[Bioc-devel] Citation info still not displayed correctly for some packages

Alexey Sergushichev alsergbox at gmail.com
Tue Aug 30 14:15:03 CEST 2016


I see.

> As a developer I would choose to write my CITATION file in a way that did
not require the package to be installed.
I agree. I would suggest then to include this into BiocCheck, so that a
warning or at least a note could be generated, when citation is not working
properly.

--
Alexey

On Tue, Aug 30, 2016 at 3:08 PM, Martin Morgan <
martin.morgan at roswellpark.org> wrote:

> On 08/30/2016 06:56 AM, Alexey Sergushichev wrote:
>
>> Martin,
>>
>> Won't it be easier to include actually installation for each package in
>> the build script, but into a separate temp folder? Thus every package
>> will be isntalled and citation() should work fine.
>>
>
> installation is computationally costly and the the build system is taxed
> for resources. We do have new hardware in the works and if that loosens the
> bottleneck then yes, we would  install all packages.
>
> It is convenient to be able to get citation information from packages that
> are not installed, and the 'meta' argument to readCitationFile() seems to
> have been introduced for exactly this reason. As a developer I would choose
> to write my CITATION file in a way that did not require the package to be
> installed.
>
> Martin
>
>
>> Best,
>> Alexey
>>
>> On Tue, Aug 30, 2016 at 1:17 AM, Martin Morgan
>> <martin.morgan at roswellpark.org <mailto:martin.morgan at roswellpark.org>>
>> wrote:
>>
>>     On 08/29/2016 05:40 PM, Ramon Diaz-Uriarte wrote:
>>
>>
>>         Dear All,
>>
>>         At least for some packages
>>         (http://bioconductor.org/packages/devel/bioc/html/OncoSimulR.html
>> <http://bioconductor.org/packages/devel/bioc/html/OncoSimulR.html>)
>>         the
>>         citation info is still not displayed correctly. I double checked
>>         installing
>>         from BioC just now, and the citation is not the one shown in the
>>         landing
>>         page on BioC;  the vignette
>>         (http://www.bioconductor.org/packages/3.4/bioc/vignettes/Onc
>> oSimulR/inst/doc/OncoSimulR.html#15_citing_oncosimulr_and_
>> other_documentation
>>         <http://www.bioconductor.org/packages/3.4/bioc/vignettes/Onc
>> oSimulR/inst/doc/OncoSimulR.html#15_citing_oncosimulr_and_
>> other_documentation>)
>>
>>         also shows the output from citation("OncoSimulR"), which again
>>         is not the
>>         same as in the BioC page.
>>
>>         I think that at least in this case the citation info has not
>>         been updated
>>         in the last two builds.
>>
>>
>>     biocViews (which processes the citations) was updated in the last
>>     build, it should be used in the next build. I'll keep an eye on it.
>>
>>     OncoSimulR seems to have a valid CITATION file.
>>
>>     For a little more information, there are a number of outstanding
>>     problems with citation processing. Citation information is added
>>     using readCitationFile() applied to the package _source_, rather
>>     than the installed package. This is because some packages (those
>>     without any reverse dependencies) are never actually installed on
>>     the build system -- they are built into a tar ball and checked, and
>>     the built tarball is used to generate citation information.
>>
>>     This means that CITATION files that do things like
>>     packageDescription("foo") fail, because "foo" is not installed. The
>>     current solution is to rely on the 'meta' argument to
>>     readCitationFile(), so CITATION files can replace lines like
>>
>>         version <- packageDescription("foo")$Version
>>
>>     with
>>
>>         version <- meta$Version
>>
>>     or more flexibly
>>
>>         if (!exists("meta") || is.null(meta))
>>             meta <- packageDescription("nlme")
>>
>>     and one can test that one has a valid CITATION file with
>>
>>         descr = packageDescription("pkg-name", "/path/to/pkg")
>>         readCitationFile("/path/to/pkg/pkg-name/inst/CITATION", descr)
>>
>>     There are related problems, e.g., with use of the knitrcitation or
>>     omitting the required field 'textVersion', and a bug (generating a
>>     warning, but apparently not compromising the citation) when a url or
>>     doi field contains an encoded url, as
>>
>>     > xx <<- "\\url{http://foo%3Fbar}"; con <- textConnection(xx);
>>     parse_Rd(con, fragment=TRUE, permissive=TRUE)
>>     Warning in withCallingHandlers(.External2(C_parseRd, tcon, srcfile,
>>     "UTF-8",  :
>>       <connection>:2: unexpected END_OF_INPUT '
>>     '
>>     \url{http://foo%3Fbar}
>>     }
>>
>>     Here is a summary of the issues I know about; the 'err' and 'warn'
>>     labels refer to how R reports the initial error; the
>>     packageDescription 'warnings' actually mean that the CITATION file
>>     may not be parsed by the build system.
>>
>>     ## > names(err)
>>     ## [1] "bioassayR"    # knitcitations()
>>     ## [2] "ChemmineR"    # knitcitations()
>>     ## [3] "clippda"      # missing textVersion
>>     ## [4] "eiR"          # knitcitations()
>>     ## [5] "fmcsR"        # knitcitations()
>>     ## [6] "metagenomeSeq"# packageVersion()
>>     ## [7] "SomatiCA"     # missing textVersion
>>     ## > names(warn)
>>     ##  [1] "ALDEx2"      # url encoding '10.1371%2'
>>     ##  [2] "aroma.light" # print(citn, style="html"); trailing } in title
>>     ##  [3] "birte"       # packageDescription()
>>     ##  [4] "BRAIN"       # packageDescription()
>>     ##  [5] "ChAMP"       # doi encoding %2F
>>     ##  [6] "ChIPpeakAnno"# url encoding %2F
>>     ##  [7] "CoGAPS"      # packageDescription()
>>     ##  [8] "derfinderHelper"  # packageDescription()
>>     ##  [9] "derfinderPlot"
>>     ## [10] "destiny"     # packageDescription()
>>     ## [11] "fabia"       # packageDescription()
>>     ## [12] "flowFP"      # packageDescription()
>>     ## [13] "GenomicAlignments"# url encoding
>>     ## [14] "GenomicFeatures"  # url encoding
>>     ## [15] "GenomicRanges"    # url encoding
>>     ## [16] "Genominator" # packageDescription()
>>     ## [17] "GOSim"       # packageDescription()
>>     ## [18] "GSRI"        # packageDescription()
>>     ## [19] "hapFabia"    # packageDescription()
>>     ## [20] "kebabs"      # packageDescription()
>>     ## [21] "manta"       # packageDescription()
>>     ## [22] "mdgsa"       # url encoding
>>     ## [23] "MEDME"       # packageDescription()
>>     ## [24] "miRLAB"      # url encoding
>>     ## [25] "networkBMA"  # packageDescription()
>>     ## [26] "pcaGoPromoter" # url encoding
>>     ## [27] "podkat"      # packageDescription()
>>     ## [28] "procoil"     # packageDescription()
>>     ## [29] "Rbowtie"     # packageDescription()
>>     ## [30] "Rcade"       # packageDescription()
>>     ## [31] "Rchemcpp"    # packageDescription()
>>     ## [32] "rfPred"      # packageDescription()
>>     ## [33] "sigaR"       # packageDescription()
>>     ## [34] "SIM"         # packageDescription()
>>     ## [35] "SWATH2stats" # url encoding
>>     ## [36] "TEQC"        # packageDescription()
>>
>>
>>
>>         Best,
>>
>>
>>         R.
>>
>>
>>
>>
>>     This email message may contain legally privileged
>> and/or...{{dropped:2}}
>>
>>
>>     _______________________________________________
>>     Bioc-devel at r-project.org <mailto:Bioc-devel at r-project.org> mailing
>> list
>>     https://stat.ethz.ch/mailman/listinfo/bioc-devel
>>     <https://stat.ethz.ch/mailman/listinfo/bioc-devel>
>>
>>
>>
>
> This email message may contain legally privileged and/or confidential
> information.  If you are not the intended recipient(s), or the employee or
> agent responsible for the delivery of this message to the intended
> recipient(s), you are hereby notified that any disclosure, copying,
> distribution, or use of this email message is prohibited.  If you have
> received this message in error, please notify the sender immediately by
> e-mail and delete this email message from your computer. Thank you.
>

	[[alternative HTML version deleted]]



More information about the Bioc-devel mailing list