<< Back to previous view

[CLJ-1515] Reify the result of range Created: 29/Aug/14  Updated: 29/Aug/14

Status: Open
Project: Clojure
Component/s: None
Affects Version/s: Release 1.7
Fix Version/s: Release 1.7

Type: Enhancement Priority: Minor
Reporter: Timothy Baldridge Assignee: Unassigned
Resolution: Unresolved Votes: 0
Labels: None

Attachments: File patch.diff    
Patch: Code and Test
Approval: Incomplete

 Description   

Currently range simply returns a lazy seq. If the return value of range were reified into a type (as it is in ClojureScript) we could optimize many functions on that resulting type. Some operations such as count and nth become O(1) in this case, while others such as reduce could receive a performance boost do to the reduced number of allocations.

Approach: this patch revives the unused (but previously existing) clojure.lang.Range class. This class acts as a lazy seq and implements several other appropriate interfaces such as Counted and Indexed. This type is implemented in Java since range is needed fairly on in core.clj before deftype is defined. The attached patch uses Numbers.* methods for all math due to the input types to range being unknown. The class also supplies a .iterator() method which allows for allocation free reducing over range.

Note: this code keeps backwards compatibility with the existing range code. This means some parts of the class (mostly relating to a step size of 0) are a bit more complex than desired, but these bits were needed to get all the tests to pass.

Note: this code does not preserve the chunked-seq nature of the original range. The fact that range used to return chunked seqs was not published in the doc strings and so it was removed to allow for simpler code in Range.java.

Patch: patch.diff



 Comments   
Comment by Alex Miller [ 29/Aug/14 3:19 PM ]

1) Not sure about losing chunked seqs - that would make older usage slower, which seems undesirable.
2) RangeIterator.next() needs to throw NoSuchElementException when walking off the end
3) I think Range should implement IReduce instead of relying on support for CollReduce via Iterable.
4) Should let _hash and _hasheq auto-initialize to 0 not set to -1. As is, I think _hasheq always would be -1?
5) _hash and _hasheq should be transient.
6) count could be cached (like hash and hasheq). Not sure if it's worth doing that but seems like a win any time it's called more than once.
7) Why the change in test/clojure/test_clojure/serialization.clj ?
8) Can you squash into a single commit?

Comment by Timothy Baldridge [ 29/Aug/14 3:40 PM ]

1) I agree, adding chunked seqs to this will dramatically increase complexity, are we sure we want this?
2) exception added
3) I can add IReduce, but it'll pretty much just duplicate the code in protocols.clj. If we're sure we want that I'll add it too.
4) fixed hash init values, defaults to -1 like ASeq
5) hash fields are now transient
6) at the cost of about 4 bytes we can cache the cost of a multiplication and an addition, doesn't seem worth it?
7) the tests in serialization.clj assert that the type of the collection roundtrips. This is no longer the case for range which starts as Range and ends as a list. The change I made converts range into a list so that it properly roundtrips. My assumption is that we shouldn't rely on all implementations of ISeq to properly roundtrip through EDN.
8) squashed.

Comment by Alex Miller [ 29/Aug/14 3:49 PM ]

6) might be useful if you're walking through it with nth, which hits count everytime, but doubt that's common
7) yep, reasonable





[CLJ-1100] Reader literals cannot contain periods Created: 02/Nov/12  Updated: 29/Aug/14

Status: Open
Project: Clojure
Component/s: None
Affects Version/s: None
Fix Version/s: Release 1.7

Type: Defect Priority: Minor
Reporter: Kevin Lynagh Assignee: Unassigned
Resolution: Unresolved Votes: 1
Labels: reader

Attachments: Text File CLJ-1100-reader-tags-with-periods.patch     Text File clj-1100-v2.patch    
Patch: Code and Test
Approval: Incomplete

 Description   

The reader tries to read a record instead of a literal if the tag contains periods.

user> (binding [*data-readers* {'foo/bar #'identity}] (read-string "#foo/bar 1"))
1
user> (binding [*data-readers* {'foo/bar.x #'identity}] (read-string "#foo/bar.x 1"))
ClassNotFoundException foo/bar.x  java.lang.Class.forName0 (Class.java:-2)

Summary of reader forms:

Kind Example Constraint Status
Record #user.Foo[1] record class name OK
Class #java.lang.String["abc"] class name OK
Clojure reader tag #uuid "c48d7d6e-f3bb-425a-abc5-44bd014a511d" not a class name, no "/" OK
Library reader tag #my/card "5H" not a class name, has "/" OK
  #my.ns/card "5H" not a class name, has "/" OK
  #my/playing.card "5H" not a class name, has "/" BROKEN - read as record

Note: reader tags should not be allowed to override the record reader.

Cause: In LispReader, CtorReader.invoke() decides between record and tagged literal based on whether the tag has a ".".

Proposed: Change the discriminator in CtorReader by doing more string inspection:

  • If name has a "/" -> readTagged (not a legal class name)
  • If name has no "/" or "." -> readTagged (records must have qualified names)
  • Else -> readRecord (also covers Java classes)

Tradeoffs: Clojure-defined data reader tags must not contain periods. Not possible to read a Java class with no package. Avoids unnecessary class loading/construction for all tags.

Patch: CLJ-1100-v2.patch

Screened by: Alex Miller

Alternatives considered:

Using class checks:

  • Attempt readRecord (also covers Java classes)
  • If failed, attempt readTagged

Tradeoffs: Clojure tags could not override Java/record constructors - not sure that's something we'd ever want to do, but this would cut that off. This alternative may attempt classloading when it would not have before.



 Comments   
Comment by Steve Miner [ 06/Nov/12 9:41 AM ]

The suggested patch (clj-1100-reader-literal-periods.patch) will break reading records when *default-data-reader-fn* is set. Try adding a test like this:

(deftest tags-containing-periods-with-default
      ;; we need a predefined record for this test so we (mis)use clojure.reflect.Field for convenience
      (let [v "#clojure.reflect.Field{:name \"fake\" :type :fake :declaring-class \"Fake\" :flags nil}"]
        (binding [*default-data-reader-fn* nil]
          (is (= (read-string v) #clojure.reflect.Field{:name "fake" :type :fake :declaring-class "Fake" :flags nil})))
        (binding [*default-data-reader-fn* (fn [tag val] (assoc val :meaning 42))]
          (is (= (read-string v) #clojure.reflect.Field{:name "fake" :type :fake :declaring-class "Fake" :flags nil})))))
Comment by Rich Hickey [ 29/Nov/12 9:36 AM ]

The problem assessment is ok, but the resolution approach may not be. What happens should be based not upon what is in data-readers but whether or not the name names a class.

Is the intent here to allow readers to circumvent records? I'm not in favor of that.

Comment by Steve Miner [ 29/Nov/12 4:01 PM ]

New patch following Rich's comments. The decision to read a record is now based on the symbol containing periods and not having a namespace. Otherwise, it is considered a data reader tag. User
defined tags are required to be qualified but they may now have periods in the name. Tests added to show that
data readers cannot override record classes. Note: Clojure-defined data reader tags may be unqualified, but they should not contain periods in order to avoid confusion with record classes.

Comment by Steve Miner [ 29/Nov/12 4:17 PM ]

I deleted my old patch and some comments referring to it to avoid confusion.

In Clojure 1.5 beta 1, # followed by a qualified symbol with a period in the name is considered a record and causes an exception for the missing record class. With the patch, only non-qualified symbols containing periods are considered records. That allows user-defined qualified symbols with periods in their names to be used as data reader tags.

Comment by Andy Fingerhut [ 07/Feb/13 9:05 AM ]

clj-1100-periods-in-data-reader-tags-patch-v2.txt dated Feb 7 2013 is identical to CLJ-1100-periods-in-data-reader-tags.patch dated Nov 29 2012, except it applies cleanly to latest master. The only change appears to be in some white space in the context lines.

Comment by Andy Fingerhut [ 07/Feb/13 12:53 PM ]

I've removed clj-1100-periods-in-data-reader-tags-patch-v2.txt mentioned in the previous comment, because I learned that CLJ-1100-periods-in-data-reader-tags.patch dated Nov 29 2012 applies cleanly to latest master and passes all tests if you use this command to apply it.

% git am --keep-cr -s --ignore-whitespace < CLJ-1100-periods-in-data-reader-tags.patch

I've already updated the JIRA workflow and screening patches wiki pages to mention this --ignore-whitespace option.

Comment by Andy Fingerhut [ 13/Feb/13 11:31 AM ]

Both of the current patches, CLJ-1100-periods-in-data-reader-tags.patch dated Nov 29 2012, and clj-1100-reader-literal-periods.patch dated Nov 6 2012, fail to apply cleanly to latest master (1.5.0-RC15) as of today, although they did last week. Given all of the changes around read / read-string and edn recently, they should probably be evaluated by their authors to see how they should be updated.

Comment by Steve Miner [ 14/Feb/13 12:23 PM ]

I deleted my patch: CLJ-1100-periods-in-data-reader-tags.patch. clj-1100-reader-literal-periods.patch is clearly wrong, but the original author or an administrator has to delete that.

Comment by Kevin Lynagh [ 14/Feb/13 1:28 PM ]

I cannot figure out how to remove my attachment (clj-1100-reader-literal-periods.patch) in JIRA.

Comment by Steve Miner [ 14/Feb/13 1:43 PM ]

Downarrow (popup) menu to the right of the "Attachments" section. Choose "manager attachments".

Comment by Kevin Lynagh [ 14/Feb/13 2:02 PM ]

Great, thanks Steve. Are you going to take another pass at this issue, or should I give it a go?

Comment by Steve Miner [ 14/Feb/13 3:04 PM ]

Kevin, I'm not planning to work on this right now as 1.5 is pretty much done. It might be worthwhile discussing the issue a bit on the dev mailing list before working on a patch, but that's up to you. I think my approach was correct, although now changes would have to be applied to both LispReader and EdnReader.

Comment by Alex Miller [ 09/Apr/14 10:29 AM ]

Updated description based on my understanding.

Comment by Steve Miner [ 22/Apr/14 3:30 PM ]

I will resurrect my old patch and update it for the changes since 1.5.

Comment by Steve Miner [ 28/Apr/14 8:21 AM ]

Added patch to allow reader tags to have periods, but only with a namespace. Added tests to confirm that it works, but does not allow overriding a record name with a data-reader.

Comment by Steve Miner [ 28/Apr/14 8:51 AM ]

The patch implements Alex's alternative 1. It's purely lexical. A tag symbol without a namespace and containing periods is handled as a record (Java class). Otherwise, it's a data-reader tag. Of course, unqualified symbols without periods are still data-reader tags.

IMHO, a Java class without a package is a pathological case which Clojure doesn't need to worry about. This patch follows the convention that Java classes are named by unqualified symbols containing dots.

I did try alternative 2, testing for an actual class, but the implementation was more complicated. Also, it would open the possibility of breaking working code by adding a record or Java class that accidentally collided with an unqualified dotted tag that had previously worked fine. It's better to follow a simple rule that unqualified dotted symbols always refer to classes. Maybe the class doesn't actually exist, but that doesn't mean the symbol might be a data-literal tag.

Comment by Alex Miller [ 13/May/14 4:49 PM ]

Added clj-1100-v2.patch - identical, just removes whitespace to simplify change.

Comment by Rich Hickey [ 29/Aug/14 9:16 AM ]

I think we should disallow this rather than enable it. We don't generally support foo/bar.x

Comment by Nicola Mometto [ 29/Aug/14 9:27 AM ]

I created http://dev.clojure.org/jira/browse/CLJ-1516 with a patch that throws an exception on `(def foo.bar)`





[CLJ-1322] doseq with several bindings causes "ClassFormatError: Invalid Method Code length" Created: 10/Jan/14  Updated: 29/Aug/14

Status: Open
Project: Clojure
Component/s: None
Affects Version/s: Release 1.5
Fix Version/s: Release 1.7

Type: Defect Priority: Major
Reporter: Miikka Koskinen Assignee: Unassigned
Resolution: Unresolved Votes: 1
Labels: None
Environment:

Clojure 1.5.1, java 1.7.0_25, OpenJDK Runtime Environment (IcedTea 2.3.10) (7u25-2.3.10-1ubuntu0.12.04.2)


Attachments: Text File doseq-bench.txt     Text File doseq.patch     File script.clj    
Patch: Code
Approval: Incomplete

 Description   

Important Perf Note the new impl is faster for collections that are custom-reducible but not chunked, and is also faster for large numbers of bindings. The original implementation is hand tuned for chunked collections, and wins for larger chunked coll/smaller binding count scenarios, presumably due to the fn call/return tracking overhead of reduce. Details are in the comments.
Screened By
Patch doseq.patch

user=> (def a1 (range 10))
#'user/a1
user=> (doseq [x1 a1 x2 a1 x3 a1 x4 a1 x5 a1 x6 a1 x7 a1 x8 a1] (do))
CompilerException java.lang.ClassFormatError: Invalid method Code length 69883 in class file user$eval1032, compiling:(NO_SOURCE_PATH:2:1)

While this example is silly, it's a problem we've hit a couple of times. It's pretty surprising when you have just a couple of lines of code and suddenly you get the code length error.



 Comments   
Comment by Kevin Downey [ 18/Apr/14 12:20 AM ]

reproduces with jdk 1.8.0 and clojure 1.6

Comment by Nicola Mometto [ 22/Apr/14 5:35 PM ]

A potential fix for this is to make doseq generate intermediate fns like `for` does instead of expanding all the code directly.

Comment by Ghadi Shayban [ 25/Jun/14 8:39 PM ]

Existing doseq handles chunked-traversal internally, deciding the
mechanics of traversal for a seq. In addition to possibly conflating
concerns, this is causing a code explosion blowup when more bindings are
added, approx 240 bytes of bytecode per binding (without modifiers).

This approach redefs doseq later in core.clj, after protocol-based
reduce (and other modern conveniences like destructuring.)

It supports the existing :let, :while, and :when modifiers.

New is a stronger assertion that modifiers cannot come before binding
expressions. (Same semantics as let, i.e. left to right)

valid: [x coll :when (foo x)]
invalid: [:when (foo x) x coll]

This implementation does not suffer from the code explosion problem.
About 25 bytes of bytecode + 1 fn per binding.

Implementing this without destructuring was not a party, luckily reduce
is defined later in core.

Comment by Andy Fingerhut [ 26/Jun/14 12:25 AM ]

For anyone reviewing this patch, note that there are already many tests for correct functionality of doseq in file test/clojure/test_clojure/for.clj. It may not be immediately obvious, but every test for 'for' defined with deftest-both is a test for 'for' and also for 'doseq'.

Regarding the current implementation of doseq: it in't simply that it is too many bytes per binding, it is that the code size doubles with each additional binding. See these results, which measures size of the macroexpanded form rather than byte code size, but those two things should be fairly linearly related to each other here:

(defn formsize [form]
  (count (with-out-str (print (macroexpand form)))))

user=> (formsize '(doseq [x (range 10)] (print x)))
652
user=> (formsize '(doseq [x (range 10) y (range 10)] (print x y)))
1960
user=> (formsize '(doseq [x (range 10) y (range 10) z (range 10)] (print x y z)))
4584
user=> (formsize '(doseq [x (range 10) y (range 10) z (range 10) w (range 10)] (print x y z w)))
9947
user=> (formsize '(doseq [x (range 10) y (range 10) z (range 10) w (range 10) p (range 10)] (print x y z w p)))
20997

Here are results for the same expressions after Ghadi's patch doseq.patch dated June 25 2014:

user=> (formsize '(doseq [x (range 10)] (print x)))
93
user=> (formsize '(doseq [x (range 10) y (range 10)] (print x y)))
170
user=> (formsize '(doseq [x (range 10) y (range 10) z (range 10)] (print x y z)))
247
user=> (formsize '(doseq [x (range 10) y (range 10) z (range 10) w (range 10)] (print x y z w)))
324
user=> (formsize '(doseq [x (range 10) y (range 10) z (range 10) w (range 10) p (range 10)] (print x y z w p)))
401

It would be good to see some performance results with and without this patch, too.

Comment by Stuart Halloway [ 28/Jun/14 2:21 PM ]

In the tests below, the new impl is called "doseq2", vs. the original impl "doseq"

(def hund (into [] (range 100)))
(def ten (into [] (range 10)))
(def arr (int-array 100))
(def s "superduper")

;; big seq, few bindings: doseq2 LOSES
(dotimes [_ 5]
  (time (doseq [a (range 100000000)])))
;; 1.2 sec

(dotimes [_ 5]
  (time (doseq2 [a (range 100000000)])))
;; 1.8 sec

;; small unchunked reducible, few bindings: doseq2 wins
(dotimes [_ 5]
  (time (doseq [a s b s c s])))
;; 0.5 sec

(dotimes [_ 5]
  (time (doseq2 [a s b s c s])))
;; 0.2 sec

(dotimes [_ 5]
  (time (doseq [a arr b arr c arr])))
;; 40 msec

(dotimes [_ 5]
  (time (doseq2 [a arr b arr c arr])))
;; 8 msec

;; small chunked reducible, few bindings: doseq2 LOSES
(dotimes [_ 5]
  (time (doseq [a hund b hund c hund])))
;; 2 msec

(dotimes [_ 5]
  (time (doseq2 [a hund b hund c hund])))
;; 8 msec

;; more bindings: doseq2 wins bigger and bigger
(dotimes [_ 5]
  (time (doseq [a ten b ten c ten d ten ])))
;; 2 msec

(dotimes [_ 5]
  (time (doseq2 [a ten b ten c ten d ten ])))
;; 0.4 msec

(dotimes [_ 5]
  (time (doseq [a ten b ten c ten d ten e ten])))
;; 18 msec

(dotimes [_ 5]
  (time (doseq2 [a ten b ten c ten d ten e ten])))
;; 1 msec
Comment by Ghadi Shayban [ 28/Jun/14 6:23 PM ]

Hmm, I cannot reproduce your results.

I'm not sure whether you are testing with lein, on what platform, what jvm opts.

Can we test using this little harness instead directly against clojure.jar? I've attached a the harness and two runs of results (one w/ default heap, the other 3GB w/ G1GC)

I added a medium and small (range) too.

Anecdotally, I see doseq2 outperform in all cases except the small range. Using criterium shows a wider performance gap favoring doseq2.

I pasted the results side by side for easier viewing.

core/doseq                          doseq2
"Elapsed time: 1610.865146 msecs"   "Elapsed time: 2315.427573 msecs"
"Elapsed time: 2561.079069 msecs"   "Elapsed time: 2232.479584 msecs"
"Elapsed time: 2446.674237 msecs"   "Elapsed time: 2234.556301 msecs"
"Elapsed time: 2443.129809 msecs"   "Elapsed time: 2224.302855 msecs"
"Elapsed time: 2456.406103 msecs"   "Elapsed time: 2210.383112 msecs"

;; med range, few bindings:
core/doseq                          doseq2
"Elapsed time: 28.383197 msecs"     "Elapsed time: 31.676448 msecs"
"Elapsed time: 13.908323 msecs"     "Elapsed time: 11.136818 msecs"
"Elapsed time: 18.956345 msecs"     "Elapsed time: 11.137122 msecs"
"Elapsed time: 12.367901 msecs"     "Elapsed time: 11.049121 msecs"
"Elapsed time: 13.449006 msecs"     "Elapsed time: 11.141385 msecs"

;; small range, few bindings:
core/doseq                          doseq2
"Elapsed time: 0.386334 msecs"      "Elapsed time: 0.372388 msecs"
"Elapsed time: 0.10521 msecs"       "Elapsed time: 0.203328 msecs"
"Elapsed time: 0.083378 msecs"      "Elapsed time: 0.179116 msecs"
"Elapsed time: 0.097281 msecs"      "Elapsed time: 0.150563 msecs"
"Elapsed time: 0.095649 msecs"      "Elapsed time: 0.167609 msecs"

;; small unchunked reducible, few bindings:
core/doseq                          doseq2
"Elapsed time: 2.351466 msecs"      "Elapsed time: 2.749858 msecs"
"Elapsed time: 0.755616 msecs"      "Elapsed time: 0.80578 msecs"
"Elapsed time: 0.664072 msecs"      "Elapsed time: 0.661074 msecs"
"Elapsed time: 0.549186 msecs"      "Elapsed time: 0.712239 msecs"
"Elapsed time: 0.551442 msecs"      "Elapsed time: 0.518207 msecs"

core/doseq                          doseq2
"Elapsed time: 95.237101 msecs"     "Elapsed time: 55.3067 msecs"
"Elapsed time: 41.030972 msecs"     "Elapsed time: 30.817747 msecs"
"Elapsed time: 42.107288 msecs"     "Elapsed time: 19.535747 msecs"
"Elapsed time: 41.088291 msecs"     "Elapsed time: 4.099174 msecs"
"Elapsed time: 41.03616 msecs"      "Elapsed time: 4.084832 msecs"

;; small chunked reducible, few bindings:
core/doseq                          doseq2
"Elapsed time: 31.793603 msecs"     "Elapsed time: 40.082492 msecs"
"Elapsed time: 17.302798 msecs"     "Elapsed time: 28.286991 msecs"
"Elapsed time: 17.212189 msecs"     "Elapsed time: 14.897374 msecs"
"Elapsed time: 17.266534 msecs"     "Elapsed time: 10.248547 msecs"
"Elapsed time: 17.227381 msecs"     "Elapsed time: 10.022326 msecs"

;; more bindings:
core/doseq                          doseq2
"Elapsed time: 4.418727 msecs"      "Elapsed time: 2.685198 msecs"
"Elapsed time: 2.421063 msecs"      "Elapsed time: 2.384134 msecs"
"Elapsed time: 2.210393 msecs"      "Elapsed time: 2.341696 msecs"
"Elapsed time: 2.450744 msecs"      "Elapsed time: 2.339638 msecs"
"Elapsed time: 2.223919 msecs"      "Elapsed time: 2.372942 msecs"

core/doseq                          doseq2
"Elapsed time: 28.869393 msecs"     "Elapsed time: 2.997713 msecs"
"Elapsed time: 22.414038 msecs"     "Elapsed time: 1.807955 msecs"
"Elapsed time: 21.913959 msecs"     "Elapsed time: 1.870567 msecs"
"Elapsed time: 22.357315 msecs"     "Elapsed time: 1.904163 msecs"
"Elapsed time: 21.138915 msecs"     "Elapsed time: 1.694175 msecs"
Comment by Ghadi Shayban [ 28/Jun/14 6:47 PM ]

It's good that the benchmarks contain empty doseq bodies in order to isolate the overhead of traversal. However, that represents 0% of actual real-world code.

At least for the first benchmark (large chunked seq), adding in some tiny amount of work did not change results signifantly. Neither for (map str [a])

(range 10000000) =>  (map str [a])
core/doseq
"Elapsed time: 586.822389 msecs"
"Elapsed time: 563.640203 msecs"
"Elapsed time: 369.922975 msecs"
"Elapsed time: 366.164601 msecs"
"Elapsed time: 373.27327 msecs"
doseq2
"Elapsed time: 419.704021 msecs"
"Elapsed time: 371.065783 msecs"
"Elapsed time: 358.779231 msecs"
"Elapsed time: 363.874448 msecs"
"Elapsed time: 368.059586 msecs"

nor for intrisics like (inc a)

(range 10000000)
core/doseq
"Elapsed time: 317.091849 msecs"
"Elapsed time: 272.360988 msecs"
"Elapsed time: 215.501737 msecs"
"Elapsed time: 206.639181 msecs"
"Elapsed time: 206.883343 msecs"
doseq2
"Elapsed time: 241.475974 msecs"
"Elapsed time: 193.154832 msecs"
"Elapsed time: 198.757873 msecs"
"Elapsed time: 197.803042 msecs"
"Elapsed time: 200.603786 msecs"

I still see reduce-based doseq ahead of the original, except for small seqs

Comment by Ghadi Shayban [ 04/Aug/14 2:55 PM ]

A form like the following will not work with this patch:

(go (doseq [c chs] (>! c :foo)))

as the go macro doesn't traverse fn boundaries. The only such code I know is core.async/mapcat*, a private fn supporting a fn that is marked deprecated.

Comment by Ghadi Shayban [ 07/Aug/14 2:09 PM ]

I see #'clojure.core/run! was just added, which has a similar limitation

Comment by Rich Hickey [ 29/Aug/14 8:19 AM ]

Please consider Ghadi's feedback, esp re: closures.





[CLJ-1297] try to catch using - instead of _ in filenames so the compiler can give a better error message for people who don't know that you need to use _ in file names Created: 19/Nov/13  Updated: 29/Aug/14

Status: Open
Project: Clojure
Component/s: None
Affects Version/s: None
Fix Version/s: Release 1.7

Type: Enhancement Priority: Major
Reporter: Kevin Downey Assignee: Unassigned
Resolution: Unresolved Votes: 11
Labels: compiler, errormsgs

Attachments: File better-error-messages-for-require.diff    
Patch: Code and Test
Approval: Incomplete

 Description   

Screener's Note: This works as advertised, but I have reservations about the approach. We could accept the patch as-is, or a much simpler patch that handles the only important (IMO) case: a-b-c to a_b_c – without generating and testing for unlikely errors like a-b_c. Please advise.

Problem: Clojure requires the files that back a namespace that has dashes in it to have the dashes replaced with underscores on the filesystem (ie a.b_c.clj for namespace a.b-c). If you require a file that has been mistakenly saved as b-c.clj instead, you will get an error message:

Exception in thread "main" java.io.FileNotFoundException: Could not locate a/b_c__init.class or a/b_c.clj on classpath:
	at clojure.lang.RT.load(RT.java:443)
	at clojure.lang.RT.load(RT.java:411)
	at clojure.core$load$fn__5018.invoke(core.clj:5530)
	at clojure.core$load.doInvoke(core.clj:5529)
	at clojure.lang.RestFn.invoke(RestFn.java:408)
	at clojure.core$load_one.invoke(core.clj:5336)
	at clojure.core$load_lib$fn__4967.invoke(core.clj:5375)
	at clojure.core$load_lib.doInvoke(core.clj:5374)
	at clojure.lang.RestFn.applyTo(RestFn.java:142)
	at clojure.core$apply.invoke(core.clj:619)
	at clojure.core$load_libs.doInvoke(core.clj:5413)
	at clojure.lang.RestFn.applyTo(RestFn.java:137)
	at clojure.core$apply.invoke(core.clj:619)
	at clojure.core$require.doInvoke(core.clj:5496)

Proposed:

  • When loading the resource-root of lib throws a FileNotFoundException, the lib is analyzed...
  • ... if the lib was a name that would be munged, it examines the combinatorial explosion of munge candidates and .clj or .class files in the classpath ...
  • ... if any of these candidates exist, it informs the user of the file's existence, and that a change to that filename would lead to that resource being loaded.
  • ... if none of these candidates exist, it throws the original exception.

It also modifies clojure.lang.RT to expose the behavior around finding clj or class files from a resource root.

Patch: better-error-messages-for-require.diff



 Comments   
Comment by Joshua Ballanco [ 20/Nov/13 12:15 AM ]

A perhaps even better solution would be to simply allow the use of dashes in *.clj[s] filenames. I can't imagine the extra disk access per-namespace would be a huge performance burden, and (since dashes aren't allowed currently) I don't think there would be any issues with backwards compatibility.

Comment by Gary Fredericks [ 20/Nov/13 8:40 AM ]

It's worth mentioning the combinatorial explosion for namespaces with multiple dashes – if I (require 'foo-bar.baz-bang), should clojure search for all four possible filenames? Does the jvm have a way to search for files by regex or similar to avoid nasty degenerate cases (like (require 'foo-------------))?

Comment by Joshua Ballanco [ 20/Nov/13 11:08 AM ]

According to the docs, the FileSystem class's "getPathMatcher" method accepts path globs, so you'd merely have to replace each instance of "-" or "_" with "{-,_}". Actual runtime characteristics would likely depend on the underlying filesystem's implementation.

Comment by Alex Miller [ 20/Nov/13 12:02 PM ]

I don't think the FileSystem stuff applies when looking up classes on the classpath. Note that Java class names cannot contain "-".

Comment by Phil Hagelberg [ 21/Nov/13 12:05 PM ]

According to the spec, Java class names can't contain dashes (though IIRC OpenJDK and Oracle's JDK accept them anyway) but the requirement that Clojure source files have names which align with their AOT'd class file eqivalents is something we've imposed upon ourselves. Introducing the disconnect between .clj files and .class files makes way more sense than disconnecting namespaces and .clj files, but arguably it's too late to fix that mistake.

In any case a check for dashed files (resulting only in a more informative compiler error, not a more permissive compiler) which only triggers when a .clj file cannot be found imposes zero overhead in the case where things are already working.

Comment by scott tudd [ 09/Dec/13 2:19 PM ]

As Clojure seems to be idiomatic to have sometimes-dashed-namespace-and-function-names as opposed to the ubiquitous camelCaseFunctionNames in java ... I agree to have the compiler automagically handle 'knowing' to look in dir_struct AND dir-struct for requisite files.

or at the least print out a nice message explaining the quirk when files "can't" be found ... WHEN there are dashes and underscores involved... anything to aid in helping things "just work" as one would think they're supposed to.

Comment by Obadz [ 12/Dec/13 5:28 AM ]

I would have saved a few hours as well.

Comment by Alexander Redington [ 14/Feb/14 2:29 PM ]

This patch changes clojure.core/load such that:

  • When loading the resource-root of lib throws a FileNotFoundException, the lib is analyzed...
  • ... if the lib was a name that would be munged, it examines the combinatorial explosion of munge candidates and .clj or .class files in the classpath ...
  • ... if any of these candidates exist, it informs the user of the file's existance, and that a change to that filename would lead to that resource being loaded.
  • ... if none of these candidates exist, it throws the original exception.

It also modifies clojure.lang.RT to expose the behavior around finding clj or class files from a resource root.

Comment by Andy Fingerhut [ 20/Mar/14 1:16 PM ]

I do not know whether it handles all of the cases proposed in this discussion, but I encourage folks to check out the filename/namespace consistency checking in the latest Eastwood release (version 0.1.1) to see if it catches the cases they would hope to catch. It does a static check based on the files in a Leiningen project, nothing at run time. https://github.com/jonase/eastwood

Of course changes to Clojure itself to give warnings about such things can still be very useful, since not everyone will be using a 3rd party tool to check for such things.

Comment by Alex Miller [ 27/Jun/14 2:24 PM ]

Re the screener's note at the top, my preference would be for the simpler approach.

Comment by Rich Hickey [ 29/Aug/14 9:48 AM ]

I see no reason to fish around in the file system at all. Why can't the message simply remind people that underscores are required and to check that they aren't using dashes?





[CLJ-1093] Empty PersistentCollections get incorrectly evaluated as their generic clojure counterpart Created: 24/Oct/12  Updated: 06/Jul/14

Status: Reopened
Project: Clojure
Component/s: None
Affects Version/s: Release 1.4, Release 1.5
Fix Version/s: Release 1.7

Type: Defect Priority: Minor
Reporter: Nicola Mometto Assignee: Unassigned
Resolution: Unresolved Votes: 4
Labels: collections, compiler

Attachments: Text File 0001-CLJ-1093-fix-compilation-of-empty-PersistentCollecti.patch     Text File clj-1093-fix-empty-record-literal-patch-v2.txt    
Patch: Code and Test
Approval: Incomplete

 Description   
user> (defrecord x [])
user.x
user> #user.x[]   ;; expect: #user.x{}
{}
user> #user.x{}   ;; expect: #user.x{}
{}
user> #clojure.lang.PersistentTreeMap[]
{}
user> (class *1)  ;; expect: clojure.lang.PersistentTreeMap
clojure.lang.PersistentArrayMap

Cause: Compiler's ConstantExpr parser returns an EmptyExpr for all empty persistent collections, even if they are of types other than the core collections (for example: records, sorted collections, custom collections). EmptyExpr reports its java class as one the classes - IPersistentList/IPersistentVector/IPersistentMap/IPersistentSet rather than the original type.

Proposed: If one of the Persistent* classes, then create EmptyExpr as before, otherwise retain the ConstantExpression of the original collection.
Since EmptyExpr is a compiler optimization that applies only to some concrete clojure collections, making EmptyExpr dispatch on concrete types rather than on generic interfaces makes the compiler behave as expected

Patch: 0001-CLJ-1093-fix-compilation-of-empty-PersistentCollecti.patch

Screened by:



 Comments   
Comment by Timothy Baldridge [ 27/Nov/12 11:41 AM ]

Unable to reproduce this bug on latest version of master. Most likely fixed by some of the recent changes to data literal readers.

Marking Not-Approved.

Comment by Timothy Baldridge [ 27/Nov/12 11:41 AM ]

Could not reproduce in master.

Comment by Nicola Mometto [ 01/Mar/13 1:23 PM ]

I just checked, and the problem still exists for records with no arguments:

Clojure 1.6.0-master-SNAPSHOT
user=> (defrecord a [])
user.a
user=> #user.a[]
{}

Admittedly it's an edge case and I see little usage for no-arguments records, but I think it should be addressed aswell since the current behaviour is not what one would expect

Comment by Herwig Hochleitner [ 02/Mar/13 8:14 AM ]

Got the following REPL interaction:

% java -jar ~/.m2/repository/org/clojure/clojure/1.5.0/clojure-1.5.0.jar
user=> (defrecord a [])
user.a
user=> (a.)
#user.a{}
user=> #user.a{}
{}
#user.a[]
{}

This should be reopened or declined for another reason than reproducability.

Comment by Nicola Mometto [ 10/Mar/13 2:18 PM ]

I'm reopening this since the bug is still there.

Comment by Andy Fingerhut [ 13/Mar/13 2:04 PM ]

Patch clj-1093-fix-empty-record-literal-patch-v2.txt dated Mar 13, 2013 is identical to Bronsa's patch 001-fix-empty-record-literal.patch dated Oct 24, 2012, except that it applies cleanly to latest master. I'm not sure why the older patch doesn't but git doesn't like something about it.

Comment by Nicola Mometto [ 26/Jun/13 8:06 PM ]

Patch 0001-CLJ-1093-fix-empty-records-literal-v2.patch solves more issues than the previous patch that was not evident to me at the time.

Only collections that are either PersistentList or PersistentVector or PersistentHash[Map|Set] or PersistentArrayMap can now be EmptyExpr.
This is because we don't want every IPersistentCollection to be emitted as a generic one eg.

user=> (class #clojure.lang.PersistentTreeMap[])
clojure.lang.PersistentArrayMap

Incidentally, this patch also solves CLJ-1187
This patch should be preferred over the one on CLJ-1187 since it's more general

Comment by Jozef Wagner [ 09/Aug/13 2:08 AM ]

Maybe this is related:

user=> (def x `(quote ~(list 1 (clojure.lang.PersistentTreeMap/create (seq [1 2 3 4])))))
#'user/x
user=> x
(quote (1 {1 2, 3 4}))
user=> (class (second (second x)))
clojure.lang.PersistentTreeMap
user=> (eval x)
(1 {1 2, 3 4})
user=> (class (second (eval x)))
clojure.lang.PersistentArrayMap

Even if the collection is not evaluated, it is still converted to the generic clojure counterpart.

Comment by Alex Miller [ 24/Apr/14 4:44 PM ]

In the change for ObjectExpr.emitValue() where you've added PersistentArrayMap to the PersistentHashMap case, should the IPersistentVector case below that be PersistentVector instead, otherwise it would snare a custom IPersistentVector that's not a PersistentVector, right?

This line: "else if(form instanceof ISeq)" at the end of the Compiler diff has different leading tabs which makes the diff slightly more confusing than it could be.

Would be nice to add a test for the sorted map case in the description.

Marking incomplete to address some of these.

Comment by Alex Miller [ 13/May/14 10:43 PM ]

bump

Comment by Nicola Mometto [ 14/May/14 4:24 AM ]

Attached patch 0001-CLJ-1093-fix-empty-collection-literal-evaluation.patch which implements your suggestions.

replacing IPersistentVector with PersistentVector in ObjectExpr.emitValue() exposes a print-dup limitation: it expects every IPersistentCollection to have a static "create" method.

This required special casing for MapEntry and APersistentVector$SubVector

Comment by Nicola Mometto [ 16/May/14 3:57 PM ]

I updated the patch adding print-dups for APersistentVector$SubVec and other IPersistentVectors rather than special casing them in the compiler

Comment by Alex Miller [ 23/May/14 4:21 PM ]

All of the checks on concrete classes in the Compiler parts of this patch don't sit well with me. I understand how you got to this point and I don't have an alternate recommendation (yet) but all of that just feels like the wrong direction.

We want to be built on abstractions such that internal collections are not special; they should conform to the same expectations as an external collection and both should follow the same paths in the compiler - needing to check for those types is a flag for me that something is amiss.

I am marking Incomplete for now based on these thoughts.

Comment by Nicola Mometto [ 06/Jul/14 10:01 AM ]

I've been thinking for a while about this issue and I've come to the conclusion that in my later patches I've been trying to incorporate fixes for 3 different albeit related issues:

1- Clojure transforms all empty IPersistentCollections in their generic Clojure counterpart

user> (defrecord x [])
user.x
user> #user.x[]   ;; expected: #user.x{}
{}
user> #user.x{}   ;; expected: #user.x{}
{}
user> #clojure.lang.PersistentTreeMap[]
{}
user> (class *1)  ;; expected: clojure.lang.PersistentTreeMap
clojure.lang.PersistentArrayMap

2- Clojure transforms all the literals of collections implementing the Clojure interfaces (IPersistentList, IPersistentVector ..) that are NOT defined with deftype or defrecord, to their
generic Clojure counterpart

user=> (class (eval (sorted-map 1 1)))
clojure.lang.PersistentArrayMap ;; expected: clojure.lang.PersistentTreeMap

3- print-dup is broken for some Clojure persistent collections

user=> (print-dup (subvec [1] 0) *out*)
#=(clojure.lang.APersistentVector$SubVector/create [1])
user=> #=(clojure.lang.APersistentVector$SubVector/create [1])
IllegalArgumentException No matching method found: create  clojure.lang.Reflector.invokeMatchingMethod (Reflector.java:53)

I'll keep this ticket regarding issue #1 and open two other tickets for issue #2 and #3

Comment by Nicola Mometto [ 06/Jul/14 10:15 AM ]

I've attached a new patch fixing only this issue, the approach is explained in the description





[CLJ-787] transient blows up when passed a vector created by subvec Created: 03/May/11  Updated: 23/May/14

Status: Open
Project: Clojure
Component/s: None
Affects Version/s: None
Fix Version/s: Release 1.7

Type: Defect Priority: Major
Reporter: Alexander Redington Assignee: Unassigned
Resolution: Unresolved Votes: 1
Labels: None

Attachments: Text File CLJ-787-p1.patch    
Patch: Code and Test
Approval: Incomplete

 Description   

Subvectors created with subvec from a PersistentVector cannot be made transient:

user=> (transient (subvec [1 2 3 4] 2))
ClassCastException clojure.lang.APersistentVector$SubVector cannot be cast to clojure.lang.IEditableCollection  clojure.core/transient (core.clj:2864)

Cause: APersistentVector$SubVector does not implement IEditableCollection

Patch: CLJ-787-p1.patch

Approach: Create a TransientSubVector based on an underlying TransientVector.

Two assumptions:

  • It's okay for TransientSubVector to delegate the ensureEditable functionality to the underlying TransientVector (sometimes explicitly, sometimes implicitly) - calling ensureEditable explicitly also requires that the field for the underlying vector be the concrete TransientVector type rather than the ITransientVector interface.
  • When an operation that would throw an exception on a PersistentVector happens from the wrong thread (or after persistent!), we throw that exception rather than the IllegalAccessError that transients throw when accessed inappropriately.


 Comments   
Comment by Stuart Sierra [ 31/May/11 9:28 AM ]

Confirmed. APersistentVector$SubVector does not implement IEditableCollection.

The current implementation of TransientVector depends on implementation details of PersistentVector, so it is not a trivial fix. The simplest fix might be to implement IEditableCollection.asTransient in SubVector by creating a new PersistentVector, but I do not know the performance implications.

Comment by Gary Fredericks [ 25/May/13 8:11 PM ]

We could get the same performance characteristics as SubVector by creating a TransientSubVector based on an underlying TransientVector, right?

Preparing a patch to that effect.

Comment by Gary Fredericks [ 25/May/13 10:58 PM ]

Text from the commit msg:

Made two assumptions:

  • It's okay for TransientSubVector to delegate the ensureEditable
    functionality to the underlying TransientVector (sometimes
    explicitely, sometimes implicitely) – calling ensureEditable
    explicitely also requires that the field for the underlying vector
    be the concrete TransientVector type rather than the
    ITransientVector interface.
  • When an operation that would throw an exception on a
    PersistentVector happens from the wrong thread (or after
    persistent!), we throw that exception rather than the
    IllegalAccessError that transients throw when accessed
    inappropriately.
Comment by Alex Miller [ 11/Oct/13 4:17 PM ]

I think there are some assumptions being made in this patch about the class structure here that do not hold. The structure is, admittedly, quite twisty.

A counter-example that highlights one of a few subtypes of APersistentVector that are not PersistentVector (like MapEntry):

user=> (transient (subvec (first {:a 1}) 0 1))
ClassCastException clojure.lang.MapEntry cannot be cast to clojure.lang.IEditableCollection  clojure.lang.APersistentVector$TransientSubVector.<init> (APersistentVector.java:592)

PersistentVector.SubVector expects to work on anything that implements IPersistentVector. Note that this includes concrete types such as MapEntry and LazilyPersistentVector, but could also be any user-implemented type IPersistentVector type too. TransientSubVector is making the assumption that the IPersistentVector in a SubVector question is also an IEditableCollection (that can be converted to be transient). Note that while PersistentVector implements TransientVector (and IEditableCollection), APersistentVector does not. To really implement this in tandem with SubVector, I think you would need to guarantee that IPersistentVector extended IEditableCollection and I don't think that's something we want to do.

I don't see an easy solution. Any time I see all these modifiers (Transient, Sub, etc) being created in different combinations, it is a clear sign that independent kinds of functionality are being remixed into single inheritance OO trees. You can see the same thing in most collection libraries (even Java's - need a ConcurrentIdentitySortedMap? too bad!).

Needs more thought.

Comment by Andy Fingerhut [ 08/Nov/13 10:17 AM ]

Patch CLJ-787-p1.patch no longer applies cleanly to latest master, but it is only because of some new tests added to the transients.clj file since the patch was created, so it is trivial to update in that sense. Not updating it now due to other more significant issues with the patch described above.

Comment by Alex Miller [ 17/Jan/14 10:19 AM ]

No good solution to consider right now, removing from 1.6.





[CLJ-1161] sources jar has bad versions.properties resource Created: 11/Feb/13  Updated: 25/Apr/14

Status: Reopened
Project: Clojure
Component/s: None
Affects Version/s: Release 1.4, Release 1.5
Fix Version/s: Release 1.6, Release 1.7

Type: Defect Priority: Minor
Reporter: Steve Miner Assignee: Stuart Halloway
Resolution: Unresolved Votes: 0
Labels: None

Attachments: Text File 0001-CLJ-1161-Remove-version.properties-from-sources-JAR.patch    
Patch: Code
Approval: Incomplete

 Description   

The "sources" jar (at least since Clojure 1.4 and including 1.5 RC) has a bad version.properties file in it. The resource clojure/version.properties is literally:

version=${version}

The regular Clojure jar has the correct version string in that resource.

I came across a problem when I was experimenting with the sources jar (as used by IDEs). I naively added the sources jar to my classpath, and Clojure died on start up. The bad clojure/versions.properties file was found first, which led to a parse error as the clojure version was being set.

Solution: patch leaves version.properties file out of sources JAR, where it causes problems for tools.



 Comments   
Comment by Steve Miner [ 11/Feb/13 10:04 AM ]

Notes from the dev mailing list:

The "sources" JAR is generated by another Maven plugin, configured here:
https://github.com/clojure/clojure/blob/clojure-1.5.0-RC15/pom.xml#L169-L181

The simplest solution might be to just exclude the file from the sources jar. It looks like maven-source-plugin has an excludes option which would do the trick:

http://maven.apache.org/plugins/maven-source-plugin/jar-mojo.html#excludes

Comment by Jeff Valk [ 21/Apr/14 8:20 AM ]

This issue is marked closed, but I'm still seeing it: the clojure-1.6.0-sources.jar, clojure-1.5.1-sources.jar, etc on Maven Central still contain the bad version.properties files. More specifically, it looks like the fix has been applied to builds in the SNAPSHOTS repository, but not to RELEASES.

Fix applied: https://oss.sonatype.org/content/repositories/snapshots/org/clojure/clojure/
Not fixed: https://oss.sonatype.org/content/repositories/releases/org/clojure/clojure/

Comment by Alex Miller [ 24/Apr/14 4:15 PM ]

Not sure what's needed here, but marking incomplete.

Comment by Jeff Valk [ 25/Apr/14 11:13 AM ]

Would a fix for this update existing sources jars (1.5.1, 1.6.0, etc) on Central? Or would any fix have to wait on the next Clojure release?

Comment by Alex Miller [ 25/Apr/14 12:37 PM ]

For all the same reasons that mutable state is undesirable, changing an existing release jar in the central Maven repository is also undesirable. While it's not technically impossible, we will not update existing releases and this will need to wait for the next. I've looked at this problem a little and I do not yet know enough to know how to fix it or why it even varies between snapshot and release. Help welcome!

In which tool do you see a resulting problem from this?

Comment by Jeff Valk [ 25/Apr/14 11:56 PM ]

Despite the way I phrased the question, I'd hoped that would be the answer. It's the right policy.

Unfortunately, this issue leaves the released sources jars essentially unusable from a tools standpoint. CIDER now has source code navigation from stacktraces – you can jump into both Clojure and Java function definitions from the error/trace. For the latter, the sources jar (for Clojure or any other Java library) needs to be on the classpath as a dev dependency. There's more host interop support in the works for CIDER too ("embrace the host platform"), but not being able to add a dependency on a stable Clojure sources jar presents a wrinkle.

Are the official Clojure releases built by Hudson? The Hudson build right before the 1.6.0 release (#532) and the one right after (#534) both show the exclusion fix, as does the git clojure-1.6.0 tag, which when I check out and build from source, is fine. The Hudson builds with release tags (e.g. 1.6 = #533, 1.6-RC1 = #512, etc), though, don't show any artifacts other than a pom.xml. This would seem to make it rather hard to audit builds...am I missing something?





Generated at Mon Sep 01 23:58:55 CDT 2014 using JIRA 4.4#649-r158309.