Closed clhunsen closed 8 years ago
I wrote Michael Rutter (the maintainer of the c2d4u
PPA with the broken package dependency) a message and he immediately fixed the issue. Thus, the issue is fixed soon, although, I think we should keep the patch (a80b39f), just to be sure for the future.
Please remove the svglite fix then.
There is no need to build the package from source when it is maintained properly in the repositories. That's effectively only wasting time during provisioning the CI and Vagrant environment.
As it seems, the dangling dependency haven't been fixed completely. I sent the package maintainer another message regarding the problem.
Problem with svglite
is solved now.
The last Travis build failed due to a random timeout, but it looks as if yet another R package just broke in addition to that:
0: 1: config.script.run({ conf <- config.from.args(positional.args = list("re 2: withCallingHandlers(expr, error = function(e) { if (!interactive()) { 3: do.ts.analysis(resdir, graphdir, conf) 4: sapply(seq(min.year, max.year), function(year) { logdevinfo(paste("Crea 5: lapply(X = X, FUN = FUN, ...) 6: FUN(X[[i]], ...) 7: ggsave(paste(graphdir, "/ts_", year, ".pdf", sep = ""), g.year, width = 16, 8: grid.draw(plot) 9: grid.draw.ggplot(plot) 10: print(x) 11: print.ggplot(x) 12: ggplot_build(x) 13: lapply(data, scales_transform_df, scales = scales) 14: FUN(X[[i]], ...) 15: unlist(lapply(scales$scales, function(s) s$transform_df(df = df)), recursiv 16: lapply(scales$scales, function(s) s$transform_df(df = df)) 17: FUN(X[[i]], ...) 18: s$transform_df(df = df) 19: f(..., self = self) 20: lapply(df[aesthetics], self$transform) 21: FUN(X[[i]], ...) 22: f(..., self = self) 23: self$trans$transform(x) 24: stop("Invalid input: date_trans works with objects of class Date only", cal
I fixed that already in commit e1cda1b.
Am 11/04/2016 um 16:44 schrieb Claus Hunsen:
I fixed that already in commit e1cda1b https://github.com/siemens/codeface/commit/e1cda1b2d5e93dca3b53a8fd3c9f92b7164b2013.
Thanks -- I have unfortunately not yet come around to reviewing your pull request, but will do so as soon as possible!
Thanks, Wolfgang
— You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub https://github.com/siemens/codeface/pull/44#issuecomment-208380756
Merged, thanks!
This pull request brings an important bugfix for the function
important.community.nodes
that computes the most important node per community, given a list of node ranks and node communities. The complete description can be found in the commit message [a37243bd0daa1b6ed90bc743f2d6f79f3470ebc8].Furthermore, I have improved the logging for the mailing-list download [a8f4bb89f3dd926237f5c7e5608dbe9acd7bccd6] and introduced logic to skip the time-series analysis for the case that too few releases are available [acbe970].
The integration scripts now set-up the
codeface_testing
database, as it is the default forcodeface test
and some users rely on it. [fd24ae9]In the end, I added some improvements and fixes for R packages [e1cda1b, ff0897d].