Add FFI Functions for Clamp Control by MichaelHuth · Pull Request #2559 · AllenInstitute/MIES

MichaelHuth · 2025-10-30T16:52:46Z

t-b · 2025-10-30T19:17:26Z

Let's do your usual appraoch here where Luke/Tim can test the FFI functions before we write the tests.

MichaelHuth · 2025-11-03T16:14:54Z

If the current clamp mode for the headstage in MIES is different from the specific calmp mode of an amplifier setting (e.g. auto bridge balance that applies only to IC) then it is ignored by MIES.

campagnola · 2025-12-10T00:21:51Z

I gave this a try today:

I need a way to get the state of the clamp for any mode, not just the currently active mode. I suggest always returning all fields for both modes regardless of the current clamp mode (much simpler, but is there a cost to assembling these structures?). Alternatively, add a mode argument to FFI_GetClampState that can optionally be -1 to request the current mode.
missing GetHeadstageActive (or is there another way to get this?)
missing SetHoldingPotentialEnable and SetBiasCurrentEnable

campagnola · 2025-12-10T00:24:20Z

If the current clamp mode for the headstage in MIES is different from the specific calmp mode of an amplifier setting (e.g. auto bridge balance that applies only to IC) then it is ignored by MIES.

That makes sense for any of the TriggerAuto* functions, but all other parameters I would expect to be settable even when the mode is not active (for example, I should be able to set the auto bias parameters before switching to IC)

t-b · 2026-01-20T20:48:57Z

I think this is on @MichaelHuth plate again.

t-b · 2026-02-25T16:32:21Z

@MichaelHuth I'm also taking over this one.

Copilot

Pull request overview

Adds new Foreign Function Interface (FFI) entry points intended for external (ACQ4/ZeroMQ) control/inspection of amplifier clamp settings, addressing issue #2547 by avoiding direct GUI element access.

Changes:

Added FFI functions to query clamp state and to set/trigger several clamp-related controls (mode, holding potential, bias current, auto bias, headstage active, and auto clamp actions).
Introduced DAP_SetClampMode as a shared helper and wired the clamp-mode checkbox handler to use it.
Exposed AI_AmpStorageControlToRowLabel for cross-module use and added a range assertion for the AutoBias Vcom parameter.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 6 comments.

File	Description
Packages/MIES/MIES_ForeignFunctionInterface.ipf	Adds clamp-control-related FFI functions, including `FFI_GetClampState` and several setters/triggers.
Packages/MIES/MIES_DAEphys.ipf	Refactors clamp-mode setting into `DAP_SetClampMode` and uses it from the GUI checkbox handler.
Packages/MIES/MIES_AmplifierInteraction.ipf	Adds AutoBias Vcom range assertion and makes `AI_AmpStorageControlToRowLabel` non-static for use by FFI.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot

Pull request overview

Copilot reviewed 3 out of 3 changed files in this pull request and generated 8 comments.

Copilot

Pull request overview

Copilot reviewed 3 out of 3 changed files in this pull request and generated 6 comments.

MichaelHuth · 2026-05-08T18:14:27Z

@campagnola I changed the following:

previous FFI_GetClampState is now FFI_GetCurrentClampState
New FFI_GetClampState returns now the Amplifier Storage Wave from MIES for the specified headstage with all amplifier settings. The wave has dimension labels and the function has docu.
FFI_SetHoldingPotential and FFI_SetBiasCurrent have an additional enable argument that can be 0 (disable) and 1 (enable). When you set enable the value is written first then it is enabled, if you disable then it is first disabled and then the value set.
FFI_SetAutoBias uses the same order for value setting and enabling as FFI_SetHoldingPotential and FFI_SetBiasCurrent
New FFI_GetHeadstageActive returns 1 or 0, depending if the given headstage is enabled or not.
FFI_SetHeadstageActive raises an assertion when attempting to activate a headstage while data acquisition is running
For FFI_SetClampMode the clamp mode change is deferred if data acquisition is running until it stopped. If that is problematic I can also add an assertion there for this case.

t-b · 2026-05-28T21:42:38Z

@campagnola Friendly ping.

this was previously done directly in the GUI event handler

In MIES the range is limited through the corresponding GUi element. However, when changed externally through a ZeroMQ function call then the range was not checked.

Added the following functions: FFI_GetCurrentClampState returns a wave with the current clamp control values FFI_GetClampState returns a wave with the all clamp control values FFI_SetClampMode FFI_SetHoldingPotential - allowed only in VC mode FFI_SetBiasCurrent - allowed only in IC mode FFI_SetAutoBias - allowed only in IC mode FFI_SetHeadstageActive - allowed only when no DAQ is running FFI_GetHeadstageActive - returns 1 is the given headstage is active, 0 otherwise FFI_TriggerAutoClampControl - wraps Pipette, Capacitance and Bridge Balance auto setting in a single call

Copilot

Pull request overview

Copilot reviewed 3 out of 3 changed files in this pull request and generated 4 comments.

Comments suppressed due to low confidence (3)

Packages/MIES/MIES_ForeignFunctionInterface.ipf:786

FFI_GetClampState says it returns a null wave for an "invalid" headstage, but invalid indices currently ASSERT in FFI_CheckValidDeviceAndHeadstage. Please align the API contract (either return null for invalid indices, or adjust docs to state that invalid input raises an error).

/// @param[in] device    Device title, e.g. "Dev1"
/// @param[in] headstage headstage number
/// @returns             wave with full clamp state, null wave for invalid/inactive headstage
Function/WAVE FFI_GetClampState(string device, variable headstage)

Packages/MIES/MIES_ForeignFunctionInterface.ipf:932

FFI_SetAutoBias currently forwards potential to the amplifier without validating it is finite and within the documented range (-99..99 mV). Adding explicit checks here provides a clearer FFI-level error (and avoids relying on downstream assertions).

	clampMode = DAG_GetHeadstageMode(device, headstage)
	ASSERT(clampMode == I_CLAMP_MODE, "Attempt to set auto bias but current clamp mode is not IC !")

Packages/MIES/MIES_ForeignFunctionInterface.ipf:618

The VC documentation table for FFI_GetCurrentClampState lists WholeCellCap twice (once in pF and again in a.u.), but the implementation filters out button-only controls via AI_AmpStorageControlToRowLabel and does not return an a.u. entry here. This duplicate label/unit is misleading for FFI consumers relying on dimension labels.

/// | PipetteOffsetVC        |                       | mV      |
/// +------------------------+-----------------------+---------+
/// | WholeCellCap           |                       | a.u.    |
/// +------------------------+-----------------------+---------+

+/// @param[in] device    Device title, e.g. "Dev1"
+/// @param[in] headstage headstage number
+/// @returns             wave with clamp state, null wave for invalid/inactive headstage
+Function/WAVE FFI_GetCurrentClampState(string device, variable headstage)


+/// @param[in] device    Device title, e.g. "Dev1"
+/// @param[in] headstage headstage number
+/// @param[in] potential holding potential in mV when current clamp mode of headstage is VC, bias current in pA when current clamp mode of headstage is IC
+/// @param[in] enable    when set to 1, enables Holding Potential after setting its value, when set to 0, disables Holding Potential before setting its values
+Function FFI_SetHoldingPotential(string device, variable headstage, variable potential, variable enable)


+
+static Function FFI_CheckValidDeviceAndHeadstage(string device, variable headstage)
+
+	ASSERT(!DAP_DeviceIsUnlocked(device), "Target device is not locked:" + device)


+Function DAP_SetClampMode(string device, variable headstage, variable mode)
+
+	NVAR dataAcqRunMode = $GetDataAcqRunMode(device)
+	if(dataAcqRunMode == DAQ_NOT_RUNNING)
+		DAP_ChangeHeadStageMode(device, mode, headstage, DO_MCC_MIES_SYNCING)


t-b · 2026-07-24T16:29:24Z

@MichaelHuth After chatting with Tim we agreed to add some tests and then merge.

MichaelHuth self-assigned this Oct 30, 2025

MichaelHuth requested review from t-b and timjarsky as code owners October 30, 2025 16:52

MichaelHuth added FFI Foreign Function Interface PatchLink labels Oct 30, 2025

MichaelHuth force-pushed the feature/2559-mh_add_ffi_clamp_control branch 2 times, most recently from 2f8e923 to d5bf06c Compare October 30, 2025 19:23

MichaelHuth assigned campagnola and unassigned MichaelHuth Oct 30, 2025

campagnola mentioned this pull request Jan 15, 2026

changes to support new MIES clamp FFI AllenInstitute/acq4#19

Open

t-b assigned MichaelHuth and unassigned campagnola Jan 20, 2026

t-b assigned t-b and unassigned MichaelHuth Feb 25, 2026

t-b assigned MichaelHuth and unassigned t-b Apr 30, 2026

MichaelHuth force-pushed the feature/2559-mh_add_ffi_clamp_control branch from d5bf06c to 93afc3b Compare May 7, 2026 15:33

Copilot AI review requested due to automatic review settings May 7, 2026 15:33

Copilot started reviewing on behalf of MichaelHuth May 7, 2026 15:34 View session

Copilot AI reviewed May 7, 2026

View reviewed changes

MichaelHuth mentioned this pull request May 7, 2026

Duplicate GUI control association to row label "WholeCellCap" in AmpStorageWave for VC #2702

Closed

MichaelHuth force-pushed the feature/2559-mh_add_ffi_clamp_control branch from 93afc3b to f2c45fb Compare May 7, 2026 17:24

Copilot AI review requested due to automatic review settings May 8, 2026 14:42

MichaelHuth force-pushed the feature/2559-mh_add_ffi_clamp_control branch from f2c45fb to c4e9cc9 Compare May 8, 2026 14:42

Copilot started reviewing on behalf of MichaelHuth May 8, 2026 14:43 View session

Copilot AI reviewed May 8, 2026

View reviewed changes

MichaelHuth force-pushed the feature/2559-mh_add_ffi_clamp_control branch from c4e9cc9 to ddddce4 Compare May 8, 2026 16:48

Copilot AI review requested due to automatic review settings May 8, 2026 17:21

MichaelHuth force-pushed the feature/2559-mh_add_ffi_clamp_control branch from ddddce4 to 3b13ea8 Compare May 8, 2026 17:21

Copilot started reviewing on behalf of MichaelHuth May 8, 2026 17:22 View session

Copilot AI reviewed May 8, 2026

View reviewed changes

MichaelHuth force-pushed the feature/2559-mh_add_ffi_clamp_control branch from 3b13ea8 to 984406a Compare May 8, 2026 17:56

MichaelHuth assigned campagnola and unassigned MichaelHuth May 8, 2026

MichaelHuth added 3 commits July 24, 2026 16:58

DAP: Refactor, Moved setting of clamp mode to own DAP function

ce80ecd

this was previously done directly in the GUI event handler

AI: Add range check for MCC_NO_AUTOBIAS_V_FUNC value write

94eff04

In MIES the range is limited through the corresponding GUi element. However, when changed externally through a ZeroMQ function call then the range was not checked.

Copilot AI review requested due to automatic review settings July 24, 2026 14:58

t-b force-pushed the feature/2559-mh_add_ffi_clamp_control branch from 984406a to 17f5faf Compare July 24, 2026 14:58

Copilot started reviewing on behalf of t-b July 24, 2026 14:59 View session

Copilot AI reviewed Jul 24, 2026

View reviewed changes

t-b assigned MichaelHuth and unassigned campagnola Jul 24, 2026


		static Function FFI_CheckValidDeviceAndHeadstage(string device, variable headstage)

		ASSERT(!DAP_DeviceIsUnlocked(device), "Target device is not locked:" + device)

Uh oh!

Conversation

MichaelHuth commented Oct 30, 2025

Uh oh!

t-b commented Oct 30, 2025

Uh oh!

MichaelHuth commented Nov 3, 2025

Uh oh!

campagnola commented Dec 10, 2025

Uh oh!

campagnola commented Dec 10, 2025

Uh oh!

t-b commented Jan 20, 2026

Uh oh!

t-b commented Feb 25, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

MichaelHuth commented May 8, 2026

Uh oh!

t-b commented May 28, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

t-b commented Jul 24, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants