update

src/util.jl

@@ -1,9 +1,10 @@
 module util
   
 export timedifference, showstracktrace, findHighestIndexKey, uuid4snakecase, replaceDictKeys,
-        findMatchingDictKey, textToDict, randstring, randstrings, timeout
+        findMatchingDictKey, textToDict, randstring, randstrings, timeout,
+        dataframeToCSV, dfToVectorDict, disintegrate_vectorDict
 
-using JSON3, DataStructures, Distributions, Random, Dates, UUIDs, MQTTClient
+using JSON3, DataStructures, Distributions, Random, Dates, UUIDs, MQTTClient, DataFrames
 
 # ---------------------------------------------- 100 --------------------------------------------- #
 
@@ -412,6 +413,131 @@ function timeout(f::Function, timeoutwindow::Integer; fargs=nothing, timeoutmsg=
     timeoutmsg
   end
 end
+
+
+
+""" Convert a dataframe into CSV.
+
+# Arguments
+  - `df::DataFrame`
+    A connection object to Postgres database
+    
+# Return
+  - `result::String`
+
+# Example
+```jldoctest
+julia> using DataFrames, GeneralUtils
+julia> df = DataFrame(A=1:3, B=5:7, fixed=1)
+julia> result = GeneralUtils.dataframeToCSV(df)
+```
+
+# Signature
+"""
+function dataframeToCSV(df::DataFrame)
+  # Create an IOBuffer to capture the output
+  io = IOBuffer()
+  CSV.write(io, df)
+  dfStr = String(take!(io))
+  return dfStr
+end
+
+""" Convert a DataFrame into a list of Dict rows.
+
+#  Arguments
+  - `df::DataFrame` 
+    The input DataFrame to be converted.
+
+#  Return
+  - `rows::Vector{Dict{String, Any}}` 
+    A vector of dictionaries, where each dictionary represents a row in a dataframe.
+
+#  Example
+  ```jldoctest
+  julia> using DataFrames, JSON3, GeneralUtils
+  julia> df = DataFrame(A = [1, 2, 3], B = ["apple", "banana", "cherry"])
+  julia> vectorDict = GeneralUtils.dfToVectorDict(df)
+  [Dict{String, Any}("B" => "apple", "A" => 1), 
+  Dict{String, Any}("B" => "banana", "A" => 2)
+  Dict{String, Any}("B" => "cherry", "A" => 3)]
+  ```
+
+#  Signature
+"""
+function dfToVectorDict(df::DataFrame)
+  vec = []
+  for row in eachrow(df)
+      d = Dict{String, Any}()
+      for col in names(df)
+          d[col] = row[col]
+      end
+      push!(vec, d)
+  end
+  return vec
+end
+
+
+
+""" Turn a large vector of dictionaries into smaller one 
+
+# Arguments
+  - `data`
+      data to be partioning
+  - `partsize`
+    how many dicts per part
+
+# Return
+  - `parts`
+    a dictionay of parts
+
+# Example
+  ```jldoctest
+  julia> using GeneralUtils, Dates, JSON3, UUIDs
+  julia> vecDict = [Dict("a" => i) for i in 1:10]
+  julia> d = GeneralUtils.disintegrate_vectorDict(vecDict, 3)
+  julia> println(d[:data])
+  Dict{Int64, Vector{Dict}} with 4 entries:
+  1 => [Dict("a"=>1), Dict("a"=>2), Dict("a"=>3)]
+  2 => [Dict("a"=>4), Dict("a"=>5), Dict("a"=>6)]
+  3 => [Dict("a"=>7), Dict("a"=>8), Dict("a"=>9)]
+  4 => [Dict("a"=>10)]
+  ```
+
+# Signature
+"""
+function disintegrate_vectorDict(data::Vector{Dict{T1, T2}}, partsize::Integer
+  ) where {T1<:Any, T2<:Any}
+  parts = Dict{Int, Vector{Dict}}()
+  for (i, dict) in enumerate(data)
+      partkey = (i - 1) ÷ partsize + 1
+      if !haskey(parts, partkey)
+        parts[partkey] = Vector{Dict}()
+      end
+      push!(parts[partkey], dict)
+  end
+  return (datatype="vector{Dict}", totalparts=length(parts), partsize=partsize, data=parts)
+end
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+