Clojure JIRA

[CLJ-113] GC Issue 109: RT.load's "don't load if already loaded" mechanism breaks ":reload-all"

Tue, 9 Aug 2011 20:31:28 -0500

Reported by scgilardi, Apr 24, 2009

What (small set of) steps will reproduce the problem?

"require" and "use" support a ":reload-all" flag that is intended to  
cause the specified libs to be reloaded along with all libs on which  
they directly or indirectly depend. This is implemented by temporarily  
binding a "loaded-libs" var to the empty set and then loading the  
specified libs.

AOT compilation added another "already loaded" mechanism to  
clojure.lang.RT.load() which is currently not sensitive to a "reload-
all" being in progress and breaks its operation in the following case:

        A, B, and C are libs
        A depends on B. (via :require in its ns form)
        B depends on C. (via :require in its ns form)
        B has been compiled (B.class is on classpath)

        At the repl I "require" A which loads A, B, and C (either from
class files or clj files)
        I modify C.clj
        At the repl I "require" A with the :reload-all flag, intending to  
pick up the changes to C
        C is not reloaded because RT.load() skips loading B: B.class
exists, is already loaded, and B.clj hasn't changed since it was compiled.


What is the expected output? What do you see instead?

I expect :reload-all to be effective. It isn't.

What version are you using?

svn 1354, 1.0.0RC1

Was this discussed on the group? If so, please provide a link to the
discussion:

http://groups.google.com/group/clojure/browse_frm/thread/9bbc290321fd895f/e6a967250021462a#e6a967250021462a

Please provide any additional information below.

I'll upload a patch soon that creates a "*reload-all*" var with a  
root binding of nil and code to bind it to true when the current  
thread has a :reload-all call pending. When *reload-all* is true,  
RT.load() will (re)load all libs from their ".clj" files even if  
they're already loaded.

The fix for this may need to be coordinated with a fix for issue #3.

[CLJ-326] add :as-of option to refer

Tue, 24 Aug 2010 10:19:00 -0500

Discussed here: http://groups.google.com/group/clojure-dev/msg/74af612909dcbe56

:as-of allows library authors to specify a known subset of vars to refer from clojure (or any other library which would use :added metadata).

(ns foo (:refer-clojure :as-of "1.1")) is equivalent to (ns foo (:refer-clojure :only [public-documented-vars-which-already-existed-in-1.1]))

[CLJ-130] Namespace metadata lost in AOT compile

Fri, 3 Dec 2010 10:48:15 -0600

For example, the namespace @clojure.contrib.def@ has metadata for doc and author.

We see this when we load the file directly from source:
<pre>
<code>
tom@goya:~/src/clj$ java -cp clojure/clojure.jar:clojure-contrib/src clojure.lang.Repl
Clojure 1.1.0-alpha-SNAPSHOT
user=> (load "clojure/contrib/def")
nil
user=> (find-ns 'clojure.contrib.def)
#<Namespace clojure.contrib.def>
user=> ^(find-ns 'clojure.contrib.def)
{:author "Stephen C. Gilardi", :doc "def.clj provides variants of def that make including doc strings and\nmaking private definitions more succinct."}
user=>
</code>
</pre>

But if we load the file from the jar where it's been compiled, the metadata is lost:
<pre>
<code>
tom@goya:~/src/clj$ java -cp clojure/clojure.jar:clojure-contrib/clojure-contrib.jar clojure.lang.Repl
Clojure 1.1.0-alpha-SNAPSHOT
user=> (use 'clojure.contrib.def)
nil
user=> (find-ns 'clojure.contrib.def)
#<Namespace clojure.contrib.def>
user=> ^(find-ns 'clojure.contrib.def)
nil
user=>
</code>
</pre>

Even if we use @load@, we don't see metadata on the item:
<pre>
<code>
tom@goya:~/src/clj$ java -cp clojure/clojure.jar:clojure-contrib/clojure-contrib.jar clojure.lang.Repl
Clojure 1.1.0-alpha-SNAPSHOT
user=> (load "clojure/contrib/def")
nil
user=> (find-ns 'clojure.contrib.def)
#<Namespace clojure.contrib.def>
user=> ^(find-ns 'clojure.contrib.def)
nil
user=>
</code>
</pre>

The jar isn't the problem, for if we use the slim jar (without the AOT
class files), we see that the metadata is fine:
<pre>
<code>
tom@goya:~/src/clj$ java -cp clojure/clojure.jar:clojure-contrib/clojure-contrib-slim.jar clojure.lang.Repl
Clojure 1.1.0-alpha-SNAPSHOT
user=> (use 'clojure.contrib.def)
nil
user=> (find-ns 'clojure.contrib.def)
#<Namespace clojure.contrib.def>
user=> ^(find-ns 'clojure.contrib.def)
{:author "Stephen C. Gilardi", :doc "def.clj provides variants of def that make including doc strings and\nmaking private definitions more succinct."}
user=>
</code>
</pre>

This seems to be true usually, but not always. For example the
metadata on the pretty print namespace is just fine from the AOT version:
<pre>
<code>
tom@goya:~/src/clj$ java -cp clojure/clojure.jar:clojure-contrib/clojure-contrib.jar clojure.lang.Repl
Clojure 1.1.0-alpha-SNAPSHOT
user=> (use 'clojure.contrib.pprint)
nil
user=> (find-ns 'clojure.contrib.pprint)
#<Namespace clojure.contrib.pprint>
user=> ^(find-ns 'clojure.contrib.pprint)
{:author "Tom Faulhaber", :doc "This module comprises two elements:\n1) A pretty printer for Clojure data structures, implemented in the function \"pprint\"\n2) A Common Lisp compatible format function, implemented as \"cl-format\" because\n Clojure is using the name \"format\" for its own format.\n\nComplete documentation is available on the wiki at the contrib google code site.", :see-also [["PrettyPrinting" "Documentation for the pretty printer"] ["CommonLispFormat" "Documentation for Common Lisp format function"]]}
user=>
</code>
</pre>

[CLJ-129] Add documentation to sorted-set-by detailing how the provided comparator may change set membership semantics

Tue, 24 Aug 2010 07:45:00 -0500

To start, let's look at some simple default sorted-set behaviour (which uses PersistentHashMap via PersistentHashSet, and therefore uses equality/hashCode to determine identity):

user=> (sorted-set [1 2] [-5 10] [1 5])
#{[-5 10] [1 2] [1 5]}
</code></pre>

sorted-set-by uses PersistentTreeMap via PersistentTreeSet though, which implies that the comparator provided to sorted-set-by will be used to determine identity, and therefore member in the set.  This can lead to (IMO) non-intuitive behaviour:

<pre><code>
user=> (sorted-set-by #(> (first %) (first %2)) [1 2] [-5 10] [1 5])
#{[1 2] [-5 10]}

Notice that because the provided comparison fn determines that [1 2] and [1 5] have the same sort order, the latter value is considered identical to the former, and not included in the set. This behaviour could be very handy, but is also likely to cause confusion when what the user almost certainly wants is to maintain the membership semantics of the original set (e.g. relying upon equality/hashCode), but only modify the ordering.

(BTW, yes, I know there's far easier ways to get the order I'm indicating above over a set of vectors thanks to vectors being comparable via the compare fn. The examples are only meant to be illustrative. The same non-intuitive result would occur, with no easy fallback (like the 'compare' fn when working with vectors) when the members of the set are non-Comparable Java object, and the comparator provided to sorted-set-by is defining a sort over some values returned by method calls into those objects.)

I'd be happy to change the docs for sorted-set-by, but I suspect that there are others who could encapsulate what's going on here more correctly and more concisely than I.

[CLJ-42] GC Issue 38: When using AOT compilation, "load"ed files are not reloaded on (require :reload 'name.space)

Tue, 24 Aug 2010 06:44:00 -0500

Reported by m...@kotka.de, Jan 07, 2009
What (small set of) steps will reproduce the problem?

1. Create a file src/foo.clj

cat >src/foo.clj <<EOF
(ns foo (:load "bar"))
EOF

2. Create a file src/bar.clj

cat >src/bar.clj <<EOF
(clojure.core/in-ns 'foo)
(def x 8)
EOF

3. Start Clojure Repl: java -cp src:classes clojure.main -r

4. Compile the namespace.

user=> (compile 'foo)
foo

5. Require the namespace
user=> (require :reload-all :verbose 'foo)
(clojure.core/load "/foo")
(clojure.core/load "/bar")

What is the expected output? What do you see instead?

6. Re-Require the namespace

user=> (require :reload-all :verbose 'foo)
(clojure.core/load "/foo")

Only the "master" file is loaded, but not the bar file.
Expected would have been to also load the bar file.
Changes to bar.clj are not reflected, and depending
on the setting (eg. using multimethods in foo from
a different namespace) code may be corrupted.

What version are you using?

SVN rev. 1195

[CLJ-445] Method/Constructor resolution does not factor in widening conversion of primitive args

Mon, 20 Feb 2012 14:04:42 -0600

Problem:
When making java calls (or inlined functions), if both args and param are primitive, no widening conversion is used to locate the proper overloaded method/constructor.

Examples:

user=> (Integer. (byte 0))
java.lang.IllegalArgumentException: No matching ctor found for class java.lang.Integer (NO_SOURCE_FILE:0)
</code></pre>
The above occurs because there is no Integer(byte) constructor, though it should match on Integer(int).
<pre><code>user=> (bit-shift-left (byte 1) 1)
Reflection warning, NO_SOURCE_PATH:3 - call to shiftLeft can't be resolved.
2

In the above, a call is made via reflection to Numbers.shiftLeft(Object, Object) and its associated auto-boxing, instead of directly to the perfectly adequate Numbers.shiftLeft(long, int).

Workarounds:
Explicitly casting to the formal type.

Ancillary benefits of fixing:
It would also reduce the amount of method overloading, e.g., RT.intCast(char), intCast(byte), intCast(short), could all be removed, since such calls would pass to RT.intCast(int).

[CLJ-322] Enhance AOT compilation process to emit classfiles only for explicitly-specified namespaces

Fri, 12 Apr 2013 09:52:24 -0500

This was originally/erroneously reported by Howard Lewis Ship in the clojure-contrib assembla:

My build file specifies the namespaces to AOT compile but if I include another namespace
(even from a JAR dependency) that is not AOT compiled, the other namespace will be compiled as well.

In my case, I was using clojure-contrib's clojure.contrib.str-utils2 namespace, and I got a bunch of
clojure/contrib/str_utils2 classes in my output directory.

I think that the AOT compiler should NOT precompile any namespaces that are transitively reached,
only namespaces in the set specified by the command line are appropriate.

As currently coded, you will frequently find unwanted third-party dependencies in your output JARs;
further, if multiple parties depend on the same JARs, this could cause bloating and duplication in the
eventual runtime classpath.

Having the option of shipping either all AOT-compiled classfiles or mixed source/AOT depending upon one's distribution requirements would make that phase of work with a clojure codebase significantly easier and less error-prone. The only question in my mind is what the default should be. We're all used to the current behaviour, but I'd guess that any nontrivial project where the form of the distributable matters (i.e. the source/AOT mix), providing as much control as possible by default makes the most sense. Given the tooling that most people are using, it's trivial (and common practice, IIUC) to provide a comprehensive list of namespaces one wishes to compile, so making that the default shouldn't be a hurdle to anyone. If an escape hatch is desired, a --transitive switch to clojure.lang.Compile could be added.

[CLJ-84] GC Issue 81: compile gen-class fail when class returns self

Fri, 10 Dec 2010 09:30:52 -0600

Reported by davidhaub, Feb 14, 2009

When attempting to compile the following program, clojure fails with a
ClassNotFoundException.  It occurs because one of the methods returns the
same class that is being generated.  If the returnMe method below is
changed to return an Object, the compile succeeds.

Beware when testing! If the classpath contains a class file (say from a
prior successful build when the returnMe method was changed to return an
object), the compile will succeed.  Always clear out the
clojure.compile.path prior to compiling.

;badgenclass.clj
(ns badgenclass
  (:gen-class
     :state state
     :methods
     [[returnMe [] badgenclass]]
     :init init))
(defn -init []
  [[] nil])

(defn -returnMe [this]
  this)

#!/bin/sh
rm -rf classes
mkdir classes
java -cp lib/clojure.jar:classes:. -Dclojure.compile.path=classes \
clojure.lang.Compile badgenclass


Comment 1 by chouser, Mar 07, 2009

Attached is a patch that accepts strings or symbols for parameter and return class
names, and generates the appropriate bytecode without calling Class/forName.  It
fixes this issue, but because 'ns' doesn't resolve :gen-class's arguments, class
names aren't checked as early anymore.  :gen-class-created classes with invalid
parameter or return types can even be instantiated, and no error will be reported
until the broken method is called.

One possible alternative would be to call Class/forName on any symbols given, but
allow strings to use the method given by this patch.  To return your own type, you'd
need a method defined like:

  [returnMe [] "badgenclass"]

Any thoughts?

[CLJ-69] GC Issue 66: Add "keyset" to Clojure; make .keySet for APersistentMap return IPersistentSet

Fri, 10 Dec 2010 08:48:40 -0600

Reported by wolfe.a.jason, Feb 04, 2009

Describe the feature/change.

Add "keyset" to Clojure; make .keySet for APersistentMap return an
IPersistentSet

Was this discussed on the group? If so, please provide a link to the
discussion:

http://groups.google.com/group/clojure/browse_thread/thread/66e708e477ae992f/ff3d8d588068b60e?hl=en#ff3d8d588068b60e

-----------------------------------------------------

A patch is attached.  Some notes:

I chose to add a "keyset" function, rather than change the existing "keys",
so as to avoid breaking anything.

The corresponding RT.keyset function just calls .keySet on the argument.
I would have liked to have "keyset" return an IPersistentSet even when
passed a (non-Clojure) java.util.Map, but this seems impossible to do in
sublinear time because of essentially the same limitation mentioned in the
above thread (the Map interface does not support getKey() or entryAt()) --
assuming, again, that "get" is supposed to return the actual (identical?)
key in a set, and not just an .equal key.

I then changed the implementation of .keySet for APersistentMap to
essentially copy APersistentSet.  A more concise alternative would have
been to extend APersistentSet and override the .get method, but that made
me a bit nervous (since if APeristentSet changed this could break). 

Anyway, this is my first patch for the Java side of Clojure, and I'm not
yet solid on the conventions and aesthetics, so
comments/questions/criticisms/requests for revisions are very welcome.

[CLJ-211] Support arbitrary functional destructuring via -> and ->>

Fri, 10 Dec 2010 08:49:03 -0600

Support arbitrary functional destructuring, that is the use of
any function in any destructuring form to help unpack data in
arbitrary ways.

The discussion began here:
http://clojure-log.n01se.net/date/2009-11-17.html#09:31c

The attached patch implements the spec described here:
http://clojure-log.n01se.net/date/2009-11-17.html#10:50a

That is, the following examples would now work:

user=> (let [(-> str a) 1] a)
"1"

user=> (let [[a (-> str b) c] [1 2]] (list a b c))
(1 "2" nil)

user=> (let [(->> (map int) [a b]) "ab"] (list a b))
(97 98)

[CLJ-291] (take-nth 0 coll) redux...

Fri, 10 Dec 2010 08:50:19 -0600

I dont seem to be able make the old ticket uninvalid so here goes
(take-nth 0 coll) causes (at least on Solaris) infinite space and time consumption
It's not a printout error as the following code causes the problem too

(let [j 0
firstprod (apply * (doall (map #(- 1 %) (take-nth j (:props mix)))))]) ; from my parameter update function

I used jvisualvm and the jvm is doing some RNI call - no clojure code is running at all
If left alone it will crash the jvm with all heap space consumed
0 is an InvalidArgument for take-nth
I wouldnt mind if it produced an infinite lazy sequence of nils even though thats wrong
It doesnt do this though it actively destroys the JVM
Its a bug nasty destructive and it took me half a day to figure out what was going on
please let someone fix it!

[CLJ-304] contrib get-source no longer works with deftype

Fri, 3 Dec 2010 11:44:01 -0600

Now that deftype creates a class (but not a var), you can't use c.c.repl-utils/get-source on a deftype. Is there something we can do on the Clojure side to help this work again?

[CLJ-196] file returns "NO_SOURCE_PATH", but the doc says it should be nil

Fri, 12 Apr 2013 10:05:40 -0500

According to http://clojure.org/api, *file* should return nil in the repl, but it returns "NO_SOURCE_PATH".

This has been true for a long time. Latest patch changes only the docstring of *file* to accurately reflect current behavior.