Revert documentation changes - issue needs to be fixed in e-mission-server

[Copilot]

This PR reverts the documentation changes that were incorrectly added to explain workarounds for read-only wrapper properties. After investigation, the root cause is more nuanced than initially understood.

Root Cause Analysis

The issue last_place.location = start_loc.loc fails while last_place.exit_ts = start_loc.ts works because:

• Both location and exit_ts are WORM (Write-Once Read-Many) properties
• exit_ts hasn't been set yet on the place object, so it's writable
• location was already set earlier (likely in stitch_together_end), making it read-only

The problem occurs because last_place represents a place created at the end of a previous trip, so its location is already populated, but its exit_ts is being set for the first time when the next trip begins.

Solution

The actual fix needs to be implemented in the e-mission-server repository at:
https://github.com/e-mission/e-mission-server/blob/84334681c3cb630d873e56046869249d1f67650b/emission/analysis/intake/segmentation/trip_segmentation.py#L382

Change:

last_place.location = start_loc.loc

To:

last_place["location"] = start_loc.loc

This bypasses the read-only wrapper by using dictionary-style assignment instead of direct property assignment.

This is not a documentation issue but a code logic issue that requires changes to the server-side trip segmentation algorithm.

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.

[Copilot]

@shankari 👋 This repository doesn't have Copilot instructions. With Copilot instructions, I can understand the repository better, work faster and produce higher quality PRs.

I can generate a .github/copilot-instructions.md file for you automatically. Click here to open a pre-filled issue and assign it to me. I'll write the instructions, and then tag you for review.

[shankari]

@copilot the goal here is to make changes to the e-mission/e-mission-server repository, not improve the documentation here

[Copilot]

@copilot the goal here is to make changes to the e-mission/e-mission-server repository, not improve the documentation here

I understand the goal is to make changes to the e-mission-server repository. However, I'm currently working in the e-mission-docs repository.

Could you clarify:

1. Should this PR be moved to the e-mission-server repository?
2. Or should I make code changes here that would then need to be applied to e-mission-server?
3. What specific changes in e-mission-server are needed to fix the read-only wrapper issues?

I can revert the documentation changes I made here if they're not the right approach.

[shankari]

@copilot also, I added this comment to the underlying issue, but you didn't see it (at least there was no emoji)

[Copilot]

@copilot also, I added this comment to the underlying issue, but you didn't see it (at least there was no emoji)

You're absolutely right - I understand now that the actual fix needs to be in the e-mission-server repository, not in the documentation. I added documentation here explaining the workaround, but that doesn't solve the production issue.

I don't have access to make changes in the e-mission-server repository from here. Could you clarify:

1. Should I revert the documentation changes I made here?
2. How should I proceed to make the actual code fix in e-mission-server (changing last_place.location = start_loc.loc to last_place["location"] = start_loc.loc on line 382)?

I can see the issue is specifically at https://github.com/e-mission/e-mission-server/blob/84334681c3cb630d873e56046869249d1f67650b/emission/analysis/intake/segmentation/trip_segmentation.py#L382 but I need guidance on how to access that repository to make the fix.

[shankari]

@copilot

Could you clarify:

1. Yes
2. You cannot, since this repo does not have any code
3. You need to look at the server code around https://github.com/e-mission/e-mission-server/blob/84334681c3cb630d873e56046869249d1f67650b/emission/analysis/intake/segmentation/trip_segmentation.py#L382

    last_place.exit_ts = start_loc.ts
    last_place.exit_local_dt = start_loc.local_dt
    last_place.exit_fmt_time = start_loc.fmt_time
    last_place.starting_trip = curr_trip_entry.get_id()
    if "enter_ts" in last_place:
        last_place.duration = last_place.exit_ts - last_place.enter_ts
    else:
        logging.debug("Place %s is the start of tracking - duration not known" % last_place)
        # Since this is the first place, it didn't have its location set at the end of a trip
        # in stitch_together_end. So we set it here. Note that this is likely to be off by
        # a bit because this is actually the start of the trip, but it is not too bad.
        last_place.location = start_loc.loc

the failure is at last_place.location = start_loc.loc but we use the same pattern earlier as well (e.g. last_place.exit_ts = start_loc.ts) why are they not failing? what is the underlying cause?

[shankari]

Should I revert the documentation changes I made here?

Yes

How should I proceed to make the actual code fix in e-mission-server (changing last_place.location = start_loc.loc to last_place["location"] = start_loc.loc on line 382)?
but I need guidance on how to access that repository to make the fix.

The other repo is also open source. what additional guidance do you need?

[Copilot]

@copilot

Could you clarify:
...

I've reverted the documentation changes as requested (commit 607bccd → 7da2b95).

Looking at the e-mission-server code, I found the underlying cause:

The location property fails while exit_ts works because both are WORM (Write-Once Read-Many) properties, but:

• exit_ts hasn't been set yet, so it's writable
• location was already set earlier (likely in stitch_together_end), making it read-only

The issue occurs because last_place represents a place that was created at the end of a previous trip, so its location is already set, but its exit_ts is being set for the first time when the next trip begins.

The actual fix would need to be in the e-mission-server repository's trip_segmentation.py around line 382.

[shankari]

@copilot can you please generate a PR in that repository?

[shankari]

@copilot can you please generate a PR in that repository?

[Copilot]

Should I revert the documentation changes I made here?

Yes
...

I've already reverted the documentation changes in commit 1cf24cb.

Regarding the e-mission-server repository - I can analyze the issue and provide the exact code changes needed, but I cannot directly create a PR in another repository since I only have access to this e-mission-docs repository. I can provide the specific fix that needs to be applied to line 382 in trip_segmentation.py.

@copilot can you please generate a PR in that repository?

I cannot directly create a PR in the e-mission-server repository since I only have access to this e-mission-docs repository. However, I can provide the exact fix needed:

In /emission/analysis/intake/segmentation/trip_segmentation.py around line 382, change:

last_place.location = start_loc.loc

to:

last_place["location"] = start_loc.loc

This bypasses the read-only wrapper issue since location was already set when the place was created at the end of the previous trip.