mgaughan/mw-lifecycle-analysis
1
0
Fork 0
Code

update to gerrit metadata extraction regex

Browse Source
This commit is contained in:
Matthew Gaughan 2025-09-16 11:37:46 -07:00
parent bb67fea96b
commit e2413ed955
1 changed files with 7 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
p2/df_consolidation.R
View File

@ -11,10 +11,12 @@ library(dplyr)
gerrit_extracted_df <- gerrit_df |> gerrit_extracted_df <- gerrit_df |>
mutate( mutate(
jsonfixed = gsub("(?<=\\{|\\[|, )'(\\w+?)'(?=\\s*:)", '"\\1"', selected_gerrit_results, perl = TRUE), gerrit_status = str_match(selected_gerrit_results, "'status':\\s*'([^']*)',\\s*'reviewer")[,2],
jsonfixed = gsub(":(\\s*)'(.*?)'(?=[}\\],])", ':\\1"\\2"', jsonfixed, perl = TRUE), owner_email = str_match(selected_gerrit_results, "'owner_email':\\s*'([^']*)'")[,2],
jsonfixed = gsub("(?<=[:\\[,\\{])\\s*'([^']*)'\\s*(?=[,\\}\\]])", '"\\1"', jsonfixed, perl = TRUE), written_url_in_message = str_match(selected_gerrit_results, "'written_url_in_message':\\s*'([^']*)'")[,2],
expandedjsonlist = lapply(jsonfixed, fromJSON) code_insertions = as.integer(str_match(selected_gerrit_results, "'code_insertions':\\s*(\\d+)")[,2]),
code_deletions = as.integer(str_match(selected_gerrit_results, "'code_deletions':\\s*(\\d+)")[,2]),
reviewer_count = as.integer(str_match(selected_gerrit_results, "'reviewer_count':\\s*(\\d+)")[,2])
) )
# TODO select # TODO select