amantahiliani/box-box
1
0
Fork 0
mirror of https://github.com/AmanTahiliani/box-box.git synced 2026-08-07 19:56:18 -04:00
Code Issues Packages Projects Releases Wiki Activity

feat(paddock): initial paddock briefing pages and store changes

Browse Source
This commit is contained in:
Aman Tahiliani
2026-05-25 15:16:00 -04:00
parent e060bcba24
commit e3453de788
19 changed files with 1354 additions and 156 deletions
Download Patch File Download Diff File Expand all files Collapse all files

148
internal/news/news.go
View File

@@ -10,30 +10,35 @@ import (
"sort"
"strings"
"time"
"golang.org/x/net/html"
)
const UserAgent = "box-box/phase-19b-rss-spike"
// Source describes a feed that can be fetched and normalized.
type Source struct {
ID string
Name string
URL string
Category string
ID string
Name string
URL string
Category string
SkipSummary bool // omit summary for sources where it's useless (e.g. YouTube)
}
// Item is the normalized shape consumed by storage and API layers.
type Item struct {
Source string
Title string
URL string
PublishedAt time.Time
Summary string
Category string
FetchedAt time.Time
Source string
Title string
URL string
PublishedAt time.Time
Summary string
Category string
FetchedAt time.Time
OGImageURL string
OGDescription string
}
// DefaultSources are free RSS/Atom feeds worth using for the Paddock Briefing spike.
// DefaultSources are free RSS/Atom feeds for the Paddock Briefing.
var DefaultSources = []Source{
{ID: "fia", Name: "FIA", URL: "https://www.fia.com/rss/news", Category: "official"},
{ID: "bbc-f1", Name: "BBC Sport F1", URL: "https://feeds.bbci.co.uk/sport/formula1", Category: "news"},
@@ -41,7 +46,7 @@ var DefaultSources = []Source{
{ID: "racefans-f1", Name: "RaceFans F1", URL: "https://www.racefans.net/category/f1-news/feed/", Category: "news"},
{ID: "guardian-f1", Name: "Guardian Formula One", URL: "https://www.theguardian.com/sport/formulaone/rss", Category: "news"},
{ID: "racer-f1", Name: "RACER F1", URL: "https://racer.com/f1/feed", Category: "news"},
{ID: "f1-youtube", Name: "Formula 1 YouTube", URL: "https://www.youtube.com/feeds/videos.xml?channel_id=UCB_qr75-ydFVKSF9Dmo6izg", Category: "video"},
{ID: "f1-youtube", Name: "Formula 1 YouTube", URL: "https://www.youtube.com/feeds/videos.xml?channel_id=UCB_qr75-ydFVKSF9Dmo6izg", Category: "video", SkipSummary: true},
}
// Fetch retrieves and parses one RSS or Atom feed with the provided HTTP client.
@@ -88,6 +93,83 @@ func Parse(source Source, r io.Reader, fetchedAt time.Time) ([]Item, error) {
return normalizeAtom(source, atom.Entries, fetchedAt), nil
}
// FetchOGMeta fetches only the <head> of a page and extracts og:image and og:description.
// It reads at most 64 KB to avoid full-page downloads.
func FetchOGMeta(ctx context.Context, client *http.Client, rawURL string) (imageURL, description string, err error) {
if client == nil {
client = &http.Client{Timeout: 8 * time.Second}
}
req, err := http.NewRequestWithContext(ctx, http.MethodGet, rawURL, nil)
if err != nil {
return "", "", err
}
req.Header.Set("User-Agent", UserAgent)
resp, err := client.Do(req)
if err != nil {
return "", "", err
}
defer resp.Body.Close()
if resp.StatusCode < 200 || resp.StatusCode >= 300 {
return "", "", fmt.Errorf("og fetch %s: status %d", rawURL, resp.StatusCode)
}
limited := io.LimitReader(resp.Body, 64<<10)
tokenizer := html.NewTokenizer(limited)
for {
tt := tokenizer.Next()
if tt == html.ErrorToken {
break
}
if tt == html.EndTagToken {
tag, _ := tokenizer.TagName()
if string(tag) == "head" {
break
}
}
if tt != html.SelfClosingTagToken && tt != html.StartTagToken {
continue
}
tag, hasAttr := tokenizer.TagName()
if !hasAttr || string(tag) != "meta" {
continue
}
var property, name, content string
for {
k, v, more := tokenizer.TagAttr()
switch string(k) {
case "property":
property = string(v)
case "name":
name = string(v)
case "content":
content = string(v)
}
if !more {
break
}
}
switch {
case property == "og:image" && imageURL == "":
imageURL = strings.TrimSpace(content)
case property == "og:description" && description == "":
description = stripHTML(content)
case name == "description" && description == "":
description = stripHTML(content)
}
if imageURL != "" && description != "" {
break
}
}
return imageURL, description, nil
}
// DeduplicateByURL keeps the newest instance of each canonical URL.
func DeduplicateByURL(items []Item) []Item {
byURL := make(map[string]Item, len(items))
@@ -157,12 +239,16 @@ func normalizeRSS(source Source, raw []rssItem, fetchedAt time.Time) []Item {
if link == "" {
link = strings.TrimSpace(entry.GUID)
}
summary := ""
if !source.SkipSummary {
summary = stripHTML(entry.Description)
}
items = append(items, Item{
Source: source.ID,
Title: cleanText(entry.Title),
URL: link,
PublishedAt: parseFeedTime(entry.PubDate),
Summary: cleanText(entry.Description),
Summary: summary,
Category: firstNonEmpty(entry.Categories, source.Category),
FetchedAt: fetchedAt,
})
@@ -177,12 +263,17 @@ func normalizeAtom(source Source, raw []atomEntry, fetchedAt time.Time) []Item {
if len(entry.Categories) > 0 {
category = firstNonEmpty([]string{entry.Categories[0].Label, entry.Categories[0].Term}, source.Category)
}
summary := ""
if !source.SkipSummary {
raw := firstNonEmpty([]string{entry.Summary, entry.Content}, "")
summary = stripHTML(raw)
}
items = append(items, Item{
Source: source.ID,
Title: cleanText(entry.Title),
URL: atomEntryURL(entry),
PublishedAt: parseFeedTime(firstNonEmpty([]string{entry.Published, entry.Updated}, "")),
Summary: cleanText(firstNonEmpty([]string{entry.Summary, entry.Content}, "")),
Summary: summary,
Category: category,
FetchedAt: fetchedAt,
})
@@ -223,6 +314,33 @@ func parseFeedTime(value string) time.Time {
return time.Time{}
}
// stripHTML removes HTML tags and decodes entities, inserting line breaks for
// block-level elements so the resulting text remains readable as prose.
func stripHTML(value string) string {
if value == "" {
return ""
}
tokenizer := html.NewTokenizer(strings.NewReader(value))
var b strings.Builder
for {
tt := tokenizer.Next()
if tt == html.ErrorToken {
break
}
switch tt {
case html.TextToken:
b.WriteString(tokenizer.Token().Data)
case html.StartTagToken, html.EndTagToken, html.SelfClosingTagToken:
tag, _ := tokenizer.TagName()
switch string(tag) {
case "p", "br", "li", "h1", "h2", "h3", "h4", "h5", "h6", "div", "blockquote", "tr":
b.WriteByte('\n')
}
}
}
return cleanText(b.String())
}
func cleanText(value string) string {
value = strings.TrimSpace(value)
value = strings.ReplaceAll(value, "\n", " ")

103
internal/news/refresh.go
View File

@@ -7,6 +7,7 @@ import (
"io"
"net/http"
"strings"
"sync"
"time"
"github.com/AmanTahiliani/box-box/internal/store"
@@ -22,12 +23,14 @@ type Store interface {
// RefreshOptions configures one local news refresh run.
type RefreshOptions struct {
Sources []Source
Client *http.Client
TTL time.Duration
DryRun bool
Now func() time.Time
Progress io.Writer
Sources []Source
Client *http.Client
TTL time.Duration
DryRun bool
Now func() time.Time
Progress io.Writer
EnrichOG bool // fetch og:image and og:description for each new item
OGParallel int // max concurrent OG fetches (default 5)
}
// RefreshResult summarizes one local news refresh run.
@@ -52,6 +55,9 @@ func Refresh(ctx context.Context, st Store, opts RefreshOptions) (RefreshResult,
if opts.TTL <= 0 {
opts.TTL = DefaultTTL
}
if opts.OGParallel <= 0 {
opts.OGParallel = 5
}
now := func() time.Time { return time.Now().UTC() }
if opts.Now != nil {
now = func() time.Time { return opts.Now().UTC() }
@@ -59,6 +65,10 @@ func Refresh(ctx context.Context, st Store, opts RefreshOptions) (RefreshResult,
var result RefreshResult
var failures []string
// Collect all items across sources for OG enrichment.
var allItems []Item
for _, source := range opts.Sources {
fetchedAt := now()
expiresAt := fetchedAt.Add(opts.TTL)
@@ -86,6 +96,7 @@ func Refresh(ctx context.Context, st Store, opts RefreshOptions) (RefreshResult,
result.ItemsFetched += len(items)
progressf(opts.Progress, "news: %s fetched %d items\n", source.ID, len(items))
if opts.DryRun {
allItems = append(allItems, items...)
continue
}
@@ -101,22 +112,40 @@ func Refresh(ctx context.Context, st Store, opts RefreshOptions) (RefreshResult,
}); err != nil {
return result, err
}
for _, item := range items {
item.FetchedAt = fetchedAt
publishedAt := timePtr(item.PublishedAt)
if err := st.UpsertNewsItem(store.NewsItem{
URL: item.URL,
Source: item.Source,
Title: item.Title,
PublishedAt: publishedAt,
Summary: item.Summary,
Category: item.Category,
FetchedAt: item.FetchedAt,
}); err != nil {
return result, err
}
result.ItemsUpserted++
for i := range items {
items[i].FetchedAt = fetchedAt
}
allItems = append(allItems, items...)
}
// OG enrichment pass: fetch og:image + og:description for each item.
if opts.EnrichOG && len(allItems) > 0 {
progressf(opts.Progress, "news: enriching %d items with OG metadata\n", len(allItems))
enrichOG(ctx, opts.Client, allItems, opts.OGParallel, opts.Progress)
}
if opts.DryRun {
return result, nil
}
// Store all items (with OG data already populated).
for _, item := range allItems {
publishedAt := timePtr(item.PublishedAt)
if err := st.UpsertNewsItem(store.NewsItem{
URL: item.URL,
Source: item.Source,
Title: item.Title,
PublishedAt: publishedAt,
Summary: item.Summary,
Category: item.Category,
FetchedAt: item.FetchedAt,
OGImageURL: item.OGImageURL,
OGDescription: item.OGDescription,
}); err != nil {
return result, err
}
result.ItemsUpserted++
}
if len(failures) > 0 {
@@ -125,6 +154,38 @@ func Refresh(ctx context.Context, st Store, opts RefreshOptions) (RefreshResult,
return result, nil
}
// enrichOG fetches og:image and og:description for each item concurrently.
func enrichOG(ctx context.Context, client *http.Client, items []Item, parallel int, progress io.Writer) {
sem := make(chan struct{}, parallel)
var mu sync.Mutex
var wg sync.WaitGroup
ogClient := &http.Client{Timeout: 8 * time.Second}
if client != nil {
ogClient = &http.Client{Timeout: 8 * time.Second, Transport: client.Transport}
}
for i := range items {
wg.Add(1)
go func(idx int) {
defer wg.Done()
sem <- struct{}{}
defer func() { <-sem }()
imgURL, desc, err := FetchOGMeta(ctx, ogClient, items[idx].URL)
if err != nil {
progressf(progress, "news: og fetch failed %s: %v\n", items[idx].URL, err)
return
}
mu.Lock()
items[idx].OGImageURL = imgURL
items[idx].OGDescription = desc
mu.Unlock()
}(i)
}
wg.Wait()
}
func progressf(w io.Writer, format string, args ...any) {
if w == nil {
return

55
internal/query/news.go
View File

@@ -8,13 +8,16 @@ import (
// NewsItem is the frontend-facing cached briefing shape.
type NewsItem struct {
Source string `json:"source"`
Title string `json:"title"`
URL string `json:"url"`
PublishedAt *time.Time `json:"published_at,omitempty"`
Summary string `json:"summary,omitempty"`
Category string `json:"category,omitempty"`
FetchedAt time.Time `json:"fetched_at"`
Source string `json:"source"`
Title string `json:"title"`
URL string `json:"url"`
PublishedAt *time.Time `json:"published_at,omitempty"`
Summary string `json:"summary,omitempty"`
Category string `json:"category,omitempty"`
FetchedAt time.Time `json:"fetched_at"`
OGImageURL string `json:"og_image_url,omitempty"`
OGDescription string `json:"og_description,omitempty"`
ReadAt *time.Time `json:"read_at,omitempty"`
}
// ListNews returns cached briefing items.
@@ -26,26 +29,36 @@ func (s *Service) ListNews(limit int, source string) ([]NewsItem, error) {
out := make([]NewsItem, 0, len(rows))
for _, row := range rows {
out = append(out, NewsItem{
Source: row.Source,
Title: row.Title,
URL: row.URL,
PublishedAt: row.PublishedAt,
Summary: row.Summary,
Category: row.Category,
FetchedAt: row.FetchedAt,
Source: row.Source,
Title: row.Title,
URL: row.URL,
PublishedAt: row.PublishedAt,
Summary: row.Summary,
Category: row.Category,
FetchedAt: row.FetchedAt,
OGImageURL: row.OGImageURL,
OGDescription: row.OGDescription,
ReadAt: row.ReadAt,
})
}
return out, nil
}
// MarkNewsRead marks a news item as read by URL.
func (s *Service) MarkNewsRead(url string) error {
return s.store.MarkNewsItemRead(url)
}
func NewsItemToStore(item NewsItem) store.NewsItem {
return store.NewsItem{
URL: item.URL,
Source: item.Source,
Title: item.Title,
PublishedAt: item.PublishedAt,
Summary: item.Summary,
Category: item.Category,
FetchedAt: item.FetchedAt,
URL: item.URL,
Source: item.Source,
Title: item.Title,
PublishedAt: item.PublishedAt,
Summary: item.Summary,
Category: item.Category,
FetchedAt: item.FetchedAt,
OGImageURL: item.OGImageURL,
OGDescription: item.OGDescription,
}
}

3
internal/store/migrations/005_news_enriched.sql Normal file
View File

@@ -0,0 +1,3 @@
ALTER TABLE news_items ADD COLUMN og_image_url TEXT;
ALTER TABLE news_items ADD COLUMN og_description TEXT;
ALTER TABLE news_items ADD COLUMN read_at INTEGER;

17
internal/store/models.go
View File

@@ -42,13 +42,16 @@ type NewsSource struct {
// NewsItem stores a normalized feed item deduplicated by URL.
type NewsItem struct {
URL string
Source string
Title string
PublishedAt *time.Time
Summary string
Category string
FetchedAt time.Time
URL string
Source string
Title string
PublishedAt *time.Time
Summary string
Category string
FetchedAt time.Time
OGImageURL string
OGDescription string
ReadAt *time.Time
}
// Meeting is a race weekend record.

45
internal/store/news.go
View File

@@ -41,18 +41,23 @@ func (s *Store) UpsertNewsSource(src NewsSource) error {
}
// UpsertNewsItem inserts or updates a normalized feed item.
// read_at is never overwritten by a re-fetch.
// og_image_url and og_description preserve an existing value when the new one is empty.
func (s *Store) UpsertNewsItem(item NewsItem) error {
_, err := s.db.Exec(`
INSERT INTO news_items (
url, source, title, published_at, summary, category, fetched_at
) VALUES (?, ?, ?, ?, ?, ?, ?)
url, source, title, published_at, summary, category, fetched_at,
og_image_url, og_description
) VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?)
ON CONFLICT(url) DO UPDATE SET
source = excluded.source,
title = excluded.title,
published_at = excluded.published_at,
summary = excluded.summary,
category = excluded.category,
fetched_at = excluded.fetched_at
source = excluded.source,
title = excluded.title,
published_at = excluded.published_at,
summary = excluded.summary,
category = excluded.category,
fetched_at = excluded.fetched_at,
og_image_url = COALESCE(excluded.og_image_url, news_items.og_image_url),
og_description= COALESCE(excluded.og_description, news_items.og_description)
`,
item.URL,
item.Source,
@@ -61,6 +66,8 @@ func (s *Store) UpsertNewsItem(item NewsItem) error {
nullString(item.Summary),
nullString(item.Category),
item.FetchedAt.Unix(),
nullString(item.OGImageURL),
nullString(item.OGDescription),
)
if err != nil {
return fmt.Errorf("upsert news item: %w", err)
@@ -68,6 +75,15 @@ func (s *Store) UpsertNewsItem(item NewsItem) error {
return nil
}
// MarkNewsItemRead sets read_at to now for the given URL.
func (s *Store) MarkNewsItemRead(url string) error {
_, err := s.db.Exec(
`UPDATE news_items SET read_at = ? WHERE url = ?`,
time.Now().Unix(), url,
)
return err
}
// ListNewsItems returns newest cached news items, optionally filtered by source.
func (s *Store) ListNewsItems(limit int, source string) ([]NewsItem, error) {
if limit <= 0 || limit > 100 {
@@ -75,7 +91,8 @@ func (s *Store) ListNewsItems(limit int, source string) ([]NewsItem, error) {
}
query := `
SELECT url, source, title, published_at, summary, category, fetched_at
SELECT url, source, title, published_at, summary, category, fetched_at,
og_image_url, og_description, read_at
FROM news_items
`
var args []any
@@ -95,8 +112,8 @@ func (s *Store) ListNewsItems(limit int, source string) ([]NewsItem, error) {
var out []NewsItem
for rows.Next() {
var item NewsItem
var published sql.NullInt64
var summary, category sql.NullString
var published, readAt sql.NullInt64
var summary, category, ogImage, ogDesc sql.NullString
var fetched int64
if err := rows.Scan(
&item.URL,
@@ -106,6 +123,9 @@ func (s *Store) ListNewsItems(limit int, source string) ([]NewsItem, error) {
&summary,
&category,
&fetched,
&ogImage,
&ogDesc,
&readAt,
); err != nil {
return nil, err
}
@@ -113,6 +133,9 @@ func (s *Store) ListNewsItems(limit int, source string) ([]NewsItem, error) {
item.Summary = summary.String
item.Category = category.String
item.FetchedAt = time.Unix(fetched, 0).UTC()
item.OGImageURL = ogImage.String
item.OGDescription = ogDesc.String
item.ReadAt = nullTimePtr(readAt)
out = append(out, item)
}
return out, rows.Err()

78
internal/web/api.go
View File

@@ -1,9 +1,11 @@
package web
import (
"context"
"database/sql"
"encoding/json"
"errors"
"fmt"
"net/http"
"sort"
"strconv"
@@ -11,6 +13,8 @@ import (
"sync"
"time"
readability "codeberg.org/readeck/go-readability/v2"
"github.com/AmanTahiliani/box-box/internal/models"
"github.com/AmanTahiliani/box-box/internal/query"
)
@@ -133,6 +137,80 @@ func (s *Server) handleNews(w http.ResponseWriter, r *http.Request) {
writeJSON(w, items)
}
// --- /api/v1/news/article ---
type articleResponse struct {
Title string `json:"title"`
Byline string `json:"byline,omitempty"`
Excerpt string `json:"excerpt,omitempty"`
ImageURL string `json:"image_url,omitempty"`
Content string `json:"content"`
SiteName string `json:"site_name,omitempty"`
}
func (s *Server) handleNewsArticle(w http.ResponseWriter, r *http.Request) {
rawURL := strings.TrimSpace(r.URL.Query().Get("url"))
if rawURL == "" {
http.Error(w, "url required", http.StatusBadRequest)
return
}
ctx, cancel := context.WithTimeout(r.Context(), 20*time.Second)
defer cancel()
withCtx := readability.RequestWith(func(req *http.Request) {
*req = *req.WithContext(ctx)
})
article, err := readability.FromURL(rawURL, 20*time.Second, withCtx)
if err != nil {
writeError(w, fmt.Errorf("article fetch: %w", err), http.StatusBadGateway, false)
return
}
var buf strings.Builder
if article.Node != nil {
if rerr := article.RenderHTML(&buf); rerr != nil {
writeError(w, rerr, http.StatusInternalServerError, false)
return
}
}
writeJSON(w, articleResponse{
Title: article.Title(),
Byline: article.Byline(),
Excerpt: article.Excerpt(),
ImageURL: article.ImageURL(),
Content: buf.String(),
SiteName: article.SiteName(),
})
}
// --- /api/v1/news/read ---
func (s *Server) handleNewsRead(w http.ResponseWriter, r *http.Request) {
if r.Method != http.MethodPost {
http.Error(w, "method not allowed", http.StatusMethodNotAllowed)
return
}
if !s.hasLocalQuery() {
w.WriteHeader(http.StatusNoContent)
return
}
var body struct {
URL string `json:"url"`
}
if err := json.NewDecoder(r.Body).Decode(&body); err != nil || body.URL == "" {
http.Error(w, "url required", http.StatusBadRequest)
return
}
if err := s.query.MarkNewsRead(body.URL); err != nil {
writeError(w, err, http.StatusInternalServerError, false)
return
}
w.WriteHeader(http.StatusNoContent)
}
// --- /api/v1/drivers ---
func (s *Server) handleDrivers(w http.ResponseWriter, r *http.Request) {

2
internal/web/server.go
View File

@@ -64,6 +64,8 @@ func (s *Server) routes() (http.Handler, error) {
mux.HandleFunc("/api/v1/race-hub", s.handleRaceHub)
mux.HandleFunc("/api/v1/seasons", s.handleSeasons)
mux.HandleFunc("/api/v1/weekend", s.handleWeekend)
mux.HandleFunc("/api/v1/news/article", s.handleNewsArticle)
mux.HandleFunc("/api/v1/news/read", s.handleNewsRead)
mux.HandleFunc("/api/v1/news", s.handleNews)
mux.HandleFunc("/api/v1/meetings", s.handleMeetings)
mux.HandleFunc("/api/v1/sessions", s.handleSessions)