[Question] The recommended way to use library

Hi guys, sorry for raising a bug. I didn't find better place to ask the question.

What's the best recommended way to use your library for reading and writing the data? String flux/measurements/anything else?
My concern is about data consistency. Look at the following code:

        Flux fluxQuery = Flux.from(influxDBProperties.getBucketChart())
                .range(-31L, ChronoUnit.DAYS)
                .filter(Restrictions.and(
                        Restrictions.measurement().equal("sentiment"),
                        Restrictions.tag("asset").equal(assetType.toString()),
                        Restrictions.tag("timeframe").equal(timeframeType.toString())
                ))
                .pivot()
                .withRowKey(new String[]{"_time"})
                .withColumnKey(new String[]{"_field"})
                .withValueColumn("_value")
                .tail(1);

I kinda worry on providing names of tags/measurements like this. I don't like to keep it as strings. Because later it's very difficult to change the code at every possible place in case of breaking change. Any small typo would cause the issue. For example, in case of JPA, there are entities and repositories and I do not worry about that. Here's a bit different case. I do not see any other choice at this moment.

Do you have any recommendation?
@bednar maybe you can take a look?

Beginne damit, das Flux-Abfragebeispiel im Issue und die dokumentierten Lese- und Schreib-APIs der Bibliothek zu prüfen. Die Arbeit ist abgeschlossen, wenn das Projekt eine klare Empfehlung dafür hat, wie sich fragile Zeichenkettennamen vermeiden lassen, während die Datenkonsistenz erhalten bleibt, und die entsprechende Nutzungsempfehlung dokumentiert ist.