|
1 | | -#' Creates filtered register csv files |
2 | | -#' |
3 | | -#' Each csv file is saved in the appropriate output_dir. |
4 | | -#' |
5 | | -#' @param filter_by A vector of strings specifying the names of the columns to filter by. |
6 | | -#' @param register A dataframe representing the register data to be filtered. |
7 | | -create_filtered_register_csvs <- function(filter_by, register){ |
8 | | - |
| 1 | +#' Creates filtered CSV files from a register based on specified filters. |
| 2 | +#' |
| 3 | +#' The function processes the register by applying filters specified in `filter_by`. |
| 4 | +#' For "codecheckers", a temporary CSV is loaded and processed as the original register.csv |
| 5 | +#' does not have the codechecker column. |
| 6 | +#' The register is then grouped by the filter column, and for each group, a CSV file is generated. |
| 7 | +#' |
| 8 | +#' @param register The register to be filtered. |
| 9 | +#' @param filter_by List of filters to apply (e.g., "venues", "codecheckers"). |
| 10 | +#' |
| 11 | +create_filtered_reg_csvs <- function(register, filter_by){ |
9 | 12 | for (filter in filter_by){ |
10 | | - column_name <- determine_filter_column_name(filter) |
11 | | - |
12 | | - # If filtered by codecheckers we replace the register with the register with codechecker |
13 | | - # columns |
14 | 13 | if (filter == "codecheckers"){ |
| 14 | + # Using the temporary codechecker register |
15 | 15 | register <- read.csv(CONFIG$DIR_TEMP_REGISTER_CODECHECKER, as.is = TRUE) |
16 | 16 | # Once the temp_register is loaded, we can remove it |
17 | 17 | file.remove(CONFIG$DIR_TEMP_REGISTER_CODECHECKER) |
18 | | - } |
19 | | - |
20 | | - unique_values <- get_unique_values_from_filter(register, column_name) |
21 | 18 |
|
22 | | - # Filtering the register |
23 | | - for (value in unique_values) { |
24 | | - # For filtering by codechecker we need to check if unique value is contained |
25 | | - # in the list which is the row value. |
26 | | - if (column_name == "Codechecker"){ |
27 | | - mask <- sapply(register$Codechecker, function(x) value %in% fromJSON(x)) |
28 | | - filtered_register <- register[mask, ] |
29 | | - |
30 | | - #! Edit depending on whether they want to keep the column |
31 | | - # Only keeping the column values specified in CONFIG$REGISTER_COLUMNS |
32 | | - filtered_register <- filtered_register[, names(filtered_register) %in% CONFIG$REGISTER_COLUMNS] |
33 | | - } |
| 19 | + # Splitting the comma-separated strings into lists |
| 20 | + register$Codechecker <- strsplit(register$Codechecker, ",") |
| 21 | + |
| 22 | + # Unnesting the files |
| 23 | + register <- register %>% tidyr::unnest(Codechecker) |
| 24 | + register$Codechecker <- unlist(register$Codechecker) |
| 25 | + } |
34 | 26 |
|
35 | | - # Else we check against the row value itself |
36 | | - else{ |
37 | | - filtered_register <- register[register[[column_name]]==value, ] |
38 | | - } |
| 27 | + filter_col_name <- CONFIG$FILTER_COLUMN_NAMES[[filter]] |
39 | 28 |
|
40 | | - output_dir <- paste0(get_output_dir(filter, value), "register.csv") |
41 | | - |
42 | | - if (!dir.exists(dirname(output_dir))) { |
43 | | - dir.create(dirname(output_dir), recursive = TRUE, showWarnings = TRUE) |
44 | | - } |
45 | | - |
| 29 | + # Creating groups of csvs |
| 30 | + # Not using the nesting functionality since we want to keep the same columns |
| 31 | + grouped_registers <- register %>% |
| 32 | + group_by(across(all_of(filter_col_name))) |
| 33 | + |
| 34 | + # Split into a list of data frames |
| 35 | + filtered_register_list <- grouped_registers %>% group_split() |
| 36 | + |
| 37 | + # Get the group names (keys) based on the filter names |
| 38 | + register_keys <- grouped_registers %>% group_keys() |
| 39 | + |
| 40 | + # Iterating through each group and generating csv |
| 41 | + for (i in seq_along(filtered_register_list)) { |
| 42 | + # Retrieving the register and its key |
| 43 | + register_key <- register_keys[[filter_col_name]][i] |
| 44 | + filtered_register <- filtered_register_list[[i]] |
| 45 | + table_details <- generate_table_details(register_key, filtered_register, filter) |
| 46 | + filtered_register <- filter_and_drop_register_columns(filtered_register, filter) |
| 47 | + output_dir <- paste0(table_details[["output_dir"]], "register.csv") |
46 | 48 | write.csv(filtered_register, output_dir, row.names=FALSE) |
47 | | - } |
48 | | - } |
49 | | -} |
50 | | - |
51 | | -#' Determines the register table's column name to filter the data by. |
52 | | -#' |
53 | | -#' @param filter The filter name |
54 | | -#' @return The column name to filter by |
55 | | -determine_filter_column_name <- function(filter) { |
56 | | - filter_column_name <- switch(filter, |
57 | | - "venues" = "Type", |
58 | | - "codecheckers" = "Codechecker", |
59 | | - NULL # Default case is set to NULL |
60 | | - ) |
61 | | - if (is.null(filter_column_name)) { |
62 | | - stop(paste("Filter", filter, "is not recognized.")) |
63 | | - } |
64 | | - |
65 | | - return(filter_column_name) |
66 | | -} |
67 | | - |
68 | | -get_unique_values_from_filter <- function(register_table, filter_column_name){ |
69 | | - # Directly retrieve from DIC_ORCID_ID_NAME |
70 | | - if (filter_column_name == "Codechecker"){ |
71 | | - unique_values <- names(CONFIG$DICT_ORCID_ID_NAME) |
72 | | - } |
73 | | - |
74 | | - else{ |
75 | | - unique_values <- unique(register_table[[filter_column_name]]) |
76 | 49 | } |
77 | | - return(unique_values) |
78 | | -} |
79 | | - |
80 | | -#' Gets the output dir depending on the filter name and the value of the filtered column |
81 | | -#' |
82 | | -#' @param filter The filter name |
83 | | -#' @param column_value The value of the column the filter applies to |
84 | | -#' @return The directory to save files to |
85 | | -get_output_dir <- function(filter, column_value) { |
86 | | - if (filter=="none"){ |
87 | | - return(paste0("docs/")) |
88 | | - } |
89 | | - |
90 | | - else if (filter=="venues"){ |
91 | | - venue_category <- determine_venue_category(column_value) |
92 | | - # In case the venue_category itself has no further subgroups we do not need subgroups |
93 | | - if (is.null(venue_category)){ |
94 | | - return(paste0("docs/", filter, "/", gsub(" ", "_", column_value), "/")) |
95 | | - } |
96 | | - |
97 | | - # Removing the venue category to obtain the venue name and replace the brackets |
98 | | - venue_name <- determine_venue_name(column_value, venue_category) |
99 | | - return(paste0("docs/", filter, "/", venue_category, "/", venue_name, "/")) } |
100 | | - |
101 | | - else if (filter=="codecheckers"){ |
102 | | - # The codechecker column is always a list of codecheckers |
103 | | - for (codechecker in column_value){ |
104 | | - return(paste0("docs/", filter, "/", gsub(" ", "_", codechecker), "/")) |
105 | | - } |
106 | | - } |
107 | | - |
108 | | - else{ |
109 | | - return(paste0("docs/", filter, "/", gsub(" ", "_", tolower(column_value)), "/")) |
110 | | - } |
111 | | -} |
112 | | - |
113 | | -#' Determines the venue category based on the venue_name |
114 | | -#' |
115 | | -#' @param venue_name The venue_name obtained from the "Type" column of the register |
116 | | -#' @return The venue category. If the venue does not belong to any category NULL is returned |
117 | | -determine_venue_category <- function(venue_name){ |
118 | | - list_venue_categories <- CONFIG$FILTER_SUBCATEGORIES[["venues"]] |
119 | | - for (category in list_venue_categories){ |
120 | | - if (grepl(category, venue_name, ignore.case=TRUE)) { |
121 | | - return(category) |
122 | | - } |
123 | | - } |
124 | | - warning(paste("Register venue", venue_name, "does not fall into any of the following venue categories:", toString(list_venue_categories))) |
125 | | - return(NULL) |
126 | | -} |
127 | | - |
128 | | -determine_venue_name <- function(unfiltered_venue_name, venue_category){ |
129 | | - if (is.null(venue_category)){ |
130 | | - return(NULL) |
131 | | - } |
132 | | - |
133 | | - venue_name <- trimws(gsub("[()]", "", gsub(venue_category, "", unfiltered_venue_name, ignore.case = TRUE))) |
134 | | - venue_name <- gsub(" ", "_", venue_name) |
135 | | - return(venue_name) |
| 50 | + } |
136 | 51 | } |
0 commit comments