This is only one example, I usually don’t use names, because that limit the use of the citation.
I do add sections, post and/or lines information when I need to.
I also have a note for each citation with a fully qualifying Bibliography/Citation to the source, as it is natively written for that “archive” (if it exist).
All information about catalogs, names of series etc. is added to the source and repositories and as Notes.
But mostly I don’t use Gramps at all for source and citation handeling, I use Zotero, so the "names/descriptions here is similar to those in Zotero, where I have total control of all source data, even though some people in this community doesn’t think it’s possible.
To be honest, I’m moving more and more away from Gramps, because of the limitations it set on the use of my data. Just a Legacy and Rootsmagic does. I do research that needs to be analyzed, to be able to find the right person, the right ship, the right house at the correct time and place.
When you i.e. research a person where you find 8-10 different individs with same name and with birth years within a span of a few years, all sailors and you find them on 7-8 different ships in a period of maybe 20-30 years, you need tools to be able to not only see the persons name in a graph, you need to be able to have timelines for all the Individs, all the Events, all the Ships, and all the Ports and Journeys the Ships has sailed on, in addition you might have to register ALL the crews names, to do search on those to see if you can find the person you are actually looking for in relation with those names other places.
If I was to make a full Citation with names for ALL of those subjects/object, I would have many thousand citations for a single research project for one person…
So as for now, and maybe for ever, or at least until Gramps 5.2 and hopefully support for Main Events - Sub-Events and a interchangeable format for sources and citations, I have stopped using Gramps for my data. it’s actually a lot easier to make discoveries of new relations and events in a simple tool like Obsidian, and with the plugin JUGGL I can create a cytoscape json file of my whole library of notes and documents, open it in Cytoscape and do more analyzing in that software, without having to write everything 2-3-4 times.
And as a bonmus I can use the markdown add-on for Zotero and add all my sources, any Notes I have and any anotation in PDF’s that I have made, to the same research, so I have all my sources in the same network grafp as I have all my research logs (one pr. object i.e. one for each person, one for each event, one for each place), all linked in markdown files with easy wiki-link, it doesn’t mater if the information is structured or unstructured, I can i a short days work link hundreds of objects save it to a json file and open it in Cytoscape if the JUGGL plugin is not enough.
I can easily add Citations an Bibliography footnotes or endnotes that is readable and identical in at least 8 different software I use including both MS Ofiice Word and Open Office Writer…
And even though there are limitations in Zotero, I have never had any real problems when it comes to the style of either citations nor bibliography.
I can even relatively easy create individual text files from an excel table and create a text file formatted as I need it to be for each row in a table, and use different cells as metadata in YAML, combine text in a body, make wiki-links of the information I want to use as linked data, etc. etc.
AND you actually see the relations and connections between multiple individs over multiple different Events or Places… and you can see where those different objects are connected through documents (sources), or through a place, vessel, a sub-event etc. etc.
Search and replace is easy… and as I say, I can use multiple tools to collect data and relatively easy also reconcile data from those tools, even if it is unstructured data.