Re: [Bioc-devel] rifiComparative

[Sarvesh Nikumbh]

Hi Loubna,

Was the error in release or the devel version?

Devel version report:
http://bioconductor.org/checkResults/3.18/bioc-LATEST/rifiComparative/
Release version report:
http://bioconductor.org/checkResults/release/bioc-LATEST/rifiComparative/

Perhaps, your push was to the devel version and resolved the error there?

Additionally, if you check the error, it is regarding package devtools'
(un)availability. Please note that using `devtools::load_all()` is not
recommended in the vignette. You should instead load your package by using
`library()` as anyone else (the end user) would do at their end.

Finally, you can always check whether your last commits were successfully
picked at two places: the build reports (links above) or also at
https://code.bioconductor.org/browse/

Hope this helps.

Best wishes,
  Sarvesh



On Sun, 21 May 2023 at 20:53, Loubna Youssar  wrote:

> Hello,
>
> I am the developer and maintainer of rifiComparative package. I was
> notified by an error and I fixed it then pushed three weeks ago.
> To my surprise I did get the same error with the last package version some
> days ago. This would mean the push did not work even I do remember no error
> was displayed. How can I be sure that my updated package with no error was
> successfully pushed? am I supposed to receive a confirmatory email?
>
> Here is the new version:
>
> Package: rifiComparative
> Title: 'rifiComparative' compares the output of rifi from two different
> conditions.
> Version: 1.1.1
>
> The error sent:
>
> Package: rifiComparative
> Version: 1.0.0
> Command: /home/biocbuild/bbs-3.17-bioc/R/bin/R CMD check
> --install=check:rifiComparative.install-out.txt
> --library=/home/biocbuild/bbs-3.17-bioc/R/site-library --timings
> rifiComparative_1.0.0.tar.gz
> StartedAt: 2023-05-20 00:15:11 -0400 (Sat, 20 May 2023)
> EndedAt: 2023-05-20 00:26:00 -0400 (Sat, 20 May 2023)
> EllapsedTime: 649.2 seconds
> RetCode: 1
> Status:   ERROR
> CheckDir: rifiComparative.Rcheck
> Warnings: NA
>
> Thanks in advance for your assistance
>
> Loubna
>
> [[alternative HTML version deleted]]
>
> ___
> Bioc-devel@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel
>


-- 

thanks!
-Sarvesh


Q: Why is this email five sentences or less?
A: http://five.sentenc.es

[[alternative HTML version deleted]]

___
Bioc-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/bioc-devel

Re: [Bioc-devel] Fixing autonomics - ssh: connect to host git.bioconductor.org port 22: Connection timed out

[Sarvesh Nikumbh]

Hi Aditya,

Can you check if this is due to the recent branch name changes: default
branch name changed from 'main' to 'devel'
This was announced on the mailing-list, and also on this blogpost:
https://bioconductor.github.io/biocblog/posts/2023-03-01-transition-to-devel/

Sarvesh

On Tue, 4 Apr 2023 at 19:38, Aditya Bhagwat via Bioc-devel <
bioc-devel@r-project.org> wrote:

> Dear BioC team,
>
> I am having the same error as many others (the missing BiocStyle
> issue) and I wanted to fix it. But I somehow am not getting into
> git.bioconductor.org anymore :
>
> $ git fetch --all
> Fetching origin
> Fetching upstream
> ssh: connect to host git.bioconductor.org port 22: Connection timed out
> fatal: Could not read from remote repository.
> Please make sure you have the correct access rights
> and the repository exists.
> error: could not fetch upstream
>
> I checked BiocCredentials and my public key is visible there.
>
> What else could this be ? I changed institutes since my last push, in
> my current setup things go through a proxyserver, could that be the
> reason?
>
> Thank you for feedback.
>
> Aditya
>
>
> --
> Aditya Bhagwat
> Statistics ∙ Bioinformatics ∙ Data Science
> Translational Proteomics ∙ Philipps-University Marburg
> Karl-von-Frisch-Str 2 ∙ 35043 Marburg ∙ Germany
> Tel.: +49 6421 28 27403
>
> ___
> Bioc-devel@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel
>


-- 

thanks!
-Sarvesh


Q: Why is this email five sentences or less?
A: http://five.sentenc.es

[[alternative HTML version deleted]]

___
Bioc-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/bioc-devel

[Bioc-devel] "Graphics API version mismatch": Devel package check on macOS failing

[Sarvesh Nikumbh]

Hi,

I see some devel packages (including my package seqArchRplus
)
failing on macOS due to "Graphics API version mismatch". The R-devel
announcement here 
has the following note under the Graphics head, which is probably the cause
of this error (as a similar error was caused on an earlier occasion)

GRAPHICS

   -

   The graphics engine version, R_GE_version, has been bumped to 16 and so
   packages that provide graphics devices should be reinstalled.



I believe re-installing Cairo on the build machine should be helpful here?
Please let us know if there is anything that developers can/should do.

Thanks and best wishes,
  Sarvesh


-- 

thanks!
-Sarvesh


Q: Why is this email five sentences or less?
A: http://five.sentenc.es

[[alternative HTML version deleted]]

___
Bioc-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/bioc-devel

Re: [Bioc-devel] download stats not accessible

[Sarvesh Nikumbh]

While I too wondered why the OP mentions unavailability "for months", now
re-reading the OP's email, may be they meant "for months" granularity which
is (now temporarily un)available! :D

Cheers,
  Sarvesh

On Mon, 20 Mar 2023 at 16:50, Hervé Pagès 
wrote:

> Not for months, only since last week. They should be back in the next
> 12-18 hours. Thanks for reporting this.
>
> Cheers,
>
> H.
>
> On 19/03/2023 11:00, Luo Weijun wrote:
> > Dear BioC team,
> > I noticed that download stats for BioC packages are not accessible for
> months:
> > https://bioconductor.org/packages/stats/bioc/pathview/
> > https://bioconductor.org/packages/stats/bioc/BiocGenerics/
> >
> > can you check on this? thanks.
> > Weijun Luo
>
> --
> Hervé Pagès
>
> Bioconductor Core Team
> hpages.on.git...@gmail.com
>
> ___
> Bioc-devel@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel
>


-- 

thanks!
-Sarvesh


Q: Why is this email five sentences or less?
A: http://five.sentenc.es

[[alternative HTML version deleted]]

___
Bioc-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/bioc-devel

[Bioc-devel] seqArchRplus build error on Linux

[Sarvesh Nikumbh]

Hi Bioc-team,

I get this build error for my package seqArchRplus on Linux/nebbiolo1.

--- re-building ‘seqArchRplus.Rmd’ using rmarkdown
Error running filter pandoc-citeproc:
Could not find executable pandoc-citeproc
Error: processing vignette 'seqArchRplus.Rmd' failed with diagnostics:
pandoc document conversion failed with error 83
--- failed re-building ‘seqArchRplus.Rmd’

SUMMARY: processing the following file failed:
  ‘seqArchRplus.Rmd’


I checked the mailing-list and see that this may be due to changing
versions of pandoc on some machines. I understand that the team is working
on it, and this should go away in a few days? Please let me know if there
is something to be done on my end.

Thanks and best wishes,
  Sarvesh



-- 

thanks!
-Sarvesh


Q: Why is this email five sentences or less?
A: http://five.sentenc.es

[[alternative HTML version deleted]]

___
Bioc-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/bioc-devel

Re: [Bioc-devel] seqArchR error on build systems

[Sarvesh Nikumbh]

Ubuntu and macOS errors are now fixed, but the palomino4/Windows error
is still there (this error is different), says, it needs OpenSSL.
Thanks in advance.

===
Collecting package metadata (current_repodata.json): ...working...
Collecting package metadata (current_repodata.json): ...working...
failed

CondaSSLError: OpenSSL appears to be unavailable on this machine.
OpenSSL is required to
download and install packages.
Exception: HTTPSConnectionPool(host='conda.anaconda.org', port=443):
Max retries exceeded with url:
/conda-forge/win-64/current_repodata.json (Caused by SSLError("Can't
connect to HTTPS URL because the SSL module is not available."))

+ "C:/Users/biocbuild/AppData/Local/r-miniconda/condabin/conda.bat"
"create" "--yes" "--prefix"
"C:/Users/biocbuild/AppData/Local/r-miniconda/envs/r-reticulate"
"python=3.8" "numpy" "--quiet" "-c" "conda-forge"
failed

CondaSSLError: OpenSSL appears to be unavailable on this machine.
OpenSSL is required to
download and install packages.

Exception: HTTPSConnectionPool(host='conda.anaconda.org', port=443):
Max retries exceeded with url:
/conda-forge/win-64/current_repodata.json (Caused by SSLError("Can't
connect to HTTPS URL because the SSL module is not available."))
===


Thanks and best,
  Sarvesh


On Tue, 25 Oct 2022 at 22:05, Hervé Pagès  wrote:
>
> Hi Sarvesh,
>
> It looks like reticulate somehow was using the wrong Python instance on
> nebbiolo2 and lconway, as reported by py_config().
>
> We're now setting RETICULATE_PYTHON on those machines to make sure that
> reticulate will use the correct Python. This change was made after the
> builds started today so it won't be reflected on tomorrow's report. But
> it should be reflected on Thursday's report.
>
> We'll do the same on the other build machines soon.
>
> Best,
>
> H.
>
> On 25/10/2022 08:53, Sarvesh Nikumbh wrote:
> > Hi Bioc-team,
> >
> > seqArchR was building fine, without any errors last week, but I am seeing
> > an error again. This time it is due to the Python module 'sklearn' not
> > available and one another error. The sklearn-related error occurs on
> > nebbiolo2 and Iconway, while palomino4 shows a different error.
> >
> > Maybe something changed on the build systems? Because these system
> > requirements were available before*.*
> >
> > Thanks and best,
> >Sarvesh
> >
> --
> Hervé Pagès
>
> Bioconductor Core Team
> hpages.on.git...@gmail.com
>


-- 

thanks!
-Sarvesh


Q: Why is this email five sentences or less?
A: http://five.sentenc.es

___
Bioc-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/bioc-devel

Re: [Bioc-devel] Package dependencies error during BiocCheck

[Sarvesh Nikumbh]

Hi Marcel,

Thanks for the reply.

Is there a way to opt for this solution on the bioc new pkg build systems?

If not, then, I guess, I will have to wait until it goes away for my bioc
submission to be reviewed further?

Thanks and best,
  Sarvesh


On Tue, 25 Oct 2022, 18:14 Marcel Ramos, 
wrote:

> Hi Sarvesh,
>
> The error has something to do with an interaction between internal base
> R code
> and checking for dependencies in BiocCheck with `loadNamespace`.
> Either way, this check in BiocCheck is redundant with R CMD check.
> It will soon get removed but for now, you can disable it by running:
>
> BiocCheck("./seqArchRplus", `no-check-dependencies` = TRUE)
>
> Best,
>
> Marcel
>
>
> On 10/25/22 12:27 PM, Sarvesh Nikumbh wrote:
> > Hi all,
> >
> > I am trying to resolve an issue with dependencies of my package which is
> > under review at the moment (Issue 2790
> > <https://github.com/Bioconductor/Contributions/issues/2790>, latest
> report
> > <
> http://bioconductor.org/spb_reports/seqArchRplus_buildreport_20221025084934.html
> >).
> > I was wondering if somebody else also encountered this issue (see details
> > below), and were successful resolving it. Any help in this regard would
> be
> > greatly appreciated.
> >
> > *The issue:*
> > Running BiocCheck throws the following error:
> >
> > Error in completeSubclasses(classDef2, class1, obj, where) :
> >trying to get slot "subclasses" from an object of a basic class
> > ("NULL") with no slots
> > Error: .onLoad failed in loadNamespace() for 'dbplyr', details:
> >call: setClass(cl, contains = c(prevClass, "VIRTUAL"), where = where)
> >error: error in contained classes ("character") for class ident;
> > class definition removed from dbplyr
> > Execution halted
> >
> > which going by info at
> > tidyverse/dbplyr#779<https://github.com/tidyverse/dbplyr/issues/779>  is
> > caused by Brobdingnag<https://github.com/RobinHankin/Brobdingnag>, and
> > resolved with version 1.2-9, which is now also available on CRAN (see
> here
> > <https://github.com/RobinHankin/Brobdingnag/issues/24>).
> >
> > I saw this error earlier on my debian machine only using bioc-devel.
> Here,
> > I could resolve this by re-installing Matrix, dbplyr and Brobdingnag.
> >
> > On bioc-side, the error can be seen on Ubuntu (latest report
> > <
> http://bioconductor.org/spb_reports/seqArchRplus_buildreport_20221025084934.html
> >).
> > Maybe, it would occur on other OSs as well, but that stage is not
> reached.
> > Thanks to Lori and Jennifer from bioc-team, the same re-installation
> > solution was attempted, but this did not resolve it there.
> >
> > Since I do not have access to a machine running Ubuntu around, I have set
> > up GHA. There, interestingly, the same error is encountered on macOS and
> > Windows, but not on Ubuntu -- details here
> > <
> https://github.com/snikumbh/seqArchRplus/actions/runs/3322212068/jobs/5490915503
> >
> > .
> >
> >
> > Any leads?
> >
> > Thanks in advance.
>
> This email message may contain legally privileged and/or...{{dropped:4}}
>
> ___
> Bioc-devel@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel
>

[[alternative HTML version deleted]]

___
Bioc-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/bioc-devel

[Bioc-devel] Package dependencies error during BiocCheck

[Sarvesh Nikumbh]

Hi all,

I am trying to resolve an issue with dependencies of my package which is
under review at the moment (Issue 2790
, latest report
).
I was wondering if somebody else also encountered this issue (see details
below), and were successful resolving it. Any help in this regard would be
greatly appreciated.

*The issue:*
Running BiocCheck throws the following error:

Error in completeSubclasses(classDef2, class1, obj, where) :
  trying to get slot "subclasses" from an object of a basic class
("NULL") with no slots
Error: .onLoad failed in loadNamespace() for 'dbplyr', details:
  call: setClass(cl, contains = c(prevClass, "VIRTUAL"), where = where)
  error: error in contained classes ("character") for class ident;
class definition removed from dbplyr
Execution halted

which going by info at
tidyverse/dbplyr#779  is
caused by Brobdingnag , and
resolved with version 1.2-9, which is now also available on CRAN (see here
).

I saw this error earlier on my debian machine only using bioc-devel. Here,
I could resolve this by re-installing Matrix, dbplyr and Brobdingnag.

On bioc-side, the error can be seen on Ubuntu (latest report
).
Maybe, it would occur on other OSs as well, but that stage is not reached.
Thanks to Lori and Jennifer from bioc-team, the same re-installation
solution was attempted, but this did not resolve it there.

Since I do not have access to a machine running Ubuntu around, I have set
up GHA. There, interestingly, the same error is encountered on macOS and
Windows, but not on Ubuntu -- details here

.


Any leads?

Thanks in advance.
-- 

thanks!
-Sarvesh

[[alternative HTML version deleted]]

___
Bioc-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/bioc-devel

[Bioc-devel] seqArchR error on build systems

[Sarvesh Nikumbh]

Hi Bioc-team,

seqArchR was building fine, without any errors last week, but I am seeing
an error again. This time it is due to the Python module 'sklearn' not
available and one another error. The sklearn-related error occurs on
nebbiolo2 and Iconway, while palomino4 shows a different error.

Maybe something changed on the build systems? Because these system
requirements were available before*.*

Thanks and best,
  Sarvesh

-- 

thanks!
-Sarvesh


Q: Why is this email five sentences or less?
A: http://five.sentenc.es

[[alternative HTML version deleted]]

___
Bioc-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/bioc-devel

Re: [Bioc-devel] Remote rejected when pushing to git

[Sarvesh Nikumbh]

Hi Ash,

Refer to Lori's email from Oct 17 apprising of the freezing of the
RELEASE_3_15 branch according to the release schedule at
https://bioconductor.org/developers/release-schedule/

As the email says, you can now only push to the master branch (devel as of
today, release candidate 3.16), and limit changes to “show-stopper” bugs
and documentation improvements as per the schedule/instructions in the
above link.

Sarvesh

-- Forwarded message -
> From: Kern, Lori 
> Date: Mon, 17 Oct 2022 at 13:06
> Subject: [Bioc-devel] RELEASE_3_15 branch frozen in a few hours
> To: bioc-devel@r-project.org 
>
>
> We are moving forward with the 3.16 release schedule.  In roughly 4 hours
> the RELEASE_3_15 will be frozen and no longer accept commits.  Any changes
> or bug corrections that are not pushed up by this time will not be in the
> Bioconductor 3.15 version of the package as we do not allow updates to
> past/frozen releases.
>
> http://bioconductor.org/developers/release-schedule/
>
> You still may push to the Bioconductor master/devel branch at any time.
>
> Cheers,
>
>
> Lori Shepherd - Kern




On Mon, 24 Oct 2022 at 10:11, Ash Sawle  wrote:

> Hi
>
> I am trying to push some changes to the upstream release 3.15 repository
> but I am getting the following error:
>
> $ git push upstream RELEASE_3_15
> Enumerating objects: 25, done.
> Counting objects: 100% (25/25), done.
> Delta compression using up to 8 threads
> Compressing objects: 100% (15/15), done.
> Writing objects: 100% (15/15), 203.52 KiB | 22.61 MiB/s, done.
> Total 15 (delta 10), reused 0 (delta 0), pack-reused 0
> remote: FATAL: W refs/heads/RELEASE_3_15 packages/qPLEXanalyzer
> AshKernow DENIED by fallthru
> remote: error: hook declined to update refs/heads/RELEASE_3_15
> To git.bioconductor.org:packages/qPLEXanalyzer
>  ![remote rejected] RELEASE_3_15 -> RELEASE_3_15 (hook declined)
> error: failed to push some refs to 'git.bioconductor.org:
> packages/qPLEXanalyzer'
>
> I am not sure what has changed as this used to work fine before. Any help
> greatly appreciated.
>
> cheers
>
> Ash
>
> [[alternative HTML version deleted]]
>
> ___
> Bioc-devel@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel
>


-- 

thanks!
-Sarvesh


Q: Why is this email five sentences or less?
A: http://five.sentenc.es

[[alternative HTML version deleted]]

___
Bioc-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/bioc-devel

Re: [Bioc-devel] Increase version of system dependency?

[Sarvesh Nikumbh]

Issue resolved. Thanks, Jennifer!

-Sarvesh

On Thu, 20 Oct 2022 at 20:18, Jennifer Wokaty 
wrote:

> Hi Sarvesh,
>
> I've installed packaging on our devel Windows builder. We should see the
> issue resolved in the next build report.
>
> Jen
> --
> *From:* Bioc-devel  on behalf of
> Sarvesh Nikumbh 
> *Sent:* Thursday, October 20, 2022 12:08 PM
> *To:* bioc-devel@r-project.org 
> *Subject:* Re: [Bioc-devel] Increase version of system dependency?
>
> ***ATTENTION: This email came from an external source. Do not open
> attachments or click on links from unknown senders or unexpected emails.***
>
> Hi bioc team,
>
> Could you please install the Python module 'packaging' on palomino4? I have
> pushed the updated SystemRequirements yesterday.  Linux and macOS seem to
> be fine.
> Thanks in advance.
>
> Best wishes,
>   Sarvesh
>
> On Wed, 19 Oct 2022 at 02:36, Hervé Pagès 
> wrote:
>
> > Hi Sarvesh,
> >
> > All I see is that seqArchR fails on palomino4 because Python module
> > 'packaging' is not available on that machine.
> >
> > The module seems to be available on the other builders though so no
> > problem there.
> >
> > Anyways if your package depends on that module (and it seems that it
> > does, via the inst/python/perform_nmf.py script), then you need to list
> > the module in your SystemRequirements.
> >
> > Then we'll make sure to install the module on all the builders.
> >
> > Thanks,
> >
> > H.
> >
> >
> > On 18/10/2022 09:31, Sarvesh Nikumbh wrote:
> > > Hi bioc team,
> > >
> > > My package seqArchR though does not error in any way, but can spit out
> > > numerous warnings (originally from Python/scikit-learn which is a
> > > dependency) depending on the version of  scikit-learn available. This
> > fills
> > > the output to the extent that it is unreadable. The cause for this is
> the
> > > deprecation/future version warning in NMF/scikitlearn
> > > <
> >
> https://urldefense.proofpoint.com/v2/url?u=https-3A__scikit-2Dlearn.org_stable_modules_generated_sklearn.decomposition.NMF.html=DwIFaQ=mRWFL96tuqj9V0Jjj4h40ddo0XsmttALwKjAEOCyUjY=FKSqKXkUOES-D4VQb2jSn9QK7Vz5lE18rLcyn73CPhA=FGhgBnQ9LlkQNoTApdaferGM3XwiCSkJ4EzVsgcMz6Kg_q8PMDzOi1eB5M6JMqxz=DzxNJLg1ADaS3XDof9AnhKWeOBenp1DvnG6qt37swGk=
>
> > >
> > > which is forced, so I cannot suppress it.
> > >
> > > I pushed a fix for this in the devel version -- where, instead of
> > > increasing the dependency version, I check the module version using
> > > packaging module from setuptools and appropriately make the python
> > function
> > > call. But this gives an error on the Windows build machine, and is fine
> > on
> > > Linux/macOS.
> > > See
> > >
> >
> https://urldefense.proofpoint.com/v2/url?u=https-3A__bioconductor.org_checkResults_3.16_bioc-2DLATEST_seqArchR_palomino4-2Dchecksrc.html=DwIFaQ=mRWFL96tuqj9V0Jjj4h40ddo0XsmttALwKjAEOCyUjY=FKSqKXkUOES-D4VQb2jSn9QK7Vz5lE18rLcyn73CPhA=FGhgBnQ9LlkQNoTApdaferGM3XwiCSkJ4EzVsgcMz6Kg_q8PMDzOi1eB5M6JMqxz=yJdY-LnSsA0-Xfh47DYFXIxLFxWOy8gQEQu4BiIwBA0=
>
> > >
> > > I expected that setuptools will be available on all machines, because
> the
> > > previous alternative, distutils, is available with vanilla python, but
> is
> > > not recommended.
> > >
> > > Would you suggest having setuptools in the SystemRequirements or using
> > > distutils?
> > > Or simply depending on higher version of scikit-learn (the latest 1.2)
> --
> > > which is the root cause of this issue anyway.
> > >
> > > Thanks and best,
> > >Sarvesh
> > >
> > --
> > Hervé Pagès
> >
> > Bioconductor Core Team
> > hpages.on.git...@gmail.com
> >
> >
>
> --
>
> thanks!
> -Sarvesh
>
> 
> Q: Why is this email five sentences or less?
> A:
> https://urldefense.proofpoint.com/v2/url?u=http-3A__five.sentenc.es=DwIFaQ=mRWFL96tuqj9V0Jjj4h40ddo0XsmttALwKjAEOCyUjY=FKSqKXkUOES-D4VQb2jSn9QK7Vz5lE18rLcyn73CPhA=FGhgBnQ9LlkQNoTApdaferGM3XwiCSkJ4EzVsgcMz6Kg_q8PMDzOi1eB5M6JMqxz=tINaCsEcKRF4gJ2k-iW6J1n_Rsucc632G8xtz784wnI=
>
>
> [[alternative HTML version deleted]]
>
> ___
> Bioc-devel@r-project.org mailing list
>
> https://urldefense.proofpoint.com/v2/url?u=https-3A__stat.ethz.ch_mailman_listinfo_bioc-2Ddevel=DwIFaQ=mRWFL96tuqj9V0Jjj4h40ddo0XsmttALwKjAEOCyUjY=FKSqKXkUOES-D4VQb2jSn9QK7Vz5lE18rLcyn73CPhA=FGhgBnQ9LlkQNoTApdaferGM3XwiCSkJ4EzVsgcMz6Kg_q8PMDzOi1eB5M6JMqxz=CNOml7Ibm5f1NLsQePoNZeV9smJCxNRI5nNd8D2JVp0=
>
>


-- 

thanks!
-Sarvesh


Q: Why is this email five sentences or less?
A: http://five.sentenc.es

[[alternative HTML version deleted]]

___
Bioc-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/bioc-devel

Re: [Bioc-devel] Increase version of system dependency?

[Sarvesh Nikumbh]

Hi bioc team,

Could you please install the Python module 'packaging' on palomino4? I have
pushed the updated SystemRequirements yesterday.  Linux and macOS seem to
be fine.
Thanks in advance.

Best wishes,
  Sarvesh

On Wed, 19 Oct 2022 at 02:36, Hervé Pagès 
wrote:

> Hi Sarvesh,
>
> All I see is that seqArchR fails on palomino4 because Python module
> 'packaging' is not available on that machine.
>
> The module seems to be available on the other builders though so no
> problem there.
>
> Anyways if your package depends on that module (and it seems that it
> does, via the inst/python/perform_nmf.py script), then you need to list
> the module in your SystemRequirements.
>
> Then we'll make sure to install the module on all the builders.
>
> Thanks,
>
> H.
>
>
> On 18/10/2022 09:31, Sarvesh Nikumbh wrote:
> > Hi bioc team,
> >
> > My package seqArchR though does not error in any way, but can spit out
> > numerous warnings (originally from Python/scikit-learn which is a
> > dependency) depending on the version of  scikit-learn available. This
> fills
> > the output to the extent that it is unreadable. The cause for this is the
> > deprecation/future version warning in NMF/scikitlearn
> > <
> https://scikit-learn.org/stable/modules/generated/sklearn.decomposition.NMF.html
> >
> > which is forced, so I cannot suppress it.
> >
> > I pushed a fix for this in the devel version -- where, instead of
> > increasing the dependency version, I check the module version using
> > packaging module from setuptools and appropriately make the python
> function
> > call. But this gives an error on the Windows build machine, and is fine
> on
> > Linux/macOS.
> > See
> >
> https://bioconductor.org/checkResults/3.16/bioc-LATEST/seqArchR/palomino4-checksrc.html
> >
> > I expected that setuptools will be available on all machines, because the
> > previous alternative, distutils, is available with vanilla python, but is
> > not recommended.
> >
> > Would you suggest having setuptools in the SystemRequirements or using
> > distutils?
> > Or simply depending on higher version of scikit-learn (the latest 1.2) --
> > which is the root cause of this issue anyway.
> >
> > Thanks and best,
> >Sarvesh
> >
> --
> Hervé Pagès
>
> Bioconductor Core Team
> hpages.on.git...@gmail.com
>
>

-- 

thanks!
-Sarvesh


Q: Why is this email five sentences or less?
A: http://five.sentenc.es

[[alternative HTML version deleted]]

___
Bioc-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/bioc-devel

Re: [Bioc-devel] Increase version of system dependency?

[Sarvesh Nikumbh]

Thanks for the reply, Hervé.
In the interest of available time for this release, I will include
'packaging' in the SystemRequirements, and look to using basilisk for the
next release.

Best,
  Sarvesh

On Wed, 19 Oct 2022 at 02:36, Hervé Pagès 
wrote:

> Hi Sarvesh,
>
> All I see is that seqArchR fails on palomino4 because Python module
> 'packaging' is not available on that machine.
>
> The module seems to be available on the other builders though so no
> problem there.
>
> Anyways if your package depends on that module (and it seems that it
> does, via the inst/python/perform_nmf.py script), then you need to list
> the module in your SystemRequirements.
>
> Then we'll make sure to install the module on all the builders.
>
> Thanks,
>
> H.
>
>
> On 18/10/2022 09:31, Sarvesh Nikumbh wrote:
> > Hi bioc team,
> >
> > My package seqArchR though does not error in any way, but can spit out
> > numerous warnings (originally from Python/scikit-learn which is a
> > dependency) depending on the version of  scikit-learn available. This
> fills
> > the output to the extent that it is unreadable. The cause for this is the
> > deprecation/future version warning in NMF/scikitlearn
> > <
> https://scikit-learn.org/stable/modules/generated/sklearn.decomposition.NMF.html
> >
> > which is forced, so I cannot suppress it.
> >
> > I pushed a fix for this in the devel version -- where, instead of
> > increasing the dependency version, I check the module version using
> > packaging module from setuptools and appropriately make the python
> function
> > call. But this gives an error on the Windows build machine, and is fine
> on
> > Linux/macOS.
> > See
> >
> https://bioconductor.org/checkResults/3.16/bioc-LATEST/seqArchR/palomino4-checksrc.html
> >
> > I expected that setuptools will be available on all machines, because the
> > previous alternative, distutils, is available with vanilla python, but is
> > not recommended.
> >
> > Would you suggest having setuptools in the SystemRequirements or using
> > distutils?
> > Or simply depending on higher version of scikit-learn (the latest 1.2) --
> > which is the root cause of this issue anyway.
> >
> > Thanks and best,
> >Sarvesh
> >
> --
> Hervé Pagès
>
> Bioconductor Core Team
> hpages.on.git...@gmail.com
>
>

-- 

thanks!
-Sarvesh


Q: Why is this email five sentences or less?
A: http://five.sentenc.es

[[alternative HTML version deleted]]

___
Bioc-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/bioc-devel

Re: [Bioc-devel] Increase version of system dependency?

[Sarvesh Nikumbh]

Hi Vince,

I did attempt to use basilisk for seqArchR, but met with some errors then.
See https://github.com/LTLA/basilisk/issues/13

If anyone has tried basilisk with BiocParallel, I will be happy to reach
out and find a way to mediate via basilisk.

I will write to you separately with more details if it is ok. I will be
glad to make basilisk+biocParallel work for seqArchR.

Best,
  Sarvesh

On Tue, 18 Oct 2022 at 17:51, Vincent Carey 
wrote:

> My suggestion is that your interfaces to python be mediated with the
> basilisk package.  Please
> contact me off line if you need assistance; see BiocSklearn for an
> example; I am sure there are
> others.
>
> On Tue, Oct 18, 2022 at 12:31 PM Sarvesh Nikumbh <
> sarvesh.niku...@gmail.com> wrote:
>
>> Hi bioc team,
>>
>> My package seqArchR though does not error in any way, but can spit out
>> numerous warnings (originally from Python/scikit-learn which is a
>> dependency) depending on the version of  scikit-learn available. This
>> fills
>> the output to the extent that it is unreadable. The cause for this is the
>> deprecation/future version warning in NMF/scikitlearn
>> <
>> https://scikit-learn.org/stable/modules/generated/sklearn.decomposition.NMF.html
>> >
>> which is forced, so I cannot suppress it.
>>
>> I pushed a fix for this in the devel version -- where, instead of
>> increasing the dependency version, I check the module version using
>> packaging module from setuptools and appropriately make the python
>> function
>> call. But this gives an error on the Windows build machine, and is fine on
>> Linux/macOS.
>> See
>>
>> https://bioconductor.org/checkResults/3.16/bioc-LATEST/seqArchR/palomino4-checksrc.html
>>
>> I expected that setuptools will be available on all machines, because the
>> previous alternative, distutils, is available with vanilla python, but is
>> not recommended.
>>
>> Would you suggest having setuptools in the SystemRequirements or using
>> distutils?
>> Or simply depending on higher version of scikit-learn (the latest 1.2) --
>> which is the root cause of this issue anyway.
>>
>> Thanks and best,
>>   Sarvesh
>>
>> --
>>
>> thanks!
>> -Sarvesh
>>
>> [[alternative HTML version deleted]]
>>
>> ___
>> Bioc-devel@r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/bioc-devel
>>
>
> The information in this e-mail is intended only for th...{{dropped:26}}

___
Bioc-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/bioc-devel

[Bioc-devel] Increase version of system dependency?

[Sarvesh Nikumbh]

Hi bioc team,

My package seqArchR though does not error in any way, but can spit out
numerous warnings (originally from Python/scikit-learn which is a
dependency) depending on the version of  scikit-learn available. This fills
the output to the extent that it is unreadable. The cause for this is the
deprecation/future version warning in NMF/scikitlearn

which is forced, so I cannot suppress it.

I pushed a fix for this in the devel version -- where, instead of
increasing the dependency version, I check the module version using
packaging module from setuptools and appropriately make the python function
call. But this gives an error on the Windows build machine, and is fine on
Linux/macOS.
See
https://bioconductor.org/checkResults/3.16/bioc-LATEST/seqArchR/palomino4-checksrc.html

I expected that setuptools will be available on all machines, because the
previous alternative, distutils, is available with vanilla python, but is
not recommended.

Would you suggest having setuptools in the SystemRequirements or using
distutils?
Or simply depending on higher version of scikit-learn (the latest 1.2) --
which is the root cause of this issue anyway.

Thanks and best,
  Sarvesh

-- 

thanks!
-Sarvesh

[[alternative HTML version deleted]]

___
Bioc-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/bioc-devel