Repository navigation
Conversation
✧ When the chat is empty, display tappable example prompts so first-time users see what kinds of questions HealthGPT can answer. Suggestions disappear once the conversation has any messages. ✧ Closes StanfordBDHG#24 ✧ Signed-off-by: Ehan Shah <ehanrsha@gmail.com>
📝 WalkthroughWalkthroughThis PR implements starter query suggestions for empty chat sessions. A new ChangesEmpty Chat Suggestion Flow
Sequence DiagramssequenceDiagram
participant User
participant SuggestionBannerView
participant cycleSuggestions
participant SampleQueries
User->>SuggestionBannerView: View appears
activate cycleSuggestions
SuggestionBannerView->>cycleSuggestions: task starts
loop Rotate suggestions
cycleSuggestions->>SuggestionBannerView: fade out (opacity = 0)
SuggestionBannerView->>SampleQueries: random(excluding previous)
SampleQueries-->>cycleSuggestions: new suggestion
cycleSuggestions->>SuggestionBannerView: update currentSuggestion
cycleSuggestions->>SuggestionBannerView: fade in (opacity = 1)
cycleSuggestions->>cycleSuggestions: wait for interval
end
User->>SuggestionBannerView: tap suggestion pill
SuggestionBannerView->>User: onSuggestionTapped(currentSuggestion)
deactivate cycleSuggestions
sequenceDiagram
participant HealthGPTView
participant chatViewContent
participant isChatEmpty
participant SuggestionBannerView
participant healthDataInterpreter
HealthGPTView->>isChatEmpty: check if chat has messages
isChatEmpty-->>chatViewContent: boolean result
chatViewContent->>chatViewContent: condition on isChatEmpty
alt Chat is empty
chatViewContent->>SuggestionBannerView: render with ZStack overlay
SuggestionBannerView->>SuggestionBannerView: wait for user tap
SuggestionBannerView->>chatViewContent: onSuggestionTapped(query)
chatViewContent->>chatViewContent: append ChatEntity to context
else Chat has messages
chatViewContent->>chatViewContent: render chat messages
end
chatViewContent->>chatViewContent: onChange messageTaskId incremented
chatViewContent->>healthDataInterpreter: queryLLM() on new non-system message
healthDataInterpreter-->>chatViewContent: LLM response or error
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~25 minutes 🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
HealthGPT/HealthGPT/HealthGPTView.swift (1)
32-83:⚠️ Potential issue | 🔴 Critical | ⚡ Quick winUnresolved merge conflict markers will break compilation.
Lines 32, 70, and 83 contain Git conflict markers (
<<<<<<< Updated upstream,=======,>>>>>>> Stashed changes). These must be resolved before merging. Based on the PR objectives, you likely want to keep the "Stashed changes" section (the newisChatEmptyproperty and refactoredbody).🐛 Proposed fix to resolve the merge conflict
Remove lines 32-70 and line 83 to keep the new implementation:
-<<<<<<< Updated upstream - var body: some View { - NavigationStack { - if let llm = self.healthDataInterpreter.llm { - let contextBinding = Binding { llm.context.chat } set: { llm.context.chat = $0 } - - ChatView(contextBinding, exportFormat: .text) - .speak(llm.context.chat, muted: !self.textToSpeech) - .speechToolbarButton(muted: !self.$textToSpeech) - .viewStateAlert(state: llm.state) - .navigationTitle("WELCOME_TITLE") - .toolbar { - ToolbarItem(placement: .primaryAction) { - self.settingsButton - } - ToolbarItem(placement: .primaryAction) { - self.resetChatButton - } - } - .onChange(of: llm.context, initial: true) { _, _ in - // Once the user enters a message in the chat, increase `messageTaskId` that triggers LLM inference - if !llm.context.isEmpty && llm.state != .generating && llm.context.last?.role != .system { - self.messageTaskId += 1 - } - } - // Triggered on every new user message in the chat via `messageTaskId` - // Automatically cancels the LLM inference once view disappears - .task(id: self.messageTaskId) { - do { - try await healthDataInterpreter.queryLLM() - } catch { - showErrorAlert = true - errorMessage = "Error querying LLM: \(error.localizedDescription)" - } - } - } else { - self.loadingChatView - } -======= /// Whether the chat has no user or assistant messages yet. /// Controls visibility of the suggestion banner overlay. private var isChatEmpty: Bool { ... } var body: some View { NavigationStack { chatViewContent ->>>>>>> Stashed changes }🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@HealthGPT/HealthGPT/HealthGPTView.swift` around lines 32 - 83, Remove the Git conflict markers and the duplicated "Updated upstream" block so the file compiles; keep the "Stashed changes" implementation (the new isChatEmpty property and the refactored var body that uses chatViewContent) and delete the entire <<<<<<< Updated upstream ... ======= ... >>>>>>> Stashed changes section (including the old ChatView-based body that references messageTaskId/llm.context). Ensure only the single isChatEmpty property and the single var body { NavigationStack { chatViewContent } } remain and no conflict markers remain in HealthGPTView.swift.
🧹 Nitpick comments (1)
HealthGPT/HealthGPT/SuggestionBannerView.swift (1)
64-65: 💤 Low valueComment/code timing mismatches.
Several comments don't match the actual timing values:
- Line 64 says "10 seconds" but line 65 uses 7 seconds
- Lines 71 and 86 say "0.4s" but the code uses
duration: 0.6📝 Proposed fix to align comments with code
- // Wait 10 seconds before cycling + // Wait 7 seconds before cycling try? await Task.sleep(for: .seconds(7)) ... - // Phase 1: Fade out (0.4s) + // Phase 1: Fade out (0.6s) withAnimation(.easeInOut(duration: 0.6)) { ... - // Phase 3: Fade in (0.4s) + // Phase 3: Fade in (0.6s) withAnimation(.easeInOut(duration: 0.6)) {Also applies to: 71-72, 86-87
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@HealthGPT/HealthGPT/SuggestionBannerView.swift` around lines 64 - 65, Comments in SuggestionBannerView.swift are out of sync with the actual timing constants; update either the comments or the code so they match. Specifically, reconcile the "Wait 10 seconds before cycling" comment with the Task.sleep call (Task.sleep(for: .seconds(7))): either change the comment to "7 seconds" or change Task.sleep to .seconds(10). Also update the two comments that say "0.4s" to reflect the actual animation duration used (duration: 0.6) or change the duration parameter to 0.4 to match the comments; the relevant symbols to inspect are the Task.sleep(for: .seconds(7)) call and the animation blocks using duration: 0.6.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@HealthGPT/HealthGPT/SuggestionBannerView.swift`:
- Line 19: SuggestionBannerView references SampleQueries.random() and
SampleQueries.random(excluding:) but that type is missing; add a small utility
type (e.g., struct SampleQueries) that exposes a static collection of sample
strings and two static helpers: random() -> String and random(excluding:
String?) -> String (or update the calls to whatever existing provider type you
already have). Implement random(excluding:) to return a random sample different
from the excluded value and call it from SuggestionBannerView where
currentSuggestion is initialized and updated so the view compiles and behavior
remains the same.
---
Outside diff comments:
In `@HealthGPT/HealthGPT/HealthGPTView.swift`:
- Around line 32-83: Remove the Git conflict markers and the duplicated "Updated
upstream" block so the file compiles; keep the "Stashed changes" implementation
(the new isChatEmpty property and the refactored var body that uses
chatViewContent) and delete the entire <<<<<<< Updated upstream ... ======= ...
>>>>>>> Stashed changes section (including the old ChatView-based body that
references messageTaskId/llm.context). Ensure only the single isChatEmpty
property and the single var body { NavigationStack { chatViewContent } } remain
and no conflict markers remain in HealthGPTView.swift.
---
Nitpick comments:
In `@HealthGPT/HealthGPT/SuggestionBannerView.swift`:
- Around line 64-65: Comments in SuggestionBannerView.swift are out of sync with
the actual timing constants; update either the comments or the code so they
match. Specifically, reconcile the "Wait 10 seconds before cycling" comment with
the Task.sleep call (Task.sleep(for: .seconds(7))): either change the comment to
"7 seconds" or change Task.sleep to .seconds(10). Also update the two comments
that say "0.4s" to reflect the actual animation duration used (duration: 0.6) or
change the duration parameter to 0.4 to match the comments; the relevant symbols
to inspect are the Task.sleep(for: .seconds(7)) call and the animation blocks
using duration: 0.6.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 8adde90d-7c69-48cd-8188-564c09e5a60f
📒 Files selected for processing (2)
HealthGPT/HealthGPT/HealthGPTView.swiftHealthGPT/HealthGPT/SuggestionBannerView.swift
| let onSuggestionTapped: (String) -> Void | ||
|
|
||
| /// The currently displayed suggestion text. | ||
| @State private var currentSuggestion: String = SampleQueries.random() |
There was a problem hiding this comment.
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash
# Search for SampleQueries definition
ast-grep --pattern 'struct SampleQueries {
$$$
}'
ast-grep --pattern 'enum SampleQueries {
$$$
}'
# Also search with rg for any SampleQueries definition
rg -n 'SampleQueries' --type swift -C3Repository: StanfordBDHG/HealthGPT
Length of output: 1306
🏁 Script executed:
#!/bin/bash
set -euo pipefail
# Locate any SampleQueries declarations/definitions and related APIs
rg -n --type swift '(^|\s)(struct|enum|class)\s+SampleQueries\b' .
# Find all references to `SampleQueries.` to inspect which APIs are expected/implemented
rg -n --type swift 'SampleQueries\.' .
# Find any occurrences of `random(` near SampleQueries usage/definitions
rg -n --type swift 'SampleQueries.*random' .
# List files that mention SampleQueries (helps locate where it's defined)
rg -l --type swift '\bSampleQueries\b' .Repository: StanfordBDHG/HealthGPT
Length of output: 48
🏁 Script executed:
#!/bin/bash
set -uo pipefail
# Broader: find ANY mentions of SampleQueries
rg -n --type swift '\bSampleQueries\b' . || true
# Find possible declaration forms (struct/enum/class/static)
rg -n --type swift '((^|\s)(struct|enum|class)\s+SampleQueries\b)|((\bstatic\s+func\b|\bfunc\b)\s+random\b.*\bSampleQueries\b)|(\bSampleQueries\s*=)' . || true
# Find expected API usages explicitly
rg -n --type swift 'SampleQueries\s*\.\s*random\s*\(' . || true
rg -n --type swift 'SampleQueries\s*\.\s*random\s*\(\s*excluding\s*:' . || true
# Find any function named random(excluding:) across repo (to correlate with SampleQueries implementation)
rg -n --type swift '\brandom\s*\(\s*excluding\s*:' . || trueRepository: StanfordBDHG/HealthGPT
Length of output: 858
🏁 Script executed:
#!/bin/bash
set -euo pipefail
# Inspect imports / context at top of the file under review
sed -n '1,60p' HealthGPT/HealthGPT/SuggestionBannerView.swift
# Search entire repo (all filetypes) for any SampleQueries declaration/definition hints
rg -n 'SampleQueries' . || true
# Specifically search for potential declaration forms across all filetypes
rg -n '(struct|enum|class)\s+SampleQueries\b|SampleQueries\s*=' . || trueRepository: StanfordBDHG/HealthGPT
Length of output: 2561
Fix missing SampleQueries type/API usage in SuggestionBannerView.
SuggestionBannerView.swift references SampleQueries.random() (line 19) and SampleQueries.random(excluding:) (line ~84), but SampleQueries is not defined anywhere in the repository (repo-wide search for SampleQueries only returns these usages). With only import SwiftUI, this will not compile unless SampleQueries is added or the calls are updated to the correct type/API.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@HealthGPT/HealthGPT/SuggestionBannerView.swift` at line 19,
SuggestionBannerView references SampleQueries.random() and
SampleQueries.random(excluding:) but that type is missing; add a small utility
type (e.g., struct SampleQueries) that exposes a static collection of sample
strings and two static helpers: random() -> String and random(excluding:
String?) -> String (or update the calls to whatever existing provider type you
already have). Implement random(excluding:) to return a random sample different
from the excluded value and call it from SuggestionBannerView where
currentSuggestion is initialized and updated so the view compiles and behavior
remains the same.
Author
|
@vishnuravi Good afternoon! Were you able to review this by any change? Do you have any feedback, or do i need to make any changes? Thank you. ✧ Created A Video Demonstrating The Addition -> https://drive.google.com/file/d/1QnNQ4cw0hT9HTUWk4DZHG_FmUZxRZUuQ/view?usp=sharing |
Member
|
Hi @ehanrsha thanks so much for the code contribution. I'll try this out and get back to you shortly! |
Author
Good evening @vishnuravi Dr. Ravi, |
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
✧ When the chat is empty, display tappable example prompts so first-time users see what kinds of questions HealthGPT can answer. Suggestions disappear once the conversation has any messages.
✧ Closes #24
✧ Signed-off-by: Ehan Shah ehanrsha@gmail.com
✧ Created A Video Demonstrating The Addition -> https://drive.google.com/file/d/1QnNQ4cw0hT9HTUWk4DZHG_FmUZxRZUuQ/view?usp=sharing