pi-wsearch 0.1.0 → 0.2.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +62 -0
- package/README.md +4 -4
- package/package.json +13 -2
- package/src/cache.ts +10 -11
- package/src/config.ts +2 -0
- package/src/engines.ts +33 -4
- package/src/fuse.ts +33 -0
- package/src/keys.ts +4 -1
- package/src/scrape.ts +395 -38
- package/src/tools.ts +159 -13
- package/src/transport.ts +229 -16
- package/src/urls.ts +20 -1
- package/justfile +0 -31
package/src/urls.ts
CHANGED
|
@@ -13,6 +13,12 @@ const JUNK_HOSTS = [
|
|
|
13
13
|
'twitch.tv',
|
|
14
14
|
'9gag.com',
|
|
15
15
|
'buzzfeed.com',
|
|
16
|
+
'twiscan.com',
|
|
17
|
+
'twstalker.com',
|
|
18
|
+
'unrollnow.com',
|
|
19
|
+
'threadreaderapp.com',
|
|
20
|
+
'nitter.net',
|
|
21
|
+
'upd.dev',
|
|
16
22
|
];
|
|
17
23
|
const JUNK_TLDS = new Set(['xyz', 'top', 'loan', 'click', 'work', 'gq', 'icu', 'rest', 'cyou']);
|
|
18
24
|
|
|
@@ -53,7 +59,11 @@ export function normalizeUrl(raw: string): string {
|
|
|
53
59
|
const pairs = Array.from(u.searchParams.entries()).sort(([a], [b]) => (a < b ? -1 : a > b ? 1 : 0));
|
|
54
60
|
const q = pairs.map(([k, v]) => `${k}=${v}`).join('&');
|
|
55
61
|
const port = u.port ? `:${u.port}` : '';
|
|
56
|
-
|
|
62
|
+
const path = u.pathname
|
|
63
|
+
.replace(/\/{2,}/g, '/')
|
|
64
|
+
.replace(/\/+$/, '')
|
|
65
|
+
.toLowerCase();
|
|
66
|
+
return `${u.protocol}//${host}${port}${path}${q ? '?' + q : ''}`;
|
|
57
67
|
} catch {
|
|
58
68
|
return raw.toLowerCase();
|
|
59
69
|
}
|
|
@@ -64,5 +74,14 @@ export function isJunk(raw: string): boolean {
|
|
|
64
74
|
if (JUNK_HOSTS.some((j) => host === j || host.endsWith('.' + j))) return true;
|
|
65
75
|
const tld = host.split('.').pop() ?? '';
|
|
66
76
|
if (JUNK_TLDS.has(tld)) return true;
|
|
77
|
+
if (/\/(?:status|thread)\/\d+(?:[/?#]|$)/i.test(pathOf(raw))) return true;
|
|
67
78
|
return false;
|
|
68
79
|
}
|
|
80
|
+
|
|
81
|
+
function pathOf(raw: string): string {
|
|
82
|
+
try {
|
|
83
|
+
return new URL(raw).pathname;
|
|
84
|
+
} catch {
|
|
85
|
+
return '';
|
|
86
|
+
}
|
|
87
|
+
}
|
package/justfile
DELETED
|
@@ -1,31 +0,0 @@
|
|
|
1
|
-
export PATH := "node_modules/.bin:" + env_var_or_default("PATH", "")
|
|
2
|
-
|
|
3
|
-
default: check
|
|
4
|
-
|
|
5
|
-
check: lint format-check typecheck knip test
|
|
6
|
-
|
|
7
|
-
lint:
|
|
8
|
-
oxlint
|
|
9
|
-
|
|
10
|
-
format:
|
|
11
|
-
oxfmt
|
|
12
|
-
|
|
13
|
-
format-check:
|
|
14
|
-
oxfmt --check
|
|
15
|
-
|
|
16
|
-
typecheck:
|
|
17
|
-
tsc --noEmit
|
|
18
|
-
|
|
19
|
-
knip:
|
|
20
|
-
knip
|
|
21
|
-
|
|
22
|
-
coverage:
|
|
23
|
-
bun test --jobs 1 test/ --coverage
|
|
24
|
-
|
|
25
|
-
test:
|
|
26
|
-
bun test --jobs 1 test/
|
|
27
|
-
|
|
28
|
-
install:
|
|
29
|
-
mkdir -p $HOME/.pi/agent/extensions/pi-wsearch/src
|
|
30
|
-
cp index.ts $HOME/.pi/agent/extensions/pi-wsearch/
|
|
31
|
-
cp src/*.ts $HOME/.pi/agent/extensions/pi-wsearch/src/
|