Add artifact paths to UI display for crashed samples, refactor report UI code for maintainability, and add documentation (#1180)

This PR displays the artifact paths to reproduce a specific crash (with
inputs and binaries if available) on the crashed sample's report page.
Both the GCS and console paths are available.

Author: myanvoos

Dockerfile

@@ -37,8 +37,8 @@ RUN echo "deb https://packages.cloud.google.com/apt cloud-sdk main" | tee -a /et
     curl https://packages.cloud.google.com/apt/doc/apt-key.gpg | apt-key add - && \
     apt-get update -y && \
     apt-get install google-cloud-cli -y
-# Set timezone to Australia/Sydney.
-ENV TZ='Australia/Sydney'
+# Set timezone to Anywhere on Earth (UTC-12).
+ENV TZ='Etc/GMT+12'
 
 
 # Install Docker for OSS-Fuzz.

ci/Dockerfile

@@ -16,7 +16,7 @@ FROM debian:12
 ENV DEBIAN_FRONTEND interactive
 
 # Set the same timezone as the main Dockerfile
-ENV TZ='Australia/Sydney'
+ENV TZ='Etc/GMT+12'
 
 # Install packages used by the Experiment. Python and Git are required for the experiment.
 # Curl, certs, and gnupg are required to install gcloud.

report/export.py

@@ -24,10 +24,15 @@
 class BaseExporter:
   """Base class for exporters."""
 
-  def __init__(self, results: Results, output_dir: str, base_url: str = ''):
+  def __init__(self,
+               results: Results,
+               output_dir: str,
+               base_url: str = '',
+               gcs_dir: str = ''):
     self._results = results
     self._output_dir = output_dir
     self._base_url = base_url.rstrip('/')
+    self._gcs_dir = gcs_dir
     self._headers = [
         "Project", "Function Signature", "Sample", "Crash Type", "Compiles",
         "Crashes", "Coverage", "Line Coverage Diff", "Reproducer Path"
@@ -51,6 +56,20 @@ def get_url_path(self) -> str:
 class CSVExporter(BaseExporter):
   """Export a report to CSV."""
 
+  def _get_reproducer_url(self, benchmark_id: str,
+                          crash_reproduction_path: str) -> str:
+    """Get the reproducer URL, using GCS bucket URL for cloud builds."""
+    if not crash_reproduction_path:
+      return ""
+
+    if self._gcs_dir:
+      return (f"https://console.cloud.google.com/storage/browser/"
+              f"oss-fuzz-gcb-experiment-run-logs/Result-reports/"
+              f"{self._gcs_dir}/results/{benchmark_id}/artifacts/"
+              f"{crash_reproduction_path}")
+    return self._get_full_url(
+        f'results/{benchmark_id}/artifacts/{crash_reproduction_path}')
+
   def generate(self):
     """Generate a CSV file with the results."""
     csv_path = os.path.join(self._output_dir, 'crashes.csv')
@@ -70,18 +89,15 @@ def generate(self):
 
         project_name = benchmark_id.split("-")[1]
 
-        project_name = benchmark_id.split("-")[1]
-
         for sample in samples:
           run_logs = self._results.get_run_logs(benchmark_id, sample.id) or ""
           parser = RunLogsParser(run_logs, benchmark_id, sample.id)
           crash_reproduction_path = parser.get_crash_reproduction_path()
 
           report_url = self._get_full_url(
               f"sample/{benchmark_id}/{sample.id}.html")
-          reproducer_path = self._get_full_url(
-              f'results/{benchmark_id}/artifacts/{sample.id}.fuzz_target-F0-01/'
-              f'{crash_reproduction_path}') if crash_reproduction_path else ""
+          reproducer_path = self._get_reproducer_url(benchmark_id,
+                                                     crash_reproduction_path)
 
           writer.writerow({
               "Project":

report/templates/README.md

@@ -0,0 +1,61 @@
+# Report Templates
+
+## Overview
+
+An experiment report consists of three main page types: Index, Benchmark, and Sample.
+
+* Index: Aggregated results across all benchmarks and projects
+* Benchmark: Results for all samples in a single benchmark
+* Sample: Detailed logs and crash analysis for a single sample
+
+All pages extend `base.html` and share common assets. Static files are injected into templates via `web.py`:
+
+* `shared_css_content` - Injected into all pages via base.html
+* `base_js_content` - Injected into all pages via base.html
+* `{page}_css_content` - Page-specific CSS (index, benchmark, sample)
+* `{page}_js_content` - Page-specific JS (index, benchmark, sample)
+
+## File Structure
+
+```
+templates/
+├── base.html           - Base layout with header, navigation, search
+├── base.js             - Search, TOC, prettifyBenchmarkName, syntax highlighting
+├── shared.css          - Common table and chart styles
+├── macros.html         - Reusable Jinja2 UI components
+├── index/              - Main experiment summary index page
+│   ├── index.html      
+│   ├── index.js        
+│   └── index.css       
+├── benchmark/          - Per-benchmark detail page
+│   ├── benchmark.html  
+│   ├── benchmark.js    
+│   └── benchmark.css   
+└── sample/             - Per-sample detail page
+    ├── sample.html     
+    ├── sample.js       
+    └── sample.css      
+
+```
+
+# Updating Template Code
+
+## Adding New Pages
+
+1. Create `{page}/{page}.html`, `{page}.js`, `{page}.css` in templates/
+2. Add `_write_{page}` method in `web.py` following existing pattern
+3. Call `self._read_static_file('{page}/{page}.js')` and pass as `{page}_js_content`
+4. Add to `generate()` method pipeline
+
+## Extending Search Functionality
+
+* To modify field weights: Adjust `BASE_SEARCH_FIELD_WEIGHTS` in `base.js`
+* To add searchable fields: Extend `fields` array in `searchSamples()` 
+* To update the scoring algorithm: Modify score calculation in `searchSamples()` loop
+* To add new fields to the data structure: Update `_build_unified_data()` in `web.py` to include new fields
+
+## Table of Contents
+
+* The table of contents is auto-generated from elements with `.toc-section`, `.toc-subsection`, `.toc-item` CSS classes.
+* The builder in `base.js:buildTOC()` creates a hierarchical tree: Sections -> Subsections -> Items. 
+* To add elements to the table of contents, you can apply the appropriate CSS class to any button/header.