Custom Collections

[tspear]

There are a few tantalizing mentions of custom collections; but I cannot find in the documentation on how to create them.
I currently use a series of smart collections in Lightroom based on the following article: http://www.beardsworth.co.uk/lightroom/workflow-smart-collections/

I am trying to replicate the same style of collections in iMatch.
Any suggestions?

Tim

[Mario]

IMatch manages a number of automatic collections for you, like "Recently Added/Updated/Viewed", Rating, Label, Pending Write-back, Protected Files, Relations, Labels, Annotations, ...Switch to the Collection View and check for yourself.
In addition to these automatically managed you have collections like bookmark, pen, flag and dot which you can use freely in the File Window.
See the help topic on Collections in the IMatch help for full details,

Looking at the page you have linked (I don't have the time to read all that text, sorry) it looks like this guy has setup a rather extensive set of collections in Lr, which group files by all kinds of stuff, from metadata to proprietary Lr editing data. Not sure how much of this stuff is really needed.

In IMatch you can do the same using data-driven categories or filters (Filter Panel), or a combination of both - depending on your workflow and how you will need to access this...and if your database has 30,000 or 500,000 files.

The smart collection this guy has built which refer to proprietary Lr settings and editing attributes may not be doable outside Lr.
If Adobe stores the data used for that only in the Lr database, you cannot access this from anywhere else. But if Adobe stores this in the XMP record they put into your image files, you can view and access this information in IMatch. You just have to figure out where it is stored, how Adobe has named the XMP field etc. This is all intentionally not documented by Adobe. Pick Flags are, for example, stored only in the Lr database. Not in the XMP.

Tip: Select one of your files in the File Window and then switch the Metadata Panel into the Browser mode. IMatch then shows you all the data it has imported for your files.

Note: IMatch by default does import frequently used metadata. It does not attempt to import all metadata in your files. In many cases, maker notes written by camera vendors (several hundred tags per file) is not useful for humans and would only blow up the database. The same is true for much of the proprietary metadata Adobe writes into files. Depending on your settings, Lr may record every brush stroke (!) in the XMP record, creating thousands of metadata tags per file. I have seen JPEG files with 3000 proprietary Adobe tags, adding several megabytes of metadata...

You can use the Edit >  Preferences > Metadata 2 : Tag Manager in IMatch to selectively enable or disable import for non-standard metadata. See the IMatch help for details.

That said,...

Things like "files without caption" is a simple value filter. I'm not sure if he means the XMP title or the XMP headline with 'Caption'. Adobe has the bad habit of renaming official XMP tags to whatever they want in their application. Let's assume you want to see all files without title in the current scope (file window).

A simple value filter with these settings will do this for you:



You can save this filter under the name "Files without title" and then recall it later with a single click.

To group files by ISO, lens and other data you would use data-driven categories. See the IMatch Standard Categories in the Category View, which already do that for frequently needed attributes like ISO, lens, camera model or location.

For data-driven categories which classify files by more complex criteria, you would use variables.
For example a data-driven category which classifies your files into files with and without a title (caption), I would create a data-driven category as follows:



The trick is the variable

{File.MD.XMP::dc\title\Title\0|hasvalue:Files with title}

it checks if the XMP tag title has a value. In that case it returns the text Files with title. Else it returns an empty value. Files without a value thus end in the catch-all "Other" category, which I renamed to "Files without title". In the Category View this then looks like this:



In the same way you would create categories for "no copyright" or whatever.

Files with a non-empty copyright: {File.MD.XMP::dc\rights\Rights\0|hasvalue:Yes}
Lr Crop: {File.MD.XMP::crs\HasCrop\HasCrop\0|is:1,Cropped,Not cropped;default:Not cropped}


Give this a try and spend an hour learning about IMatch data-driven categories and what they can do.

Tip: If you are concerned about having files without a copyright or title, create a Metadata Template which fills in this information, and then configure IMatch to automatically apply this template when it imports new files (Edit > Preferences > Indexing). This way you can ensure the data is always filles.

[Jingo]

Great answer and explanation Mario... WE really appreciate you taking the time to respond is such detail as it helps us create our own data driven items using your examples.. of course the help if very detailed as well - but sometimes seeing specific examples in the forum helps immensely! 

[sinus]

Great answer and explanation Mario... WE really appreciate you taking the time to respond is such detail as it helps us create our own data driven items using your examples.. of course the help if very detailed as well - but sometimes seeing specific examples in the forum helps immensely!

+1 Exactly! 

[tspear]

Mario,

Thanks. I got it sort of working.
I need to read up on how your fields work, and tags.
FYI, besides Adobe, a bunch of other companies are calling things "Smart Collections", "Dynamic Collections", Intelligent Collections"....
So you may want to figure a way as part of the collections overview send the user to categories instead.

Tim

[Mario]

I have added a few more keywords and a paragraph linking to data-driven categories to the collection topic.
It's not always easy to figure out in advance how people will look for things in the help, or which search keywords they will use.