hh-shipments 0.3.11 → 0.3.12
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/package.json +1 -1
- package/public/app.js +1 -0
- package/public/index.html +1 -0
- package/server.js +2 -0
- package/src/extract.js +7 -3
package/package.json
CHANGED
package/public/app.js
CHANGED
package/public/index.html
CHANGED
package/server.js
CHANGED
|
@@ -60,6 +60,8 @@ function checkIntakeToken(req, res, next) {
|
|
|
60
60
|
next();
|
|
61
61
|
}
|
|
62
62
|
|
|
63
|
+
app.get('/api/version', (req, res) => res.json({ version: require('./package.json').version }));
|
|
64
|
+
|
|
63
65
|
app.get('/api/setup-needed', (req, res) => {
|
|
64
66
|
const n = db.prepare('SELECT COUNT(*) c FROM users').get().c;
|
|
65
67
|
res.json({ needed: n === 0 });
|
package/src/extract.js
CHANGED
|
@@ -54,7 +54,9 @@
|
|
|
54
54
|
}
|
|
55
55
|
// WhatsApp export style: "[10/12/26, 9:41 AM] John Smith: message"
|
|
56
56
|
var m = text.match(/^\[?[^\]\n]*\]?\s*([^:\[\]\n]{2,40}):\s/m);
|
|
57
|
-
|
|
57
|
+
// ...plus a run of 3+ lowercase letters: real names have them ("Marcus
|
|
58
|
+
// Webb"), OCR fragments like "Eg NTT" don't.
|
|
59
|
+
if (m && !/^(you|me|http)/i.test(m[1].trim()) && looksLikeName(m[1]) && /[a-z]{3,}/.test(m[1])) out.seller_name = m[1].trim();
|
|
58
60
|
// "from John", "seller: John"
|
|
59
61
|
if (!out.seller_name) {
|
|
60
62
|
var m2 = text.match(/(?:from|seller)\s*:\s*([A-Z][a-z]+(?:\s+[A-Z][a-z]+)?)/);
|
|
@@ -67,8 +69,10 @@
|
|
|
67
69
|
var labelWords = function (s) { return String(s).trim().split(/\s+/); };
|
|
68
70
|
var looksLikeLabelName = function (s) {
|
|
69
71
|
var words = labelWords(s);
|
|
70
|
-
if (words.length < 2 || words.length >
|
|
71
|
-
|
|
72
|
+
if (words.length < 2 || words.length > 3) return false;
|
|
73
|
+
// every word 4+ letters: OCR garbage ("Eg NTT", "EE SS") never
|
|
74
|
+
// qualifies; real label names ("MOHAMMED SALEM") do.
|
|
75
|
+
for (var i = 0; i < words.length; i++) if (!/^[A-Za-z]{4,}$/.test(words[i])) return false;
|
|
72
76
|
if (/\b(UPS|FEDEX|USPS|DHL|GROUND|EXPRESS|PRIORITY|BILLING|SHIP\b|FROM|TRACKING|POSTAGE|LABEL)\b/i.test(s)) return false;
|
|
73
77
|
return true;
|
|
74
78
|
};
|