Debian Bug report logs - #877288
r-base: fix calling methods on dots generics

Toggle useless messages

Message #5 received at submit@bugs.debian.org (full text, mbox, reply):

From: Graham Inggs <ginggs@debian.org>

To: Debian Bug Tracking System <submit@bugs.debian.org>

Subject: r-base: Rebuilding r-bioc-biocgenerics with new R changes behaviour

Date: Sat, 30 Sep 2017 09:13:10 +0200

Source: r-base
Version: 3.4.1.20170921-1
Affects: r-bioc-biocgenerics r-bioc-iranges r-bioc-variantannotation
User: ubuntu-devel@lists.ubuntu.com
Usertags: origin-ubuntu artful autopkgtest

Hi Dirk

Rebuilding the arch:all package r-bioc-biocgenerics against r-base-dev
3.4.1.20170921-1 or 3.4.2-1 seems to cause the eventual failure of
autopkgtests in r-bioc-iranges and r-bioc-variantannotation.

IRanges RUnit Tests - 91 test functions, 2 errors, 0 failures
ERROR in test_AtomicList_general: Error in match.arg(method) : 'arg'
must be of length 1
ERROR in test_AtomicList_numerical: Error in match.arg(method) : 'arg'
must be of length 1

VariantAnnotation RUnit Tests - 80 test functions, 1 error, 0 failures
ERROR in test_VRanges_vcf: Error in match.arg(method) : 'arg' must be
of length 1

On a system with all of the packages built against r-base-dev 3.4.2-1,
force-installing only the old r-bioc-biocgenerics package makes the
"'arg' must be of length 1" error disappear and the tests pass.
Similarly, on a system with all of the old packages, force-installing
only r-bioc-biocgenerics built against r-base-dev 3.4.2-1 causes the
error to appear.

Any ideas what could be causing this or how I can debug this further,
please?  Might r-bioc-biocgenerics/r-bioc-iranges/r-bioc-variantannotation
need some changes to be compatible with R 3.4.2?

Regards
Graham

Message #16 received at 877288@bugs.debian.org (full text, mbox, reply):

From: Charles Plessy <plessy@debian.org>

To: Graham Inggs <ginggs@debian.org>, 877288@bugs.debian.org

Subject: Re: Bug#877288: r-base: Rebuilding r-bioc-biocgenerics with new R changes behaviour

Date: Sat, 30 Sep 2017 16:52:32 +0900

Le Sat, Sep 30, 2017 at 09:13:10AM +0200, Graham Inggs a écrit :
> 
> Rebuilding the arch:all package r-bioc-biocgenerics against r-base-dev
> 3.4.1.20170921-1 or 3.4.2-1 seems to cause the eventual failure of
> autopkgtests in r-bioc-iranges and r-bioc-variantannotation.
> 
> IRanges RUnit Tests - 91 test functions, 2 errors, 0 failures
> ERROR in test_AtomicList_general: Error in match.arg(method) : 'arg'
> must be of length 1
> ERROR in test_AtomicList_numerical: Error in match.arg(method) : 'arg'
> must be of length 1

Hi Graham,

this reminds me the transition from R 3.3.3 to 3.4.0, although the
error message is not the same.  (See <https://bugs.debian.org/858183>).

The order of builds may matter (because of subtelties of the S4 system).
Have you tried rebuilding r-bioc-iranges in a system after the rebuilt
r-bioc-biocgenerics package was installed ?

Also, I note that Upstream has not switched to R 3.4.2 yet (as of 2017-09-28)
(<http://bioconductor.org/checkResults/devel/bioc-LATEST/BiocGenerics/malbec1-checksrc.html>)
Therefore, the issue may still be unknown upstream.

If the problems persists, I suggest reporting them at <https://support.bioconductor.org>.

Cheers,

Charles

-- 
Charles Plessy
Debian Med packaging team,
http://www.debian.org/devel/debian-med
Tsurumi, Kanagawa, Japan

Message #21 received at 877288@bugs.debian.org (full text, mbox, reply):

From: Graham Inggs <ginggs@debian.org>

To: Charles Plessy <plessy@debian.org>

Cc: 877288@bugs.debian.org

Subject: Re: Bug#877288: r-base: Rebuilding r-bioc-biocgenerics with new R changes behaviour

Date: Sat, 30 Sep 2017 10:04:58 +0200

On 30 September 2017 at 09:52, Charles Plessy <plessy@debian.org> wrote:
> The order of builds may matter (because of subtelties of the S4 system).
> Have you tried rebuilding r-bioc-iranges in a system after the rebuilt
> r-bioc-biocgenerics package was installed ?

I did rebuild r-bioc-iranges after r-bioc-biocgenerics, but will try
rebuilding in different orders, thanks.

> Also, I note that Upstream has not switched to R 3.4.2 yet (as of 2017-09-28)
> (<http://bioconductor.org/checkResults/devel/bioc-LATEST/BiocGenerics/malbec1-checksrc.html>)
> Therefore, the issue may still be unknown upstream.
>
> If the problems persists, I suggest reporting them at <https://support.bioconductor.org>.

OK, thanks!

Message #26 received at 877288@bugs.debian.org (full text, mbox, reply):

From: Charles Plessy <plessy@debian.org>

To: 877288@bugs.debian.org

Cc: Graham Inggs <ginggs@debian.org>

Subject: Re: Bug#877288: r-base: Rebuilding r-bioc-biocgenerics with new R changes behaviour

Date: Sat, 30 Sep 2017 17:06:38 +0900

Le Sat, Sep 30, 2017 at 04:52:32PM +0900, Charles Plessy a écrit :
> The order of builds may matter (because of subtelties of the S4 system).
> Have you tried rebuilding r-bioc-iranges in a system after the rebuilt
> r-bioc-biocgenerics package was installed ?

Sorry, I did not see that the binNMU was already done for r-bioc-iranges (it
does not show on the tracker).

> If the problems persists, I suggest reporting them at <https://support.bioconductor.org>.

I think that it is definitely worth a report.

Cheers,

-- 
Charles

Message #33 received at submit@bugs.debian.org (full text, mbox, reply):

From: Dirk Eddelbuettel <edd@debian.org>

To: Graham Inggs <ginggs@debian.org>, 877288@bugs.debian.org

Cc: Debian Bug Tracking System <submit@bugs.debian.org>

Subject: Re: Bug#877288: r-base: Rebuilding r-bioc-biocgenerics with new R changes behaviour

Date: Sat, 30 Sep 2017 07:30:39 -0500

On 30 September 2017 at 09:13, Graham Inggs wrote:
| Source: r-base
| Version: 3.4.1.20170921-1

Out of date. We have 3.4.2 now.

| Affects: r-bioc-biocgenerics r-bioc-iranges r-bioc-variantannotation
| User: ubuntu-devel@lists.ubuntu.com
| Usertags: origin-ubuntu artful autopkgtest

Also wrong package. r-base, really r-base-core, provides the engine.

But I see no evidence of anything at fault at r-base-core.  I think you have
a case of wrongly ordered builds of dependencies here _in the client package_
so please reassign the bug report.


Dirk

-- 
http://dirk.eddelbuettel.com | @eddelbuettel | edd@debian.org

Message #43 received at 877288@bugs.debian.org (full text, mbox, reply):

From: Graham Inggs <ginggs@debian.org>

To: 877288@bugs.debian.org, Debian Med Project List <debian-med@lists.debian.org>

Subject: Re: Bug#877288: r-base: Rebuilding r-bioc-biocgenerics with new R changes behaviour

Date: Wed, 4 Oct 2017 22:44:57 +0200

Control: reassign -1 src:r-bioc-s4vectors 0.14.3-1
Control: affects -1 = r-bioc-iranges r-bioc-variantannotation
Control: tags -1 + fixed-upstream
Control: clone -1 -2
Control: reassign -2 src:r-bioc-iranges 2.10.2-1


On 30 September 2017 at 14:30, Dirk Eddelbuettel <edd@debian.org> wrote:
> Out of date. We have 3.4.2 now.

3.4.1.20170921-1 is the version where the regression appeared.

> But I see no evidence of anything at fault at r-base-core.  I think you have
> a case of wrongly ordered builds of dependencies here _in the client package_
> so please reassign the bug report.

Upstream have released workarounds in S4Vectors 0.14.6 and IRanges
2.10.4, so reassigning.
See commit from S4Vectors below:

diff --git a/R/List-comparison.R b/R/List-comparison.R
index c096a00..6f788fb 100644
--- a/R/List-comparison.R
+++ b/R/List-comparison.R
@@ -43,7 +43,16 @@ setMethod("pcompareRecursively", "list", function(x) TRUE)
         return(OP1_Vector_method(x, ...))
     }
     compress_ans <- !is(x, "SimpleList")
-    ANS_CONSTRUCTOR(lapply(x, OP1, ...), compress=compress_ans)
+    ## Note that we should just be able to do
+    ##   y <- lapply(x, OP1, ...)
+    ## instead of the extremely obfuscated form below (which, in a bug-free
+    ## world, should be equivalent to the simple form above).
+    ## However, because of a regression in R 3.4.2, using the simple form
+    ## above doesn't seem to work properly if OP1 is a generic function with
+    ## dispatch on ... (e.g. order()). The form below seems to work though,
+    ## so we use it as a temporary workaround.
+    y <- lapply(x, function(xi) do.call(OP1, list(xi, ...)))
+    ANS_CONSTRUCTOR(y, compress=compress_ans)
 }

 ### Apply a binary operator.

Message #60 received at 877288@bugs.debian.org (full text, mbox, reply):

From: Graham Inggs <ginggs@debian.org>

To: 877288@bugs.debian.org, Dirk Eddelbuettel <edd@debian.org>

Subject: r-base: fix calling methods on dots generics

Date: Fri, 13 Oct 2017 10:47:57 +0200

Control: reassign -1 src:r-base 3.4.1.20170921-1
Control: affects -1 =
Control: retitle -1 r-base: fix calling methods on dots generics

Hi Dirk

r-bioc-biocgenerics, r-bioc-s4vectors and r-bioc-iranges all got 
workarounds for this and autopkgtests are passing again [1][2].

The regression was fixed in R-3-4-branch r73465.  I have no idea if 
anything else in the archive is affected as not all packages have 
working autopkgtests.  Would you please consider uploading 'R 3.4.2 
Patched' from svn >= r73465?  r73475 includes a test.

Regards
Graham


[1] https://ci.debian.net/packages/r/r-bioc-iranges/unstable/amd64/
[2] 
https://ci.debian.net/packages/r/r-bioc-variantannotation/unstable/amd64/

Message #75 received at 877288@bugs.debian.org (full text, mbox, reply):

From: Dirk Eddelbuettel <edd@debian.org>

To: Graham Inggs <ginggs@debian.org>

Cc: 877288@bugs.debian.org, Dirk Eddelbuettel <edd@debian.org>

Subject: Re: r-base: fix calling methods on dots generics

Date: Fri, 13 Oct 2017 06:24:48 -0500

On 13 October 2017 at 10:47, Graham Inggs wrote:
| Control: reassign -1 src:r-base 3.4.1.20170921-1
| Control: affects -1 =
| Control: retitle -1 r-base: fix calling methods on dots generics
| 
| Hi Dirk
| 
| r-bioc-biocgenerics, r-bioc-s4vectors and r-bioc-iranges all got 
| workarounds for this and autopkgtests are passing again [1][2].
| 
| The regression was fixed in R-3-4-branch r73465.  I have no idea if 
| anything else in the archive is affected as not all packages have 
| working autopkgtests.  Would you please consider uploading 'R 3.4.2 
| Patched' from svn >= r73465?  r73475 includes a test.

Could do.  We never needed a patch release in all those years.

Some years we end up with three, four, or even five point releases before the
next annual one.  I'd be inclined to just wait.

Dirk

-- 
http://dirk.eddelbuettel.com | @eddelbuettel | edd@debian.org

Message #80 received at 877288@bugs.debian.org (full text, mbox, reply):

From: Dirk Eddelbuettel <edd@debian.org>

To: Dirk Eddelbuettel <edd@debian.org>

Cc: Graham Inggs <ginggs@debian.org>, 877288@bugs.debian.org

Subject: Re: r-base: fix calling methods on dots generics

Date: Fri, 13 Oct 2017 07:07:11 -0500

On 13 October 2017 at 06:24, Dirk Eddelbuettel wrote:
| 
| On 13 October 2017 at 10:47, Graham Inggs wrote:
| | Control: reassign -1 src:r-base 3.4.1.20170921-1
| | Control: affects -1 =
| | Control: retitle -1 r-base: fix calling methods on dots generics
| | 
| | Hi Dirk
| | 
| | r-bioc-biocgenerics, r-bioc-s4vectors and r-bioc-iranges all got 
| | workarounds for this and autopkgtests are passing again [1][2].
| | 
| | The regression was fixed in R-3-4-branch r73465.  I have no idea if 
| | anything else in the archive is affected as not all packages have 
| | working autopkgtests.  Would you please consider uploading 'R 3.4.2 
| | Patched' from svn >= r73465?  r73475 includes a test.
| 
| Could do.  We never needed a patch release in all those years.
| 
| Some years we end up with three, four, or even five point releases before the
| next annual one.  I'd be inclined to just wait.

Today's r-devel update [1] reveals another candidate: Java fixes I may need
for rJava.

[1] http://developer.r-project.org/blosxom.cgi/R-devel/2017/10/13#n2017-10-13

Dirk

-- 
http://dirk.eddelbuettel.com | @eddelbuettel | edd@debian.org

Message #85 received at 877288@bugs.debian.org (full text, mbox, reply):

From: Dirk Eddelbuettel <edd@debian.org>

To: Charles Plessy <plessy@debian.org>, 877288@bugs.debian.org

Cc: Graham Inggs <ginggs@debian.org>

Subject: Re: Bug#877288: r-base: Rebuilding r-bioc-biocgenerics with new R changes behaviour

Date: Sun, 19 Nov 2017 10:08:43 -0600

Graham, Charles,

R 3.4.3 is scheduled for Nov 30.

Am I correct in assuming that it addresses this issue for BioConductor?

Thanks,  Dirk

-- 
http://dirk.eddelbuettel.com | @eddelbuettel | edd@debian.org

Message #90 received at 877288@bugs.debian.org (full text, mbox, reply):

From: Graham Inggs <ginggs@debian.org>

To: Dirk Eddelbuettel <edd@debian.org>, Charles Plessy <plessy@debian.org>, 877288@bugs.debian.org

Subject: Re: Bug#877288: r-base: Rebuilding r-bioc-biocgenerics with new R changes behaviour

Date: Mon, 20 Nov 2017 17:06:46 +0200

Hi Dirk

On 19/11/2017 18:08, Dirk Eddelbuettel wrote:
> Am I correct in assuming that it addresses this issue for BioConductor?

I had a look at what I assumed is the first beta [1] for 3.4.3.  I see 
the following in NEWS:

"Fix calling of methods on S4 generics that dispatch on ... when the 
call contains ..."

I believe this is the fix for this bug.

Regards
Graham


[1] 
https://cran.r-project.org/src/base-prerelease/R-beta_2017-11-19_r73753.tar.gz

Message #95 received at 877288@bugs.debian.org (full text, mbox, reply):

From: Dirk Eddelbuettel <edd@debian.org>

To: Graham Inggs <ginggs@debian.org>, 877288@bugs.debian.org

Cc: Dirk Eddelbuettel <edd@debian.org>, Charles Plessy <plessy@debian.org>

Subject: Re: Bug#877288: r-base: Rebuilding r-bioc-biocgenerics with new R changes behaviour

Date: Mon, 20 Nov 2017 09:33:25 -0600

Hi Graham,

On 20 November 2017 at 17:06, Graham Inggs wrote:
| Hi Dirk
| 
| On 19/11/2017 18:08, Dirk Eddelbuettel wrote:
| > Am I correct in assuming that it addresses this issue for BioConductor?
| 
| I had a look at what I assumed is the first beta [1] for 3.4.3.  I see

I am actually building that [1] as we speak. (Well, started, had issues with
getting to versions of the 3.4.2~20171120-1 form, and am now at work ...)

| the following in NEWS:
| 
| "Fix calling of methods on S4 generics that dispatch on ... when the 
| call contains ..."
| 
| I believe this is the fix for this bug.

Thanks for checking!  In general the BioC-catching-up-to-R chain should be
well honed now.

Dirk

-- 
http://dirk.eddelbuettel.com | @eddelbuettel | edd@debian.org

Message #100 received at 877288-done@bugs.debian.org (full text, mbox, reply):

From: Graham Inggs <ginggs@debian.org>

To: 877288-done@bugs.debian.org

Subject: Re: Bug#877288: r-base: Rebuilding r-bioc-biocgenerics with new R changes behaviour

Date: Thu, 7 Dec 2017 11:55:53 +0200

Version: 3.4.2.20171120-1

The first R 3.4.3 beta included this fix.

Send a report that this bug log contains spam.