Skip to content

[Bioc-devel] Trigger a new build after version bump

10 messages · Jakob Theorell, Shepherd, Lori, Ahmed Metwally +1 more

Ahmed Metwally
# 
Dear Bioc devel team,

OmicsLonDA package is failing to build on release and devel branches:
http://bioconductor.org/checkResults/release/bioc-LATEST/OmicsLonDA/
http://bioconductor.org/checkResults/devel/bioc-LATEST/OmicsLonDA/


I have fixed the issue and pushed the changes to RELEASE_3_10, and dev-master branch. I bumped the package version on the RELEASE_3_10 to v1.2.1 instead of v1.2.0. I did the same for the dev branch, bumped the version to v1.3.3 instead of v1.3.0. I was following instructions listed here: https://bioconductor.org/developers/how-to/git/sync-existing-repositories/

However, my changes didn't trigger a new build. In the past, when I was submitting the OmicsLonDA to Bioc, whenever I do version bump, I receive an email says that I successfully triggered a new build. This time I didn't receive any emails. Also, the build error and data don't change in the above-mentioned URLs to reflect a new build!


Please let me know what I'm missing here.
Thanks,
Jakob Theorell
# 
Dear Ahmed,
I am not in the core team, but I still wonder: how long have you waited since you pushed? During the review process you get an immediate update of the package upon push with a version bump, but with accepted packages it takes 24-48 hours for the changes to trickle through to the build report. And you have to go in and check the build reports yourself on the BioConductor sites related to the release and devel branches.
Best regards
Jakob Theorell, MD/PhD
Autoimmune Neurology Group
Nuffield Department of Clinical Neurosciences
University of Oxford
Ahmed Metwally
# 
Hi Jakob,

Thanks for your explanation. I was looking for any statement about how long should I wait but couldn't find any. I pushed the updates around 6 hours ago. I'll wait for 24-48h as you suggested and will check back.

Best,
Shepherd, Lori
# 
Jakob is correct.   To expand a bit more:
The results will not be present on today's build report but should be there on tomorrows.   We do a git pull around 4:45 PM EST for the daily builds.  If commits are made after that (which it seems like in this case) it would be on the following days.
For What Its worth I do see your changes on git.bioconductor.org so you were able to push them correctly.

Cheers,


Lori Shepherd

Bioconductor Core Team

Roswell Park Comprehensive Cancer Center

Department of Biostatistics & Bioinformatics

Elm & Carlton Streets

Buffalo, New York 14263
Ahmed Metwally
# 
Great! Thank you, Lori

Ahmed A. Metwally
Sent from iPhone
On Dec 17, 2019, at 4:21 AM, Shepherd, Lori <Lori.Shepherd at RoswellPark.org<mailto:Lori.Shepherd at RoswellPark.org>> wrote:
Jakob is correct.   To expand a bit more:
The results will not be present on today's build report but should be there on tomorrows.   We do a git pull around 4:45 PM EST for the daily builds.  If commits are made after that (which it seems like in this case) it would be on the following days.
For What Its worth I do see your changes on git.bioconductor.org<http://git.bioconductor.org> so you were able to push them correctly.

Cheers,


Lori Shepherd

Bioconductor Core Team

Roswell Park Comprehensive Cancer Center

Department of Biostatistics & Bioinformatics

Elm & Carlton Streets

Buffalo, New York 14263
2 days later
Ahmed Metwally
# 
Hi,


The built is has been done on the updates I pushed. Here you see that the release version has been bumped to 1.2.1 instead of 1.2.0 and dev version has been bumped to 1.3.3 instead of 1.3.0.

However, I still see the same error as before. Here is the OS built results for the release branch: http://bioconductor.org/checkResults/release/bioc-LATEST/OmicsLonDA/merida1-buildsrc.html

Command output

##############################################################################
##############################################################################
###
### Running command:
###
###   /Library/Frameworks/R.framework/Versions/Current/Resources/bin/R CMD build --keep-empty-dirs --no-resave-data OmicsLonDA
###
##############################################################################
##############################################################################


* checking for file ?OmicsLonDA/DESCRIPTION? ... OK
* preparing ?OmicsLonDA?:
* checking DESCRIPTION meta-information ... OK
* installing the package to build vignettes
* creating vignettes ... ERROR
--- re-building ?OmicsLonDA.Rmd? using rmarkdown
Bioconductor version 3.10 (BiocManager 1.30.10), R 3.6.2 (2019-12-12)
Installing package(s) 'OmicsLonDA'
trying URL 'https://bioconductor.org/packages/3.10/bioc/bin/macosx/el-capitan/contrib/3.6/OmicsLonDA_1.2.0.tgz'
Content type 'application/x-gzip' length 5219591 bytes (5.0 MB)

In the above snippet, it says "trying URL 'https://bioconductor.org/packages/3.10/bioc/bin/macosx/el-capitan/contrib/3.6/OmicsLonDA_1.2.0.tgz'
"

v1.2.0 is the old version. The new version is 1.2.1. Should I update anything to make sure BioC checks the new version "1.2.1"?

Thanks,
Martin Morgan
# 
It seems like in your vignette you are trying to install your own package. But if you needed to install the package, then the vignette wouldn't be available to install it from... ;) In this code chunk

```{r results='hide',message=FALSE,warning=FALSE}
if (!requireNamespace("BiocManager", quietly = TRUE))
    install.packages("BiocManager", repos = "http://cran.us.r-project.org")
BiocManager::install("OmicsLonDA")
```

Change the options to

```{r, eval = FALSE}

Martin

?On 12/19/19, 2:22 PM, "Bioc-devel on behalf of Ahmed Metwally" <bioc-devel-bounces at r-project.org on behalf of ametwall at stanford.edu> wrote:

    Hi,
    
    
    The built is has been done on the updates I pushed. Here you see that the release version has been bumped to 1.2.1 instead of 1.2.0 and dev version has been bumped to 1.3.3 instead of 1.3.0.
    
    However, I still see the same error as before. Here is the OS built results for the release branch: http://bioconductor.org/checkResults/release/bioc-LATEST/OmicsLonDA/merida1-buildsrc.html
    
    Command output
    
    ##############################################################################
    ##############################################################################
    ###
    ### Running command:
    ###
    ###   /Library/Frameworks/R.framework/Versions/Current/Resources/bin/R CMD build --keep-empty-dirs --no-resave-data OmicsLonDA
    ###
    ##############################################################################
    ##############################################################################
    
    
    * checking for file ?OmicsLonDA/DESCRIPTION? ... OK
    * preparing ?OmicsLonDA?:
    * checking DESCRIPTION meta-information ... OK
    * installing the package to build vignettes
    * creating vignettes ... ERROR
    --- re-building ?OmicsLonDA.Rmd? using rmarkdown
    Bioconductor version 3.10 (BiocManager 1.30.10), R 3.6.2 (2019-12-12)
    Installing package(s) 'OmicsLonDA'
    trying URL 'https://bioconductor.org/packages/3.10/bioc/bin/macosx/el-capitan/contrib/3.6/OmicsLonDA_1.2.0.tgz'
    Content type 'application/x-gzip' length 5219591 bytes (5.0 MB)
    
    In the above snippet, it says "trying URL 'https://bioconductor.org/packages/3.10/bioc/bin/macosx/el-capitan/contrib/3.6/OmicsLonDA_1.2.0.tgz'
    "
    
    v1.2.0 is the old version. The new version is 1.2.1. Should I update anything to make sure BioC checks the new version "1.2.1"?
    
    Thanks,
1 day later
Ahmed Metwally
# 
Thank you, Martin!

It worked exactly as you mentioned. Now the package has been built successfully, except for windows on the release branch. Here is the error: http://bioconductor.org/checkResults/release/bioc-LATEST/OmicsLonDA/tokay1-buildsrc.html


* installing the package to build vignettes
      -----------------------------------
ERROR: dependencies 'SummarizedExperiment', 'BiocParallel' are not available for package 'OmicsLonDA'
* removing 'C:/Users/biocbuild/bbs-3.10-bioc/tmpdir/RtmpULajPf/Rinst420c6eb71a2e/OmicsLonDA'

I have 'SummarizedExperiment' and  'BiocParallel' in the import section of the DESCRIPTION file:
https://github.com/aametwally/OmicsLonDA/blob/master/DESCRIPTION

Any help?
Martin Morgan
# 
I think this is because of build failures of other packages on the Windows builder, as mentioned by Lori

  https://stat.ethz.ch/pipermail/bioc-devel/2019-December/015962.html

For instance, your package can?t find SummarizedExperiment, because when the build system tried to install SummarizedExperiment it failed

  http://bioconductor.org/checkResults/release/bioc-LATEST/SummarizedExperiment/tokay1-install.html


? because DelayedArray failed

  http://bioconductor.org/checkResults/release/bioc-LATEST/DelayedArray/tokay1-install.html

? because BiocParallel failed

  http://bioconductor.org/checkResults/release/bioc-LATEST/BiocParallel/tokay1-install.html

? because the CRAN package ?snow? failed to install, for reasons that are still under investigation.

There is nothing for you to do until this has been fixed on our end.

Martin

From: Ahmed Metwally <ametwall at stanford.edu>
Date: Friday, December 20, 2019 at 5:08 PM
To: Martin Morgan <mtmorgan.bioc at gmail.com>, "Shepherd, Lori" <Lori.Shepherd at RoswellPark.org>
Cc: "bioc-devel at r-project.org" <bioc-devel at r-project.org>
Subject: Re: [Bioc-devel] Trigger a new build after version bump

Thank you, Martin!

It worked exactly as you mentioned. Now the package has been built successfully, except for windows on the release branch. Here is the error: http://bioconductor.org/checkResults/release/bioc-LATEST/OmicsLonDA/tokay1-buildsrc.html


* installing the package to build vignettes

      -----------------------------------

ERROR: dependencies 'SummarizedExperiment', 'BiocParallel' are not available for package 'OmicsLonDA'

* removing 'C:/Users/biocbuild/bbs-3.10-bioc/tmpdir/RtmpULajPf/Rinst420c6eb71a2e/OmicsLonDA'

I have 'SummarizedExperiment' and  'BiocParallel' in the import section of the DESCRIPTION file:
https://github.com/aametwally/OmicsLonDA/blob/master/DESCRIPTION

Any help?
Ahmed Metwally
# 
I see! Thanks for letting me know, Martin.

Best,