Object oriented programming
Dr. Alexander Fisher
Duke University
January 25, 2023
Review of types
R’s type hierarchy
- from most to least complex: class \(\rightarrow\) mode \(\rightarrow\) type
| value | typeof() |
mode() |
class() |
|---|---|---|---|
TRUE |
logical | logical | logical |
1 |
double | numeric | numeric |
1L |
integer | numeric | integer |
"A" |
character | character | character |
NULL |
NULL | NULL | NULL |
list(1, "A") |
list | list | list |
factor("A") |
integer | numeric | factor |
function(x) x^2 |
closure | function | function |
\+ |
builtin | function | function |
\[ |
special | function | function |
objects have class even if no formal class attribute (seen with attributes())
S3 Object System
What is S3?
S3 is R’s core object oriented implementation
S3 is the only OO system used in the base and stats packages, and it’s the most commonly used system in CRAN packages.
S3 is more flexible (read also: easier to break) than traditional object oriented programming you may have seen before (e.g. Java).
Working model: S3 is a system where we attach metadata to an object via the class attribute.
One of the most compelling reasons to use object oriented programming (OOP) is polymorphism. Polymorphism means that a developer can consider a function’s interface separately from its implementation.
S3 class specialization
Other examples
Note
The ... (dot-dot-dot) argument means the function can take any number of additional arguments. This is called a “variable argument” and it is commonly used so you pass variables to additional nested functions. For example:
Generics (what’s going on!?)
- An S3 object behaves differently from its underlying base type whenever it’s passed to a generic function
The easiest way to tell if a function is a generic is to use sloop::ftype() and look for “generic” in the output, e.g.
Fundamentally, the generic function dispatches to a specialized version of the function depending on the class of its input. UseMethod() tells a generic to look for the specialized version.
The syntax for specialized functions is: <generic>.<class>
Examples
We can see all the specialized versions of the generic using the methods() function.
Methods of print
[1] print.acf*
[2] print.AES*
[3] print.anova*
[4] print.aov*
[5] print.aovlist*
[6] print.ar*
[7] print.Arima*
[8] print.arima0*
[9] print.AsIs
[10] print.aspell*
[11] print.aspell_inspect_context*
[12] print.bibentry*
[13] print.Bibtex*
[14] print.browseVignettes*
[15] print.by
[16] print.changedFiles*
[17] print.check_bogus_return*
[18] print.check_code_usage_in_package*
[19] print.check_compiled_code*
[20] print.check_demo_index*
[21] print.check_depdef*
[22] print.check_details*
[23] print.check_details_changes*
[24] print.check_doi_db*
[25] print.check_dotInternal*
[26] print.check_make_vars*
[27] print.check_nonAPI_calls*
[28] print.check_package_code_assign_to_globalenv*
[29] print.check_package_code_attach*
[30] print.check_package_code_data_into_globalenv*
[31] print.check_package_code_startup_functions*
[32] print.check_package_code_syntax*
[33] print.check_package_code_unload_functions*
[34] print.check_package_compact_datasets*
[35] print.check_package_CRAN_incoming*
[36] print.check_package_datalist*
[37] print.check_package_datasets*
[38] print.check_package_depends*
[39] print.check_package_description*
[40] print.check_package_description_encoding*
[41] print.check_package_license*
[42] print.check_packages_in_dir*
[43] print.check_packages_used*
[44] print.check_po_files*
[45] print.check_pragmas*
[46] print.check_Rd_line_widths*
[47] print.check_Rd_metadata*
[48] print.check_Rd_xrefs*
[49] print.check_RegSym_calls*
[50] print.check_S3_methods_needing_delayed_registration*
[51] print.check_so_symbols*
[52] print.check_T_and_F*
[53] print.check_url_db*
[54] print.check_vignette_index*
[55] print.checkDocFiles*
[56] print.checkDocStyle*
[57] print.checkFF*
[58] print.checkRd*
[59] print.checkRdContents*
[60] print.checkReplaceFuns*
[61] print.checkS3methods*
[62] print.checkTnF*
[63] print.checkVignettes*
[64] print.citation*
[65] print.cli_ansi_html_style*
[66] print.cli_ansi_string*
[67] print.cli_ansi_style*
[68] print.cli_boxx*
[69] print.cli_diff_chr*
[70] print.cli_doc*
[71] print.cli_progress_demo*
[72] print.cli_rule*
[73] print.cli_sitrep*
[74] print.cli_spark*
[75] print.cli_spinner*
[76] print.cli_tree*
[77] print.codoc*
[78] print.codocClasses*
[79] print.codocData*
[80] print.colorConverter*
[81] print.compactPDF*
[82] print.condition
[83] print.connection
[84] print.CRAN_package_reverse_dependencies_and_views*
[85] print.data.frame
[86] print.Date
[87] print.default
[88] print.dendrogram*
[89] print.density*
[90] print.difftime
[91] print.dist*
[92] print.Dlist
[93] print.DLLInfo
[94] print.DLLInfoList
[95] print.DLLRegisteredRoutines
[96] print.document_context*
[97] print.document_position*
[98] print.document_range*
[99] print.document_selection*
[100] print.dummy_coef*
[101] print.dummy_coef_list*
[102] print.ecdf*
[103] print.eigen
[104] print.factanal*
[105] print.factor
[106] print.family*
[107] print.fileSnapshot*
[108] print.findLineNumResult*
[109] print.formula*
[110] print.fseq*
[111] print.ftable*
[112] print.function
[113] print.getAnywhere*
[114] print.glm*
[115] print.hclust*
[116] print.help_files_with_topic*
[117] print.hexmode
[118] print.HoltWinters*
[119] print.hsearch*
[120] print.hsearch_db*
[121] print.htest*
[122] print.html*
[123] print.html_dependency*
[124] print.htmltools.selector*
[125] print.htmltools.selector.list*
[126] print.infl*
[127] print.integrate*
[128] print.isoreg*
[129] print.json*
[130] print.key_missing*
[131] print.kmeans*
[132] print.knitr_kable*
[133] print.Latex*
[134] print.LaTeX*
[135] print.libraryIQR
[136] print.listof
[137] print.lm*
[138] print.loadings*
[139] print.loess*
[140] print.logLik*
[141] print.ls_str*
[142] print.medpolish*
[143] print.method_table*
[144] print.MethodsFunction*
[145] print.mtable*
[146] print.NativeRoutineList
[147] print.news_db*
[148] print.nls*
[149] print.noquote
[150] print.numeric_version
[151] print.object_size*
[152] print.octmode
[153] print.packageDescription*
[154] print.packageInfo
[155] print.packageIQR*
[156] print.packageStatus*
[157] print.pairwise.htest*
[158] print.person*
[159] print.POSIXct
[160] print.POSIXlt
[161] print.power.htest*
[162] print.ppr*
[163] print.prcomp*
[164] print.princomp*
[165] print.proc_time
[166] print.quosure*
[167] print.quosures*
[168] print.raster*
[169] print.Rd*
[170] print.recordedplot*
[171] print.restart
[172] print.RGBcolorConverter*
[173] print.rlang_box_done*
[174] print.rlang_box_splice*
[175] print.rlang_data_pronoun*
[176] print.rlang_dict*
[177] print.rlang_dyn_array*
[178] print.rlang_envs*
[179] print.rlang_error*
[180] print.rlang_fake_data_pronoun*
[181] print.rlang_lambda_function*
[182] print.rlang_message*
[183] print.rlang_trace*
[184] print.rlang_warning*
[185] print.rlang_zap*
[186] print.rlang:::list_of_conditions*
[187] print.rle
[188] print.rlib_bytes*
[189] print.rlib_error_3_0*
[190] print.rlib_trace_3_0*
[191] print.roman*
[192] print.scalar*
[193] print.sessionInfo*
[194] print.shiny.tag*
[195] print.shiny.tag.env*
[196] print.shiny.tag.list*
[197] print.shiny.tag.query*
[198] print.simple.list
[199] print.smooth.spline*
[200] print.socket*
[201] print.srcfile
[202] print.srcref
[203] print.stepfun*
[204] print.stl*
[205] print.StructTS*
[206] print.subdir_tests*
[207] print.summarize_CRAN_check_status*
[208] print.summary.aov*
[209] print.summary.aovlist*
[210] print.summary.ecdf*
[211] print.summary.glm*
[212] print.summary.lm*
[213] print.summary.loess*
[214] print.summary.manova*
[215] print.summary.nls*
[216] print.summary.packageStatus*
[217] print.summary.ppr*
[218] print.summary.prcomp*
[219] print.summary.princomp*
[220] print.summary.table
[221] print.summary.warnings
[222] print.summaryDefault
[223] print.table
[224] print.tables_aov*
[225] print.terms*
[226] print.ts*
[227] print.tskernel*
[228] print.TukeyHSD*
[229] print.tukeyline*
[230] print.tukeysmooth*
[231] print.undoc*
[232] print.vignette*
[233] print.warnings
[234] print.xfun_raw_string*
[235] print.xfun_rename_seq*
[236] print.xfun_strict_list*
[237] print.xgettext*
[238] print.xngettext*
[239] print.xtabs*
see '?methods' for accessing help and source codeFind the documentation
Write the name of the function to see its definition.
function (x, quote = FALSE, max.levels = NULL, width = getOption("width"),
...)
{
ord <- is.ordered(x)
if (length(x) == 0L)
cat(if (ord)
"ordered"
else "factor", "(0)\n", sep = "")
else {
xx <- character(length(x))
xx[] <- as.character(x)
keepAttrs <- setdiff(names(attributes(x)), c("levels",
"class"))
attributes(xx)[keepAttrs] <- attributes(x)[keepAttrs]
print(xx, quote = quote, ...)
}
maxl <- if (is.null(max.levels))
TRUE
else max.levels
if (maxl) {
n <- length(lev <- encodeString(levels(x), quote = ifelse(quote,
"\"", "")))
colsep <- if (ord)
" < "
else " "
T0 <- "Levels: "
if (is.logical(maxl))
maxl <- {
width <- width - (nchar(T0, "w") + 3L + 1L +
3L)
lenl <- cumsum(nchar(lev, "w") + nchar(colsep,
"w"))
if (n <= 1L || lenl[n] <= width)
n
else max(1L, which.max(lenl > width) - 1L)
}
drop <- n > maxl
cat(if (drop)
paste(format(n), ""), T0, paste(if (drop)
c(lev[1L:max(1, maxl - 1)], "...", if (maxl > 1) lev[n])
else lev, collapse = colsep), "\n", sep = "")
}
if (!isTRUE(val <- .valid.factor(x)))
warning(val)
invisible(x)
}
<bytecode: 0x7fa01974f030>
<environment: namespace:base>Missing print functions?
There is no method to print integers specifically:
- When a method for a specific class is not found, the default will be called:
function (x, digits = NULL, quote = TRUE, na.print = NULL, print.gap = NULL,
right = FALSE, max = NULL, width = NULL, useSource = TRUE,
...)
{
args <- pairlist(digits = digits, quote = quote, na.print = na.print,
print.gap = print.gap, right = right, max = max, width = width,
useSource = useSource, ...)
missings <- c(missing(digits), missing(quote), missing(na.print),
missing(print.gap), missing(right), missing(max), missing(width),
missing(useSource))
.Internal(print.default(x, args, missings))
}
<bytecode: 0x7fa01a9c3f50>
<environment: namespace:base>What methods are specialized for a class?
[1] [ [[ [[<- [<- all.equal
[6] as.character as.data.frame as.Date as.list as.logical
[11] as.POSIXlt as.vector c coerce droplevels
[16] format initialize is.na<- length<- levels<-
[21] Math Ops plot print relevel
[26] relist rep show slotsFromS3 summary
[31] Summary xtfrm
see '?methods' for accessing help and source codeMake your own class and class-specific methods
Create an object of a new class
cloud = structure(c(1, 2, 3), # object called cloud with some values
class = "first_class") # new class type
cloud[1] 1 2 3
attr(,"class")
[1] "first_class"Create a new method for first_class objects
Exercise 1
Create a class called “accounting”. If a numeric vector has this class, function print() should print the vector with a $ in front of each number and display values up to two decimals. Create a method for this class and test it on the accounting object expenses below
[1] 500.50 750.00 200.30 305.11
attr(,"class")
[1] "accounting"Hint
- Use
format(500, digits = 2, nsmall = 2)as a template to round to specific significant digits. - Use
pasteorpaste0to piece strings together.
Defining a new S3 Generic
Exercise 2
Below is an S3 method called report, it is designed to return a message about the type/mode/class of an object passed to it.
Create an implementation for classes integer, double and numeric. The functions should print to the screen “I’m an integer”, “I’m a double”, and “I’m a numeric” respectively.
Next,
- run
report(1)andreport(1L). - run
rm("report.integer")and re-run the functions. What has changed? - look at
class(1),mode(1),typeof(1). Does this surprise you?
- Integers and doubles violate the typical
classsearching rule.
Some classes
Date class
Date vectors are built on top of double vectors. They have class “Date” and no other attributes:
The dawn of time
The date is measured in unix time, that is, the number of days that have passed since January 1st, 1970. Seconds are counted similarly, startin from the beginning of time, 1970 00:00:00 UTC.
Examples
Date-times
R offers two ways to store date-time info. POSIXct, and POSIXlt
POSIX stands for “Portable Operating Systems Interface” X (“across”) platforms
ct: calendar timelt: local time
ct is the simplest, is built on top of an atomic vector, and is most appropriate for use in data frames.
Exercise 3
- what class is
today? - how can you find all the different functions that take the class of
today? - how many days until the end of class? Note: the last day of class is April 26. Use
difftime
Matrices
R supports the creation of 2d data structures (rows and columns) of atomic vector types. Generally these are formed via a call to matrix().
Matrix structure
Matrices (and arrays) are just atomic vectors with a dim attribute attached (they do not have a class attribute, but they do have a class).
Arrays
Arrays are just an \(n\)-dimensional extension of matrices and are defined by adding the appropriate dimension sizes.
Data frames
Data frames are built on top of lists with attributes: names, row.names, and class. Here the class is data.frame.
[1] "list"$names
[1] "speed" "dist"
$class
[1] "data.frame"
$row.names
[1] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
[26] 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50List of 2
$ speed: num [1:50] 4 4 7 7 8 9 10 10 10 11 ...
$ dist : num [1:50] 2 10 4 22 16 10 18 26 34 17 ...
- attr(*, "row.names")= int [1:50] 1 2 3 4 5 6 7 8 9 10 ...A data frame is how R handles heterogeneous tabular data (i.e. rows and columns) and is one of the most commonly used data structure in R.
Build your own data frame
