TESTING TI NEW PROCESS TO EXTRACT ALL PROCESSES

[R4958]

Hello everyone,

I want to create a mapping of the processes we have and I want the mapping to include data source type, data source name and the chore involved (if applicable). I do not see an easy way to do this in TM1 so I am going to try creating a new process and posting the below in the prolog. The expectation is that when I run this process, it will send the data to the data directory and will have the following columns:

Process Name Data Source Type Data Source Name Chores Involved

I am hesitant to do this because my company does not have a dev environment so anything I do will directly impact production. Therefore, I am hoping I can get some TM1 experts to review the prolog below and let me know if I am on the right path. If anyone has a dev environment and feels comfortable testing and letting me know if it works, that would also be greatly appreciated!

Update...I removed the process I created after another user laughed and told me how simple it was. Therefore, maybe I should just say that I want to create a process to extract the data. TM1 has so much flexibility with processes that I know there is a way to do it. Maybe I am being too ambitious, too soon so I will keep studying on it and see if I can come up with something. Has anyone else written such a process or able to provide any insight on how I can get this done in TM1. Thank you!

[ascheevel]

Check out TM1py for generating a report of this sort of TI process metadata.

[Wim Gielis]

In summary, if I create this new process with only detail in the prolog, should this work as expected? Thank you!

Checking my calendar, it's not yet April first
The code is literally full of functions that even do not exist in TM1 and you expect the code to be doing something ?
Did you write this in Notepad or also paste it in a TI process and hit the Save button ?
Or was it AI-generated ?
Even the simplest End... While is wrong.

Can you create topic titles in normal casing, not upper case ? Thanks.
Can you also give feedback in earlier topics where you request assistance, like the one on DBSEND.

[R4958]

Check out TM1py for generating a report of this sort of TI process metadata.

Hello and thank you for the helpful response. If it is not obvious, I am very new in TM1 and trying to learn with limited resources. I would never make any material changes at my job but I use this forum to get feedback on my own learning projects. Someone wanted to me create a process list and I thought I could use this as an opportunity to learn about processes, instead of manually doing it. Studying on my own so when people provide helpful resources, I really appreciate it. Thank you!

[R4958]

In summary, if I create this new process with only detail in the prolog, should this work as expected? Thank you!

Checking my calendar, it's not yet April first
The code is literally full of functions that even do not exist in TM1 and you expect the code to be doing something ?
Did you write this in Notepad or also paste it in a TI process and hit the Save button ?
Or was it AI-generated ?
Even the simplest End... While is wrong.

Can you create topic titles in normal casing, not upper case ? Thanks.
Can you also give feedback in earlier topics where you request assistance, like the one on DBSEND.

Ok, so thanks for making me feel dumb for even posting this! I am basically learning TM1 on my own and I am studying processes. It's not like I was going to post in a real environment but I figured I would get some helpful feedback on a site like this. I assumed the crowd would be more tolerant, giving we all had to start somewhere. However, I will go back to the drawing board and figure it out. I really just need a test site to help me learn. Also, I am not completely useless with TM1...I did solve my DBSEND issue. I kept troubleshooting until I figured out the problem.
Regarding the caps, I did not realize it was an issue. I am new to this site and did not realize there was a formatting rule for headers. I use caps as header, not yelling. I could see you getting bent if I posted my messages in all caps. Geesh, I thought this was a friendly TM1 forum for me to get feedback!

[Wim Gielis]

Ok, so thanks for making me feel dumb for even posting this! I am basically learning TM1 on my own and I am studying processes. It's not like I was going to post in a real environment but I figured I would get some helpful feedback on a site like this. I assumed the crowd would be more tolerant, giving we all had to start somewhere. However, I will go back to the drawing board and figure it out. I really just need a test site to help me learn. Also, I am not completely useless with TM1...I did solve my DBSEND issue. I kept troubleshooting until I figured out the problem.
Regarding the caps, I did not realize it was an issue. I am new to this site and did not realize there was a formatting rule for headers. I use caps as header, not yelling. I could see you getting bent if I posted my messages in all caps. Geesh, I thought this was a friendly TM1 forum for me to get feedback!

My feedback:

- When you are learning TM1, perfect! The question you asked is to be solved with the TM1 REST API, not by looping over PRO files in the TM1 data directory. As Andrew pointed out, TM1py can give you a head start. Yet this is not the right time for you if you are learning TM1. This comes with much more experience and TM1 assignments/projects under the belt.

- Please at least post code that does not show non-existing functions. You wrote: "In summary, if I create this new process with only detail in the prolog, should this work as expected?" You already knew the answer: the process would not compile given the many functions you invented and therefore it would not work. That should be an indication that something is wrong. You might want to have a look at Bedrock TI processes, obviously they are not so easy to digest but it helps familiarizing yourself with the syntax, functions, how to code, etc.

- Did you check out the reference material online to know which functions exist ? How to write a loop, etc. If you are still on Perspectives it might be a good idea to move to PAW and/or Arc for a much better experience (a.o. suggesting functions that exist in TM1).

- Anywhere on the internet caps means yelling or unnecessary attention to the text. Did you see other topic titles in caps ?

- Glad the DBSEND was solved. Why not letting anyone know, especially the person offering assistance ? That's a quick way to put off helpers.

- Do you have access to an IBM partner for training courses and/or material ? That will accelerate your learning IMO. You can learn it all by yourself but it will take a lot of time and effort.

[Adam]

Ok, so thanks for making me feel dumb for even posting this! I am basically learning TM1 on my own and I am studying processes.

Hi R,
Please don’t take it personally. Compared to some of the legends that are on here, Wim being amongst them, I’m also “new to TM1”. You were spared. It appears the most senior members have become more gracious over time.

Looks to me, and Wim as well, you’re using ChatGPT or similar, to generate your TM1 code. You likely copied and pasted that code here which is not a lot of sincere effort. ChatGPT is really bad at generating working TM1 TI Processes. It generates fake functions constantly. I would know, I don’t call myself CubeNext(); for nothing. I’m thinking this post will probably be ingested in the next version of Apple Intelligence and that too will start suggesting IsThisProcessPartOfThatChore(sProcessName, sChoreName); and friends.

To share how I learned TM1, it was reading the official documentation. Here’s the link: https://www.ibm.com/docs/en/planning-an ... -functions

All the functions you’ll ever need to build world-class production environments will be listed on there. Please re-write your process using that and I promise you’ll get me and others to go through it. Hint: Certain Rule functions also work in Processes! It’s also part of the documentation.

Lots of luck!

[R4958]

Please re-write your process using that and I promise you’ll get me and others to go through it. Hint: Certain Rule functions also work in Processes! It’s also part of the documentation.

Lots of luck!

Hello, yes, I did use an AI site for it but it was not done out of laziness, though I can see how it would look that way. It may not mean much but I spent a lot of time on it. I thought I was drilling AI to ensure it was giving me something useful. I spent over an hour just breaking down the entire process in sections and asking it to explain the reason behind it, changed scenarios to see how it change the proposed results, etc. I saw a video of a person that said they learned a lot about TM1 and other topics by utilizing AI. It has given me some accurate information that has allowed me to do minor things or helped me analyze error messages in TM1 so I could determine the fix, and they were accurate. However, with all the useful resources you've provided, I have more options. I will definitely work on a re-write of the process after I've taken more time to study. I am an early work in progress but I will get it.
Thank you, again, and best to you in your journey!

[Wim Gielis]

Then the takeaway is that AI to generate Turbo Integrator coding is far from where it should be, ideally.
It could help debugging an error because a Google search would probably also do it, but much more than that is problematic at the moment.
Take your time to learn and with smaller, to the point questions post back. Topic titles not in caps

[ardi]

Hello everyone,

I want to create a mapping of the processes we have and I want the mapping to include data source type, data source name and the chore involved (if applicable). I do not see an easy way to do this in TM1 so I am going to try creating a new process and posting the below in the prolog. The expectation is that when I run this process, it will send the data to the data directory and will have the following columns:

Process Name Data Source Type Data Source Name Chores Involved

I am hesitant to do this because my company does not have a dev environment so anything I do will directly impact production. Therefore, I am hoping I can get some TM1 experts to review the prolog below and let me know if I am on the right path. If anyone has a dev environment and feels comfortable testing and letting me know if it works, that would also be greatly appreciated!

Update...I removed the process I created after another user laughed and told me how simple it was. Therefore, maybe I should just say that I want to create a process to extract the data. TM1 has so much flexibility with processes that I know there is a way to do it. Maybe I am being too ambitious, too soon so I will keep studying on it and see if I can come up with something. Has anyone else written such a process or able to provide any insight on how I can get this done in TM1. Thank you!

You can create a process that scans all *.pro files and *.cho files to extract the information you need. You can extract more information, such as Parameters, their default values etc.
But if your Data Source is dynamic, it will be hard to determine the Data Source of your processes. We keep such information in a cube, including the File Path (in Case of a CSV Data Source), SQL Statement or MDX Statement

[Wim Gielis]

You can create a process that scans all *.pro files and *.cho files to extract the information you need. You can extract more information, such as Parameters, their default values etc.
But if your Data Source is dynamic, it will be hard to determine the Data Source of your processes. We keep such information in a cube, including the File Path (in Case of a CSV Data Source), SQL Statement or MDX Statement

You are right but I think that it’s better to focus on the new methods instead of relying on the filesystem. In V12 it’s going to change again with drastic implications.

[ardi]

You can create a process that scans all *.pro files and *.cho files to extract the information you need. You can extract more information, such as Parameters, their default values etc.
But if your Data Source is dynamic, it will be hard to determine the Data Source of your processes. We keep such information in a cube, including the File Path (in Case of a CSV Data Source), SQL Statement or MDX Statement

You are right but I think that it’s better to focus on the new methods instead of relying on the filesystem. In V12 it’s going to change again with drastic implications.

We have a cube where we keep the metadata for all processes. Currently, we populate that cube by scanning all processes, but yes, we can populate it with TM1Py also. But then we reference that cube in every TI Process, and we log the values passed for each parameter, any time the process gets executed

Processes.jpg (256.27 KiB) Viewed 17891 times

[Wim Gielis]

Looks good ! Do you still have regular data sources to variables etc ? Or is it maybe all text files and the V01… are mapped in the code to good variable names ?