosen for the summer.
</p>

<p>
A set of large case studies is one possible summer task.  The most common
choice is case studies of a recently-written type system, to determine its
usability.  Another choice is to annotate popular libraries for an existing
type system, to make it more usable.
</p>

<p>
Here are a few suggestions, but a case study of any type system
distributed with the Checker Framework is of value.
</p>


<h2 id="library-annotations">Library annotations</h2>

<p>
When type-checking a method call, the Checker Framework uses the method
declaration's annotations.
This means that in order to type-check code that uses a library, the
Checker Framework needs an annotated version of the library.
</p>

<p>
The Checker Framework comes with a
few <a href="https://search.maven.org/search?q=annotatedlib">annotated
libraries</a>.  Increasing this number will make the Checker Framework even
more useful, and easier to use.
</p>

<p>
After you have <a href="#choose-a-library">chosen a library</a>,
fork the library's source code, adjust
its <a href="https://checkerframework.org/manual/#external-tools">build
system</a> to run the Checker Framework, and add annotations to it until
the type-checker issues no warnings.
</p>

<p>
Before you get started, be sure to read
<a href="https://checkerframework.org/manual/#get-started-with-legacy-code">How
to get started annotating legacy code</a>.  More generally, read the
<a href="https://checkerframework.org/manual/#how-to-read-this-manual">relevant
sections of the Checker Framework manual</a>.
</p>


<h3 id="choose-a-library">Choosing a library to annotate</h3>

<p>
There are several ways to <b>choose a library</b> to annotate:
</p>
<ul>
  <li>
    Choose a <a href="https://docs.google.com/spreadsheets/d/17x_jKkGquEFq7LBQhS9HGXiG7iIl2AlXoPGfB6N5_bw">popular
      Java library</a>.
  </li>
  <li>
    Choose a library that you need annotations for.  That is, choose a
    library that is used by a program (or another library) that you have
    tried to annotate.
  </li>
  <li>
    Choose one of the suggestions below.
  </li>
</ul>

<p>
  Whatever library you choose, you will need to deeply understand its
  source code.  You will find it easier to work with a library that is
  well-designed and well-documented.
</p>

<p>
  You should choose a library that is
  not <a href="https://github.com/typetools/checker-framework/tree/master/checker/lib">already
    annotated</a>.  There are two exceptions to this.
</p>
<ul>
  <li>
    A library might be annotated for one type system, but you add
    annotations for a different type system.  One advantage of this is that
    the library's build system is already set up to run the Checker
    Framework.  You can tell which type systems a library is annotated for
    by examining its source code.
  </li>
  <li>
    A library might be annotated, but the annotations have not been
    verified by running the type-checker on the library source code.  You
    would verify that the annotations in the library are correct.
  </li>
</ul>


<h2 id="case-study-signature">Signature strings</h2>

<p>
Show that the <a href="https://asm.ow2.org/">ASM library</a>, or
the <a href="https://commons.apache.org/proper/commons-bcel/">BCEL
library</a>, properly handles signature strings (or find bugs in them).
</p>

<p>
  To get started:
</p>

<ul>
  <li>Fork https://github.com/typetools/commons-bcel.git</li>
  <li>Clone your new fork</li>
  <li><code>git checkout typecheck-signature</code></li>
  <li>mvn verify</li>
</ul>

<p>
   Some challenging aspects of this case study are:
</p>
<ul>
  <li>
    The Signature String Checker needs to be enhanced to precisely handle code like
        <code>someString.replace('.', '/')</code>
      which converts from <code>@ClassGetName</code> to <code>@FieldDescriptor</code>.  It also converts from <code>@FullyQualifiedName</code> to <code>@BinaryName</code>, but only for non-anonymous classes.
      The full rules for that, and for other calls such
      as <code>someString.replace('/', '.')</code>, need to be worked out
        and implemented.
  </li>
  <li>
    ASM and BCEL define their own new signature string formats (!), which need to be
    defined.  If they are not defined, then warnings need to be suppressed
    and the guarantee is weaker.  (Defining them could be a second part of
    the case study.)
  </li>
  <li>
    Some of ASM's and BCEL's documentation is incorrect, and in other cases the
    string format is not defined.
  </li>
  <li>
    The BCEL codebase has been partially annotated, but we are not sure the
    annotations are 100% correct.
  </li>
</ul>


<h2 id="case-study-android-support">Android support annotations</h2>

<p>
Android uses its own annotations that are similar to some in the Checker
Framework.  Examples include the
<a href="https://tips.seebrock3r.me/annotations-to-support-your-contracts-609ff259d5df">Android
  Studio support annotations</a>,
  including <code>@NonNull</code>, <code>@IntRange</code>, <code>@IntDef</code>,
  and others.
</p>

<p>
The goal of this project is to implement support for these annotations.
That is probably as simple as creating aliased annotations
by calling method <code>addAliasedAnnotation()</code>
in <a href="https://checkerframework.org/api/org/checkerframework/framework/type/AnnotatedTypeFactory.html">AnnotatedTypeFactory</a>.
</p>

<p>
  Then, do a case study to show the utility (or not) of
  pluggable type-checking, by comparison with how Android Studio currently
  checks the annotations.
</p>


<h2 id="case-study-signedness">Signed and unsigned numbers</h2>

<p>
  The <a href="https://checkerframework.org/manual/#signedness-checker">Signedness
  Checker</a> ensures that you do not misuse unsigned values, such as
  by mixing signed and unsigned values in a computation or by performing a
  meaningless operation.
</p>

<p>
  Perform a case study of the Signedness Checker, in order to detect errors
  or guarantee that code is correct.
</p>

<p>
  You will need to find Java projects that use unsigned arithmetic, or
  that <em>could</em> use unsigned arithmetic but do not.
  When doing the case study, it is important to type-check both a library and
  a client that uses it.  Type-checking the client will ensure that the
  library annotations are accurate.
</p>

<p>
  Here are some libraries that you could annotate (some are already
  annotated for you).  You would need to find client code that uses the
  signedness-sensitive routines.
</p>
<ul>
  <li>In the JDK's <code>Integer</code>
  and <code>Long</code>, these include
   <code>compareUnsigned</code>,
   <code>divideUnsigned</code>,
   <code>parseUnsignedInt</code>,
   <code>remainderUnsigned</code>, and
   <code>toUnsignedLong</code>.
   <br/>
   Classes like <code>DataInputStream</code>, <code>ObjectInputStream</code>,
   and <code>RandomAccessFile</code> have <code>readUnsignedByte</code>.
   <br/>
   <code>Arrays</code> has <code>compareUnsigned</code>.
  </li>
  <li>
    In Guava, see
  its <a href="https://github.com/google/guava/wiki/PrimitivesExplained#unsigned-support">unsigned
  support</a>, such
  as <a href="https://google.github.io/guava/releases/snapshot-jre/api/docs/com/google/common/primitives/UnsignedBytes.html">UnsignedBytes</a>,
  <a href="https://google.github.io/guava/releases/snapshot-jre/api/docs/com/google/common/primitives/UnsignedLong.html">UnsignedLong</a>,
  <a href="https://google.github.io/guava/releases/snapshot-jre/api/docs/com/google/common/primitives/UnsignedLongs.html">UnsignedLongs</a>,
  etc.
  </li>
  <li>The <a href="https://github.com/jOOQ/jOOU">jOOU</a> library consists of support for unsigned
      integers.</li>
</ul>

<p>
  Here are some other possible case studies; you would need to determine
  whether this code is the library, the client, or both:
</p>
<ul>
  <li><a href="https://github.com/bcgit/bc-java">bc-java</a>.  This is
  over 200 KLOC of code.</li>

  <!-- Not a good choice because this is just example code, not an actively-maintained project.
  <li>Project
  Nayuki: <a href="https://www.nayuki.io/page/forcing-a-files-crc-to-any-value">CRC</a>, <a href="https://www.nayuki.io/page/notepadcrypt-format-decryptor-java">crypt</a>, <a href="https://www.nayuki.io/page/native-hash-functions-for-java">hash</a>.</li>
  -->

  <!-- Not a good choice because it is not under active development:
  <li><a href="https://bytonic.de/html/jake2.html">Jake2</a></li>
   -->
</ul>

<p>
  Your case studies will show the need for enhancements to the Signedness
  Checker.  For example, the Signedness Checker does not currently handle
  boxed integers and BigInteger; these haven't yet come up in case studies
  but could be worthwhile enhancements.  There may also be the need to
  write more annotations for libraries such as the JDK.
</p>


<h2 id="optional-case-study">Java's Optional class</h2>

<p>
Java 8 introduced the
<a href="https://docs.oracle.com/en/java/javase/11/docs/api/java.base/java/util/Optional.html"><code>Optional</code></a>
class, a container that is either empty or contains a non-null value.
It is intended to solve the problem of null
pointer exceptions.  However, <code>Optional</code> has <a href="https://homes.cs.washington.edu/~mernst/advice/nothing-is-better-than-optional.html">its own problems</a>.
</p>

<p>
Because of <code>Optional</code>'s problems, many commentators advise programmers to use
<code>Optional</code> only in limited ways.
</p>

<p>
The goal of this project is to evaluate
the <a href="https://checkerframework.org/manual/#optional-checker">Optional
    Checker</a>, which warns programmers who
have misused <code>Optional</code>.
Another goal is to extend the Optional Checker to make it more precise or
to detect other mis-uses of Optional.
</p>


<h2 id="case-study-nullness">Null pointer exceptions</h2>

<p>
These are just some suggestions; many other libraries need annotations.
</p>


<h3 id="case-study-nullness-guava">Guava library</h3>

<p>
  Guava is already partially annotated with nullness annotations &mdash; in
  part by Guava's developers, and in part by the Checker Framework team.
  However, Guava does not yet type-check without errors.  Doing so could
  find more errors (the Checker Framework has found nullness and indexing
  errors in Guava in the past) and would be a good case study to learn the
  limitations of the Nullness Checker.
</p>


<h3 id="case-study-nullness-bazel">Bazel tool</h3>

<!-- John Field of Google is interested in this. -->

<p>
This project is related to the
<a href="https://bazel.build/">Bazel</a> build system, and was
proposed by its development manager.
</p>

<p>
The Bazel codebase contains 1586 occurrences of the <code>@Nullable</code>
annotation.  This annotation indicates that a variable may hold a null
value.  This is valuable documentation and helps programmers avoid null
pointer exceptions that would crash Bazel.  However, these annotations are
not checked by any tool.  Instead, programmers have to do their best to
obey the <code>@Nullable</code> specifications in the source code.  This is
a lost opportunity, since documentation is most useful when it is
automatically processed and verified.  (For several years, Google tried
using <a href="http://findbugs.sourceforge.net/">FindBugs</a>, but they
eventually abandoned it:  its analysis is too weak, suffering too many
false positives and false negatives.)
</p>

<p>
Despite the programmers' best efforts, null pointer exceptions do still
creep into the code, impacting users.  The Bazel developers would like to
prevent these.  They want a guarantee, at compile time, that no null
pointer exceptions will occur at run time.
</p>

<p>
Such a tool already exists:  the
<a href="https://checkerframework.org/manual/#nullness-checker">Nullness
Checker</a> of the <a href="https://checkerframework.org/">Checker
Framework</a>.  It runs as a compiler plug-in, and it issues a warning at
every possible null pointer dereference.  If it issues no warnings, the
code is guaranteed not to throw a <code>NullPointerException</code> at run time.
</p>

<p>
The goal of this project is to do a large-scale case study of the Nullness
Checker on Bazel.  The main goal is to understand how the Nullness Checker
can be used on a large-scale industrial codebase.  How many lurking bugs
does it find?  What
<a href="https://checkerframework.org/releases/1.9.13/api/org/checkerframework/checker/nullness/qual/Nullable.html"><code>@Nullable</code></a>
annotations are missing from the codebase because the developers failed to
write them?  What are its limitations, such as code patterns that it cannot
recognize as safe?  (You might create new analyses and incorporating them
into the Nullness Checker, or you might just reporting bugs to the Nullness
Checker developers for fixing.)  What burdens does it place on users?  Is
the cost-benefit tradeoff worth the effort &mdash; that is, should Google
adopt this tool more broadly?  How should it be improved?  Are the most
needed improvements in the precision of the analysis, or in the UI of the
tooling?
</p>


<h3 id="case-study-nullness-bcel">BCEL library</h3>

<p>
  Annotate the BCEL library to express its contracts with respect to nullness.
  Show that the BCEL library has no null pointer exceptions (or find bugs
  in BCEL).  There are
  already <a href="https://github.com/apache/commons-bcel/compare/trunk...typetools:trunk?expand=1">some
  annotations</a> in BCEL, but they have not been verified as correct by
  running the Nullness Checker on BCEL.  (Currently, those annotations are
  trusted when type-checking clients of BCEL.)
</p>

<p>
  To get started:
</p>

<ul>
  <li>Fork https://github.com/typetools/commons-bcel.git</li>
  <li>Clone your new fork</li>
  <li><code>git checkout typecheck-nullness</code></li>
  <li>mvn verify</li>
</ul>

<p>
   Some challenging aspects of this case study are:
</p>

<ul>
  <li>
    There is some poor design that needs to be resolved in discussions with
    the BCEL maintainers.  For example, consider the <code>copy()</code>
    method.  Some implementations of <code>copy()</code> return null, but
    are not documented to do so.  In addition, some implementations
    of <code>copy()</code> catch and ignore exceptions.  I think it would
    be nicest to change the methods to never return null, but to throw an
    exception instead.  (This is no more burdensome to users, who currently
    have to check for null.)  Alternately, the methods could all be
    documented to return null.
  </li>
</ul>


<h2 id="index-errors">Improving error messages</h2>

<p>
Compiler writers have come to realize that clarity of error
messages is as important as the speed of the executable
(<a href="http://www.brettbecker.com/wp-content/uploads/2016/06/Becker-Effective-2016-SIGCSE.pdf">1</a>, <a href="https://www.mville.edu/sites/default/files/p53-munson_1.pdf">2</a>,
<a href="http://se.ethz.ch/~meyer/publications/teaching/compiler-errors.pdf">3</a>
<a href="http://static.barik.net/barik/publications/icse2017/PID4655707.pdf">4</a>).  This is especially true when the language or type system has rich features.
</p>

<p>
The goal of this project is to improve a compiler's error messages.  One
example (not the only possible one) is the Checker Framework.  Here are
some distinct challenges:
</p>
<ul>
  <li>
    Some type errors can be more concisely or clearly expressed than the
    standard "found type A, expected type B" message.
  </li>
  <li>
    Some types are complex.  The error message could explain them, or link
    to the manual, or give suggested fixes.
  </li>
  <li>
    Compiler messages currently show
    the <a href="https://checkerframework.org/manual/#effective-qualifier">effective
    type</a>, which may be different than what the user wrote due to
    defaulting, inference, and syntactic sugar.  For example, a user-written
    <code>@IndexFor("a")</code> annotation is syntactic sugar for
    <code>@NonNegative @LTLengthOf("a")</code>, and those types are
    the ones that currently appear in error messages.
    It would be good to show simpler types or ones that the user wrote.
  </li>
  <li>
    Some checkers combine multiple cooperating type systems;
    the <a href="https://checkerframework.org/manual/#nullness-checker">Nullness
    Checker</a> and
    the <a href="https://checkerframework.org/manual/#index-checker">Index
    Checker</a> are examples.  If there is a problem with a variable's
    lower bound type, then its upper bound type should not be shown in the
    error message.  This will make the message shorter and more specific,
    and avoid distracting the user with irrelevant information.
  </li>
  <li>
    When a checker has multiple type systems, a type error or the lack of one may depend on facts from multiple type systems, and this should be expressed to the user.
  </li>
</ul>

<p>
It would be reasonable to start by improving the
<a href="https://checkerframework.org/manual/#index-checker">Index
  Checker</a>'s error messages, which frequently stymie users.  Then,
  generalize the results to other type systems.
</p>

<h2 id="sound-by-default">Sound checking by default</h2>

<p>
By default, the Checker Framework is
<a href="https://checkerframework.org/manual/#unsound-by-default">unsound
in several circumstances</a>.  &ldquo;Unsound&rdquo; means that the Checker Framework
may report no warning even though the program can misbehave at run time.
</p>

<p>
The reason that the Checker Framework is unsound is that we believe that
enabling these checks would cause too many false positive warnings:
warnings that the Checker Framework issues because it cannot prove that the
code is safe (even though a human can see that the code is safe).  Having
too many false positive warnings would irritate users and lead them not to
use the checker at all, or would force them to simply disable those checks.
</p>

<p>
We would like to do studies of these command-line options to see whether
our guess is right.  Is it prohibitive to enable sound checking?  Or can we
think of enhancements that would let us turn on those checks that are
currently disabled by default?
</p>


<h2 id="compare-other-tools">Comparison to other tools</h2>

<p>
  Many other tools exist for prevention of programming errors, such as
  Error Prone, NullAway, FindBugs, JLint, PMD, and IDEs such as Eclipse and
  IntelliJ.  These tools
  are not as powerful as the Checker Framework (some are bug finders rather
  than verification tools, and some perform a shallower analysis), but they
  may be easier to use.
  Programmers who use these tools wonder, "Is it worth my time to switch to
  using the Checker Framework?"
</p>

<p>
  The goal of this project is to perform a head-to-head comparison of as
  many different tools as possible.  You will quantify:
</p>

<ul>
  <li>the number of annotations that need to be written</li>
  <li>the number of bugs detected</li>
  <li>the number of bugs missed</li>
  <li>the number of false positive warnings</li>
</ul>

<p>
  This project will help programmers to choose among the different tools
  &mdash; it will show when a programmer should or should not use the
  Checker Framework.
  This project will also indicate how each tool should be improved.
</p>

<p>
  One place to start would be with an old version of a program that is
  known to contain bugs.  Or, start with the latest version of the program
  and re-introduce fixed bugs.  (Either of these is more realistic than
  introducing artificial bugs into the program.)  A possibility would be to
  use the Lookup program that has been used in previous case studies.
</p>


<h1 id="new-type-systems">New type systems</h1>

<p>
The Checker Framework is shipped with <a href="https://checkerframework.org/manual/#introduction">about 20 type-checkers</a>.  Users can
<a href="https://checkerframework.org/manual/#creating-a-checker">create a
  new checker</a> of their own.  However, some users don't want to go to
that trouble.  They would like to have more type-checkers packaged with the
Checker Framework for easy use.
</p>

<p>
Each of these projects requires you to design a <a href="https://checkerframework.org/manual/#creating-a-checker">new type system</a>,
implement it, and perform case studies to demonstrate that it is both
usable and effective in finding/preventing bugs.
</p>


<h2 id="non-empty-checker">Non-Empty Checker for precise handling of Queue.peek() and poll()</h2>

<p>
The Nullness Checker issues a false positive warning for this code:
</p>

<pre>
import java.util.PriorityQueue;
import org.checkerframework.checker.nullness.qual.NonNull;

public class MyClass {
    public static void usePriorityQueue(PriorityQueue&lt;@NonNull Object&gt; active) {
        while (!(active.isEmpty())) {
            @NonNull Object queueMinPathNode = active.peek();
        }
    }
}
</pre>

<p>
The Checker Framework does not determine that <code>active.peek()</code> returns a non-null value in this context.
</p>

<p>
The contract of <code>peek()</code> is that it returns a non-null value if the queue is not empty and the queue contains no null values.
</p>

<p>
To handle this code precisely, the Nullness Checker needs to know, for each queue, whether it is empty.
This is analogous to how the Nullness Checker tracks whether a particular value is a key in a map.
</p>

<p>
It should be handled the same way:  by adding a new subchecker, called the
  Nonempty Checker, to the Nullness Checker.  Its types are:
</p>
<ul>
<li><code>@UnknownNonEmpty</code> &mdash; the queue might or might not be empty
<li><code>@NonEmpty</code> &mdash; the queue is definitely non-empty
</ul>

<p>
There is a start at this type-checker in branch <code>nonempty-checker</code>.  It:
</p>
<ul>
  <li>defines the annotations
  </li>
  <li>creates the integration into the Nullness Checker
  </li>
</ul>
<p>
However, it is not done.  (In fact, it doesn't even compile.)
For information about what needs to be done, see <a href="https://github.com/typetools/checker-framework/issues/399">issue #399</a>.
</p>

<p>
When you are done, the Nullness Checker should issue only the <code>// ::</code> diagnostics from <code>checker/tests/nullness/IsEmptyPoll.java</code> &mdash; no more and no fewer.
You can test that by running the Nullness Checker on the file, and when you are done you should delete the <code>// @skip-test</code> line so that the file is run as part of the Checker Framework test suite.
</p>


<h2 id="Make_programs_deterministic">Make programs deterministic</h2>

<p>
Programs are easier to use and debug if their output is deterministic.  For
example, it is easier to test a deterministic program, because
nondeterminism can lead to flaky tests that sometimes succeed and sometimes
fail.  As another example, it is easier for a user or programmer to compare
two deterministic executions than two nondeterministic executions.
</p>

<p>
A number of Java methods return nondeterministic results, making any
program that uses them potentially nondeterministic.  Here are a few
examples:
</p>
<dl>
<dt>iteration over <code>HashMap</code>s and <code>HashSet</code>s</dt>
<dd>solution:  sort, or iterate over <code>LinkedHashMaps</code> and <code>LinkedHashSets</code></dd>
<dt><code>File.list()</code></dt>
<dd>solution:  sort the resulting list</dd>
<dt><code>Object.toString()</code>, <code>Object.hashCode()</code></dt>
<dd>solution:  use overriding implementations</dd>
<dt><code>new Random()</code></dt>
<dd>solution:  call it with a fixed seed</dd>
<dt>time and date functions</dt>
<dd>solution:  don't make the program result depend on these</dd>
</dl>

<p>
You can find more examples of non-deterministic specifications,
and suggestions for how to avoid them, in the
<a href="https://randoop.github.io/randoop/manual/index.html#nondeterminism">Randoop
manual</a> and in the ICST 2016 paper
<a href="http://mir.cs.illinois.edu/marinov/publications/ShiETAL16NonDex.pdf">Detecting
assumptions on deterministic implementations of non-deterministic
specifications</a> by A. Shi, A. Gyori, O. Legunsen, and D. Marinov, which
presents the NonDex tool.
</p>

<p>
The NonDex tool works dynamically, which means that it cannot detect all
user-visible nondeterminism nor give a guarantee of correctness &mdash; a
guarantee that the program is deterministic from the user's point of
view.
</p>

<p>
The goal of this project is to create a tool, based on a type system, that
gives a guarantee.  The tool would report to the user all possible
nondeterminism in a program, so that the user can fix the program before it
causes problems during testing or in the field.
</p>

<p>
More concretely, this problem can be handled by creating two simple type
systems that indicate whether a given value is deterministic.
In each diagram, the supertype appears above the subtype.
</p>

<pre>
    @PossiblyNonDeterministic            @PossiblyNonDeterministicOrder
              |                                       |
        @Deterministic                       @DeterministicOrder
</pre>

<p>
The programmer would annotate routines that are expected to take
deterministic inputs.  (An example could be all printing routines.)  Then,
the type system would issue a warning whenever one of those routines is
called on a possibly non-deterministic value.
</p>

<p>
The standard library would have annotations for
</p>
<ul>
  <li>methods that return non-deterministic results</li>
  <li>methods like sort that convert nondeterministically-ordered collections into deterministically-ordered ones</li>
</ul>

<p>
You can find
a <a href="https://checkerframework.org/determinism-checker-manual/manual.html#determinism-checker">draft
manual chapter</a> that documents a possible design for a Determinism
Checker.  It differs slightly from the above proposal, for instance by
having a single type hierarchy instead of two.  That type system
is <a href="https://github.com/t-rasmud/checker-framework/tree/nondet-checker">implemented</a>,
so your best choice for this project is to do a case study of it, which
could lead to design work to improve it.
</p>


<h2 id="custom-tainting-checking">Custom tainting checking</h2>

<p>
The Checker Framework comes with
a <a href="https://checkerframework.org/manual/#tainting-checker">Tainting
  Checker</a> that is so general that it is not good for much of anything.
In order to be useful in a particular domain, a user must customize it:
</p>
<ul>
  <li>
    rename the <code>@Tainted</code> and <code>@Untainted</code> qualifiers
    to something more specific (such as <code>@Private</code>
    or <code>@PaymentDetails</code> or <code>@HtmlQuoted</code>), and
  <li>
    annotate libraries.
</ul>

<p>
The first part of this project is to make this customization easier to do
&mdash; preferably, a user does not have to change any code in the Checker
Framework, as is currently the case for the
<a href="https://checkerframework.org/manual/#subtyping-checker">Subtyping
Checker</a>.
As part of making customization easier, a user should be able to specify
multiple levels of taint &mdash; many information classification hierarchies
have more than two levels (for example, the US government separates classified
information into three categories: Confidential, Secret, and Top Secret).
</p>

<p>
The second part of this project is to provide several examples, and do case
studies showing the utility of compile-time taint checking.
</p>

<p>
  Possible examples include:
</p>
<ul>
  <li>SQL injection
  <li>OS command injection
  <li>the <code>@PrivacySource</code> and <code>@PrivacySink</code>
    annotations used by the Facebook <a href="http://fbinfer.com/">Infer
    static analyzer</a>.
  <li>information flow
  <li>many of the <a href="http://cwe.mitre.org/top25/">CWE/SANS most
  dangerous software programming errors</a> (and the "on the cusp" ones too)
  <!-- More details appear in these files:
      ~/research/games/notes/notes
      ~/prof/grants/2012-02-darpa-verigames/proposal/top25-as-types.pdf
  -->
</ul>

<p>
For some microbenchmarks, see the Juliette test suite for Java from CWE.
</p>




<h2 id="Bounded-size_strings">Bounded-size strings</h2>

<!-- John Field of Google is interested in this. -->

<p>
Windows cannot run command lines longer than 8191 characters.  Creating a
too-long command line causes failures when the program is run on Windows.
These failures are irritating when discovered during testing, and
embarrassing or worse when discovered during deployment.  The same command
line would work on Unix, which has longer command-line limits, and as a
result developers may not realize that their change to a command can cause
such a problem.
</p>

<p>
Programmers would like to enforce that they don't accidentally pass a
too-long string to the <code>exec()</code> routine.  The goal of this
project is to give a compile-time tool that provides such a guarantee.
</p>

<p>
Here are two possible solutions.
</p>

<p>
<b>Simple solution:</b>
For each array and list, determine whether its length is known at compile
time.  The routines that build a command line are only allowed to take such
constant-length lists, on the assumption that if the length is constant,
its concatenation is probably short enough.
</p>

<p>
<b>More complex solution:</b>
For each String, have a compile-time estimate of its maximum length.  Only
permit <code>exec()</code> to be called on strings whose estimate is no more than 8191.
String concatenation would return a string whose estimated size is the sum
of the maximums of its arguments, and likewise for concatenating an array
or list of strings.
</p>


<h2 id="overflow">Overflow checking</h2>

<p>
Overflow is when 32-bit arithmetic differs from ideal arithmetic.  For
example, in Java the <code>int</code> computation 2,147,483,647 + 1 yields
a negative number, -2,147,483,648.  The goal of this project is to detect
and prevent problems such as these.
</p>

<p>
One way to write this is as an extension of the Constant Value Checker,
which already keeps track of integer ranges.  It even already
<a href="https://checkerframework.org/manual/#value-checker-overflow">checks
  for overflow</a>, but it never issues a warning when it discovers
  possible overflow.  Your variant would do so.
</p>

<p>
This problem is so challenging that there has been almost no previous
research on static approaches to the problem.  (Two relevant papers are
<a href="https://www.researchgate.net/publication/221655385_IntScope_Automatically_Detecting_Integer_Overflow_Vulnerability_in_X86_Binary_Using_Symbolic_Execution">IntScope:
Automatically Detecting Integer Overflow Vulnerability in x86 Binary Using
Symbolic Execution</a> and
<a href="https://dl.acm.org/citation.cfm?id=3136872">Integer Overflow
Vulnerabilities Detection in Software Binary Code</a>.)  Researchers are
concerned that users will have to write a lot of annotations indicating the
possible ranges of variables, and that even so there will be a lot of false
positive warnings due to approximations in the conservative analysis.
For example, will every loop that contains <code>i++</code> cause a warning that <code>i</code> might overflow?
That would not be acceptable:  users would just disable the check.
</p>

<p>
You can convince yourself of the difficulty by manually analyzing programs
to see how clever the analysis has to be, or manually simulating your
proposed analysis on a selection of real-world code to learn its
weaknesses.  You might also try it
on <a href="https://ai.googleblog.com/2006/06/extra-extra-read-all-about-it-nearly.html">good
  and bad binary search code</a>.
</p>

<p>
One way to make the problem tractable is to limit its scope:  instead of
being concerned with all possible arithmetic overflow, focus on a specific
use case.
As one concrete application,
the <a href="https://checkerframework.org/manual/#index-checker">Index
Checker</a> is currently unsound in the presence of integer overflow.  If
an integer <code>i</code> is known to be <code>@Positive</code>, and 1 is
added to it, then the Index Checker believes that its type
remains <code>@Positive</code>. If <code>i</code> was
already <code>Integer.MAX_VALUE</code>, then the result is negative &mdash;
that is, the Index Checker's approximation to it is unsound.
</p>

<p>
This project involves removing this unsoundness by implementing a type system to track when an
integer value might overflow &mdash but this only matters for values that
are used as an array index.
That is, checking can be restricted to computations that involve an operand
of type <code>@IntRange</code>).
Implementing such an analysis would permit the Index Checker
to extend its guarantees even to programs that might overflow.
</p>

<p>
This analysis is important for some indexing bugs in practice.
Using the Index Checker, we found 5 bugs in Google
Guava related to overflow.  Google marked these as high priority and
fixed them immediately.  In practice, there would be a run-time exception
only for an array of size approximately <code>Integer.MAX_INT</code>.
</p>

<p>
You could write an extension of the Constant Value Checker, which already
keeps track of integer ranges and
even <a href="https://checkerframework.org/manual/#value-checker-overflow">determines
when overflow is possible</a>.  It doesn't issue a warning, but your
checker could record whether overflow was possible (this could be a
two-element type system) and then issue a warning, if the value is used as
an array index.
Other implementation strategies may be possible.
</p>

<p>
Here are some ideas for how to avoid the specific problem
of issuing a warning about potential overflow for every <code>i++</code> in
a loop (but maybe other approaches are possible):
</p>
<ul>
  <li>
    The loop checks whether <code>i == Integer.MAX_VALUE</code> before
    incrementing.  This wide-scale, disruptive code change is not
    acceptable.
  </li>
  <li>
    Make the default array size (the length of an unannotated array) be
    <code>@ArrayLenRange(0, Integer.MAX_VALUE-1)</code> rather
    than <code>@UnknownVal</code>, which is equivalent
    to <code>@ArrayLenRange(0, Integer.MAX_VALUE-1)</code>.  Now, every
    array construction requires the client to establish that the length is
    not <code>Integer.MAX_VALUE</code>.  I don't have a feel for whether
    this would be unduly burdensome to users.
  </li>
</ul>


<h2 id="lock-ordering">Lock ordering</h2>

<p>
The <a href="https://checkerframework.org/manual/#lock-checker">Lock
Checker</a> prevents race conditions by ensuring that locks are held when
they need to be.  It does not prevent deadlocks that can result from locks
being acquired in the wrong order.  This project would extend the Lock
Checker to address deadlocks, or create a new checker to do so.
</p>

<p>
Suppose that a program contains two different locks.  Suppose that one
thread tries to acquire lockA then lockB, and another thread tries to
acquire lockB then lockA, and each thread acquires its first lock.  Then
both locks will wait forever for the other lock to become available.  The
program will not make any more progress and is said to
be <a href="https://en.wikipedia.org/wiki/Deadlock">deadlocked</a>.
</p>

<p>
If all threads acquire locks in the same order &mdash; in our example, say
lockA then lockB &mdash; then deadlocks do not happen.  You will extend the
Lock Checker to verify this property.
</p>


<h2 id="index-checker-mutable-length">Index checking for mutable length data structures</h2>

<p>
The <a href="https://checkerframework.org/manual/#index-checker">Index
Checker</a> is currently restricted to fixed-size data structures.
A fixed-size data structure is one whose length cannot be changed once it is created;
examples of fixed-size data structures are arrays and Strings.
This limitation prevents the Index Checker from verifying indexing operations on mutable-size data structures,
like Lists, that have add or remove methods. Since these kind of collections are common
in practice, this is a severe limitation for the Index Checker.
</p>

<p>
The limitation is caused by the Index Checker's use of types that are dependent on the length of data structures,
like <code>@LTLengthOf("data_structure")</code>. If <code>data_structure</code>'s length could change,
then the correctness of this type might change.
</p>

<p>
A naive solution would be to invalidate these types any time a method is called on <code>data_structure</code>.
Unfortunately, aliasing makes this still unsound. Even more, a great solution to this problem would keep
the information in the type when a method like add or remove is called on <code>data_structure</code>.
A more complete solution might involve some special annotations on List that permit the information to be persisted.
</p>

<p>
This project would involve designing and implementing a solution to this problem.
</p>


<h2 id="nullness-bug-detector">Nullness bug detector</h2>

<p>
Verifying a program to be free of errors can be a daunting task.  When
starting out, a user may be more interested in
<a href="https://checkerframework.org/manual/#other-tools">bug-finding</a>
than verification.  The goal of this project is to create a nullness bug
detector that uses the powerful analysis of the Checker Framework and its
Nullness Checker, but omits some of its more confusing or expensive
features.  The goal is to create a fast, easy-to-use bug detector.  It
would enable users to start small and advance to full verification in the
future, rather than having to start out doing full verification.
</p>

<p>
This could be structured as a new NullnessLight Checker, or as a
command-line argument to the current Nullness Checker.  Here are some
differences from the real Nullness checker:
</p>
<ul>
  <li>No initialization analysis; the checker assumes that every value is
  initialized.</li>
  <li>No map key analysis; assume that, at every call to
    <code>Map.get</code>, the given key appears in the map.</li>
  <li>No invalidation of dataflow facts.  Assume all method calls are pure,
  so method calls do not invalidate dataflow facts.  Assume there is no
    aliasing, so field updates do not invalidate dataflow facts.
    </li>
  <li>Assume that boxing of primitives is <code>@Pure</code>: it returns
  the same value on every call.</li>
  <li>If the Checker Framework cannot infer a type argument, assume that
  the type argument is <code>@NonNull</code>.</li>
</ul>
<p>
  Each of these behaviors should be controlled by its own command-line
  argument, as well as being enabled in the NullnessLight Checker.
</p>

<p>
  The implementation may be relatively straightforward, since in most cases
  the behavior is just to disable some functionality of existing checkers.
</p>

<p>Tools such as FindBugs, NullAway, NullnessLight, and the Nullness
  Checker form a spectrum from easy-to-use bug detectors to sound
  verification.  NullnessLight represents a new point in the design space.
  It will be interesting to compare these checkers:
</p>

<ul>
    <li>How much easier is it to use?  For example, how many fewer
      annotations need to be written?</li>
    <li>
      How many more fewer true positives does it report &mdash; in other
      words, how many more false negatives does it suffer?
    </li>
    <li>
      How many fewer false positives does it report?
    </li>
</ul>

<p>
  Uber's <a href="https://github.com/uber/NullAway">NullAway</a> tool is also
  an implementation of this idea (that is, a fast, but incomplete and
  unsound, nullness checker).  NullAway doesn't let the user specify Java
  Generics:  it assumes that every type parameter is <code>@NonNull</code>.
  Does Uber's tool provide users a good
  introduction to the ideas that a user can use to transition to a nullness
  type system later?
</p>



<h2 id="typestate">Stateful type systems</h2>

<p>
This project is to improve support for
<a href="https://checkerframework.org/manual/#typestate-checker">typestate checking</a>
</p>

<p>
Ordinarily, a program variable has
the same type throughout its lifetime from when the variable is declared
until it goes out of scope. “Typestate”
permits the type of an object or variable to <em>change</em> in a controlled way.
Essentially, it is a combination of standard type systems with dataflow
analysis. For instance, a file object changes from unopened, to opened, to
closed; certain operations such as writing to the file are only permitted
when the file is in the opened typestate. Another way of saying this is
that <code>write</code> is permitted after <code>open</code>, but not after <code>close</code>.
Typestate
is applicable to many other types of software properties as well.
</p>

<p>
Two <a href="https://checkerframework.org/manual/">typestate checking frameworks</a>
exist for the Checker Framework.  Neither is being maintained; a new one
needs to be written.
</p>


<h2 id="your-own-new-type-system">Invent your own new type system</h2>

<p>
We also welcome your ideas for new type systems.  For example, any run-time
failure can probably be prevented at compile time with the right
analysis.  Can you come up with a way to fix your pet peeve?
</p>

<p>
It is easiest, but not required, to choose an existing type system from the
literature, since that means you can skip the design stage and go right to
implementation.
</p>

<p>
This task can be simple or very
challenging, depending on how ambitious the type system is.  Remember to
focus on what helps a software developer most!
</p>



<h1 id="cf-other">Type system enhancements and tools</h1>


<h2 id="annotated-jdk">JDK annotations</h2>

<p>
The JDK is the most important library to annotate, and the Checker
Framework ships with partial annotations for it.
These are scattered in multiple locations:  in multiple subdirectories of
<a href="https://github.com/typetools/checker-framework/tree/master/checker/jdk"><code>checker/jdk/</code></a>,
and in files <code>jdk.astub</code> under
<a href="https://github.com/typetools/checker-framework/tree/master/checker/src/org/checkerframework/checker"><code>checker/src/org/checkerframework/checker/</code></a>
</p>

<p>
The goal of this project is to put the annotations in a single place:  a
new clone of the JDK repository.  The effort can be partially annotated
&mdash; say, by using
the <a href="https://checkerframework.org/annotation-file-utilities/">Annotation
File Utilities</a> to move annotations around, or enhancing the Annotation
File Utilities if they are not up to the task) and partially done manually.
</p>


<h2 id="flow-expression-parser">Flow expression parser</h2>

<p>
A number of type annotations take, as an
argument, <a href="https://checkerframework.org/manual/#java-expressions-as-arguments">a
Java expression</a>.  The parser for these is a hack.  The goal of this
project is to replace it by calls
to <a href="https://github.com/javaparser/javaparser">JavaParser</a>.
For example, the <code>FlowExpressions.Receiver</code> class, which
represents an AST, should be replaced by the JavaParser AST.
</p>

<p>
This task
should be straightforward, since JavaParser is already used in other parts
of the Checker Framework.
</p>


<h2 id="asm">Upgrade to a newer version of ASM</h2>

<p>
  The
  <a href="https://checkerframework.org/annotation-file-utilities/">Annotation
  File Utilities</a>, or AFU, insert annotations into, and extract
  annotations from, <code>.java</code> files, <code>.class</code> files,
  and text files.  These programs were written before the
  <a href="https://asm.ow2.org/">ASM</a> bytecode library supported Java 8's
  type annotations.  Therefore, the AFU has its own custom version of ASM
  that supports type annotations.  Now that ASM 6 has been released and it
  supports type annotations, the AFU needs to be slightly changed to use
  the official ASM 6 library instead of its own custom ASM variant.
</p>

<p>
  This project is a good way to learn about <code>.class</code> files and
  Java bytecodes:  how they are stored, and how to manipulate them.
</p>


<h2 id="analysis_diffs">Tool for analysis diffs</h2>
<!--
  This project idea is duplicated at
  ~mernst/public_html/uw-only/research/potential-research-projects.html .
-->

<p>
   Many program analyses are too verbose for a person to read their entire
   output.  However, after a program change, the analysis results may
   change only slightly.  An "analysis diff" tool could show the
   difference between the analysis run on the old code and the analysis run
   on the new code.
</p>
<ul>
  <li>
    The analysis diffs may help the programmer to better understand the
    changes.
  </li>
  <li>
    Bug detection tools. such
    as <a href="http://findbugs.sourceforge.net/">FindBugs</a> or
    the <a href="https://checkerframework.org/">Checker Framework</a>, have
    extremely verbose output when first run on a program.  Programmers
    could examine and fix only the warnings about code they have changed
    (and that they are currently thinking about).
  </li>
  <li>
    Tools that always have large output, such as inference tools, could
    become manageable to users if output is shown in small doses.
  </li>
  <li>
    You can probably think of other uses.
  </li>
</ul>

<p>
  The analysis diff tool would take as input two analysis results (the
  previous and the current one).  It would output only the new parts of its
  second input.  (It could optionally output a complete diff between two
  analysis results.)
</p>

<p>
  One challenge is dealing with changed line numbers and other analysis
  output differences between runs.
</p>

<p>
It would be nice to integrate the tool with git pre-commit hooks or GitHub
pull requests, to enable either of the following functionality (for either
commits to master or for pull requests):
</p>
<ul>
  <li>
    Permit only those commits/pulls that do not add any new analysis warnings.
  </li>
  <li>
    Permit only those commits/pulls that are "clean" &mdash; the analysis
    issues no warnings for any changed line.
  </li>
</ul>

<p>
  A concrete example of an analysis diff tool
  is <a href="https://github.com/plume-lib/checklink/blob/master/checklink-persistent-errors">checklink-persistent-errors</a>;
  see the documentation at the top of the file.  That tool only works for
  one particular analysis, the W3C Link Checker.
  An analysis diff tool also appears to be built into FindBugs.
  The goal of this project is to build a general-purpose tool that is easy
  to apply to new analyses.
</p>


<h2 id="Whole-program_type_inference">Whole-program type inference</h2>

<p>
A type system is useful because it prevents certain errors.  The downside
of a type system is the effort required to write the types.  Type inference
is the process of writing the types for a program.
</p>

<p>
Type-<em>checking</em> is a modular, or local, analysis.  For example, given a
procedure in which types have been written, a type-checker can verify the
procedure's types without examining the implementation of any other
procedure.
</p>

<p>
By contrast, type <em>inference</em> is a non-local, whole-program analysis.  For
example, to determine what type should be written for a procedure's formal
parameter, it is necessary to examine the type of the argument at every
call to that procedure.  At every call, to determine the type of some argument A, it
may be necessary to know the types of the formal parameters to the
procedure that contains A, and so forth.  It is possible to resolve this
seemingly-infinite regress, but only by examining the entire program in the
worst case.
</p>

<p>
The differences between type checking and type inference means that they
are usually written in very different ways.  Type inference is usually done
by first collecting all of the constraints for the entire program, then
passing them to a specialized solver.  Writing a type inference tool is
harder.  Worst of all, it's annoying to encode all the type rules twice in
different ways:  once for the type checker and once for the type inference.
</p>

<p>
As a result, many type systems have a type checker but no type inference
tool.  This makes programmers reluctant to use these type systems, which denies
programmers the benefits of type-checking.
</p>

<p>
The goal of this project is to automatically create type inference tools
from type-checking tools, so that it is not necessary for the type system
designer to implement the type system twice in different ways.
</p>

<p>
A key insight is that the type-checker already encodes all knowledge about
what is a legal, well-typed program.  How can we exploit that for the
purpose of type inference as well as type-checking?  The idea is to
iteratively run the type-checker, multiple times, observing what types are
passed around the program and what errors occur.  Each iteration collects
more information, until there is nothing more to learn.
</p>

<p>
This approach has some disadvantages:  it is theoretically slower, and
theoretically less accurate, than a purpose-built type inference tool for
each type system.  However, it has the major advantage that it requires no
extra work to implement a type inference tool.  Furthermore, maybe it works
well enough in practice.
</p>

<p>
A prototype implementation of this idea already exists, but it needs to be
evaluated in order to discover its flaws, improve its design, and discover
how accurate it is in practice.
</p>

<p>
A valuable project all by itself would be to compare heavy-weight and
light-weight type inference this whole-program inference vs. Checker
Framework Inference vs. Julia, to understand when each one is worth using.
</p>


<h2 id="dataflow">Dataflow enhancements</h2>

<p>
The Checker
Framework's <a href="https://checkerframework.org/manual/#creating-dataflow">dataflow
    framework</a>
(<a href="https://checkerframework.org/manual/checker-framework-dataflow-manual.pdf">manual
    here</a>) implements flow-sensitive type refinement (local type
inference) and other features.  It is used in the Checker
Framework and also in <a href="http://errorprone.info/">Error Prone</a>,
<a href="https://github.com/uber/NullAway">NullAway</a>, and elsewhere.
</p>

<p>
There are a number
of <a href="https://github.com/typetools/checker-framework/issues?q=is%3Aopen+is%3Aissue+label%3ADataflow">open
issues</a> &mdash; both bugs and feature requests &mdash; related to the
dataflow framework.  The goal of this project is to address as many of
those issues as possible, which will directly improve all the tools that
use it.
</p>


<h2 id="Purity_analysis">Purity (side effect) analysis</h2>

<p>
A program analysis technique makes estimates about the current values of
expressions.  When a method call occurs, the analysis has to throw away
most of its estimates, because the method call might change any variable.
If the method is known to have no side effects, then the analysis doesn't
need to throw away its estimates, and the analysis is more precise.
</p>

<p>
For example, the Checker Framework unsoundly trusts but does not check
<a href="https://checkerframework.org/manual/#type-refinement-purity">purity annotations</a>.  This makes
the system vulnerable to programmer mistakes when writing annotations.  The
Checker Framework contains a sound checker for immutability annotations,
but it suffers too many false positive warnings and thus is not usable.  A
better checker is necessary.  It will also incorporate aspects of an escape
analysis.
</p>

<p>
Choosing an algorithm from the literature is the best choice, but there
still might be research work to do:  in the past, when implementing
algorithms from research papers, we have sometimes found that they did not
work as well as claimed, and we have had to enhance them.  One challenge is
that any technique used by pluggable type-checking to verify immutability
must be modular, but many side effect analyses require examining the whole
program.  The system should require few or no method annotations within
method bodies.  I'm not sure whether such a system already exists or we
need to design a new one.
</p>

<p>
Perhaps one of these existing side effect analyses could be used:
 https://github.com/Sable/soot/wiki/Using-Side-Effect-Attributes
 http://www2.informatik.uni-freiburg.de/~geffken/GeffkenST14.pdf
</p>

<!--
For design and implementation ideas, see:
 * $qn/notes-purity
 * ~/prof/grants/2011-09-darpa-apac/phase3-extension-proposal
-->


<h2 id="javadoc">Javadoc support</h2>

<p>
Currently, type annotations are only displayed in Javadoc if they are
explicitly written by the programmer.  However, the Checker Framework
provides flexible defaulting mechanisms, reducing the annotation overhead.
This project will integrate the Checker Framework defaulting phase with
Javadoc, showing the signatures after applying defaulting rules.
</p>

<p>
There are other type-annotation-related improvements to Javadoc that can be
explored, e.g. using JavaScript to show or hide only the type annotations
currently of interest.
</p>


<h2 id="performance">Performance improvements</h2>

<p>
  The Checker Framework runs much slower than the standard javac compiler
  &mdash; often 20 times slower!  This is not acceptable as part of a
  developer's regular process, so we need to speed up the Checker
  Framework.  This project involves determining the cause of slowness in
  the Checker Framework, and correcting those problems.
</p>

<p>
This is a good way to learn about performance tuning for Java applications.
</p>

<p>
  Some concrete tasks include:
</p>

<ul>
  <li>Profile the Checker Framework.  Run a profiler such as
  <a href="https://www.yourkit.com/java/profiler/">YourKit</a> to determine
  which parts of the Checker Framework consume the most CPU time and memory.
  </li>

  <li>Perhaps compare a profile of the Checker Framework against a profile
  of regular javac.  This probably is not necessary because the Checker
    Framework is so much slower than regular javac.
  </li>

  <li>Consider interning string values.  The Checker Framework does a fair
  amount of string manipulation, in part because it reads from resources
  such as stub files that do not produce <code>Element</code>s.  Interning
  could save time when doing comparisons.  You can verify the correctness
  of the optimization by running the
  <a href="https://checkerframework.org/manual/#interning-checker">Interning
    Checker</a> on the Checker Framework code.  Compare the run time of the
  Checker Framework before and after this optimization.
  </li>

  <li> Based on profiling results, devise other optimizations, implement
    them, and evaluate them.
  </li>
</ul>


<h2 id="run-time-checking">Run-time checking</h2>

<p>
Implement run-time checking to complement compile-time checking.  This will
let users combine the power of static checking with that of dynamic
checking.
</p>

<p>
Every type system is too strict: it rejects some programs that never go
wrong at run time. A human must insert a type loophole to make such a
program type-check.  For example, Java takes this approach with its
cast operation (and in some other places).
</p>

<p>
When doing type-checking, it is desirable to automatically insert run-time
checks at each operation that the static checker was unable to verify.
(Again, Java takes exactly this approach.)  This guards against mistakes by
the human who inserted the type loopholes.  A nice property of this
approach is that it enables you to prevent errors in a program
with no type annotations:  whenever the static checker is unable
to verify an operation, it would insert a dynamic check.  Run-time checking
would also be useful in verifying whether the suppressed warnings are
correct &mdash; whether the programmer made a mistake when writing them.
</p>

<p>
The annotation processor (the pluggable type-checker) should automatically
insert the checks, as part of the compilation process.
</p>

<p>
There should be various modes for the run-time checks:
</p>
<ul>
  <li>fail immediately.</li>
  <li>logging, to permit post-mortem debugging without crashing the program.</li>
</ul>

<p>
The run-time penalty should be small:  a run-time check is necessary only
at the location of each cast or suppressed warning.  Everywhere that the
compile-time checker reports no possible error, there is no need to insert a
check.  But, it will be an interesting project to determine how to minimize
the run-time cost.
</p>

<p>
Another interesting, and more challenging, design question is whether you need to add and maintain
a run-time representation of the property being tested.  It's easy to test
whether a particular value is null, but how do you test whether it is
tainted, or should be treated as immutable?  For a more concrete example,
see the discussion of the (not yet implemented)
[Javari run-time checker](http://pag.csail.mit.edu/pubs/ref-immutability-oopsla2005-abstract.html).
Adding this run-time support would be an interesting and challenging project.
</p>

<p>
We developed a prototype for the
<a href="https://ece.uwaterloo.ca/~wdietl/publications/pubs/EnerJ11-abstract.html">EnerJ runtime system</a>.
That code could be used as starting point, or you could start afresh.
</p>

<p>
In the short term, this could be prototyped as a source- or
bytecode-rewriting approach; but integrating it into the type checker is a
better long-term implementation strategy.
</p>



<h2 id="ide-support">IDE and build system support</h2>

<p>
The Checker Framework comes with support for
<a href="https://checkerframework.org/manual/#external-tools">external tools</a>,
including both IDEs (such as an Eclipse plug-in) and build tools
(instructions for Maven, etc.).
</p>

<p>
  These plug-ins and other integration should be improved.
  We have a number of concrete ideas, but you will also probably come up
  with some after a few minutes of using the existing IDE plugins!
</p>

<p>
  This is only a task for someone who is already an expert, such as someone
  who has built IDE plugins before or is very familiar with the build system.
  One reason is that these tools tend to be complex, which can lead to
  subtle problems.
  Another reason is that we don't want to be stuck maintaining code written by
  someone who is just learning how to write an IDE plugin.
</p>

<p>
  Rather than modifying the Checker Framework's existing support or
  building new support from scratch, it may be better to adapt some other
  project's support for build systems and IDEs.  For instance, you might
  make <a href="https://github.com/coala/coala">coala</a> support the
  Checker Framework, or you might adapt the tool integration provided
  by <a href="http://errorprone.info/">Error Prone</a>.
</p>


<!--
Integrate the Checker Framework with an IDE such as Eclipse, IntelliJ, IDEA or NetBeans,
which will make pluggable type-checking easier to use and more attractive to developers.

Some specific projects include:
  * Create an IDE plug-in that invokes the checker and reports any errors to the developer, just as the IDE currently does for type errors.
  * Improve the existing Eclipse plug-in.
  * Add a button to the IDE that hides/shows all annotations of a given variety, to reduce clutter.  This would be useful beyond type annotations.
  * Implement quick fixes for common errors.
  * Integrate type inference with the IDE, to make it easier for a developer to add annotations to code.
  * Improve an IDE's existing refactoring support so that it is aware of, and properly retains, type annotations.
  * Highlight defaulted types and flow-sensitive type refinements, making error messages easier to understand.
  * Highlight the parts of the code that influenced flow-sensitive type refinements, again to make errors easier to comprehend.

IDEs that build on the OpenJDK Java compiler could benefit from even tighter integration with the Checker Framework.

This project may entail the following:

  * familiarity with the IDE plug-in development environment, Eclipse PDT or NetBeans plug-in support:  By the end of the project, the developer should be familiar with building reasonably complex IDE plug-ins.

  * UI/UX design:  The developer would design a UI experience that is appealing to developers and integrates well with the developer's workflow!
-->


<h2 id="exhaustive-testing">Model checking of a type system</h2>

<p>
Design and implement an algorithm to check type soundness of a type system
by exhaustively verifying the type checker on all programs up to a certain
size. The challenge lies in efficient enumeration of all programs and
avoiding redundant checks, and in knowing the expected outcome of the
tests.  This approach is related to bounded exhaustive
testing and model checking; for a reference, see
[Efficient Software Model Checking of Soundness of Type Systems](http://www.eecs.umich.edu/~bchandra/publications/oopsla08.pdf).
</p>



</body>
</html>

<!--  LocalWords:  GSoC uwplse codespecs randoop typetools blogosphere Shi
 -->
<!--  LocalWords:  Marks's Bikesheds ICST Gyori Legunsen Marinov NonDex bc
 -->
<!--  LocalWords:  PossiblyNonDeterministic PossiblyNonDeterministicOrder
 -->
<!--  LocalWords:  DeterministicOrder Bazel PossbilyPropagatable Lazar tex
 -->
<!--  LocalWords:  NotPropagatable JavaParser contravariantly mortem EnerJ
 -->
<!--  LocalWords:  unsoundnesses nondeterministic nondeterministically ASM
 -->
<!--  LocalWords:  deterministically prototyped Stateful plugins plugin
 -->
<!--  LocalWords:  PDFs Stubparser JLint NullnessLight bytecodes YourKit
 -->
<!--  LocalWords:  timeline Uber's NullAway Uber coala BCEL mvn BCEL's AFU
 -->
<!--  LocalWords:  Signedness Jake2 README Lookup ASM's Nayuki CRC mis CWE
 -->
<!--  LocalWords:  BigInteger lockB lockA signedness microbenchmarks pre
 -->
<!--  LocalWords:  checklink subparts UnsignedBytes runtime MyClass jOOU
 -->
<!--  LocalWords:  usePriorityQueue PriorityQueue queueMinPathNode
 -->
<!--  LocalWords:  subchecker IntScope UnsignedLong UnsignedLongs
 -->