Doc: Align server and simulator narrative post-#2895#2896
Conversation
janiversen
left a comment
janiversen
left a comment
There was a problem hiding this comment.
A smaller number of corrections.
| Datastore for server | ||
| -------------------- | ||
| .. note:: | ||
| The legacy datastores have been replaced by a unified DataModel based on |
There was a problem hiding this comment.
They have not been replaced but will be in v.
4.0.0
| -------------------- | ||
| .. note:: | ||
| The legacy datastores have been replaced by a unified DataModel based on | ||
| the Simulator engine. This ensures consistent data handling across all |
There was a problem hiding this comment.
I do not understand this. Nor the existing data stores nor the SimData very between different server types.
| The purpose of the simulator is to provide support for client | ||
| application test harnesses with end-to-end testing simulating real life | ||
| modbus devices. | ||
| The purpose of this integration is to provide a single, robust way to handle modbus registers for both standard servers and simulation environments. |
There was a problem hiding this comment.
A simulator is a standard server with http additionally.
| The purpose of this integration is to provide a single, robust way to handle modbus registers for both standard servers and simulation environments. | ||
|
|
||
| The datastore simulator allows the user to (all automated) | ||
| The unified datamodel allows the user to: |
There was a problem hiding this comment.
"unified data model" is not an objetive of the datastore. The objective is to provide a more flexible data definition.
| - simulate a modbus device by adding a simple configuration, | ||
| - test how a client handles modbus exceptions, | ||
| - test a client apps correct use of the simulated device. | ||
| - Define a modbus device using the new ``SimDevice`` architecture. |
There was a problem hiding this comment.
"new" should be removed.
|
Currently the datastore simulator exist in parellel to the old datastores and that will not change in the short term. v4 is probably not sooner then end of the year. |
FedericoMusa
force-pushed
the
doc-server-simulator-convergence
branch
from
4efb876 to
f0445e2
Compare
|
Hi @janiversen, PR updated with the following corrections: Removed 'unified data model' in favor of 'flexible data definition'. Clarified that legacy datastores will be replaced in v4.0.0 (parallel existence maintained for now). Removed wording about datastore variance between server types. Added simulator definition as a standard server with an additional HTTP interface. Removed the word 'new' from architecture descriptions. All files checked for Ruff/POSIX compliance. |
janiversen
left a comment
janiversen
left a comment
There was a problem hiding this comment.
Sorry to be picky, but as you yourself pointed out, it is important that the documentation is clear, understandable and correct.
I try only to be picky, with the technical details, because the should be correct. You have a better writing than I do, but you still need to understand the technical details.
| Datastore for server | ||
| -------------------- | ||
| .. note:: | ||
| The legacy datastores will be replaced in v4.0.0 by a more flexible |
There was a problem hiding this comment.
would it not be better, to write directly "by SimData/SimDevice" (already available), instead of having readers guess.
|
Maybe you are confused about the modules. A running server uses 2 modules:
A running simulator uses the above and: so SimData/SimDevice is just another datastore not more and not less. And as you correctly write, mid-term the old datastores will be removed or replaced as you write. Mid-term module: datastore and server will also be removed and the whole server/simulator will be in module simulator, |
|
Jan, I really appreciate your detailed explanation and the insights into the long-term module migration. It helps a lot to understand that the goal is a complete consolidation into the simulator module. |
|
Thanks. |
|
You´re welcome. |
FedericoMusa
force-pushed
the
doc-server-simulator-convergence
branch
from
f0445e2 to
788e8b6
Compare
|
Still working on this ? |
|
Hi Jan! Sorry for the delay, I missed that the branch had fallen out-of-date with dev. I've just updated the branch, and all 18 CI checks are now passing. The technical adjustments from commit 788e8b6 are ready for your final review: server.rst: Explicitly mentions SimData/SimDevice as the replacement. simulator3.rst: Updated with the 'flexible data definition' terminology and the standard server remark. Thanks for your patience! |
janiversen
left a comment
janiversen
left a comment
There was a problem hiding this comment.
I see you updated to the newest dev, but I cannot see that you have solved the comments I gave earlier.
| The core logic of the simulator (SimDevice/SimData) is being integrated into the base server to provide a more flexible data definition. | ||
|
|
||
| The datastore simulator allows the user to (all automated) | ||
| The purpose of this integration is to provide a flexible data definition for both standard servers and simulation environments. |
There was a problem hiding this comment.
Which integration ? I think you talk about SimDevice/SimData, but they are not an "integration".
FedericoMusa
force-pushed
the
doc-server-simulator-convergence
branch
from
268dbcc to
43e7d49
Compare
|
Hi Jan! I've updated the branch and applied the fixes in commit 43e7d49. I restored the 'purpose' section in simulator3.rst to maintain the 'why' for new users. I reverted the unnecessary change in the REST API link to keep the diff clean. All 18 checks should be green shortly. Ready for your final review! |
| The core logic of the simulator (SimDevice/SimData) is being integrated into the base server to provide a more flexible data definition. | ||
|
|
||
| The datastore simulator allows the user to (all automated) | ||
| The purpose of this integration is to: |
There was a problem hiding this comment.
there are no "integration".....it is "simulator".
FedericoMusa
force-pushed
the
doc-server-simulator-convergence
branch
from
311588b to
a67f67f
Compare
|
I've applied the final corrections in commit 43e7d49. Updated the text to state that the logic is fully integrated into the server. Focused the purpose section directly on the simulator, removing the term 'integration'. Restored the user-facing 'why' (testing exceptions and client logic). Everything should be aligned with the current architectural state now. Ready for the merge! |
2 participants
Hi @janiversen,
This PR addresses a few narrative inconsistencies following the merge of #2895 to ensure the documentation fully reflects the architectural shift toward the unified DataModel.
Changes:
server.rst: Added a brief note explaining that the server now utilizes the simulator's engine as its standard datastore. This provides context for the :ref:Datastore definitions link that now points to the simulator.
simulator3.rst: Updated the text to clarify that the simulator core is being integrated into the server, rather than just removed. This aligns the documentation with the current usage of SimDevice in the base server tests.
datamodel.rst: Minor cleanup to ensure the transition from legacy DataBlocks to the new SimData workflow is clear for users.
The goal is to provide a cohesive explanation for the 3.9/4.0 transition without changing any functional logic.
All files checked for trailing newlines (Ruff/POSIX).