---
title: "Reducing file bloat"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{Reducing file bloat}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)
```

## Change dates

All top level records in a GEDCOM file can record the date and time they were last modified. The `tidyged` package (the main package for creating and summarising GEDCOM files) includes change dates (today's date) by default every time a record is created or modified. Since the time is very unlikely to be useful in such a context, the package ignores this by default. We illustrate by loading the `tidyged` and `tidyged.utils` packages, and creating an example object.

```{r}
library(tidyged)
library(tidyged.utils)

gedcom(subm("Me")) |> 
  knitr::kable()
```

See row `r which(gedcom(subm("Me"))$tag == "CHAN")` and the row after for the change date for the submitter record.

For GEDCOM files with thousands of records, including change dates can add considerable bloat. For this reason it is possible to remove all change date structures with the `remove_change_dates()` function:

```{r}
gedcom(subm("Me")) |> 
  remove_change_dates() |> 
  knitr::kable()
```


## Unreferenced records

If there are any records that are not referenced anywhere else, they can be found with the `identify_unused_records()` function. In the example below we create 6 family group records, half with members, half without, and also an unreferenced Repository record:

```{r}
some_unref <- gedcom(subm("Me")) |> 
  add_indi(qn = "Tom Smith") |> 
  add_indi(qn = "Tammy Smith") |> 
  add_indi(qn = "Alice White") |> 
  add_indi(qn = "Phil Brown")

tom_xref <- find_indi_name(some_unref, "Tom")
tammy_xref <- find_indi_name(some_unref, "Tammy")
phil_xref <- find_indi_name(some_unref, "Phil")
alice_xref <- find_indi_name(some_unref, "Alice")

some_unref <- some_unref |>
  add_famg(husband = tom_xref, wife = tammy_xref) |> 
  add_famg() |> 
  add_famg(husband = phil_xref) |> 
  add_famg() |> 
  add_famg(children = alice_xref) |> 
  add_famg() |> 
  add_repo("Test repo") 
  
identify_unused_records(some_unref)
```

We can find out more about these xrefs by using the `describe_records()` function from the `tidyged` package:

```{r}
identify_unused_records(some_unref) |> 
  describe_records(gedcom = some_unref)
```