I have a package I am developing with some notebooks I would like to let users play with, and I am using myBinder.org as online computational environment.
All fine, until I replaced in the [compat] section of Project.toml the entry with Julia with multiple values:
julia = "1.3.1,1.3,1.4,1.5"
Now I receive the following error in myBinder:
Error during build: invalid literal for int() with base 10: '1,1'
If this is a bug, where should I report it ?
EDIT:
This has been addressed in the jupyter discourse forum and resulted in a new issue being created on their repository: https://discourse.jupyter.org/t/julia-mybinder-build-error-invalid-literal-for-int-with-base-10/4755/3
Related
I am writing a vignette in RMarkdown and I want to point out a programming mistake.
listofnumbers <- as.list(1:4)
listofnumbers[2] + listofnumbers[3]
This code snippet produces an error:
Error in listofnumbers[2] + listofnumbers[3] : non-numeric argument
to binary operator
However, I am working on a system localised to German. The error message I get is in German:
Error in listofnumbers[2] + listofnumbers[3]: nicht-numerisches Argument für binären Operator
If anyone renders the vignette themselves that may be good. But to pre-render it with the package I would need the English message.
Also, I do not want to change my system settings permanently to English.
I investigated using an output hook. But I do not know how to translate the German message to English.
I am working on Windows 10.
If the environment variable LANGUAGE is set to en, error messages will be displayed in English. This needs to be in place at the time R starts.
If you are using RStudio, then it starts a new copy of R to process vignettes when you click on Knit, so you can do this in the current session using
Sys.setenv(LANGUAGE="en")
That should also work if the vignette is processed when you build the package as long as you start a new session to do the build. By default devtools::build() and related functions do that.
If you are running RMarkdown within the current session (e.g. by calling rmarkdown::render()) that won't work, because it happens too late. In that case you can use OS features to set the environment variable, or set it in the .Renviron file in your home directory as
LANGUAGE=en
Since you say you are on Windows, you have to do this carefully: Windows doesn't like filenames like .Renviron, and may try to change it to something else.
This method should work if you use R CMD build to build your packages outside of RStudio.
I'm new to Julia and I want to use COBRA Package.
For adding COBRA I use command :
Pkg.add("COBRA")
But when running, I get these errors:
INFO: Building WinRPM
WARNING: skipping repodata/repomd.xml, not in cache -- call WinRPM.update() to download
WARNING: skipping repodata/repomd.xml, not in cache -- call WinRPM.update() to download
INFO: Downloading https://cache.julialang.org/http://download.opensuse.org/repositories/windows:/mingw:/win32/openSUSE_Leap_42.2/repodata/repomd.xml
WARNING: Unknown download failure, error code: 2148270086
WARNING: Retry 1/5 downloading: https://cache.julialang.org/http://download.opensuse.org/repositories/windows:/mingw:/win32/openSUSE_Leap_42.2/repodata/repomd.xml
until:
WARNING: Unknown download failure, error code: 2148270086
WARNING: Retry 5/5 downloading: https://cache.julialang.org/http://download.opensuse.org/repositories/windows:/mingw:/win32/openSUSE_Leap_42.2/repodata/repomd.xml
WARNING: received error 0 while downloading
https://cache.julialang.org/http://download.opensuse.org/repositories/windows:/mingw:/win32/openSUSE_Leap_42.2/repodata/repomd.xml
After these errors, I have checked the link and I've found out the source is not there, anymore.
So then I ran this command as said above:
WinRPM.update()
But it has thrown this error:
ERROR: UndefVarError: WinRPM not defined
Please Help me to add COBRA to Julia. I'm using version 0.6.4 because Cobra does not work on the next versions.(Windows 10)
Any help would be appreciated.
The problem, as you've mentioned, is that the original sources used by the package aren't valid URLs anymore. The proper way to solve this is to open a pull request with the package to change the URLs (which I've now done here ). As a stopgap measure, here's a hacky way to work around this issue for now:
Do a Pkg.add("WinRPM"). This might trigger the same warnings as above, ignore these.
Next, using WinRPM to load the package we added
Now, the source URLs are read from the sources.list file (linked above) into a global WinRPM.sources variable. We're going to edit this variable's contents to point to new working URLs. (This is generally a terrible idea, to directly poke into the internals of a module and change things.)
julia> WinRPM.sources[:] = replace.(WinRPM.sources, "Leap_42.3" => "Leap_15.3")
2-element Vector{String}:
"https://cache.julialang.org/http://download.opensuse.org/repositories/windows:/mingw:/win32/openSUSE_Leap_15.3"
"https://cache.julialang.org/http://download.opensuse.org/repositories/windows:/mingw:/win64/openSUSE_Leap_15.3"
A WinRPM.update() should now work, and download things into a local cache.
I'm using version 0.6.4 because Cobra does not work on the next versions.
I hope you mean Julia 1.6.4 here, or version 0.6.4 of something else in the stack. If you mean Julia version 0.6.4, you might face further problems down the road and find it hard to get support for such an old version.
I am trying to setup the working directory-using the url-"https:\..."
Code:
box_setwd(url)
Error:-
Error in box_id(dir_id) :
box.com API ids must be (coercible to) 64-bit integers.
I am exploring bit64 package meanwhile it will be great to have a workaround.
I hope the details are enough.
the dir and other ids pertaining to the boxr commands are the numeric extensions in the url- it's available in the documentation
Im facing some difficulties while using Tesseract.
First let me present my actual configuration.
Im working on Windows and have this Tesseract version :
tesseract 3.05.00dev
leptonica-1.73
libgif 4.1.6(?) : libjpeg 8d (libjpeg-turbo 1.4.2) : lib
.0.6 : zlib 1.2.8 : libwebp 0.4.3 : libopenjp2 2.1.0
I have added every languages available with the msi installation.
Unfortunately, i am not able to run a conversion with dutch, italian, spanish, and french languages.
Im using the cmd to execute tesseract.
To show you what kind of error i keep having, im going to use a pretty simple example.
When I run that :
tesseract MainMenu.tif result -l deu
I keep having this error endlessly :
Params model::Incomplete line
That is pretty weird and i am unable to spot whats going wrong.
I can convert in any other languages without any trouble.
I have tried to dl the latest .traineddata from github but it doesnt change anyhing.
I have uninstalled and re-installed the soft too.
Using tesseract --list-langs i am able to see deu, fra, spa and ita.
I have been reading the source code from Github, that makes this error message pop. But I cant find what I am doing wrong.
As I said it works just fine with any languages, except those 4.
Hopefully you will have a lead on this one
I got this problem too.
In my case it is because of using the wrong version of trained data file.
I have use https://github.com/tesseract-ocr/tessdata (master).
This is for tesseract 4 (for this date 15/1/2017).
The solution is use this https://github.com/tesseract-ocr/tessdata/tree/3.04.00.
hope this helps.
When I download files (Windows 7) using
tryCatch(download.file(paste0(url_bv,arq), paste0(dir_bv,arq))
,error=function(cond) message(paste('erro:',arq,'não encontrado'))
,warning=function(cond) message(paste('warning:',arq,'não encontrado')))
I can´t delete files, getting the msg in the question title.
If I just do
download.file(paste0(url_bv,arq), paste0(dir_bv,arq))
there is no problem.
How can I release the folder/file in R?
I think this is likely a bug in R (windows version). I can faithfully reproduce it in R-3.2.5 (win10_64). I suggest you file a bug report (read R FAQ 9.2 for clear direction).
I tried it with several variations:
different url schemes ("http://" and "https://")
different download.file(..., method=...) options ("wininet", "internal", and "auto" all behaved in this manner; "libcurl" did not, but it merely downloaded the 404.html without telling me there was a problem)
triggers whenever warning=... or condition=... clauses are given to tryCatch and something weird happens in expr
does not trigger with error=... or finally=... arguments
32bit and 64bit variants
not in RStudio
as R --vanilla
All attempted on ubuntu-14.04 (R-3.2.3-64bit) as well, it would not trigger.