[[ mechanism should handle original names

[dankelley]

This relates a bit to #103, the practical answer to which is to choose the order (say temperature, temperature2, temperature3 or whatever) based on column order but to ensure that [[ can handle original names so the user can work with that. Users are likely to know which they prefer.

Right now, [[ does not work on original names. From the code, it looks like it never did. (This package borrows ideas from oce but does not use it at a deep level. That's because of fundamental differences, e.g. a ctd object has no equivalent to glider and payload streams.)

[dankelley]

The first step was to get the original names stored! That was not being done right. The problem is the separation of things into 'glider' and 'payload1', I think. Anyway, I now have that working, as below. This is not pushed to GH. I won't push until I clear the issues opened lately.

PS. note that salinity does not have an original name listed. That's because here, as in oce, we construct salinity if we have conductivity. I am a bit divided on whether we should do that, so I'll think about whether to also include salinity as from the file, and then this constructed one with no original name. The idea is that salinity might not have been printed with sufficient digits and maybe (who knows) the formula was not correct.

library(oceglider)
#> Loading required package: oce
#> Loading required package: gsw
directory <- system.file("extdata/seaexplorer/sub", package = "oceglider")
g <- read.glider.seaexplorer.realtime(directory = directory, yo = 2, progressBar = FALSE)
summary(g)
#> Glider Summary
#> --------------
#> 
#> * Input file:
#>     /Library/Frameworks/R.framework/Versions/4.4-x86_64/Resources/library/oceglider/extdata/seaexplorer/sub/sea021.71.gli.sub.2.gz;/Library/Frameworks/R.framework/Versions/4.4-x86_64/Resources/library/oceglider/extdata/seaexplorer/sub/sea021.71.pld1.raw.2.gz
#> * Type:    seaexplorer
#> * Subtype: realtime
#> * Time ranges from 2023-04-11 13:56:32 to 2023-04-11 14:26:55 with 1763 samples and mean increment 1.034259 s
#> * Data Overview (of the "payload1" stream):
#>                                    Min.       Mean       Max. Dim.         OriginalName
#>     backscatter              0.00018149 0.00043433 0.00091013 1763 FLBBCD_BB_700_SCALED
#>     backscatterCount                116     210.73        389 1763  FLBBCD_BB_700_COUNT
#>     cdom                         0.5418     1.1278      1.806 1763   FLBBCD_CDOM_SCALED
#>     cdomCount                        51     57.489         65 1763    FLBBCD_CDOM_COUNT
#>     chlorophyll                  0.1936    0.73339     1.7424 1763    FLBBCD_CHL_SCALED
#>     chlorophyllCount                 56     100.61        184 1763     FLBBCD_CHL_COUNT
#>     conductivity                 2.7353      2.761     2.8098 1763   GPCTD_CONDUCTIVITY
#>     conductivity2                27.306     27.603     28.101 1763  LEGATO_CONDUCTIVITY
#>     conductivityTemperature      1.6931      2.184     2.5327 1763      LEGATO_CONDTEMP
#>     latitude [°N]                44.491     44.492     44.493 1763         NAV_LATITUDE
#>     LEGATO_TEMPERATURE           1.5852     2.0128      2.229 1763   LEGATO_TEMPERATURE
#>     longitude [°E]              -63.387    -63.386    -63.386 1763        NAV_LONGITUDE
#>     navState                        100     114.96        118 1763         NAV_RESOURCE
#>     oxygenFrequency              3259.4     3886.4     4335.9 1763            GPCTD_DOF
#>     pressure [dbar]                1.15     16.045      47.22 1763       GPCTD_PRESSURE
#>     pressure2                    0.8606     15.863     47.059 1763      LEGATO_PRESSURE
#>     pressureNav                    -0.1     16.722       45.9 1763            NAV_DEPTH
#>     salinity [PSS-78]            30.869     31.025     31.463 1763                    -
#>     salinity2                    30.495     31.017     31.467 1763      LEGATO_SALINITY
#>     temperature [°C, ITS-90]     1.5838     2.0129     2.2243 1763    GPCTD_TEMPERATURE
#>     yoNumber                          2          2          2 1763             yoNumber
#> 
#> * Data-quality Flag Scheme
#> 
#>     name    "IOOS"
#>     mapping list(pass=1, not_evaluated=2, suspect=3, fail=4, missing=9)
#>     default NULL
#> 
#> * Data-quality Flags
#> 
#>     time:                    "2" 1763
#>     navState:                "2" 1763
#>     longitude:               "2" 1763
#>     latitude:                "2" 1763
#>     pressureNav:             "2" 1763
#>     conductivity2:           "2" 1763
#>     LEGATO_TEMPERATURE:      "2" 1763
#>     pressure2:               "2" 1763
#>     salinity2:               "2" 1763
#>     conductivityTemperature: "2" 1763
#>     chlorophyllCount:        "2" 1763
#>     chlorophyll:             "2" 1763
#>     backscatterCount:        "2" 1763
#>     backscatter:             "2" 1763
#>     cdomCount:               "2" 1763
#>     cdom:                    "2" 1763
#>     conductivity:            "2" 1763
#>     temperature:             "2" 1763
#>     pressure:                "2" 1763
#>     oxygenFrequency:         "2" 1763
#>     yoNumber:                "2" 1763
#>     salinity:                "2" 1763
#> 
#> * Processing Log
#> 
#>     - 2024-06-21 13:08:59 UTC: `create 'glider' object`
#>     - 2024-06-21 13:08:59 UTC: `initializeFlagScheme(object, name="IOOS", mapping=list(pass=1,not_evaluated=2,suspect=3,fail=4,missing=9)), default=c())`
#>     - 2024-06-21 13:08:59 UTC: `read.glider.seaexplorer.realtime(directory="/Library/Frameworks/R.framework/Versions/4.4-x86_64/Resources/library/oceglider/extdata/seaexplorer/sub",yo=c(2),missingValue=9999)`

^{Created on 2024-06-21 with reprex v2.1.0}

[dankelley]

This works now in "develop" commit 2b56efaa992166758c21a59015c940313a6eebb5, as shown below. Note that commit does not pass the test suite so I don't recommend using it. Most likely the test suite will fail until Monday. Also, the vignette will need rewriting. Some of this is due to a change in the built-in dataset but part is also due to what might be confusion on my part about just what realtime datasets contain.

PS. the graph shows the mismatch between pressures from two instruments. I suspect if I colour coded for the nav-state, the pattern would be clarified. But my goal right now is to clean up some issues so I can use the main code next week.

library(oceglider)
#> Loading required package: oce
#> Loading required package: gsw
directory <- system.file("extdata/seaexplorer/sub", package = "oceglider")
g <- read.glider.seaexplorer.realtime(directory, progressBar = FALSE)
plot(g[["pressure"]], g[["pressure"]] - g[["LEGATO_PRESSURE"]], cex = 0.5)

^{Created on 2024-06-22 with reprex v2.1.0}

[dankelley]

"develop" commit 1f59e6d5cf6c55cc459a87a2440f38f27d06f8ea now has a test suite that checks this, so I'm closing the issue now.

See a note at the end of the test suite about navState confusion between glider and payload1. I don't have the understanding of file formats to speculate on the difference. I'm not even sure these files are 'realtime' ... but that's another fish to fry.