difficulty loading previous ecodata versions

[sgaichas]

I've tried devtools::install_github("NOAA-EDAB/[email protected]") to get back the commercial engagement indicator from the 2021 report (a time series of med-high and high), but it appears to install the current ecodata when I do this, and the newer version of the indicator is still there (engagement by community).

Entirely likely I'm doing it wrong but here is what I get:

✔  checking for file ‘/private/var/folders/5x/skybg7p1061160n07vnw9cq40000gn/T/RtmpP81H7F/remotes1cad71c22b7f/NOAA-EDAB-ecodata-933e882/DESCRIPTION’ ...
─  preparing ‘ecodata’:
✔  checking DESCRIPTION meta-information ...
─  checking for LF line-endings in source and make files and shell scripts
─  checking for empty or unneeded directories
─  building ‘ecodata_0.1.0.tar.gz’
   
* installing *source* package ‘ecodata’ ...
** using staged installation
** R
** data
*** moving datasets to lazyload DB
** inst
** byte-compile and prepare package for lazy loading
** help
*** installing help indices
** building package indices
** installing vignettes
** testing if installed package can be loaded from temporary location
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* DONE (ecodata)
> library(ecodata)
> eng<-ecodata::engagement %>% 
+   filter(!Var == "med.high.scores")
Error in `filter()`:
! Problem while computing `..1 = !Var == "med.high.scores"`.
Caused by error in `mask$eval_all_filter()`:
! object 'Var' not found

[sgaichas]

It seems to install the same tar.gz when I build current:

> remotes::install_github("NOAA-EDAB/ecodata")
Downloading GitHub repo NOAA-EDAB/ecodata@HEAD
✔  checking for file ‘/private/var/folders/5x/skybg7p1061160n07vnw9cq40000gn/T/RtmpP81H7F/remotes1cad277675b6/NOAA-EDAB-ecodata-cbbff94/DESCRIPTION’ ...
─  preparing ‘ecodata’: (475ms)
✔  checking DESCRIPTION meta-information ...
─  checking for LF line-endings in source and make files and shell scripts
─  checking for empty or unneeded directories
─  building ‘ecodata_0.1.0.tar.gz’
   Warning in utils::tar(filepath, pkgname, compression = compression, compression_level = 9L,  :
     storing paths of more than 100 bytes is not portable:
     ‘ecodata/docs/images/Figure 5_ PSP samples percentage  exceeded the toxicity threshold in Maine 2005-2019 - David Nelson - NOAA Federal.jpg’
   Warning in utils::tar(filepath, pkgname, compression = compression, compression_level = 9L,  :
     storing paths of more than 100 bytes is not portable:
     ‘ecodata/docs/images/Figure 8_PSP samples percentage exceeded the threshold in New Hampshire 2000-2019 - David Nelson - NOAA Federal.jpg’
   Warning in utils::tar(filepath, pkgname, compression = compression, compression_level = 9L,  :
     storing paths of more than 100 bytes is not portable:
     ‘ecodata/docs/images/Figure11_PSP_samples_percentage_exceeding_threshold_in_Massachusetts_1972-2019 - David Nelson - NOAA Federal.jpg’
   Warning in utils::tar(filepath, pkgname, compression = compression, compression_level = 9L,  :
     using GNU extension for long pathname
   
* installing *source* package ‘ecodata’ ...
** using staged installation
** R
** data
*** moving datasets to lazyload DB
** inst
** byte-compile and prepare package for lazy loading
** help
*** installing help indices
** building package indices
** installing vignettes
** testing if installed package can be loaded from temporary location
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* DONE (ecodata)
* ```

[andybeet]

What you have done should work. not sure why it hasnt.
As a work around you could just download the zip file found on the [release page]
You'd need to create a project copy the files then build it

[kimberly-bastille]

It looks like the time series version of the engagement indicator was included in the 2020 SOEs (Version 1.1.0).

If that install still doesn't work, the raw data is available here.

[sgaichas]

Bummer, same thing happens when I try 1.1.0. I will do one of the workarounds mentioned above. Thanks @andybeet and @kimberly-bastille

[slucey]

Linked to issue #69

[slucey]

Description file still says version 0.1.0 ... possible with switch to pkgdown this might be fixed.

[andybeet]

@sgaichas So i can install v1.1.0 (and other versions) and access the engagement data

devtools::install_github("NOAA-EDAB/[email protected]", force = T)
#> Downloading GitHub repo NOAA-EDAB/[email protected]
#>          checking for file 'C:\Users\andrew.beet\AppData\Local\Temp\2\RtmpAPmDSV\remotes4db05852ec1\NOAA-EDAB-ecodata-2d173f3/DESCRIPTION' ...  ✔  checking for file 'C:\Users\andrew.beet\AppData\Local\Temp\2\RtmpAPmDSV\remotes4db05852ec1\NOAA-EDAB-ecodata-2d173f3/DESCRIPTION'
#>       ─  preparing 'ecodata': (1.3s)
#>    checking DESCRIPTION meta-information ...     checking DESCRIPTION meta-information ...   ✔  checking DESCRIPTION meta-information
#>       ─  checking for LF line-endings in source and make files and shell scripts
#>       ─  checking for empty or unneeded directories
#>       ─  building 'ecodata_0.1.0.tar.gz'
#>      
#> 

^{Created on 2023-10-05 by the reprex package (v2.0.1)}

And i also see that the tarball has the same name for any install (i think this is because the DESCRIPTION file is v0.1.0 for every release, as @slucey noted. We need to update that). This is also why packageVersion("ecodata") always reports v0.1.0

library(magrittr) 
    ecodata::engagement %>% tibble::as_tibble()
#> # A tibble: 150 × 5
#>    EPU    Time Var              Value Units   
#>    <chr> <dbl> <chr>            <dbl> <chr>   
#>  1 NE     2004 %Low            0.801  unitless
#>  2 NE     2004 %Moderate       0.101  unitless
#>  3 NE     2004 %Medium High    0.0461 unitless
#>  4 NE     2004 %High           0.0524 unitless
#>  5 NE     2004 med.high.scores 0.677  unitless
#>  6 NE     2005 %Low            0.782  unitless
#>  7 NE     2005 %Moderate       0.111  unitless
#>  8 NE     2005 %Medium High    0.0587 unitless
#>  9 NE     2005 %High           0.0482 unitless
#> 10 NE     2005 med.high.scores 0.673  unitless
#> # ℹ 140 more rows

^{Created on 2023-10-05 by the reprex package (v2.0.1)}

I did need to restart R after the install. If that doesn't work then you might want to remove.packages("ecodata") and reinstall

We also might want to consider adding a NEWS.md file now too to report changes in package betwen versions

[BBeltz1]

@BBeltz1 check that old ecodata can be installed