We are using the default Unity Localization package in our project with the Dialogue System asset.
It's also worth noting that we're using the Google Sheets extension for String Tables in order to export from unity to google sheets.
However, we are running into a few issues when it comes to export our Databases into a String Table using the "DS To Loc" window, both "bugs" and usability problems, so I wanted to reach out and ask what can we do better.
Before I begin, here's how we're using it:
1. Synchronized Actors are being exported for each Database
All Databases shown in the image above after the "Main Database", are using the "Sync From" option in the Actors tab to sync from the Main Database. The following image is from "Act1 Datbase"
When we export all those databases to a String Table, the actors are exported 6 times (one for each database).
Suppose we have 3 actors in the Main Database, like
Actor1
Actor2
Actor3
What we end up in the String Table looks like:
Actor1
Actor2
Actor3
Actor1
Actor2
Actor3
Actor1
Actor2
Actor3
Actor1
Actor2
Actor3
Actor1
Actor2
Actor3
Actor1
Actor2
Actor3
All with different Guid/keys.
After that, the conversations appear.
2. Too Many Blanks
Every node that includes only Sequence commands or Scripts appear as blanks.
Feels like over half of the entire String Table are empty fields.
In the following example, the highlighted node has only a Sequence command and a Script, no text, but it shows up at the export anyway.
As you can see, there's even more empty cells below from other dialogues.
Given that we can't see where a conversation ends in the String Table, the blanks make it a bit more difficult to follow conversations when we are editing the text from a google spreadsheet, and speaking of which...
3. Conversation Names
Below is an image a Google Sheet created by the String Table extension. That is what we intend to send to translators later on.
There's the Guid being the Key, the blanks previously mentioned and messages from 2 different conversations (the highlighted line is the last Node from the previous example in the Blanks section).
Suppose a translator finds something weird that we have to double check inside Unity, if we knew the name of the conversation that would be easy, but since we only have the text and Guid, I have no idea how to find the conversation where the line comes from.
Is it possible to change the Unique Field Title in the DS To Loc to the conversation name plus some id? Or some other way to make a spreadsheet easier to follow?
4. Old Conversations
Ok, I admit that this one is more of an production issue, but it doesn't hurt to ask.
If we delete a conversation from the Database, the String Table will still have all the translations for the deleted nodes in there, is there something we can do besides manually removing those lines from the String Table?
We are having to produce many Demos before we finish the dialogue, so we ended up cluttering the String Table with lines that were later removed from the game but are still there.
For now those are the most important issues that we have come across and we wanted some guidance about how can we better use the asset.
Thank you in advance.
