I'm trying to make a 'Priority' graph that will read my tags.
A tag of PR1 will fill 20% of the graph. PR2 = 40% etc. Can this be done with an evaluator column?
This pic is when this Priority column I'm making is reading the files' Ratings data. But I want to use tag data instead, or if there is a different metadata field that would work?
Maybe I could use labels.
A filter label would match the tag PR1 and make a PR1 label.
Then that should trigger a fill of the graph, from 0 - 100, from what I've gathered.
This is my best effort. I tried to learn it and tried a bunch of different syntax.
Of course it won't work. How do you people do this all day? I'm ready for violence at this point.
Hold on to your Zulu sword, it isn't all that hard to do.
Try this to start with.
<?xml version="1.0"?>
<evalcolumn align="0" attrrefresh="no" autorefresh="yes" customgrouping="no" foldertype="all" header="Priority" keyword="TJPriority" maxstars="5" namerefresh="no" reversesort="yes" title="TJ Priority" type="9">x = keywords;
If(keywords=="PR1")
{
return(20);
}
elseif(keywords=="PR2")
{
return(40);
}
elseif(keywords=="PR3")
{
return(60);
}
elseif(keywords=="PR4")
{
return(80);
}
elseif(keywords=="PR5")
{
return(100);
}
else
{
return(0);
}
</evalcolumn>
Edit Note:
Edit 2: If files have multiple tags, you'll need to modify this. It is probably easiest to keep the PR#s as the first tag. Then you can use StrCmp() to isolate the PR# tag for testing.
Short Claymore. 
It's only filling the PR1 20% fill for me, regardless of tag. I can't figure out why.
I forgot to mention if no PR tag is found then have an empty bar.
Only one tag will be problematic. I could possibly assign the PR tag to the first tag slot always, if I learn how to dictate what order the tags are added and removed. Good thing is files can only have one priority at a time. So your evaluator will only have to look at the first tag.
If it's easier for us I don't mind going back to using the Ratings attribute instead. But I will have to fix that evaluator bug somehow. I would want to use those nice stars and use the Set Ratings button.
The first line was missing, sorry.
I see you did the else if = 0, for the no tag empty fill.
Changed my code to be exact to your example pic and now the bars are there but all blank.
Tried to test the tags directly but that doesn't work.
OK, try this instead. This will allow multiple tags, but looks for PR# as being the first tag.
I commented out the return(0) line. If you want the empty tag graph, just uncomment it.
<?xml version="1.0"?>
<evalcolumn align="0" attrrefresh="no" autorefresh="yes" customgrouping="no" foldertype="all" header="Priority" keyword="TJPriority" maxstars="5" namerefresh="no" reversesort="yes" title="TJ Priority" type="9">x = keywords;
If(StrCmp(x,"PR1",3)==0)
{
return(20);
}
elseif(StrCmp(x,"PR2",3)==0)
{
return(40);
}
elseif(StrCmp(x,"PR3",3)==0)
{
return(60);
}
elseif(StrCmp(x,"PR4",3)==0)
{
return(80);
}
elseif(StrCmp(x,"PR5",3)==0)
{
return(100);
}
else
{
//return(0);
}
</evalcolumn>
Decent.
That's very good, thank you.
I changed the tag from PR to A, so that I can sort tags alphabetically and so I don't need to worry about what order I add the tags when using buttons to change the priority of a file. I might use one of the script tools on this forum as well to make things easier.
Thanks so much.
Here's a fuzzy metadata kitty for you. 
cute kitty
You could replace If(StrCmp(x,"PR1",3)==0) with If(InStr(x,"PR1")>= 0), but it is problematic.
InStr() returns the position "PR1" is found in the string. The first position is 0.
But if "PR1" is part of any other tags, you'll get a false positive result.
But if that is understood, it does save you the trouble of sorting tags.
Oh. That's good too.
Think that I'm fine with A# as my tag. A for ACTION TIME!
Having the tag system sort them alphabetically doesn't hurt either. It's easier to find them.
I wanted a subgroup tag system so that if I'm looking for a certain tag it's in the same spot every time, but that's not a big deal.
I made my change priority menu.
Using a replacement of the A# tag.
@nodeselect
SetAttr META tags:-(tags seperated by a semicolon)
SetAttr META tags:+(tag)
So now changing the tag, the domino effect on the lister updates the graph, and max priority things get a status icon and a red row background. This is now independent of the Ratings, which I followed your advice and setting the rating by directly clicking in the column is nice.
I can run searches anywhere, looking for priority, by using the tags.
Also made hide/show and pinned buttons for selections. One of the buttons I made is pin the max priorities to the top and hide everything else. You know, like an adult. 
I might do a separate post for that.
After thinking about it some, I changed mine to this.
It uses #[0-5]# as the priority tag.
That should ensure that no false positives are matched.
I tinkered a while with evaluator stings and regular expressions attempting to prove the tag match, but in the end, this was a good solution.
To use #[0-5]# as a tag Finding all priority tags, you will need to escape # with an apostrophe as # is also wildcard syntax.
Find HERE CLEAR RECURSE FILTERDEF tags match '#[0-5]'#
And then the Evaluator Column is this:
<?xml version="1.0"?>
<evalcolumn align="0" attrrefresh="no" autorefresh="no" customgrouping="no" foldertype="all" graphcol="#ffff00" header="Priority" keyword="Priority" maxstars="5" namerefresh="no" reversesort="no" title="Priority Graph" type="9">//Priority Tags have the form of "#[0-5]#
x = keywords;
If(InStr(x,"#1#")>= 0)
{
return(20);
}
elseif(InStr(x,"#2#")>= 0)
{
return(40);
}
elseif(InStr(x,"#3#")>= 0)
{
return(60);
}
elseif(InStr(x,"#4#")>= 0)
{
return(80);
}
elseif(InStr(x,"#5#")>= 0)
{
return(100);
}
elseif(InStr(x,"#0#")>= 0)
{
return(0);
}
else
{
//return(0);
}</evalcolumn>
The number is where the tags are in that file's list of tags?
Tag index? 0 - 5?
Find HERE CLEAR RECURSE FILTERDEF tags match '#[0-5]'#
This finds all files/folders with a tag that matches #0# or #1# or, #2#, or #3#, or #4#, or #5#.
This finds all priorities instead just one priority.
You can change it to just '#3'#" for instance to find only one of them if you want.
Adding nowild eliminates the need for the escape apostrophe, but then [0-5] won't work to find them all.
This works:
Find HERE CLEAR RECURSE FILTERDEF tags match A1 or tags match A2 or tags match A3 or tags match A4 or tags match A5How do I use the source tab name as the search results tab name? {path} doesn't work.
...
I found something. That looks like a recurse thing to me, but haven't tried it.
foldercontent match (
tags match BUS
)
Also, is there a command that tells the Find tool not to make the collection folder permanent?
As in: "When you close the tab, delete the collection". ?
Yeah, COLLNAME={sourcepath} doesn't work.
Maybe we could use Evaluator code, but I'm not familiar enough with this tonight.
I'm thinking of trying to inject Evaluator code lister_title to fix this.
But I'm not quite sure how to do it at this time nor do I know if it will work.
I'll sleep on it, but I welcome other input on this.
Edit: I tried COLLNAME {=lister_title=} but that doesn't work either.
Later I discovered that I can just call a Find preset. And it will do all this work for me.
Find PRESET="DM Today"Sooooo I basically went full circle and ended up back where Leo told me to use Find instead of Select. 
But I still don't like it. If the file collections just keeps piling on new search folders every time I search for something ... like what the hell is that. Who would want a permanent folder made cause they did a search? It should be temp by default and you have to save the search if you want to keep it.
Another thing that really sucks.
Is my add metadata button; for filetypes that don't support an Author it parse errors out and cancels the operation. So I can only put the info on the comment I guess. ¯(°_o)/¯ The only text field that is common to all file types. I don't know how companies manage copyright protections etc. on their databases. I'm going to have to look into it.
You can use the same results collection each time so it overwrites the old one. Instead of making a new one. Or clean up the old ones after a while. It’s up to you.
If you can’t imagine how being able to keep multiple find results is useful, that’s a problem with you rather than the design. 
You do this a lot where you disparage the notion of anything you don’t personally need or understand.
What should it do instead? Writing an author metadata field to a plain text file isn’t possible. What do you want the program to do when asked to do an impossible thing?
You could deal with that in various ways using scripting, depending on exactly how you want to handle the impossible operation (e.g. only set the Author field on specific types that can have such a field and skip the othet files entirely).
No it's not.
The design forces a particular workflow and that's what I'm disparaging.
If you search for something it must become a file collection. Why? You are calling it better but it seems to me I always have like thirty file collections hanging around that I'll never use more than once that I have to clean up. Which is more work than not cleaning up. Don't get me wrong, file collections are great but saved searches aren't always.
You told on yourself when you said they get overwritten. So these searches are not as important as you described, now are they.
I suggest an option for the Find system; to make a search either a temp or file collection. And what subfolder to put the new collection. So the user can define the type of search by how they are using it. When I'm using the find tool only for the pattern matching to filter things (shoring up this void on the filter bar), then I don't need a file collection. Many times people search for things to look at them. They are looking for a particular comparison data and that is only for the moment.
If we're being honest it should have been dealt with by now. Computer devs that have solved some of the most difficult logic puzzles in the world can make it possible.
But I've learned that 'Author' and 'Artist' are considered to be the same thing. Which solves that particular issue for me, but you've caught me before I was able to come back and declare it.
Probably because you're talking to a Dank Meme Lord wondering why I'm abrasive, but disparaging is the name of the game sir. It's a good time. Just another day in the office.
In the midst of giving you grey hair I've been constructing a More Than Slightly Extremely Sweet Filter Array. I should trademark that name. It's almost done. The normies are gonna wanna get their hands on this bad boy.
If you want to have an antagonistic relationship with the people you’re asking to do work and change things for you, and who admin the forum you’re posting on, it’s not going to get you any positive results.
