Add targets for uploading to S3 + repo cleanup #46
Conversation
|
Once this is merged, I plan to delete the now unused files in S3. I don't want to do this right away, though, because I believe that we will need to build beta and prod or they will break?
|
| # Convert to `data.table` for fast reshaping, then back | ||
| # to tibble. Note that `reshape2::dcast()` is deprecated. | ||
| setDT() %>% |
There was a problem hiding this comment.
dcast has been one of my most used functions, I'm resistant to the change
There was a problem hiding this comment.
Can still use dcast() but has to be from the data.table package not rehsape2. The warning I got said the next update would turn it into an error!
|
|
||
| THE ANALYSIS IN THIS VIZ REPO (AS IT CURRENTLY STANDS 12/23/2020) IS NOT MEANT TO BE FINAL. JUST USED AS AN EXAMPLE OF HOW TO BUILD A VIDEO-BASED DATAVIZ USING SCIPIPER | ||
| ## Generate the data behind the visualization and push to S3 | ||
|
|
There was a problem hiding this comment.
Maybe add a little more context here, like:
This animation is built using an R-based pipeline for data processing, and Vue.js + D3.js to create an animated site. The R-based pipeline (1) calculates daily percentiles for each well based on the historic record at each site, and (2) writes an svg map for the base of the animation. This pipeline leverages an internal package, library(scipiper) to automate the workflow. You do not need to run the pipeline to be able to build the app locally.
README.md
Outdated
| ## Build the visualization locally | ||
|
|
||
| The data (all CSVs and SVGs) needed for this visualization are already publicly available through `labs.waterdata.usgs.gov/visualizations/data/[FILENAME]`. You should be able open this repo, run the following code chunk, and see the visualization locally (http://localhost:8080/) without needed to build the data pipeline: | ||
|
|
||
| ``` | ||
| npm install | ||
| npm run serve |
There was a problem hiding this comment.
and then move this up before we get into how to build the data? I suggest that because it is simpler, but the data part might be what people visiting the repo are more interested in.
…he instructions are shown for the app vs pipeline
Fixes #41. Also helps with #27 but doesn't complete it. Still need to figure out the dynamic-naming piece & how to inject SVG into Vue from a URL.
What this PR does do:
.indfiles for our S3 pushes)reshape2::dcast()deprecationassets/until @ceenell is able to inject SVG from a URL into Vue.Results in the current state of the app from a local build to look like this: