Compare commits

..

1 Commits

Author SHA1 Message Date
u80864958
ac5ff7aeeb feat(pierre): add diff chunking and configurable review settings 2026-02-13 17:04:22 +01:00
13 changed files with 95 additions and 628 deletions

216
AGENTS.md
View File

@@ -1,171 +1,77 @@
---
layout: default
---
# Pierre Bot
# AGENTS.md Repository Guidelines for Pierre Bot
Pierre Bot is an intelligent, AI-powered code review assistant designed for Bitbucket Server/Data Center. It automates the initial pass of code review by analyzing Pull Request diffs and identifying potential bugs, logic errors, and style issues using modern LLMs (Google Gemini 2.0 Flash or Ollama).
> This document is consulted by autonomous coding agents (including the
> OpenSource `opencode` agent) when they read, modify, test, or lint the
> codebase. It contains the canonical commands for building, testing, and
> linting the project as well as a concise style guide that all Go code must
> follow. The file is deliberately kept around **150lines** to be readable for
> both humans and LLMs.
## Project Overview
---
* **Type:** Go CLI Application
* **Core Function:** Fetches PR diffs from Bitbucket -> Sends to LLM -> Prints structured review comments.
* **Key Technologies:**
* **Language:** Go (1.25+)
* **AI SDKs:** `google/generative-ai-go`, `ollama/ollama`
* **CLI Framework:** `alecthomas/kong`
## 1⃣ Build / Run Commands
## Architecture
| Command | Description |
| --------------------------------------------- | --------------------------------------------------- |
| `go build -o bin/pierre ./cmd/pierre/main.go` | Build the binary into `./bin/pierre`. |
| `go run ./cmd/pierre/main.go --help` | Run the CLI with the help flag (fast sanity check). |
| `./bin/pierre --version` | Verify the built binary reports its version. |
The project follows a standard Go project layout:
> **Tip for agents** always prefer the `./bin/pierre` binary over invoking `go run` in CI or tests; it guarantees the same compilation flags.
* **`cmd/pierre/`**: Contains the `main.go` entry point. It handles configuration parsing (flags, env vars, file), initializes adapters, and orchestrates the application flow.
* **`internal/pierre/`**: Contains the core business logic.
* `judge.go`: Defines the `JudgePR` function which prepares the system prompt and context for the LLM.
* **`internal/chatter/`**: Abstraction layer for LLM providers.
* `gemini.go`: Implements the `ChatAdapter` interface for Google Gemini. notably includes **dynamic JSON schema generation** via reflection (`schemaFromType`) to enforce structured output from the model.
* `ollama.go`: Implements the `ChatAdapter` for Ollama (local models).
* **`internal/gitadapters/`**: Abstraction for Version Control Systems.
* `bitbucket.go`: Client for fetching PR diffs from Bitbucket Server.
---
## Building and Running
## 2⃣ Test Commands
### Prerequisites
The project uses the standard Go testing framework.
* Go 1.25 or later
* Access to a Bitbucket Server instance
* API Key for Google Gemini (or a running Ollama instance)
| Command | Use case |
| ------------------------------------- | -------------------------------------------------------------------- |
| `go test ./...` | Run **all** unit and integration tests. |
| `go test ./... -run ^TestJudgePR$` | Run a **single** test (replace `TestJudgePR` with the desired name). |
| `go test -cover ./...` | Run tests and emit a coverage report. |
| `go test -run ^TestName$ -count=1 -v` | Verbose, noncached run for debugging a single test. |
| `go test ./... -bench .` | Execute benchmarks (if any). |
> **Agents should never** invoke `go test` with `-parallel` flags; the default parallelism is sufficient and ensures deterministic ordering for our tabledriven tests.
---
## 3⃣ Lint / Static Analysis
We rely on two lightweight builtin tools plus optional `golangci-lint` when
available.
| Tool | Command | Description |
| -------------------------- | ------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `go vet ./...` | `go vet ./...` | Detect obvious bugs, misuse of `fmt.Printf`, etc. |
| `staticcheck ./...` | `staticcheck ./...` | Deeper static analysis (must be installed via `go install honnef.co/go/tools/cmd/staticcheck@latest`). |
| `golangci-lint` (optional) | `golangci-lint run` | Runs a configurable suite of linters. Install with `brew install golangci-lint` or `go install github.com/golangci/golangci-lint/cmd/golangci-lint@latest`. |
**Agent tip** when `golangci-lint` is present, run it after `go vet` to catch style
issues early. The CI pipeline (see `.github/workflows/*.yml` if added later) will
use the same commands.
---
## 4⃣ CodeStyle Guidelines (Go)
All changes must satisfy the following conventions. They are enforced by
`go fmt`, `go vet`, and the optional `golangci-lint` config. The code does not have
to be backwards compatible.
### 4.1 Formatting & Imports
- Run `go fmt ./...` before committing. The repository uses **tabindented** Go code.
- Imports are grouped in three blocks, separated by a blank line:
1. **Standard library**
2. **External dependencies** (e.g., `github.com/alecthomas/kong`)
3. **Internal packages** (e.g., `github.com/yourorg/pierre/internal/...`)
- Within each block, imports are sorted alphabetically.
- Use `goimports` (or `golangcilint --fast`) to automatically fix import ordering.
### 4.2 Naming Conventions
| Entity | Rule |
| --------------------- | ---------------------------------------------------------------------- |
| Packages | lowercase, no underscores or hyphens. |
| Files | snake_case (`*_test.go` for tests). |
| Types / Structs | PascalCase, descriptive (e.g., `ReviewConfig`). |
| Interfaces | End with `er` when appropriate (e.g., `ChatAdapter`). |
| Variables / Constants | camelCase for locals, PascalCase for exported. |
| Functions | PascalCase if exported, camelCase if private. |
| Constants | Use `CamelCase` for groups, `CONST_NAME` only for truly global values. |
| Test Functions | `TestXxx` and optionally `BenchmarkXxx`. |
### 4.3 Error Handling
- Errors are **never** ignored. Use the blank identifier only when the value is truly irrelevant.
- Wrap contextual information using `fmt.Errorf("context: %w", err)` or `errors.Wrap` if the project imports `github.com/pkg/errors` (currently not used).
- Return errors **as soon as** they are detected "guard clause" style.
- In public APIs, prefer error values over panics. Panics are limited to unrecoverable
programming errors (e.g., nilpointer dereference in `init`).
### 4.4 Documentation & Comments
- Exported identifiers **must** have a preceding doc comment beginning with the name (e.g., `// JudgePR reviews a PR and returns comments.`).
- Inline comments should be sentencecase and end with a period.
- Use `//go:generate` directives sparingly; they must be accompanied by a test that ensures the generated file is uptodate.
### 4.5 Testing Practices
- Keep tests **tabledriven** where possible; this yields concise, readable test suites.
- Use the `testing` package only; avoid thirdparty test frameworks.
- Prefer `t.Fatalf` for fatal errors and `t.Errorf` for nonfatal assertions.
- When comparing complex structs, use `github.com/google/go-cmp/cmp` (already in `go.mod`).
- Test coverage should be **80%** for new code.
- All tests must be **deterministic** no reliance on external services; use mocks or fakes.
### 4.6 Dependency Management
- All dependencies are managed via Go modules (`go.mod`).
- Run `go mod tidy` after adding/removing imports.
- Do not commit `vendor/` unless a vendoring strategy is explicitly adopted.
### 4.7 Logging & Output
- Use the standard library `log` package for uservisible output.
- For structured logs (debug mode), wrap with a small helper that respects the `--debug` flag.
- CLI output must be **machineparsable** when `--json-output` is set (future feature).
---
## 5⃣ Git & PullRequest Workflow (for agents)
1. **Branch naming** `feature/<shortdesc>` or `bugfix/<shortdesc>`.
2. **Commits** One logical change per commit, with a concise subject line (<50chars) and an optional body explaining _why_ the change was needed.
3. **CI** Every PR must pass `go test ./...`, `go vet ./...`, and `golangcilint run` (if configured).
4. **Review** Agents should add comments only via the LLM; do not edit code generated by the LLM unless a test fails.
5. **Rebasing** Keep a linear history; use `git rebase -i` locally before merging.
---
## 6⃣ Cursor / Copilot Rules (None Present)
The repository does not contain a `.cursor` directory or a `.github/copilotinstructions.md`
file. If such files are added in the future, agents should read them and incorporate
any additional constraints into this document.
---
## 7⃣ Quick Reference CheatSheet (for agents)
### Build
```bash
# Build
go build -o bin/pierre ./cmd/pierre/main.go
# Run single test (replace TestFoo)
go test ./... -run ^TestFoo$ -v
# Lint & Vet
go vet ./...
staticcheck ./...
# optional
golangci-lint run
# Format & Imports
go fmt ./...
goimports -w .
# Module tidy
go mod tidy
go build -o pierre ./cmd/pierre/main.go
```
---
### Configuration
_End of AGENTS.md_
Configuration is handled via `kong` and supports a hierarchy: **Flags > Env Vars > Config File**.
**1. Environment Variables:**
* `BITBUCKET_URL`: Base URL of the Bitbucket instance.
* `BITBUCKET_TOKEN`: Personal Access Token (HTTP) for Bitbucket.
* `LLM_PROVIDER`: `gemini` or `ollama`.
* `LLM_API_KEY`: API Key for Gemini.
* `LLM_MODEL`: Model name (e.g., `gemini-2.0-flash`).
**2. Configuration File (`config.yaml`):**
See `config.example.yaml` for a template. Place it in the current directory or `~/.pierre.yaml`.
### Usage
Run the bot against a specific Pull Request:
```bash
# Syntax: ./pierre [flags] <PROJECT_KEY> <REPO_SLUG> <PR_ID>
./pierre --llm-provider=gemini --llm-model=gemini-2.0-flash MYPROJ my-repo 123
```
## Development Conventions
* **Structured Output:** The bot relies on the LLM returning valid JSON matching the `Comment` struct. This is enforced in `internal/chatter/gemini.go` by converting the Go struct definition into a `genai.Schema`.
* **Dependency Injection:** Adapters (`gitadapters`, `chatter`) are initialized in `main` and passed to the core logic, making testing easier.
* **Error Handling:** strict error checks are preferred; the bot will exit if it cannot fetch the diff or initialize the AI.
## Key Files
* **`cmd/pierre/main.go`**: Application entry point and config wiring.
* **`internal/pierre/judge.go`**: The "brain" that constructs the prompt for the AI.
* **`internal/chatter/gemini.go`**: Gemini integration logic, including the reflection-based schema generator.
* **`config.example.yaml`**: Reference configuration file.

Binary file not shown.

View File

@@ -3,10 +3,8 @@ package main
import (
"context"
"log"
"log/slog"
"os"
"path/filepath"
"strings"
"git.schreifuchs.ch/schreifuchs/pierre-bot/internal/chatter"
"git.schreifuchs.ch/schreifuchs/pierre-bot/internal/gitadapters/bitbucket"
@@ -45,12 +43,9 @@ type ReviewConfig struct {
MaxChunkSize int `help:"Maximum diff chunk size in bytes" default:"60000"`
Guidelines []string `help:"Project guidelines to prepend" sep:","`
DisableComments bool `help:"Disable posting comments (dry run)"`
SanityCheck bool `help:"Run sanitycheck LLM prompts per comment" default:"true"`
}
type Config struct {
LogLevel string `help:"Log verbosity: debug, info, warn, error" default:"info"`
// Embedding ReviewConfig with a prefix changes flag names to `--review-…`.
// Existing configuration files using the old flag names will need to be updated.
// Consider keeping backwards compatibility if required.
@@ -80,24 +75,6 @@ func main() {
kong.Configuration(kongyaml.Loader, "config.yaml", defaultConfig),
)
// Configure global slog logger based on the --log-level flag
lvl := slog.LevelInfo
switch strings.ToLower(cfg.LogLevel) {
case "debug":
lvl = slog.LevelDebug
case "info":
lvl = slog.LevelInfo
case "warn":
lvl = slog.LevelWarn
case "error":
lvl = slog.LevelError
}
// Logs are sent to stderr so that stdout can be safely piped.
// The review output (fmt.Printf) stays on stdout unchanged.
logger := slog.New(slog.NewTextHandler(os.Stderr, &slog.HandlerOptions{Level: lvl}))
slog.SetDefault(logger)
// Auto-detect provider
provider := cfg.GitProvider
if provider == "" {
@@ -153,7 +130,6 @@ func main() {
}
pierreService := pierre.New(ai, git, cfg.Review.MaxChunkSize, cfg.Review.Guidelines, cfg.Review.DisableComments)
pierreService.SetSanityCheck(cfg.Review.SanityCheck)
if err := pierreService.MakeReview(context.Background(), cfg.Repo.Owner, cfg.Repo.Repo, cfg.Repo.PRID); err != nil {
log.Fatalf("Error during review: %v", err)
}

View File

@@ -1,80 +0,0 @@
package bitbucket
import (
"context"
"net/http"
"net/http/httptest"
"testing"
)
func TestBitbucketGetFileContentSuccess(t *testing.T) {
const expected = "file content"
server := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
// Verify path structure
if r.URL.Path != "/rest/api/1.0/projects/owner/repos/repo/raw/path/to/file.go" {
t.Fatalf("unexpected URL path: %s", r.URL.Path)
}
w.WriteHeader(http.StatusOK)
w.Write([]byte(expected))
}))
defer server.Close()
// Trim trailing slash handling done in NewBitbucket
adapter := NewBitbucket(server.URL, "")
content, err := adapter.GetFileContent(context.Background(), "owner", "repo", "path/to/file.go", "")
if err != nil {
t.Fatalf("unexpected error: %v", err)
}
if content != expected {
t.Fatalf("expected %q, got %q", expected, content)
}
}
func TestBitbucketGetFileContentError(t *testing.T) {
server := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
w.WriteHeader(http.StatusNotFound)
w.Write([]byte("not found"))
}))
defer server.Close()
adapter := NewBitbucket(server.URL, "")
_, err := adapter.GetFileContent(context.Background(), "owner", "repo", "missing.go", "")
if err == nil {
t.Fatalf("expected error for non200 response")
}
}
func TestBitbucketGetPRHeadSHASuccess(t *testing.T) {
const sha = "deadbeef"
server := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
if r.URL.Path != "/rest/api/1.0/projects/owner/repos/repo/pull-requests/42" {
t.Fatalf("unexpected URL: %s", r.URL.Path)
}
w.WriteHeader(http.StatusOK)
w.Write([]byte(`{"toRef":{"latestCommit":"` + sha + `"}}`))
}))
defer server.Close()
adapter := NewBitbucket(server.URL, "")
got, err := adapter.GetPRHeadSHA(context.Background(), "owner", "repo", 42)
if err != nil {
t.Fatalf("unexpected error: %v", err)
}
if got != sha {
t.Fatalf("expected sha %s, got %s", sha, got)
}
}
func TestBitbucketGetPRHeadSHAError(t *testing.T) {
server := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
w.WriteHeader(http.StatusInternalServerError)
w.Write([]byte("error"))
}))
defer server.Close()
adapter := NewBitbucket(server.URL, "")
_, err := adapter.GetPRHeadSHA(context.Background(), "owner", "repo", 1)
if err == nil {
t.Fatalf("expected error for non200 response")
}
}

View File

@@ -5,7 +5,6 @@ import (
"encoding/json"
"fmt"
"io"
"log/slog"
"net/http"
"strconv"
"strings"
@@ -14,7 +13,6 @@ import (
)
func (b *BitbucketAdapter) GetDiff(ctx context.Context, projectKey, repositorySlug string, pullRequestID int) (diff io.ReadCloser, err error) {
slog.Debug("Bitbucket GetDiff start", "project", projectKey, "repo", repositorySlug, "pr", pullRequestID)
r, err := b.CreateRequest(
ctx,
http.MethodGet,
@@ -41,7 +39,6 @@ func (b *BitbucketAdapter) GetDiff(ctx context.Context, projectKey, repositorySl
}
func (b *BitbucketAdapter) GetPR(ctx context.Context, projectKey, repositorySlug string, pullRequestID int) (pr PullRequest, err error) {
slog.Debug("Bitbucket GetPR start", "project", projectKey, "repo", repositorySlug, "pr", pullRequestID)
r, err := b.CreateRequest(
ctx,
http.MethodGet,
@@ -50,23 +47,17 @@ func (b *BitbucketAdapter) GetPR(ctx context.Context, projectKey, repositorySlug
)
response, err := http.DefaultClient.Do(r)
defer response.Body.Close() // Add this
if err != nil {
return
}
defer response.Body.Close() // Add this
err = json.NewDecoder(response.Body).Decode(&pr)
if err != nil {
slog.Error("Bitbucket GetPR decode error", "err", err)
return
}
slog.Info("Bitbucket GetPR success", "id", pullRequestID)
return
}
func (b *BitbucketAdapter) AddComment(ctx context.Context, owner, repo string, prID int, comment pierre.Comment) (err error) {
slog.Debug("Bitbucket AddComment start", "owner", owner, "repo", repo, "pr", prID, "file", comment.File, "line", comment.Line)
// pr, err := b.GetPR(ctx, owner, repo, prID)
// if err != nil {
// return
@@ -95,18 +86,15 @@ func (b *BitbucketAdapter) AddComment(ctx context.Context, owner, repo string, p
}
response, err := http.DefaultClient.Do(r)
defer response.Body.Close() // Add this
if err != nil {
return err
}
defer response.Body.Close() // Add this
if response.StatusCode >= 300 || response.StatusCode < 200 {
sb := &strings.Builder{}
io.Copy(sb, response.Body)
err = fmt.Errorf("error while creating comment staus %d, body %s", response.StatusCode, sb.String())
slog.Error("Bitbucket AddComment failed", "status", response.StatusCode, "err", err)
} else {
slog.Info("Bitbucket AddComment succeeded", "pr", prID)
}
return err

View File

@@ -1,10 +1,6 @@
package bitbucket
import (
"context"
"fmt"
"io"
"net/http"
"strings"
"git.schreifuchs.ch/schreifuchs/pierre-bot/internal/gitadapters/baseadapter"
@@ -22,45 +18,3 @@ func NewBitbucket(baseURL string, bearerToken string) *BitbucketAdapter {
Rest: baseadapter.NewRest(baseURL, bearerToken),
}
}
// GetFileContent returns the raw file content at the given ref (commit SHA) or HEAD if ref is empty.
func (b *BitbucketAdapter) GetFileContent(ctx context.Context, projectKey, repositorySlug, path, ref string) (string, error) {
// Use the Rest helper to build the base URL, then add the "at" query param if needed.
r, err := b.CreateRequest(ctx, http.MethodGet, nil,
"/projects/", projectKey, "repos", repositorySlug, "raw", path)
if err != nil {
return "", err
}
// Ensure raw file retrieval
r.Header.Set("Accept", "application/octet-stream")
if ref != "" {
q := r.URL.Query()
q.Set("at", ref)
r.URL.RawQuery = q.Encode()
}
resp, err := http.DefaultClient.Do(r)
if err != nil {
return "", err
}
defer resp.Body.Close()
if resp.StatusCode != http.StatusOK {
sb := &strings.Builder{}
io.Copy(sb, resp.Body)
return "", fmt.Errorf("error fetching file %s status %d, body %s", path, resp.StatusCode, sb.String())
}
content, err := io.ReadAll(resp.Body)
if err != nil {
return "", err
}
return string(content), nil
}
// GetPRHeadSHA fetches the PR and returns the SHA of the source (to) branch.
func (b *BitbucketAdapter) GetPRHeadSHA(ctx context.Context, projectKey, repositorySlug string, pullRequestID int) (string, error) {
pr, err := b.GetPR(ctx, projectKey, repositorySlug, pullRequestID)
if err != nil {
return "", err
}
return pr.ToRef.LatestCommit, nil
}

View File

@@ -3,10 +3,7 @@ package gitea
import (
"bytes"
"context"
"fmt"
"io"
"log/slog"
"net/http"
"code.gitea.io/sdk/gitea"
"git.schreifuchs.ch/schreifuchs/pierre-bot/internal/pierre"
@@ -27,19 +24,15 @@ func New(baseURL, token string) (*Adapter, error) {
}
func (g *Adapter) GetDiff(ctx context.Context, owner, repo string, prID int) (io.ReadCloser, error) {
slog.Debug("Gitea GetDiff start", "owner", owner, "repo", repo, "pr", prID)
g.client.SetContext(ctx)
diff, _, err := g.client.GetPullRequestDiff(owner, repo, int64(prID), gitea.PullRequestDiffOptions{})
if err != nil {
return nil, err
}
rc := io.NopCloser(bytes.NewReader(diff))
slog.Info("Gitea GetDiff success", "owner", owner, "repo", repo, "pr", prID, "bytes", len(diff))
return rc, nil
return io.NopCloser(bytes.NewReader(diff)), nil
}
func (g *Adapter) AddComment(ctx context.Context, owner, repo string, prID int, comment pierre.Comment) error {
slog.Debug("Gitea AddComment start", "owner", owner, "repo", repo, "pr", prID, "file", comment.File, "line", comment.Line)
g.client.SetContext(ctx)
opts := gitea.CreatePullReviewOptions{
State: gitea.ReviewStateComment,
@@ -51,43 +44,6 @@ func (g *Adapter) AddComment(ctx context.Context, owner, repo string, prID int,
},
},
}
_, resp, err := g.client.CreatePullReview(owner, repo, int64(prID), opts)
if err != nil {
slog.Error("Gitea AddComment failed", "err", err)
return err
}
if resp != nil && resp.StatusCode != http.StatusCreated {
return fmt.Errorf("unexpected status %d creating comment", resp.StatusCode)
}
slog.Info("Gitea AddComment succeeded", "pr", prID)
return nil
}
// GetFileContent returns the file content at a given path and ref (commit SHA).
func (g *Adapter) GetFileContent(ctx context.Context, owner, repo, path, ref string) (string, error) {
slog.Debug("Gitea GetFileContent start", "owner", owner, "repo", repo, "path", path, "ref", ref)
g.client.SetContext(ctx)
// The SDK's GetFile returns the raw bytes of the file.
data, _, err := g.client.GetFile(owner, repo, ref, path)
if err != nil {
return "", err
}
slog.Info("Gitea GetFileContent success", "bytes", len(data))
return string(data), nil
}
// GetPRHeadSHA fetches the pull request and returns the head commit SHA.
func (g *Adapter) GetPRHeadSHA(ctx context.Context, owner, repo string, prID int) (string, error) {
slog.Debug("Gitea GetPRHeadSHA start", "owner", owner, "repo", repo, "pr", prID)
g.client.SetContext(ctx)
// GetPullRequest returns the detailed PR information.
pr, _, err := g.client.GetPullRequest(owner, repo, int64(prID))
if err != nil {
return "", err
}
if pr == nil || pr.Head == nil {
return "", fmt.Errorf("pull request %d has no head information", prID)
}
slog.Info("Gitea GetPRHeadSHA success", "sha", pr.Head.Sha)
return pr.Head.Sha, nil
_, _, err := g.client.CreatePullReview(owner, repo, int64(prID), opts)
return err
}

View File

@@ -1,70 +0,0 @@
package pierre
import (
"fmt"
"strings"
"testing"
)
func TestParseGuidelinesFromStringValid(t *testing.T) {
md := "# Rule One\n\n - Item A \n\n# Rule Two\n"
lines, err := parseGuidelinesFromString(md)
if err != nil {
t.Fatalf("unexpected error: %v", err)
}
expected := []string{"# Rule One", "- Item A", "# Rule Two"}
if got, want := fmt.Sprint(lines), fmt.Sprint(expected); got != want {
t.Fatalf("expected %v, got %v", expected, lines)
}
}
func TestParseGuidelinesFromStringEmpty(t *testing.T) {
md := "\n \n"
lines, err := parseGuidelinesFromString(md)
if err != nil {
t.Fatalf("unexpected error: %v", err)
}
if len(lines) != 0 {
t.Fatalf("expected empty slice, got %d elements", len(lines))
}
}
func TestParseGuidelinesFromStringTooManyLines(t *testing.T) {
// generate 1001 non-empty lines
var sb strings.Builder
for i := 0; i < 1001; i++ {
sb.WriteString(fmt.Sprintf("Line %d\n", i))
}
_, err := parseGuidelinesFromString(sb.String())
if err == nil {
t.Fatalf("expected error for exceeding line limit, got nil")
}
}
func TestWithGuidelinesSuccess(t *testing.T) {
svc := &Service{}
md := "First line\nSecond line\n"
if err := svc.WithGuidelines(md); err != nil {
t.Fatalf("unexpected error: %v", err)
}
expected := []string{"First line", "Second line"}
if got, want := fmt.Sprint(svc.guidelines), fmt.Sprint(expected); got != want {
t.Fatalf("expected guidelines %v, got %v", expected, svc.guidelines)
}
}
func TestWithGuidelinesError(t *testing.T) {
svc := &Service{guidelines: []string{"old"}}
var sb strings.Builder
for i := 0; i < 1001; i++ {
sb.WriteString("x\n")
}
err := svc.WithGuidelines(sb.String())
if err == nil {
t.Fatalf("expected error, got nil")
}
// ensure old guidelines unchanged
if len(svc.guidelines) != 1 || svc.guidelines[0] != "old" {
t.Fatalf("guidelines should remain unchanged on error")
}
}

View File

@@ -4,14 +4,12 @@ import (
"context"
"fmt"
"io"
"log/slog"
"strings"
"git.schreifuchs.ch/schreifuchs/pierre-bot/internal/chatter"
)
// DefaultChunkSize is the fallback maximum size (in bytes) for a diff chunk when no explicit value is configured.
const DefaultChunkSize = 60000
const defaultChunkSize = 60000
type Comment struct {
File string `json:"file"`
@@ -20,7 +18,6 @@ type Comment struct {
}
func (s *Service) judgePR(ctx context.Context, diff io.Reader) (comments []Comment, err error) {
slog.Info("judgePR started")
diffBytes, err := io.ReadAll(diff)
if err != nil {
return nil, fmt.Errorf("failed to read diff: %w", err)
@@ -29,41 +26,36 @@ func (s *Service) judgePR(ctx context.Context, diff io.Reader) (comments []Comme
// Determine chunk size (use default if not set)
maxSize := s.maxChunkSize
if maxSize <= 0 {
maxSize = DefaultChunkSize // default 60KB ~ 15k tokens
maxSize = defaultChunkSize // default 60KB ~ 15k tokens
}
chunks := splitDiffIntoChunks(diffBytes, maxSize)
allComments := []Comment{}
// Build optional guidelines text (added as a separate section with a clear delimiter)
// Build optional guidelines text
guidelinesText := ""
if len(s.guidelines) > 0 {
// Two newlines ensure the guidelines start on a fresh paragraph.
guidelinesText = "\n\nProject guidelines:\n"
guidelinesText = "Project guidelines:\n"
for _, g := range s.guidelines {
guidelinesText += "- " + g + "\n"
}
}
// System prompt that instructs the LLM precisely.
baseSystem := strings.TrimSpace(`
You are a strict senior software architect.
Only comment on newly added or modified lines in the diff; ignore deletions, pure formatting, or reordering that does not change behavior.
For each issue output a JSON object with fields "file", "line", and "message" (message should be concise, ≤2 sentences, and actionable).
If project guidelines are provided, treat them as hard rules that must be respected.`) + guidelinesText
You are a very strict senior software architect.
You review **only** newly added or modified lines in a unified diff, together with the immediate hunk context.
You do **not** report issues that appear **solely** in deleted lines (“-") or that have already been fixed by the change.
No comments are made on pure formatting/whitespace changes or reordering that does not alter the programs behavior.`) + guidelinesText
for i, chunk := range chunks {
// Include the chunk identifier in the system message only if there are multiple chunks.
systemContent := baseSystem
if len(chunks) > 1 {
systemContent = fmt.Sprintf("%s\nChunk %d of %d.", baseSystem, i+1, len(chunks))
}
userContent := chunk
// Add a small header so the model knows this is a fragment
header := fmt.Sprintf("\n--- Chunk %d of %d ---\n", i+1, len(chunks))
userContent := fmt.Sprintf("Hello please review my PR. Write comments where improvements are necessary in new lines.%s\nHere is the git diff of it: %s", header, chunk)
var chunkComments []Comment
err = s.chat.GenerateStructured(ctx, []chatter.Message{{
Role: chatter.RoleSystem,
Content: systemContent,
Content: baseSystem,
}, {
Role: chatter.RoleUser,
Content: userContent,
@@ -83,7 +75,6 @@ If project guidelines are provided, treat them as hard rules that must be respec
for _, v := range unique {
comments = append(comments, v)
}
slog.Info("judgePR finished", "comments", len(comments))
return
}
@@ -91,10 +82,6 @@ If project guidelines are provided, treat them as hard rules that must be respec
// It tries to split on file boundaries ("diff --git") first, then on hunk boundaries (@@),
// and finally on a hard byte limit.
func splitDiffIntoChunks(diff []byte, maxSize int) []string {
// Preserve the file header for each chunk when a single file's diff is split across multiple chunks.
// The header is the portion before the first hunk marker "@@" (including the "diff --git" line).
// When we need to split by hunks, we prepend this header to every resulting subchunk.
if len(diff) <= maxSize {
return []string{string(diff)}
}
@@ -114,31 +101,22 @@ func splitDiffIntoChunks(diff []byte, maxSize int) []string {
current.Reset()
}
if len(seg) > maxSize {
// Determine if there is a hunk marker. If not, fall back to simple sizebased chunking.
headerEnd := strings.Index(seg, "\n@@ ")
if headerEnd == -1 {
// No hunk marker split purely by size.
remaining := seg
for len(remaining) > maxSize {
chunks = append(chunks, remaining[:maxSize])
remaining = remaining[maxSize:]
// Split further by hunks
hunks := strings.Split(seg, "\n@@ ")
for j, h := range hunks {
var hseg string
if j == 0 {
// First hunk segment already contains the preceding content (including any needed newline)
hseg = h
} else {
// Subsequent hunks need the leading newline and "@@ " marker restored
hseg = "\n@@ " + h
}
current.WriteString(remaining)
continue
}
// Preserve the header up to the first hunk.
header := seg[:headerEnd+1] // include newline before "@@"
// Split the rest of the segment into hunks (excluding the header part).
hunks := strings.Split(strings.TrimPrefix(seg, header), "\n@@ ")
for _, h := range hunks {
// Reconstruct each hunk with its header and "@@ " prefix.
hseg := header + "@@ " + h
if current.Len()+len(hseg) > maxSize && current.Len() > 0 {
chunks = append(chunks, current.String())
current.Reset()
}
if len(hseg) > maxSize {
// If a single hunk exceeds maxSize, split it further.
for len(hseg) > maxSize {
chunks = append(chunks, hseg[:maxSize])
hseg = hseg[maxSize:]

View File

@@ -37,15 +37,6 @@ func (g *mockGit) AddComment(ctx context.Context, owner, repo string, prID int,
return nil
}
func (g *mockGit) GetFileContent(ctx context.Context, owner, repo, path, ref string) (string, error) {
// For tests, return a simple placeholder content.
return "package main\n\nfunc placeholder() {}", nil
}
func (g *mockGit) GetPRHeadSHA(ctx context.Context, owner, repo string, prID int) (string, error) {
return "dummysha", nil
}
func TestSplitDiffIntoChunks(t *testing.T) {
cases := []struct {
name string

View File

@@ -1,44 +0,0 @@
package pierre
import (
"context"
"testing"
"git.schreifuchs.ch/schreifuchs/pierre-bot/internal/chatter"
)
type overlapChat struct{ callCount int }
func (m *overlapChat) GenerateStructured(ctx context.Context, msgs []chatter.Message, target interface{}) error {
m.callCount++
if cSlice, ok := target.(*[]Comment); ok {
// Return two comments with same file and line to test deduplication
*cSlice = []Comment{{File: "dup.go", Line: 10, Message: "first"}, {File: "dup.go", Line: 10, Message: "second"}}
return nil
}
return nil
}
func (m *overlapChat) GetProviderName() string { return "mock" }
func TestJudgePR_DeduplicationOverlap(t *testing.T) {
chat := &overlapChat{}
svc := &Service{
maxChunkSize: 1000,
guidelines: nil,
git: &mockGit{},
chat: chat,
}
diffReader, err := svc.git.GetDiff(context.Background(), "", "", 0)
if err != nil {
t.Fatalf("failed to get diff: %v", err)
}
defer diffReader.Close()
comments, err := svc.judgePR(context.Background(), diffReader)
if err != nil {
t.Fatalf("judgePR error: %v", err)
}
if len(comments) != 1 {
t.Fatalf("expected 1 deduplicated comment, got %d", len(comments))
}
}

View File

@@ -2,9 +2,7 @@ package pierre
import (
"context"
"fmt"
"io"
"strings"
"git.schreifuchs.ch/schreifuchs/pierre-bot/internal/chatter"
)
@@ -16,68 +14,25 @@ import (
// elsewhere in the future.
type Service struct {
maxChunkSize int
guidelines []string // stored as slice of lines; legacy, see WithGuidelines
skipSanityCheck bool // if true, skip LLM sanitycheck prompts per comment
guidelines []string
disableComments bool
git GitAdapter
chat ChatAdapter
}
func New(chat ChatAdapter, git GitAdapter, maxChunkSize int, guidelines []string, disableComments bool) *Service {
// Existing constructor retains slice based guidelines for backward compatibility.
return &Service{
git: git,
chat: chat,
maxChunkSize: maxChunkSize,
guidelines: guidelines,
skipSanityCheck: false,
disableComments: disableComments,
}
}
// WithGuidelines parses a raw Markdown string (or any multiline string) into
// individual guideline lines, validates the linecount (max 1000 nonempty lines),
// and stores the result in the Service. It returns an error if validation fails.
// This is a convenience mutator for callers that have the guidelines as a
// single string.
func (s *Service) WithGuidelines(md string) error {
lines, err := parseGuidelinesFromString(md)
if err != nil {
return err
}
s.guidelines = lines
return nil
}
// parseGuidelinesFromString splits a markdown string into trimmed, nonempty
// lines and ensures the total number of lines does not exceed 1000.
func (s *Service) SetSanityCheck(enabled bool) {
s.skipSanityCheck = !enabled
}
// parseGuidelinesFromString splits a markdown string into trimmed, nonempty
func parseGuidelinesFromString(md string) ([]string, error) {
var result []string
// Split on newline. Handles both \n and \r\n because TrimSpace removes \r.
rawLines := strings.Split(md, "\n")
for _, l := range rawLines {
trimmed := strings.TrimSpace(l)
if trimmed == "" {
continue
}
result = append(result, trimmed)
}
if len(result) > 1000 {
return nil, fmt.Errorf("guidelines exceed 1000 lines (found %d)", len(result))
}
return result, nil
}
type GitAdapter interface {
GetDiff(ctx context.Context, owner, repo string, prID int) (io.ReadCloser, error)
AddComment(ctx context.Context, owner, repo string, prID int, comment Comment) error
GetFileContent(ctx context.Context, owner, repo, path, ref string) (string, error)
GetPRHeadSHA(ctx context.Context, owner, repo string, prID int) (string, error)
}
type ChatAdapter interface {

View File

@@ -3,9 +3,7 @@ package pierre
import (
"context"
"fmt"
"log/slog"
"git.schreifuchs.ch/schreifuchs/pierre-bot/internal/chatter"
"log"
)
func (s *Service) MakeReview(ctx context.Context, organisation string, repo string, prID int) error {
@@ -22,62 +20,21 @@ func (s *Service) MakeReview(ctx context.Context, organisation string, repo stri
return fmt.Errorf("error judging PR: %w", err)
}
// ---------- Sanitycheck step ----------
headSHA, err := s.git.GetPRHeadSHA(ctx, organisation, repo, prID)
if err != nil {
slog.Warn("could not fetch PR head SHA", "error", err)
} else if !s.skipSanityCheck {
filtered := []Comment{}
for _, c := range comments {
// Retrieve full file content at the PR head
fileContent, fErr := s.git.GetFileContent(ctx, organisation, repo, c.File, headSHA)
if fErr != nil {
slog.Warn("failed to fetch file", "path", c.File, "error", fErr)
filtered = append(filtered, c)
continue
}
// Build a simple sanitycheck prompt
systemPrompt := `You are a senior software architect. Given the full source code of a file and a review comment that refers to it, decide whether the comment is useful. Return JSON with fields "useful" (bool) and "reason" (short explanation, ≤2 sentences).`
userPrompt := fmt.Sprintf("File content:\n%s\n\nComment:\n%s", fileContent, c.Message)
type sanityResult struct {
Useful bool `json:"useful"`
Reason string `json:"reason"`
}
var res sanityResult
if err := s.chat.GenerateStructured(ctx, []chatter.Message{{Role: chatter.RoleSystem, Content: systemPrompt}, {Role: chatter.RoleUser, Content: userPrompt}}, &res); err != nil {
slog.Error("sanity check error", "file", c.File, "line", c.Line, "error", err)
filtered = append(filtered, c)
continue
}
if res.Useful {
// Optionally annotate the comment with the reason for debugging
c.Message = fmt.Sprintf("%s (Reason: %s)", c.Message, res.Reason)
filtered = append(filtered, c)
} else {
slog.Info("comment discarded", "file", c.File, "line", c.Line, "reason", res.Reason)
}
}
comments = filtered
}
fmt.Printf("Analysis complete. Found %d issues.\n---\n", len(comments))
model := s.chat.GetProviderName()
for _, c := range comments {
c.Message = fmt.Sprintf("%s (Generated by: %s)", c.Message, model)
// Normal mode: print to stdout and post the comment to the VCS.
fmt.Printf("File: %s\nLine: %d\nMessage: %s\n%s\n",
c.File, c.Line, c.Message, "---")
if !s.disableComments {
if err := s.git.AddComment(ctx, organisation, repo, prID, c); err != nil {
slog.Error("failed to add comment", "error", err)
}
if s.disableComments {
log.Printf("dry-run: not posting comment for %s:%d", c.File, c.Line)
} else if err := s.git.AddComment(ctx, organisation, repo, prID, c); err != nil {
log.Printf("Failed to add comment: %v", err)
}
}
return nil