Files
comitium/fetch/gemini.go
T
2021-03-23 08:15:36 -06:00

149 lines
3.8 KiB
Go

// Copyright (C) 2021 nytpu
// SPDX-License-Identifier: AGPL-3.0-or-later
// For more license details, see LICENSE or <https://www.gnu.org/licenses/agpl-3.0.html>.
// This file incorporates code from Drew DeVault's gemreader:
// https://git.sr.ht/~sircmpwn/gemreader/tree/master/item/feeds/fetch.go
// Original source is Copyright 2021 Drew DeVault and is licensed under the
// terms of the GNU Affero General Public License Version 3.0.
// https://www.gnu.org/licenses/agpl-3.0-standalone.html
package fetch
import (
"context"
"crypto/sha256"
"fmt"
"io"
"mime"
"net/url"
"strings"
"time"
"git.sr.ht/~adnano/go-gemini"
"github.com/mmcdole/gofeed"
"golang.nytpu.com/comitium/core"
)
// fetchGemini will make a request for a gemini resource
func fetchGemini(remote string) (*gemini.Response, error) {
ctx, cancel := context.WithTimeout(context.Background(), time.Duration(core.Timeout)*time.Second)
defer cancel()
client := &gemini.Client{}
resp, err := client.Get(ctx, remote)
if err != nil {
return nil, err
}
if resp.Status != gemini.StatusSuccess {
return nil, fmt.Errorf("Unexpected Gemini response: %d %s", resp.Status, resp.Meta)
}
return resp, nil
}
// geminiFeed will fetch and insert a new core.Feed into a core.FullData given
// a gemini url (or update if it's preexisting)
func geminiFeed(data *core.FullData, remote string, title string) error {
resp, err := fetchGemini(remote)
if err != nil {
return err
}
defer resp.Body.Close()
mimetype, _, err := mime.ParseMediaType(resp.Meta)
if err != nil {
return fmt.Errorf("Unintelligible content type: %s", resp.Meta)
}
reader := io.LimitReader(resp.Body, 1073741824) // 1 GiB
switch mimetype {
case "text/gemini":
var feed core.Feed
feed.Title = title // will be set later if given title is ""
feed.Link = remote
feed.FeedLink = remote
feed.Updated = time.Now()
// we know that remote is a good url because it's been parsed previously
parsedRemote, _ := url.ParseRequestURI(remote)
text, err := gemini.ParseText(resp.Body)
if err != nil {
return err
}
for _, line := range text {
switch line := line.(type) {
case gemini.LineHeading1:
if feed.Title == "" {
feed.Title = strings.TrimLeft(line.String(), "# ")
}
case gemini.LineLink:
if line.Name == "" || len(line.Name) < 10 {
continue
}
name := strings.TrimLeft(line.Name[:10], " -—")
date, err := time.Parse("2006-01-02", name)
if err != nil {
continue
}
link, err := url.Parse(line.URL)
if err != nil {
continue
}
link = parsedRemote.ResolveReference(link)
item := &core.Item{}
item.Title = strings.TrimLeft(line.Name[10:], " :-–—")
item.Published = date
item.Link = link.String()
feed.Items = append(feed.Items, item)
}
}
data.InsertFeed(&feed, remote)
return nil
case "text/xml",
"application/rss+xml",
"application/atom+xml",
"application/xml",
"application/json":
gofeed, err := gofeed.NewParser().Parse(reader)
if err != nil {
return err
}
feed, err := core.GofeedConvert(gofeed, title)
if err != nil {
return err
}
data.InsertFeed(feed, remote)
return nil
default:
return fmt.Errorf("Cannot interpret %s as a feed", resp.Meta)
}
}
// geminiPage will fetch and insert a new core.Page into a core.FullData given
// a gemini url (or update if it's preexisting)
func geminiPage(data *core.FullData, remote string, title string) error {
resp, err := fetchGemini(remote)
if err != nil {
return err
}
defer resp.Body.Close()
reader := io.LimitReader(resp.Body, 1073741824) // 1 GiB
var page core.Page
page.Title = title
page.Link = remote
h := sha256.New()
if _, err := io.Copy(h, reader); err != nil {
return err
}
newHash := fmt.Sprintf("%x", h.Sum(nil))
page.Hash = newHash
page.Updated = time.Now()
data.InsertPage(&page, remote)
return nil
}