Dear preCICE team，I have found that there is very little information available about the Fluent adapter, and I am very interested in learning more about it. I have read the Fluent code on GitHub, but still feel like I only have a partial understanding. For example, I know that initialization is required before running the simulation, and that this can be done either by selecting “Read → scheme” or by using fsi_init(Domain* domain) in FSI.C. Could you explain the differences between these two initialization methods? Also, are there any learning resources available for the Fluent adapter, such as documentation or videos? I have not been able to find much information on this topic, and would greatly appreciate any help. Thank you
Hi @LJR,
the reason you cannot find much information about the Fluent adapter is because we don’t have much about it. It is generally unmaintained at the moment, or in the best case considered experimental. It is tricky for us to work on adapters for proprietary codes, especially codes that we don’t ourselves use.
The last people still involved that had some touch to the Fluent adapter were @IshaanDesai and @BenjaminRodenberg.
In case you want to develop this adapter further, your contribution would be very welcome and helpful. 
General documentation on writing adapters is available on Couple your code | preCICE - The Coupling Library
Okay, Makis，thank you very much for your answer.
By the way, I moved the Fluent and COMSOL adapters to the “Third party” list, as it was previously a bit misleading on the state of things overall.
1 Like
Hi @LJR
The adapter is indeed not fully maintained right now. A good starting point is the branch: GitHub - precice/fluent-adapter at fluent_21_2_0. I would recommend reading the README on that branch, it has a lot of useful information.
okay，Thank you，IshaanDesai，I will read carfully
Hi,IshaanDesai,I have some doubts about the “0.2 Build requirements” section in the README. I am not sure if I understand it correctly：My understanding is that if preCICE functions are used, then the UDF needs to be compiled outside of the Fluent GUI, and a specific directory structure that conforms to the GUI compilation needs to be manually created. I would like to know, if not compiled through the GUI, how are the Makefile, user.udf, and udf_names.c files generated?
Hi @LJR
As far as I remember, the files Makefile, user.udf and udf_names.c are generated manually and are part of the adapter. So these files exist from before. You just need to have the specific folder structure and be able to compile the UDFs. Please keep in mind that the folder structure given in the adapter repository is valid for Fluent 21.2.0. For a newer Fluent version, you might have to make some changes based on the new documentation.