Dear fellow R-developers, I received an error when submitting my package MHCtools v. 1.4.3 to CRAN, and I?m simply blank on what to do. Please see: <https://CRAN.R-project.org/web/checks/check_results_MHCtools.html> It appears to concern the pdf version of the manual, which is generated by the CRAN servers from the markdown docs in the package folder ?man?, which in turn are generated by Roxygen from the R code files. If I undestand the CRAN package check results correctly, the problem seems to be some characters, that are not translated by the package inputenc in the CRAN server: ! Package inputenc Error: Unicode char ?<c2><80><c2><90> (U+2010) (inputenc) not set up for use with LaTeX. This was also listed as existing problems for the previous version of MHCtools (1.4.2), and I?m equally baffled about that, because that passed CRAN submission without errors on May 23rd this year. I have browsed through all my code files and all the markdown files in the package folder ?man?, and I don?t find any odd characters. Interestingly, I don?t get the error when running check_win_devel() or check_rhub(). I hope someone can shed some light on what is going on here. Kind regards, Jacob Roved Postdoc GLOBE Institute University of Copenhagen
[R-pkg-devel] Inputenc error when submitting package
3 messages · Jacob Roved, Duncan Murdoch, Sebastian Meyer
On 08/08/2022 8:58 a.m., Jacob Roved wrote:
Dear fellow R-developers, I received an error when submitting my package MHCtools v. 1.4.3 to CRAN, and I?m simply blank on what to do. Please see: <https://CRAN.R-project.org/web/checks/check_results_MHCtools.html> It appears to concern the pdf version of the manual, which is generated by the CRAN servers from the markdown docs in the package folder ?man?, which in turn are generated by Roxygen from the R code files. If I undestand the CRAN package check results correctly, the problem seems to be some characters, that are not translated by the package inputenc in the CRAN server: ! Package inputenc Error: Unicode char ?<c2><80><c2><90> (U+2010) (inputenc) not set up for use with LaTeX.
U+2010 is the hyphen character. It looks a lot like the ascii "-" (code 2D), which Unicode calls "hyphen-minus". You can find occurrences of it (and other non-ASCII chars) in your files using the tools::showNonASCIIfile() function, e.g. files <- list.files(recursive = TRUE) for (f in files) tools::showNonASCIIfile(f) Duncan Murdoch
This was also listed as existing problems for the previous version of MHCtools (1.4.2), and I?m equally baffled about that, because that passed CRAN submission without errors on May 23rd this year. I have browsed through all my code files and all the markdown files in the package folder ?man?, and I don?t find any odd characters. Interestingly, I don?t get the error when running check_win_devel() or check_rhub(). I hope someone can shed some light on what is going on here. Kind regards, Jacob Roved Postdoc GLOBE Institute University of Copenhagen [[alternative HTML version deleted]]
______________________________________________ R-package-devel at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-package-devel
Am 09.08.22 um 10:30 schrieb Duncan Murdoch:
On 08/08/2022 8:58 a.m., Jacob Roved wrote:
Dear fellow R-developers, I received an error when submitting my package MHCtools v. 1.4.3 to CRAN, and I?m simply blank on what to do. Please see: <https://CRAN.R-project.org/web/checks/check_results_MHCtools.html> It appears to concern the pdf version of the manual, which is generated by the CRAN servers from the markdown docs in the package folder ?man?, which in turn are generated by Roxygen from the R code files. If I undestand the CRAN package check results correctly, the problem seems to be some characters, that are not translated by the package inputenc in the CRAN server: ???? ! Package inputenc Error: Unicode char ?<c2><80><c2><90> (U+2010) ???? (inputenc) not set up for use with LaTeX.
U+2010 is the hyphen character.? It looks a lot like the ascii "-" (code 2D), which Unicode calls "hyphen-minus".
In most fonts, "\u2010" is indistinguishable from the ASCII "hyphen-minus". LaTeX/inputenc gained support for the Unicode hyphen in the March 2016 release (substituting an ASCII hyphen), so you need a rather old OS to reproduce the error. Among CRAN's check systems, r-oldrel-windows seems to be the only one using a sufficiently old LaTeX version. FWIW, Debian oldoldstable (stretch) is at TeX Live 2016, so supports U+2010. Sebastian Meyer
You can find occurrences of it (and other non-ASCII chars) in your files using the tools::showNonASCIIfile() function, e.g. ? files <- list.files(recursive = TRUE) ? for (f in files) tools::showNonASCIIfile(f) Duncan Murdoch
This was also listed as existing problems for the previous version of MHCtools (1.4.2), and I?m equally baffled about that, because that passed CRAN submission without errors on May 23rd this year. I have browsed through all my code files and all the markdown files in the package folder ?man?, and I don?t find any odd characters. Interestingly, I don?t get the error when running check_win_devel() or check_rhub(). I hope someone can shed some light on what is going on here. Kind regards, Jacob Roved Postdoc GLOBE Institute University of Copenhagen ????[[alternative HTML version deleted]]
______________________________________________ R-package-devel at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-package-devel
______________________________________________ R-package-devel at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-package-devel