Fuzzy, self-updating list filter workflow template (working in 12.3)

[dfay]

# fuzzylist
Fuzzy, self-updating list filter workflow for Alfred 3 & 4

This is a workflow template - it does nothing as is.

 

## Usage:
- create a csv file like you would for an Alfred List Filter, with an optional fourth field containing the path to an icon file for that result.
- name the file *list.csv* and add it to the workflow directory

On the initial run, the workflow will create a file list.json for output to the fuzzy search.  If list.csv is modified, it will update list.json .  

 

## Credits
- uses fuzzy.py by @deanishe - https://github.com/deanishe/alfred-fuzzy

 

## Download

 

workflow at https://github.com/derickfay/fuzzylist/blob/master/Fuzzy List Filter.alfredworkflow (v. 0.3 as of 2022-03-30)

 

## Working on 12.3

 

v. 0.3 now uses Python 3 and works on MacOS 12.3

Edited March 31, 2022 by dfay

[evanfuchs]

This has been really helpful for me with a workflow I've been running several times a day for a couple years. Thank you!

 

I duplicated the entire workflow for an another application, but how I would go about having multiple lists within the single workflow, each with their own their own trigger? I understand list.json is generated from list.csv, so I imagine the solution would be creating another json from a different csv, but I dont know anything about python.

 

Any help would be appreciated

[evanfuchs]

19 hours ago, evanfuchs said:

how I would go about having multiple lists within the single workflow, each with their own their own trigger?

 

I know this can be done by iterating through multiple csv files in the original script, but I don't know how to do that. Trying to help myself and anyone who might stumble across this, this works:

 

• Add another csv file (list2.csv) to the workflow directory
• Duplicate fuzzylist.py (fuzzylist2.py), and change the following line:
• Change theFile="list.csv" to theFile="list2.csv"
• Duplicate the Script Filter Object and change the reference to fuzzylist.py to fuzzylist2.py in its script.

[deanishe]

4 hours ago, evanfuchs said:

• Duplicate fuzzylist.py (fuzzylist2.py), and change the following line:
• Change theFile="list.csv" to theFile="list2.csv"
• Duplicate the Script Filter Object and change the reference to fuzzylist.py to fuzzylist2.py in its script.

 

A more sensible approach would be to change fuzzylist.py to accept a filename as an argument and use the same script for every Script Filter.

Basically, theFile = sys.argv[1]

Edited December 3, 2020 by deanishe

[evanfuchs]

48 minutes ago, deanishe said:

A more sensible approach would be to change fuzzylist.py to accept a filename as an argument and use the same script for every Script Filter.

Basically, theFile = sys.argv[1]

 

That makes sense. When I change theFile = "list.csv" to theFile = sys.argv[1]  I get an error NameError: name 'sys' is not defined whenever I try to provide any argument.

 

As I said, I don't know what I am doing but I am trying

 

 

 

[dfay]

You just need to move that line down to below import sys - that error is occurring because sys hasn't been loaded yet.

 

Everything else stays the same.  (Sorry about my sloppy python!)

 

I tried to post the code here but got a server error so it's at https://github.com/derickfay/fuzzylist/blob/master/fuzzylist-mf.py

 

 

Edited December 4, 2020 by dfay

[evanfuchs]

34 minutes ago, dfay said:

You just need to move that line down to below import sys 

 

Thanks @dfay - I did try that, and I get another error. I cant post this as code so here is a screenshot of the debugger.

 

 

I am running the the script filter with a space followed by the file name like this: Fuzzy List Filter list.csv

 

Is that correct?

 

 

[dfay]

Here's what your log should show:

 

 

It looks like you're passing the file name as the query - is that what you're trying to do?  I had thought you were adding separate files but not using the workflow to select among them.  

Here's what my script filters look like:

 

I've added several with unique keywords for each and different csv files in the command above.

 

Edited December 4, 2020 by dfay

[evanfuchs]

5 hours ago, dfay said:

Here's what my script filters look like

 

That's perfect. Thank you. I didn't know that's how you pass the argument 😬

 

I had also tried changing the script filter to "with input as query" and passing the file name in the Alfred window, but I got an error either way. This was super helpful.

 

Thanks

[Chris Messina]

@dfay I don't suppose there's a way to specify the icon to use for items imported from the CSV?

[dfay]

Yeah I added that per your request - took way longer than expected because of issues building the json dict but it's done.  The download link is the same.

[jhinden]

Hi there, is it possible to get the search working for Title and Subtile etc.? Not much of a programmer, but tried editing the queries with no luck...
Thanks.  

[deanishe]

7 hours ago, jhinden said:

Hi there, is it possible to get the search working for Title and Subtile etc.?

 

Yes, but you would need to edit the script so it generates different JSON. Specifically, adding a match field to each JSON item that contains both its title and subtitle should get you filtering across both fields.

 

[jhinden]

I see, thanks. Had a look at the files and docs on "match" and think above my pay grade. Will try two lists as described above and just invert the the title/subtile and manage two cvs. 

[dfay]

You don't need two lists.  Just create one with your desired searchable text duplicated in the Title and the Subtitle fields.  e.g.

 

myTitle,mySubtitle,myArg

mySubtitle,myTitle,myArg

[jhinden]

Ah, yes! Much simpler. Thank you.  

[yellowsoar]

I create a workflow for web search based on your work. @dfay

https://github.com/yellowsoar/alfred_web_search_with_fuzzy_list

[dfay]

Here's a version of fuzzylist.py that works in Python 3:

#!/usr/bin/env python3
# -*- coding: utf-8 -*-

import csv
import sys
import json
import os

theFile=sys.argv[1]
fieldnames=["title","subtitle","arg","iconfile"]

json_filename = theFile.split(".")[0]+".json"


def convert(csv_filename, json_filename, fieldnames):
	f=open(csv_filename, 'r')
	csv_reader = csv.DictReader(f,fieldnames)

	jsonf = open(json_filename,'w') 
	jsonf.write('{"items":[')

	data=""

	for r in csv_reader:
		r['uid']=r['arg']
		r['icon']={"path":r['iconfile']}
		data = data+json.dumps(r)+",\n"
	
	jsonf.write(data[:-2]) 

	jsonf.write(']}')
	f.close()
	jsonf.close()

if (not os.path.isfile(json_filename)) or (os.path.getmtime(theFile) > os.path.getmtime(json_filename)) :
	convert(theFile, json_filename, fieldnames)

with open(json_filename, 'r') as fin:
	print(fin.read(), end="")

and a link to an updated version of fuzzy.py:

https://github.com/deanishe/alfred-fuzzy/issues/3