Apologies if this is covered by another post, but my initial research on importing CSV files didn’t make anything super obvious to me.
I have a CSV file with over 2000 lines, which I would like to import as separate markdown notes into a new Database or Group, with one of the fields used as a markdown text, another used to create the file name, another with comma-separated tags which I would like to be converted into DEVONthink tags. There might be a couple of other fields to convert into other metadata entries in DT. Does anyone have a starter script or suggested workflow to help me get this done?
‘content’ would be the content of the file, ‘title’ could become the title of the markdown file, although I may look at trying to wrangle the ‘reference_start’ and ‘reference_end’ entries to create the title, but that’s my problem… the main thing would be also to make sure any comma-separated values in the ‘tag’ column become DT tags. Maintaining the date_created field as a DT Date Created metadata would be useful too.
Why? Apart from the fact that the product you linked to is not free – the sample they show is just converting CSV data into an MD table (something a one-liner could do with awk or sed, for free, btw).
I guess if the OP wanted to have a table, they could import their CSV into Excel. And they said that they have “over 2000 lines, which I would like to import as separate markdown notes”. That might mean “one line – one note”.
All assumptions, of course. If it were me, I’d script it. @BLUEFROG might come up with something that does not require coding.
I use a custom applescript to process .csv files; generating separate Devonthink notes
My use-case is .csv transaction files downloaded from the bank
The script “wrangles” the .csv data; formats the note title, content, and assigns tags
Click for sample script
tell application id "DNtp"
set theFilingGroup to get record with uuid "878AADEA-A15F-4DBD-B9B9-BF59099F5554"
set selctedFiles to selected records
repeat with theBankcsvFile in selctedFiles -------------------------------Process each .csv file; extract the contents
set theBankcsvName to name of the theBankcsvFile
set theBankcsvPath to path of theBankcsvFile as string
set theBankcsvText to paragraphs of (read theBankcsvPath as «class utf8»)
my process_Rows(theBankcsvText)
set theNote to move record theBankcsvFile to theFilingGroup
set name of theNote to "StatementCSV [" & theBankcsvName & "] " & theDate & " zcType-StatementCSV " & theVendor
set tags of theNote to tags of theNote & {"zcType-StatementCSV", theVendor}
end repeat
end tell
on process_Rows(theBankcsvText)
repeat with theBankcsvRow in theBankcsvText --------------------------- Process each row
set theNoteTitle to my Process_Columns(theBankcsvRow) -------------- Parse the row content by columns
set theNoteTags to Process_Tags(theNoteTitle)
set theoffset to offset of "]" in theNoteTitle
set theDate to date (text (theoffset + 7) thru (theoffset + 11) of theNoteTitle & "-" & text (theoffset + 2) thru (theoffset + 5) of theNoteTitle)
tell application id "DNtp" ----------------------------------------------- Create Note
set theNewRecord to create record with {name:theNoteTitle, type:txt, content:theBankcsvRow, tags:theNoteTags} in theFilingGroup
set creation date of theNewRecord to theDate
end tell
end repeat
end process_Rows
on Process_Columns(theBankcsvRow) ---------------------------------- Process Clolumns
set theBankcsvColumns to ParceCSV(theBankcsvRow, ",")
...
set noteTitle to "Banking [" & detailContent & "] " & dateContent & " $" & amountContent
return noteTitle
end Process_Columns
Because creating an awk or sed or a custom AppleScript is for some folks also not “free” or as unambiguously obvious or as immediately error-free as it may be for you to do.
