Download tags with counts?

[amit]

Hi! Newbie to Obsidian, only 2 weeks in, already hooked!

I’m wondering if there’s a way to download a csv or json of all tags with their counts. Basically, the tag sidebar, but downloaded.

What I’m trying to do

I have imported notes from a variety of sources and am left w/ a lot of tags that only refer to one file. I’d like to remove all of these since they no longer make sense within the context of a “second brain”.

If I could download them, I could set up an R script to programmatically remove the # from tags w/ a count less than 2 or 3 or whatever.

Things I have tried

• copypasting from the sidebar, no joy
• Downloading plugins like Tag Wrangler, but couldn’t find any that have the functionality I want
• Could a dataview work?

[anon45210282]

Hello.

Would this work for you?

```dataview
TABLE WITHOUT ID (tag + " – " + length(rows.file.link) + "") AS Tags
FROM ""
WHERE file.tags 
FLATTEN file.tags AS tag 
GROUP BY tag
SORT length(rows.file.link) DESC
```

Reference: Creating a tag index table with Dataview? - #11 by mnvwvnm

[holroy]

I’d rather use file.etags in such a query, as your query would split and count every part of a nested tag.

Look at the output of this query:

```dataview 
TABLE file.tags, file.etags
WHERE contains(string(file.etags), "/")
LIMIT 5
```

Notice how the first variants splits all the nested tags.

---

So here is a similar query showing the tags with their count:

```dataview
LIST count
FLATTEN file.etags as tag
GROUP BY tag
FLATTEN length(rows) AS count
WHERE count > 10
```

Here you could change which count of tags you want to see in the last line.

Also note that if you truly want a static count of tags, you could consider calling this query from within a Templater template, and insert the result into a note somewhere.

[anon45210282]

Very nice. Don’t use nested tags, but useful to know. Prefer a table, so for me:

```dataview
// table of tags and quantity //
TABLE Number
FLATTEN file.etags as Tags
GROUP BY Tags
FLATTEN length(rows) AS Number
SORT length(rows.file.link) DESC
```

[amit]

Thanks both! Those solutions both work! I guess from there it’s just a matter of copying into a txt and converting to csv? Or is there a good way to output to csv from a dataview?

Both of your answers look great, Ill accept holroy’s because of the nested tags.

Thanks!