COCOMO cost estimates

library(glockr)
library(gt)

scc includes a COCOMO 81 model that estimates project effort and cost from SLOC. scc only emits COCOMO output via its tabular formatter — never in its JSON output — so glockr opts in by running scc a second time in tabular mode, scraping the COCOMO block, and parsing it into a tibble.

Opting in with `cocomo = TRUE`

By default scc() returns the per-language tibble and ignores all COCOMO-related arguments. Pass cocomo = TRUE to also compute and return the COCOMO summary:

rlang_cocomo <- scc(rlang_path, cocomo = TRUE)
#  # A tibble: 11 × 10
#     language   files lines  code comments blanks complexity weighted_complexity
#     <chr>      <int> <int> <int>    <int>  <int>      <int>               <dbl>
#   1 R            156 43171 27043    10926   5202       2239                8.28
#   2 Markdown      52 13437 11344        0   2093          0                0   
#   3 C             69 13173 10016      864   2293       1827               18.2 
#   4 C Header      70  8272  5322     1750   1200        677               12.7 
#   5 YAML           8   635   505       26    104          0                0   
#   6 C++            2    25    21        0      4          0                0   
#   7 C++ Header     1    26    21        0      5          0                0   
#   8 SVG           10    10    10        0      0          0                0   
#   9 Makefile       1    11     7        0      4          2               28.6 
#  10 License        1     2     2        0      0          0                0   
#  11 TOML           1     0     0        0      0          0                0   
#  # ℹ 2 more variables: bytes <int>, uloc <int>
#  # A tibble: 3 × 3
#    metric                    project_type value       
#    <chr>                     <chr>        <chr>       
#  1 Estimated Cost to Develop organic      $1,790,851  
#  2 Estimated Schedule Effort organic      17.16 months
#  3 Estimated People Required organic      9.27

The returned value is now a list (returned invisibly) with two tibbles:

names(rlang_cocomo)
#  [1] "scc"    "cocomo"

rlang_cocomo$cocomo |> 
  gt::gt() |> 
  gt::tab_header(
    title = "COCOMO Estimates", 
    subtitle = "rlang package", 
    preheader = "rlang package")

metric	project_type	value
COCOMO Estimates
rlang package
Estimated Cost to Develop	organic	$1,790,851
Estimated Schedule Effort	organic	17.16 months
Estimated People Required	organic	9.27

Customize the COCOMO model

The customization arguments only take effect when cocomo = TRUE:

avg_wage: annual salary in local currency
cocomo_project_type: “organic”, “semi-detached”, or “embedded”
eaf: effort adjustment factor
overhead: corporate overhead multiplier
currency_symbol: symbol shown in the value column
auto_print_scc: skip auto-printing the language tibble

embedded <- scc(rlang_path,
    cocomo              = TRUE,          # required to compute COCOMO
    avg_wage            = 120000L,        
    cocomo_project_type = "embedded",     
    eaf                 = 1.1,            
    overhead            = 2.0,            
    currency_symbol     = "$",            
    auto_print_scc      = FALSE)         
#  # A tibble: 3 × 3
#    metric                    project_type value       
#    <chr>                     <chr>        <chr>       
#  1 Estimated Cost to Develop embedded     $9,558,694  
#  2 Estimated Schedule Effort embedded     18.00 months
#  3 Estimated People Required embedded     26.55

embedded$cocomo |> 
  gt::gt() |> 
  gt::tab_header(
    title = "COCOMO", 
    subtitle = "embedded project", 
    preheader = "rlang package")

metric	project_type	value
COCOMO
embedded project
Estimated Cost to Develop	embedded	$9,558,694
Estimated Schedule Effort	embedded	18.00 months
Estimated People Required	embedded	26.55

organic <- scc(rlang_path,
    cocomo              = TRUE,
    avg_wage            = 140000L,
    cocomo_project_type = "organic",
    eaf                 = 1.1,
    overhead            = 2.0,
    currency_symbol     = "$",
    auto_print_scc      = FALSE)
#  # A tibble: 3 × 3
#    metric                    project_type value       
#    <chr>                     <chr>        <chr>       
#  1 Estimated Cost to Develop organic      $4,083,383  
#  2 Estimated Schedule Effort organic      17.80 months
#  3 Estimated People Required organic      9.83

organic$cocomo |> 
  gt::gt() |> 
  gt::tab_header(
    title = "COCOMO", 
    subtitle = "organic project", 
    preheader = "rlang package")

metric	project_type	value
COCOMO
organic project
Estimated Cost to Develop	organic	$4,083,383
Estimated Schedule Effort	organic	17.80 months
Estimated People Required	organic	9.83

Quieting the auto-print

When cocomo = TRUE, scc() prints both tibbles to the console as a side effect. Set auto_print_scc = FALSE and/or auto_print_cocomo = FALSE to suppress either one and just keep the returned list:

silent <- scc(rlang_path,
              cocomo            = TRUE,
              auto_print_scc    = FALSE,
              auto_print_cocomo = FALSE)

SLOCCount-compatible format

sloccount_format = TRUE reformats the upstream COCOMO block in SLOCCount-compatible form (useful for tooling that parses that output).

sloccount <- scc(rlang_path,
                 cocomo           = TRUE,
                 sloccount_format = TRUE,
                 auto_print_scc   = FALSE)

Total Physical Source Lines of Code (SLOC)                     = 54,291
Development Effort Estimate, Person-Years (Person-Months)      = 13.26 (159.10)
 (Basic COCOMO model, Person-Months = 2.40*(KSLOC**1.05)*1.00)
Schedule Estimate, Years (Months)                              = 1.43 (17.16)
 (Basic COCOMO model, Months = 2.50*(person-months**0.38))
Estimated Average Number of Developers (Effort/Schedule)       = 9.27
Total Estimated Cost to Develop                                = $1,790,851
 (average salary = $56,286/year, overhead = 2.40)

The block doesn’t fit the standard metric / project_type / value shape, so it’s printed verbatim and $cocomo is NULL:

is.null(sloccount$cocomo)
#  [1] TRUE

Why not use COCOMO ???

scc()only gives us raw structural numbers (lines, code, complexity, bytes, etc.), but COCOMO II needs 22 judgement inputs we can’t derive from source code:

The current COCOMO project types ("organic" / "semi-detached" / "embedded") don’t translate to COCOMO II
COCOMO II uses 5 scale factors (PREC, FLEX, RESL, TEAM, PMAT) for organizational/process attributes and 17 effort multipliers (RELY, DATA, CPLX, RUSE, DOCU, TIME, STOR, PVOL, ACAP, PCAP, PCON, APEX, PLEX, LTEX, TOOL, SITE, SCED) on product, platform, personnel, and project context
Every one of these is a human rating (VL / L / N / H / VH / XH) about the team, requirements stability, reuse strategy, tool maturity, schedule pressure, etc.

The only one that might map would be CPLX (product complexity), but COCOMO II’s CPLX is architectural/algorithmic complexity, not McCabe’s cyclomatic.

COCOMO cost estimates