knitr::opts_chunk$set( collapse = TRUE, comment = "#>" )
This article explains conventions in googlesheets4 around:
Almost all functions in googlesheets4 have one of these prefixes:
gs4_
, referring variously to the googlesheets4 package, v4 of the Google
Sheets API, or to operations on one or more (spread)Sheetssheet_
, referring to an operation on one or more (work)sheetsrange_
, referring to an operation on a range of cellsThis table summarizes what the gs4_
, sheet_
, range_
mean conceptually and which arguments you can expect to see in the function signature.
| prefix | ss
| sheet
| range
| scope |
|--------|-------|---------|---------|------------------|
| gs4_ | yes | no | no | a (spread)Sheet |
| sheet_ | yes | yes | no | a (work)sheet |
| range_ | yes | yes | yes | a range of cells |
Take this table with a grain of salt.
There are a few violations of it in function signatures, when justified, but it's true in broad strokes.
And remember that gs4_
is also used for general, package-level functions.
sheets_
prefixWhen googlesheets4 first appeared on CRAN and up to v0.1.1 (released 2020-03-21), it had a nearly universal sheets_
prefix.
In version 0.2.0 (released 2020-05-08), googlesheets4 gained a tremendous amount of writing and editing functionality.
At that time, it became clear that the sheets_
scheme was an impediment to generating descriptive, predictable function names.
For example, you can delete a (spread)Sheet, a (work)sheet, or a cell range.
Which one of these functions will be named sheets_delete()
?
How do we name the others?
There is no good answer to this, which is why we adopted the 3 prefix strategy.
Any function that existed in v0.1.1 was formally deprecated in v0.2.0 and was removed in v1.0.0. Below are tables documenting the name changes, that also cover functions that only existed in dev versions, but that may have seen use in the community. As time goes on, this article becomes more of historical note.
The dev version of googlesheets4 was bumped from 0.1.1.9000 to 0.1.1.9100 when the sheets_
naming scheme was abandoned.
The last commit under the old scheme was also tagged as "v0.1.1.9000", which means one can install from that specific state via devtools::install_github("tidyverse/googlesheets4@v0.1.1.9000")
.
dat <- tibble::tribble( ~ "<= v0.1.1.9000", ~ ">= v0.1.1.9100", "`sheets_api_key()` (*)", "`gs4_api_key()`", "`sheets_auth()` (*)", "`gs4_auth()`", "`sheets_auth_configure()` (*)", "`gs4_auth_configure()`", "`sheets_deauth()` (*)", "`gs4_deauth()`", "`sheets_endpoints()` (*)", "`gs4_endpoints()`", "`sheets_has_token()` (*)", "`gs4_has_token()`", "`sheets_oauth_app()` (*)", "`gs4_oauth_app()`", "`sheets_token()` (*)", "`gs4_token()`", "`sheets_user()` (*)", "`gs4_user()`" ) knitr::kable(dat)
(*) indicates functions in the CRAN versions <= v0.1.1.
dat <- tibble::tribble( ~ "<= v0.1.1.9000", ~ ">= v0.1.1.9100", "`sheets_browse()` (*)", "`gs4_browse()`", "`sheets_create()`", "`gs4_create()`", "`sheets_find()` (*)", "`gs4_find()`", "`sheets_fodder()`", "`gs4_fodder()`", "`sheets_formula()`", "`gs4_formula()`", "`sheets_example()` (*)", "`gs4_example()`", "`sheets_examples()` (*)", "`gs4_examples()`", "`sheets_get()` (*)", "`gs4_get()`", "`sheets_random()`", "`gs4_random()`" ) knitr::kable(dat)
(*) indicates functions in the CRAN versions <= v0.1.1.
dat <- tibble::tribble( ~ "<= v0.1.1.9000", ~ ">= v0.1.1.9100", "`sheets_append()`", "`sheet_append()`", "`sheets_sheet_add()`", "`sheet_add()`", "`sheets_sheet_copy()`", "`sheet_copy()`", "`sheets_sheet_delete()`", "`sheet_delete()`", "`sheets_sheet_names()`", "`sheet_names()`", "`sheets_sheet_properties()`", "`sheet_properties()`", "`sheets_sheet_relocate()`", "`sheet_relocate()`", "`sheets_sheet_rename()`", "`sheet_rename()`", "`sheets_sheet_resize()`", "`sheet_resize()`", "`sheets_sheets()` (*)", "`sheet_names()`", "`sheets_write()`", "`sheet_write()`" ) knitr::kable(dat)
(*) indicates functions in the CRAN versions <= v0.1.1.
dat <- tibble::tribble( ~ "<= v0.1.1.9000", ~ ">= v0.1.1.9100", "`sheets_auto_resize_dims()`", "`range_autofit()`", "`sheets_cells()` (*)", "`range_read_cells()`", "`sheets_clear()`", "`range_clear()`", "`sheets_edit()`", "`range_write()`", "`sheets_flood()`", "`range_flood()`", "`sheets_read()` (*)", "`range_read()`", "`sheets_speedread()`", "`range_speedread()`" ) knitr::kable(dat)
(*) indicates functions in the CRAN versions <= v0.1.1.
Note: "range" can mean two things in the Sheets API:
A1:B3
(a cell range), Africa
(a worksheet name), Africa!A1:B3
(a sheet-qualified cell range), arts_data
(a named range). Some API endpoints require this, believe it or not.GridRange
. Most API endpoints use this.Fun fact: The "cell rectangle" type of range is almost a sub-case of the A1-style range, except there are rectangles open on one or more sides that can be described via GridRange
that cannot be expressed as an A1-range string.
The mostly-developer-facing article Range specification documents all this and how I approach it internally.
The range_
prefix encompasses both types of ranges and each function has to indicate what sorts of range
it supports, which is determined by the behaviour of the associated API endpoint.
Any user facing class that is related to a schema should be named like googlesheets4_schema_name
, where the schema name is in lower_snake_case.
The internal schema-derived classes should be like googlesheets4_schema_SchemaName
/ googlesheets4_schema
/ list
. Use the googlesheets4_schema
prefix and retain the API's UpperCamelCase.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.