bibi/src/merge-wos.jl

### A Pluto.jl notebook ###
# v0.19.27

using FileIO
using CSV
using TidierData
using DataFrames
using RCall
import Pandas.read_csv as pdread_csv

thewd = joinpath(pwd(), "crawled-data", "wos", "out")
# cd(thewd)

csv_files = filter(x -> endswith(x, ".csv"), readdir(thewd, join = true))
combined_df = DataFrame()


for file_path in csv_files
    # 讀取CSV檔案
	# try
    	# df = pdread_csv(file_path) |> DataFrame
	df = CSV.read(file_path, DataFrame)
	# 從檔名中提取Cluster名稱
	cluster_name = split(file_path, ".")[1]  # 假設檔名是"ClusterName.csv"
	# println(cluster_name)
	# 新增Cluster欄位
	df[!, "Cluster"] .= cluster_name
	# 合併DataFrame
	combined_df = @bind_rows(combined_df, df)
	println("$(cluster_name) done")
	
	# catch e
	# 	# println("$(file_path)發生錯誤: $e")
	# 	println("出錯了！！")
	# end
end

select!(combined_df, :Cluster, :)
exportpth = joinpath(dirname(thewd), "binded.csv")
CSV.write(exportpth, combined_df)

## 比對檔案
df = @chain exportpth begin
	CSV.read(DataFrame)
end


srda = @chain pwd() begin
	joinpath("backend-DB/SRDA-BibData.csv")
	CSV.read(DataFrame)
end 

srda[!, :原始題目]

l0 = uppercase.(df[!, :TI])
l1 = uppercase.(srda[!, :原始題目])

bol = l0 .∉ Ref(l1)

xx = df[bol, :]

CSV.write("ts.csv", xx)