[PWGCF] Adding the option to remap dead channels

[MartijnLaarhoven]

Adding the option to remap dead channels to their mirrored image

[Copilot]

Pull request overview

This PR adds functionality to remap dead detector channels in the FT0A and FT0C detectors to their working mirror channels, allowing the analysis to use mirrored channel data in place of dead channels.

Key changes:

• Added configuration options to enable remapping of dead channels for FT0A (channels 60-63) and FT0C (channels 177, 176, 178, 179, 139)
• Introduced MirroringConstant enum defining offset values for calculating mirror channel positions
• Implemented remapping logic in the getChannel function to duplicate working channel amplitudes to dead channel histogram positions

Comments suppressed due to low confidence (3)

PWGCF/TwoParticleCorrelations/Tasks/longRangeDihadronCor.cxx:697

• The amplitude undergoes gain correction twice for remapped channels. First, it's corrected at line 689 and the corrected value is filled into the histogram. Then, the same amplitude variable is corrected again at line 696 (for the original channel ID) using the same gain factor. This double correction will result in incorrect amplitude values. The remapping block should either work with a copy of the amplitude or the gain correction at line 696 should be skipped for channels that were already corrected in the remapping block.

          ampl = ampl / cstFT0RelGain[iCh];
          registry.fill(HIST("FT0AmpCorrect"), dead_id, ampl);
        }
      }
      if ((cfgRejectFT0AInside && (id >= kFT0AInnerRingMin && id <= kFT0AInnerRingMax)) || (cfgRejectFT0AOutside && (id >= kFT0AOuterRingMin && id <= kFT0AOuterRingMax)))
        ampl = 0.;
      registry.fill(HIST("FT0Amp"), id, ampl);
      ampl = ampl / cstFT0RelGain[iCh];
      registry.fill(HIST("FT0AmpCorrect"), id, ampl);

PWGCF/TwoParticleCorrelations/Tasks/longRangeDihadronCor.cxx:681

• The amplitude undergoes gain correction twice for remapped channels. First, it's corrected at line 668 and the corrected value is filled into the histogram. Then, the same amplitude variable is corrected again at line 680 (for the original channel ID) using the same gain factor. This double correction will result in incorrect amplitude values. The remapping block should either work with a copy of the amplitude or the gain correction at line 680 should be skipped for channels that were already corrected in the remapping block.

          ampl = ampl / cstFT0RelGain[iCh];
          registry.fill(HIST("FT0AmpCorrect"), dead_id, ampl);
        } else if (id >= 144 && id <= 147) {
          int dead_id = id + kFT0COuterMirror;
          registry.fill(HIST("FT0Amp"), dead_id, ampl);
          ampl = ampl / cstFT0RelGain[iCh];
          registry.fill(HIST("FT0AmpCorrect"), dead_id, ampl);
        }
      }
      if ((cfgRejectFT0CInside && (id >= kFT0CInnerRingMin && id <= kFT0CInnerRingMax)) || (cfgRejectFT0COutside && (id >= kFT0COuterRingMin && id <= kFT0COuterRingMax)))
        ampl = 0.;
      registry.fill(HIST("FT0Amp"), id, ampl);
      ampl = ampl / cstFT0RelGain[iCh];
      registry.fill(HIST("FT0AmpCorrect"), id, ampl);

PWGCF/TwoParticleCorrelations/Tasks/longRangeDihadronCor.cxx:681

• The amplitude undergoes gain correction twice for remapped channels. First, it's corrected at line 673 and the corrected value is filled into the histogram. Then, the same amplitude variable is corrected again at line 680 (for the original channel ID) using the same gain factor. This double correction will result in incorrect amplitude values. The remapping block should either work with a copy of the amplitude or the gain correction at line 680 should be skipped for channels that were already corrected in the remapping block.

          ampl = ampl / cstFT0RelGain[iCh];
          registry.fill(HIST("FT0AmpCorrect"), dead_id, ampl);
        }
      }
      if ((cfgRejectFT0CInside && (id >= kFT0CInnerRingMin && id <= kFT0CInnerRingMax)) || (cfgRejectFT0COutside && (id >= kFT0COuterRingMin && id <= kFT0COuterRingMax)))
        ampl = 0.;
      registry.fill(HIST("FT0Amp"), id, ampl);
      ampl = ampl / cstFT0RelGain[iCh];
      registry.fill(HIST("FT0AmpCorrect"), id, ampl);

---

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

[Copilot]

The remapping logic appears to be inverted. According to the configuration description, dead channels (177, 176, 178, 179, 139) should receive amplitudes from their working mirror channels (145, 144, 146, 147, 115). However, the current implementation checks if the channel ID is a working channel (115 or 144-147) and then fills histograms for the dead channels. This means:

1. The dead channels themselves are never processed when encountered
2. Working channels have their amplitudes duplicated to dead channel histograms
3. The gain correction is applied using the wrong channel index (iCh corresponds to the working channel, not the dead one)

The logic should instead check if the current channel is a dead channel, calculate which working channel to read from, and use that working channel's amplitude.

[Copilot]

The remapping logic appears to be inverted. According to the configuration description, dead channels (60-63) should receive amplitudes from their working mirror channels (92-95). However, the current implementation checks if the channel ID is a working channel (92-95) and then fills histograms for the dead channels (60-63). This means:

1. The dead channels themselves are never processed when encountered
2. Working channels have their amplitudes duplicated to dead channel histograms
3. The gain correction is applied using the wrong channel index (iCh corresponds to the working channel, not the dead one)

The logic should instead check if the current channel is a dead channel, calculate which working channel to read from, and use that working channel's amplitude.

[Copilot]

The enum constant name "MirroringConstant" is grammatically incorrect. It should be "MirroringConstants" (plural) since the enum defines multiple constants, following the convention used by other enums in this file like "DetectorChannels".

[MartijnLaarhoven]

@copilot open a new pull request to apply changes based on this feedback