[Bioc-devel] Error : object 'lengths' is not exported by 'namespace:BiocGenerics'

Kasper Daniel Hansen k@@perd@nielh@n@en @ending from gm@il@com
Wed Nov 14 15:08:58 CET 2018 

It seems the short answer is:
  yes, base::lengths() works and should be fast
If it either doesn't work or is slow, you should report it.

On Tue, Nov 13, 2018 at 3:33 PM Pages, Herve <hpages using fredhutch.org> wrote:

> I'm going to try to provide some details, at the risk to confuse you even
> more.
>
>
> In BioC 3.8 / R 3.5 base::lengths() was not just slow on some Bioconductor
> objects like IRanges or GRanges, it didn't work:
>
>
>   > ir <- IRanges(1:21, 20)
>   > base::lengths(ir)
>   Error in getListElement(x, i, ...) :
>     IRanges objects don't support [[, as.list(), lapply(), or unlist() at
>     the moment
>
>
> Now it works (in BioC 3.9 / R 3.6):
>
>
>   > ir <- IRanges(1:21, 20)
>   > base::lengths(ir)
>    [1] 20 19 18 17 16 15 14 13 12 11 10  9  8  7  6  5  4  3  2  1  0
>
>
> The change I announced in my original post is about base::lengths() being
> now an S4 generic. A standard S4 generic like BiocGenerics::lengths() or
> base::lengths() doesn't do any work beyond dispatching to the right method.
> What happened in R 3.6 is that base::lengths() is now an S4 generic. This
> means that it knows how to take care of dispatching to the right method.
> The various "lengths" methods defined in Bioconductor are the ones doing
> the real work and they are fast. They have not changed and their speed
> should not be affected by this change in base::lengths().
>
>
> Furthermore: in R 3.6, even though you can't see it, there is a default
> "lengths" method defined in base (it's used as a fallback when no other
> method applies). In my 1st code chunk above (i.e. when base::lengths() was
> not a generic), I'm calling directly this default method (note that this
> was the default method of the BiocGenerics::lengths() generic). This
> default method is indeed slow or doesn't work on Bioconductor objects. By
> calling BiocGenerics::lengths() (in BioC 3.8) or base::lengths() (in BioC
> 3.9), dispatch selects the fast method so nothing has changed from that
> perspective.
>
>
> Hope this helps,
>
> H.
>
>
> On 11/13/18 08:59, Håkon Tjeldnes wrote:
> Just to be sure, you are saying that now base::lengths is as fast as
> BiocGenerics::lengths was ? Else we would still need the Bioc version.
> Because the old base implementation was terrible, I will try to verify this
> if you are unsure.
> ________________________________
> Fra: Michael Lawrence <lawrence.michael using gene.com><mailto:
> lawrence.michael using gene.com>
> Sendt: fredag 2. november 2018 23.50
> Til: hauken_heyken using hotmail.com<mailto:hauken_heyken using hotmail.com>
> Kopi: bioc-devel; Hervé Pagès
> Emne: Re: [Bioc-devel] Error : object 'lengths' is not exported by
> 'namespace:BiocGenerics'
>
> Yes, please follow Hervé's instructions.
>
> On Fri, Nov 2, 2018 at 3:41 PM Håkon Tjeldnes <hauken_heyken using hotmail.com
> <mailto:hauken_heyken using hotmail.com>> wrote:
> Just a quick question, in 3.8 we used biogenerics::lengths for ORFik,
> since it was much faster than base, for our dataset 1 second vs 30 minutes.
> We should change to S4 version of lengths now ?
>
> Get Outlook for Android<https://aka.ms/ghei36<
> https://urldefense.proofpoint.com/v2/url?u=https-3A__eur02.safelinks.protection.outlook.com_-3Furl-3Dhttps-253A-252F-252Faka.ms-252Fghei36-26data-3D02-257C01-257C-257Ce617c34c45584b70cf7008d6411590f9-257C84df9e7fe9f640afb435aaaaaaaaaaaa-257C1-257C0-257C636767958210704172-26sdata-3DUl3jkwHfZdrwQjkzzBQTKrXdQ5NXQ1vWRrx7vojZEAw-253D-26reserved-3D0&d=DwMFAw&c=eRAMFD45gAfqt84VtBcfhQ&r=BK7q3XeAvimeWdGbWY_wJYbW0WYiZvSXAJJKaaPhzWA&m=VkR31AqDi_FLujiTJm39v-jaLNkPJov4_ZlFjFit9UI&s=C8hC46CIc0bUw9WUxrvaekaCsN1NAlLcyxiuFKgRjCs&e=
> >>
>
> ________________________________
> From: Bioc-devel <bioc-devel-bounces using r-project.org<mailto:
> bioc-devel-bounces using r-project.org>> on behalf of Pages, Herve <
> hpages using fredhutch.org<mailto:hpages using fredhutch.org>>
> Sent: Friday, November 2, 2018 6:08:13 PM
> To: bioc-devel using r-project.org<mailto:bioc-devel using r-project.org>
> Subject: [Bioc-devel] Error : object 'lengths' is not exported by
> 'namespace:BiocGenerics'
>
> Hi developers,
>
> Starting with R 3.6, base::lengths() can now be treated as an S4 generic
> function (like base::length() or base::names()) so there is no need for
> us to define our own lengths() generic function in BiocGenerics.
> Therefore we've removed the lengths() generic from BiocGenerics (in
> version 0.29.1). Note that you can see the S4 generic associated with
> base::lengths() by calling getGeneric("lengths") in a fresh R session:
>
>    > getGeneric("lengths")
>    standardGeneric for "lengths" defined from package "base"
>
>    function (x, use.names = TRUE)
>    standardGeneric("lengths")
>    <bytecode: 0x3548c90>
>    <environment: 0x3540928>
>    Methods may be defined for arguments: x
>    Use  showMethods("lengths")  for currently available ones.
>
> This used to return NULL in R < 3.6.
>
> Because of these changes, packages that currently explicitly import
> lengths or a lengths method in their NAMESPACE are failing to install
> with an error that looks like this:
>
>    ** byte-compile and prepare package for lazy loading
>    Error : object 'lengths' is not exported by 'namespace:BiocGenerics'
>    ERROR: lazy loading failed for package 'BubbleTree'
>
> The packages currently affected by this are: BubbleTree, csaw,
> GA4GHclient, matter, NADfinder, and SomaticSignatures. The fix is to
> simply remove any explicit import of lengths from the NAMESPACE.
>
> Note that many other software packages currently fail to install in BioC
> 3.9 because they depend on one of the above packages.
>
> Please let us know if you have questions or need assistance with this.
>
> H.
>
> --
> Hervé Pagès
>
> Program in Computational Biology
> Division of Public Health Sciences
> Fred Hutchinson Cancer Research Center
> 1100 Fairview Ave. N, M1-B514
> P.O. Box 19024
> Seattle, WA 98109-1024
>
> E-mail: hpages using fredhutch.org<mailto:hpages using fredhutch.org>
> Phone:  (206) 667-5791
> Fax:    (206) 667-1319
>
> _______________________________________________
> Bioc-devel using r-project.org<mailto:Bioc-devel using r-project.org> mailing list
>
> https://nam05.safelinks.protection.outlook.com/?url=https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fbioc-devel&data=02%7C01%7C%7C7c4bb0f70b934115a93708d6410fbb9d%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636767933158117589&sdata=0UfC1T6Q52n1V%2FcKb6vR4RT0h5m0DXuWv4C6EwJ2YCA%3D&reserved=0
> <
> https://urldefense.proofpoint.com/v2/url?u=https-3A__eur02.safelinks.protection.outlook.com_-3Furl-3Dhttps-253A-252F-252Fstat.ethz.ch-252Fmailman-252Flistinfo-252Fbioc-2Ddevel-26data-3D02-257C01-257C-257Ce617c34c45584b70cf7008d6411590f9-257C84df9e7fe9f640afb435aaaaaaaaaaaa-257C1-257C0-257C636767958210704172-26sdata-3DnWJ1xrPRmDExBtqAQt4QnYojC8btTQCDKbNHNtGXEKg-253D-26reserved-3D0&d=DwMFAw&c=eRAMFD45gAfqt84VtBcfhQ&r=BK7q3XeAvimeWdGbWY_wJYbW0WYiZvSXAJJKaaPhzWA&m=VkR31AqDi_FLujiTJm39v-jaLNkPJov4_ZlFjFit9UI&s=X1qstfX2NwHi1UEDdonTJAg-XpXfkWfEjftcE9o8S3c&e=
> >
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> Bioc-devel using r-project.org<mailto:Bioc-devel using r-project.org> mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel<
> https://urldefense.proofpoint.com/v2/url?u=https-3A__eur02.safelinks.protection.outlook.com_-3Furl-3Dhttps-253A-252F-252Fstat.ethz.ch-252Fmailman-252Flistinfo-252Fbioc-2Ddevel-26data-3D02-257C01-257C-257Ce617c34c45584b70cf7008d6411590f9-257C84df9e7fe9f640afb435aaaaaaaaaaaa-257C1-257C0-257C636767958210704172-26sdata-3DnWJ1xrPRmDExBtqAQt4QnYojC8btTQCDKbNHNtGXEKg-253D-26reserved-3D0&d=DwMFAw&c=eRAMFD45gAfqt84VtBcfhQ&r=BK7q3XeAvimeWdGbWY_wJYbW0WYiZvSXAJJKaaPhzWA&m=VkR31AqDi_FLujiTJm39v-jaLNkPJov4_ZlFjFit9UI&s=X1qstfX2NwHi1UEDdonTJAg-XpXfkWfEjftcE9o8S3c&e=
> >
>
> --
> Hervé Pagès
>
> Program in Computational Biology
> Division of Public Health Sciences
> Fred Hutchinson Cancer Research Center
> 1100 Fairview Ave. N, M1-B514
> P.O. Box 19024
> Seattle, WA 98109-1024
>
> E-mail: hpages using fredhutch.org<mailto:hpages using fredhutch.org>
> Phone:  (206) 667-5791
> Fax:    (206) 667-1319
>
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> Bioc-devel using r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel
>

	[[alternative HTML version deleted]]

More information about the Bioc-devel mailing list