| <html> | |
| <head> | |
| <script type="text/javascript" src="https://www.gstatic.com/charts/loader.js"></script> | |
| <script type="text/javascript"> | |
| // This is a quick hack to be able to see Gantt-like Timelines of gitlab build pipelines. | |
| // It makes it easier to diagnose the running time of the pipelines and understand resource usage. | |
| // Ideally, gitlab would directly provide such a view in its frontend. | |
| google.charts.load('current', {'packages':['gantt']}); |
| library(shiny) | |
| library(shinyjs) | |
| if (!dir.exists('www/')) { | |
| dir.create('www') | |
| } | |
| download.file( | |
| url = 'https://cdn.jsdelivr.net/npm/js-cookie@2/src/js.cookie.min.js', | |
| destfile = 'www/js.cookie.js' |
| library(shiny) | |
| # Also uses parallel, shinyjs, tools | |
| # Create a long-running task, executed in a forked process. (Doesn't work on Windows) | |
| # | |
| # The return value is a promise-like object with three | |
| # methods: | |
| # - completed(): FALSE initially, then TRUE if the task succeeds, | |
| # fails, or is cancelled. Reactive, so when the state changes | |
| # any reactive readers will invalidate. |
This depends on biom version >= 2.1.5, < 2.2.0 and vsearch >= 1.7.0.
Please note that all of this is highly experimental. I'm keeping these notes as I work with this approach. For published work, I still recommend using standard pipelines, such as those in QIIME 1.9.1.
$ biom --version
biom, version 2.1.5
$ vsearch --version
vsearch v1.7.0_osx_x86_64, 16.0GB RAM, 8 cores