redcap_read() Error in inherits(ds, "data.frame") : object 'ds' not found

[dlemas]

I am trying to create a script that queries physical activity data in BEACH redcap project. When I called the data using redcap_read()- I received the following error:

redcap_read() Error in inherits(ds, "data.frame") : object 'ds' not found

does this happen for anyone else? I have tried several scripts.

BEACH>reports>data_inventory.Rmd

[XinsongDu]

Greetings Dr. Lemas:

I tried the code just now, and I was able to pull the data without the error if I changed the "ds" to "dat". I think the error might be because "ds" has never been defined as a variable in the code? Following is part of my code:

> # Create connections
> rcon <- redcapConnection(url=uri, token=beach_token)
> desired_fields=c("test_id","redcap_event_name",
+                  "beach_part_drop_out",
+                  "mompa_todays_date","mompa_study_visit",
+                  "mompa_walk_slow","mompa_walk_quick",
+                  "mompa_walk_hills","mompa_jog",
+                  "mompa_prenatal_exer","mompa_swim",
+                  "mompa_dance","mompa_other_exer")
> desired_fields=c("test_id","redcap_event_name",
+                  "beach_part_drop_out",
+                  "mompa_todays_date","mompa_study_visit",
+                  "mompa_walk_slow","mompa_walk_quick",
+                  "mompa_walk_hills","mompa_jog",
+                  "mompa_prenatal_exer","mompa_swim",
+                  "mompa_dance","mompa_other_exer")
> dropped.records.v1=c("BLS006A",
+                      "BLS010A",
+                      "BLS015A",
+                      "BLS026A",
+                      "BLS029A",
+                      "BLS037A",
+                      "BLS039A",
+                      "BLS050A",
+                      "BLS054A")
> dat<- redcap_read(
+     batch_size=300,
+     records= dropped.records.v1,
+     redcap_uri = uri, 
+     token      = beach_token, 
+     fields     = desired_fields
+ )$data
> head(dat)
  test_id     redcap_event_name redcap_repeat_instrument redcap_repeat_instance beach_part_drop_out mompa_study_visit mompa_walk_slow mompa_walk_quick
1 BLS006A        baseline_arm_1                       NA                     NA                   1                NA              NA               NA
2 BLS006A third_trimester_arm_1                       NA                     NA                   1                 1               2                2
3 BLS010A        baseline_arm_1                       NA                     NA                   1                NA              NA               NA
4 BLS010A third_trimester_arm_1                       NA                     NA                   0                 1               6                2
5 BLS010A        two_week_arm_1                       NA                     NA                  NA                 2               2                2
6 BLS010A       two_month_arm_1                       NA                     NA                  NA                 3               3                2
  mompa_walk_hills mompa_jog mompa_prenatal_exer mompa_swim mompa_dance
1               NA        NA                  NA         NA          NA
2                2         1                   1          1           1
3               NA        NA                  NA         NA          NA
4                1         1                   1         NA           1
5                1         1                   2          1           1
6                1         1                   1          1           1
> inherits(dat, "data.frame")
[1] TRUE

[luran1]

that happens to me sometimes, A workaround is declaring an empty data frame with the same variable name before the redcap_read().

[ChemiKyle]

Hi guys! Not sure if this issue is still causing you trouble, but we (UF CTS-IT) came across this issue in a project we're running and traced the problem to non-UTF-8 characters in the section headers for the REDCap project we were reading data from.

redcap_read crashes along the way because the rogue character upsets the redcap_read_metadata function (which is not used in redcap_read_oneshot, so that still works, so it'd be a useful litmus test for the metadata failure cascade).