warning if pdf of vignette is missing
4
0
Entering edit mode
@manuela-hummel-1423
Last seen 9.7 years ago
Hello, I am working on building a package under Windows. Everything works well; only if I do not provide the pdf version of the vignette in the 'inst/doc' directory I get the following warning when running Rcmd check: * checking package vignettes in 'inst/doc' ... WARNING Package vignettes without corresponding PDF Also then the pdf is not built with Rcmd build. I have seen on the daily check results page of the development packages that many packages get this warning. How can I avoid this warning or can I ignore it? Or do I have to provide the pdf (which is not recommended)? Thank you very much Best regards Manuela Hummel
• 2.5k views
ADD COMMENT
0
Entering edit mode
Seth Falcon ★ 7.4k
@seth-falcon-992
Last seen 9.7 years ago
Hi Manuela, On 15 Sep 2005, hummel at ibe.med.uni-muenchen.de wrote: > I am working on building a package under Windows. Everything works > well; only if I do not provide the pdf version of the vignette in > the 'inst/doc' directory I get the following warning when running > Rcmd check: > > * checking package vignettes in 'inst/doc' ... WARNING > Package vignettes without corresponding PDF Yes, R CMD check wants to see a pdf corresponding to each Rnw file in the inst/doc package location. > Also then the pdf is not built with Rcmd build. If the proper tools (MikTeX) are installed, R CMD build foo/ should "build" the package's vignettes, that is, create the pdf. > I have seen on the daily check results page of the development > packages that many packages get this warning. How can I avoid this > warning or can I ignore it? Or do I have to provide the pdf (which > is not recommended)? The warnings are because the pdf isn't getting generated. In many cases, the pdf isn't getting generated because there are problems during the vignette processing --- so the best solution is to make it so that the vignettes can be built on Windows (this does work for many packages, so we know that it is possible). Hope that helps some, + seth PS: Next time, send this sort of question to bioc-devel. If you are writing packages for BioC, you should subscribe to this list and post development related questions there.
ADD COMMENT
0
Entering edit mode
Seth Falcon ★ 7.4k
@seth-falcon-992
Last seen 9.7 years ago
On 16 Sep 2005, huber at ebi.ac.uk wrote: > as far as I understand, you can safely ignore this Warning. I have a different opinion. Vignettes are an important component of BioC packages. If R CMD build isn't building them, that usually indicates a problem that needs fixing. If we can help get the vignette building, post some details about what is going on. Best, + seth
ADD COMMENT
0
Entering edit mode
> I have a different opinion. Vignettes are an important component of > BioC packages. If R CMD build isn't building them, that usually > indicates a problem that needs fixing. It depends on the situation, that warning can mean various things. It is entirely likely that the vignettes can and do build when one gets that warning (and this is the situation that I think Wolfgang was referring to). A few versions of R back 'check' started looking to make sure a .pdf was *already* present in /inst/doc for every vignette file. Personally I don't really agree w/ that behavior but I don't make the rules. :) That was actually part of the reason behind all the pre-building/etc before checks are run in the daily package checking code (which I'm guessing has changed radically since I last saw it, but also guessing the general flow of activity is similar in this regard). -J
ADD REPLY
0
Entering edit mode
@wolfgang-huber-3550
Last seen 29 days ago
EMBL European Molecular Biology Laborat…
Hi Manuela, as far as I understand, you can safely ignore this Warning. It seems there are different opinions around what should constitute a warning, ranging from "look at this interesting factoid" to "here's something that is not acceptable for a bioconductor package", which are not the same. However, "R CMD build" should produce the PDF file for the vignette - why does this go wrong? Do have details? Best wishes Wolfgang Manuela Hummel wrote: > Hello, > > I am working on building a package under Windows. Everything works well; > only if I do not provide the pdf version of the vignette in the > 'inst/doc' directory I get the following warning when running Rcmd check: > > * checking package vignettes in 'inst/doc' ... WARNING > Package vignettes without corresponding PDF > > Also then the pdf is not built with Rcmd build. > I have seen on the daily check results page of the development packages > that many packages get this warning. > How can I avoid this warning or can I ignore it? Or do I have to provide > the pdf (which is not recommended)? > > Thank you very much > Best regards > Manuela Hummel > ------------------------------------- Wolfgang Huber European Bioinformatics Institute European Molecular Biology Laboratory Cambridge CB10 1SD England Phone: +44 1223 494642 Fax: +44 1223 494486 Http: www.ebi.ac.uk/huber
ADD COMMENT
0
Entering edit mode
Hi Wolfgang, thank you for your reply. So this warning should not be the problem - but the fact that no vignette can be built... When I build a zip file everything is normal except that no pdf is created. When building a tar.gz file there is something strange: Z:\GlobalAncova>Rcmd build z:/GA_Subversion/GlobalAncova * checking for file 'z:/GA_Subversion/GlobalAncova/DESCRIPTION' ... OK * preparing 'z:/GA_Subversion/GlobalAncova': * creating vignettes ... OK * removing junk files * building 'GlobalAncova_1.2.tar.gz' tar: GlobalAncova/inst/doc/Rplots.ps: file changed as we read it tar: Error exit delayed from previous errors Do you know what can be the reason? Best regards Manuela Wolfgang Huber schrieb: > Hi Manuela, > > as far as I understand, you can safely ignore this Warning. It seems > there are different opinions around what should constitute a warning, > ranging from "look at this interesting factoid" to "here's something > that is not acceptable for a bioconductor package", which are not the > same. > > However, "R CMD build" should produce the PDF file for the vignette - > why does this go wrong? Do have details? > > Best wishes > Wolfgang > > Manuela Hummel wrote: > >> Hello, >> >> I am working on building a package under Windows. Everything works >> well; only if I do not provide the pdf version of the vignette in the >> 'inst/doc' directory I get the following warning when running Rcmd >> check: >> >> * checking package vignettes in 'inst/doc' ... WARNING >> Package vignettes without corresponding PDF >> >> Also then the pdf is not built with Rcmd build. >> I have seen on the daily check results page of the development >> packages that many packages get this warning. >> How can I avoid this warning or can I ignore it? Or do I have to >> provide the pdf (which is not recommended)? >> >> Thank you very much >> Best regards >> Manuela Hummel >> > > > ------------------------------------- > Wolfgang Huber > European Bioinformatics Institute > European Molecular Biology Laboratory > Cambridge CB10 1SD > England > Phone: +44 1223 494642 > Fax: +44 1223 494486 > Http: www.ebi.ac.uk/huber > -------------------------------------
ADD REPLY
0
Entering edit mode
Seth Falcon ★ 7.4k
@seth-falcon-992
Last seen 9.7 years ago
On 19 Sep 2005, hummel at ibe.med.uni-muenchen.de wrote: > When I build a zip file everything is normal except that no pdf is > created. When building a tar.gz file there is something strange: > > Z:\GlobalAncova>Rcmd build z:/GA_Subversion/GlobalAncova > * checking for file 'z:/GA_Subversion/GlobalAncova/DESCRIPTION' ... > OK > * preparing 'z:/GA_Subversion/GlobalAncova': > * creating vignettes ... OK > * removing junk files > * building 'GlobalAncova_1.2.tar.gz' > tar: GlobalAncova/inst/doc/Rplots.ps: file changed as we read it > tar: Error exit delayed from previous errors The first thing I would try is running R CMD build again on a clean source dir. Make sure all generated files are removed first. If this is 100% reproducible, then my guess would be that your Rnw file is doing something funny with figures. Do you have Rgraphviz installed for Windows? + seth
ADD COMMENT
0
Entering edit mode
I recognized that I forgot updating the system path to the new R version... So I guess this was the reason something went wrong (for example with Rgraphviz). Now the pdf is created correctly. Thank you, kind regards Manuela Seth Falcon schrieb: >On 19 Sep 2005, hummel at ibe.med.uni-muenchen.de wrote: > > >>When I build a zip file everything is normal except that no pdf is >>created. When building a tar.gz file there is something strange: >> >>Z:\GlobalAncova>Rcmd build z:/GA_Subversion/GlobalAncova >>* checking for file 'z:/GA_Subversion/GlobalAncova/DESCRIPTION' ... >> OK >>* preparing 'z:/GA_Subversion/GlobalAncova': >>* creating vignettes ... OK >>* removing junk files >>* building 'GlobalAncova_1.2.tar.gz' >>tar: GlobalAncova/inst/doc/Rplots.ps: file changed as we read it >>tar: Error exit delayed from previous errors >> >> > >The first thing I would try is running R CMD build again on a clean >source dir. Make sure all generated files are removed first. > >If this is 100% reproducible, then my guess would be that your Rnw >file is doing something funny with figures. Do you have Rgraphviz >installed for Windows? > >+ seth > >_______________________________________________ >Bioconductor mailing list >Bioconductor at stat.math.ethz.ch >https://stat.ethz.ch/mailman/listinfo/bioconductor > >
ADD REPLY

Login before adding your answer.

Traffic: 975 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6