search: add a query language

copyparty/httpcli.py

@@ -600,8 +600,9 @@ class HttpCli(object):
             taglist = {}
         else:
             # search by query params
-            self.log("qj: " + repr(body))
+            q = body["q"]
-            hits, taglist = idx.search(vols, body)
+            self.log("qj: " + q)
+            hits, taglist = idx.search(vols, q)
             msg = len(hits)
 
         idx.p_end = time.time()

copyparty/u2idx.py

@@ -51,7 +51,7 @@ class U2idx(object):
         uv = [wark[:16], wark]
 
         try:
-            return self.run_query(vols, uq, uv, {})[0]
+            return self.run_query(vols, uq, uv)[0]
         except Exception as ex:
             raise Pebkac(500, repr(ex))
 
@@ -67,36 +67,121 @@ class U2idx(object):
         self.cur[ptop] = cur
         return cur
 
-    def search(self, vols, body):
+    def search(self, vols, uq):
         """search by query params"""
         if not HAVE_SQLITE3:
             return []
 
-        qobj = {}
+        q = ""
-        _conv_sz(qobj, body, "sz_min", "up.sz >= ?")
+        va = []
-        _conv_sz(qobj, body, "sz_max", "up.sz <= ?")
+        joins = ""
-        _conv_dt(qobj, body, "dt_min", "up.mt >= ?")
+        is_key = True
-        _conv_dt(qobj, body, "dt_max", "up.mt <= ?")
+        is_size = False
-        for seg, dk in [["path", "up.rd"], ["name", "up.fn"]]:
+        is_date = False
-            if seg in body:
+        kw_key = ["(", ")", "and ", "or ", "not "]
-                _conv_txt(qobj, body, seg, dk)
+        kw_val = ["==", "=", "!=", ">", ">=", "<", "<=", "like "]
+        ptn_mt = re.compile(r"^\.?[a-z]+$")
+        mt_ctr = 0
+        mt_keycmp = "substr(up.w,1,16)"
+        mt_keycmp2 = None
 
-        uq, uv = _sqlize(qobj)
+        while True:
+            uq = uq.strip()
+            if not uq:
+                break
 
-        qobj = {}
+            ok = False
-        if "tags" in body:
+            for kw in kw_key + kw_val:
-            _conv_txt(qobj, body, "tags", "mt.v")
+                if uq.startswith(kw):
+                    is_key = kw in kw_key
+                    uq = uq[len(kw) :]
+                    ok = True
+                    q += kw
+                    break
 
-        if "adv" in body:
+            if ok:
-            _conv_adv(qobj, body, "adv")
+                continue
+
+            v, uq = (uq + " ").split(" ", 1)
+            if is_key:
+                is_key = False
+
+                if v == "size":
+                    v = "up.sz"
+                    is_size = True
+
+                elif v == "date":
+                    v = "up.mt"
+                    is_date = True
+
+                elif v == "path":
+                    v = "up.rd"
+
+                elif v == "name":
+                    v = "up.fn"
+
+                elif v == "tags":
+                    v = "mt.v"
+
+                elif ptn_mt.match(v):
+                    mt_ctr += 1
+                    mt_keycmp2 = "mt{}.w".format(mt_ctr)
+                    joins += "inner join mt mt{} on {} = {} ".format(
+                        mt_ctr, mt_keycmp, mt_keycmp2
+                    )
+                    v = "mt{0}.k = '{1}' and mt{0}.v".format(mt_ctr, v)
+
+                else:
+                    raise Pebkac(400, "invalid key [" + v + "]")
+
+                q += v + " "
+                continue
+
+            head = ""
+            tail = ""
+
+            if is_date:
+                is_date = False
+                v = v.upper().rstrip("Z").replace(",", " ").replace("T", " ")
+                while "  " in v:
+                    v = v.replace("  ", " ")
+
+                for fmt in [
+                    "%Y-%m-%d %H:%M:%S",
+                    "%Y-%m-%d %H:%M",
+                    "%Y-%m-%d %H",
+                    "%Y-%m-%d",
+                ]:
+                    try:
+                        v = datetime.strptime(v, fmt).timestamp()
+                        break
+                    except:
+                        v = None
+
+            elif is_size:
+                is_size = False
+                v = int(float(v) * 1024 * 1024)
+
+            else:
+                if v.startswith("*"):
+                    head = "'%'||"
+                    v = v[1:]
+
+                if v.endswith("*"):
+                    tail = "||'%'"
+                    v = v[:-1]
+
+            q += "{}?{} ".format(head, tail)
+            va.append(v)
+            is_key = True
 
         try:
-            return self.run_query(vols, uq, uv, qobj)
+            return self.run_query(vols, joins + q, va)
         except Exception as ex:
             raise Pebkac(500, repr(ex))
 
-    def run_query(self, vols, uq, uv, targs):
+    def run_query(self, vols, uq, uv):
-        self.log("qs: {} {} ,  {}".format(uq, repr(uv), repr(targs)))
+        self.log("qs: {} {}".format(uq, repr(uv)))
 
         done_flag = []
         self.active_id = "{:.6f}_{}".format(
@@ -112,33 +197,12 @@ class U2idx(object):
         thr.daemon = True
         thr.start()
 
-        if not targs:
+        if not uq:
-            if not uq:
+            q = "select * from up"
-                q = "select * from up"
+            v = ()
-                v = ()
-            else:
-                q = "select * from up where " + uq
-                v = tuple(uv)
         else:
-            q = "select up.* from up"
+            q = "select * from up where " + uq
-            keycmp = "substr(up.w,1,16)"
+            v = tuple(uv)
-            where = []
-            v = []
-            ctr = 0
-            for tq, tv in sorted(targs.items()):
-                ctr += 1
-                tq = tq.split("\n")[0]
-                keycmp2 = "mt{}.w".format(ctr)
-                q += " inner join mt mt{} on {} = {}".format(ctr, keycmp, keycmp2)
-                keycmp = keycmp2
-                where.append(tq.replace("mt.", keycmp[:-1]))
-                v.append(tv)
-
-            if uq:
-                where.append(uq)
-                v.extend(uv)
-
-            q += " where " + (" and ".join(where))
 
         # self.log("q2: {} {}".format(q, repr(v)))
 
@@ -204,78 +268,3 @@ def _open(ptop):
     db_path = os.path.join(ptop, ".hist", "up2k.db")
     if os.path.exists(db_path):
         return sqlite3.connect(db_path).cursor()
-
-
-def _conv_sz(q, body, k, sql):
-    if k in body:
-        q[sql] = int(float(body[k]) * 1024 * 1024)
-
-
-def _conv_dt(q, body, k, sql):
-    if k not in body:
-        return
-
-    v = body[k].upper().rstrip("Z").replace(",", " ").replace("T", " ")
-    while "  " in v:
-        v = v.replace("  ", " ")
-
-    for fmt in ["%Y-%m-%d %H:%M:%S", "%Y-%m-%d %H:%M", "%Y-%m-%d %H", "%Y-%m-%d"]:
-        try:
-            ts = datetime.strptime(v, fmt).timestamp()
-            break
-        except:
-            ts = None
-
-    if ts:
-        q[sql] = ts
-
-
-def _conv_txt(q, body, k, sql):
-    for v in body[k].split(" "):
-        inv = ""
-        if v.startswith("-"):
-            inv = "not"
-            v = v[1:]
-
-        if not v:
-            continue
-
-        head = "'%'||"
-        if v.startswith("^"):
-            head = ""
-            v = v[1:]
-
-        tail = "||'%'"
-        if v.endswith("$"):
-            tail = ""
-            v = v[:-1]
-
-        qk = "{} {} like {}?{}".format(sql, inv, head, tail)
-        q[qk + "\n" + v] = u8safe(v)
-
-
-def _conv_adv(q, body, k):
-    ptn = re.compile(r"^(\.?[a-z]+) *(==?|!=|<=?|>=?) *(.*)$")
-
-    parts = body[k].split(" ")
-    parts = [x.strip() for x in parts if x.strip()]
-
-    for part in parts:
-        m = ptn.match(part)
-        if not m:
-            p = html_escape(part)
-            raise Pebkac(400, "invalid argument [" + p + "]")
-
-        k, op, v = m.groups()
-        qk = "mt.k = '{}' and mt.v {} ?".format(k, op)
-        q[qk + "\n" + v] = u8safe(v)
-
-
-def _sqlize(qobj):
-    keys = []
-    values = []
-    for k, v in sorted(qobj.items()):
-        keys.append(k.split("\n")[0])
-        values.append(v)
-
-    return " and ".join(keys), values

copyparty/web/browser.css

@@ -529,6 +529,17 @@ input[type="checkbox"]:checked+label {
 	height: 1em;
 	margin: .2em 0 -1em 1.6em;
 }
+#tq_raw {
+	width: calc(100% - 2em);
+	margin: .3em 0 0 1.4em;
+}
+#tq_raw td+td {
+	width: 100%;
+}
+#op_search #q_raw {
+	width: 100%;
+	display: block;
+}
 #files td div span {
 	color: #fff;
 	padding: 0 .4em;

copyparty/web/browser.js

@@ -1026,6 +1026,7 @@ document.onkeydown = function (e) {
 	for (var a = 0; a < trs.length; a += 2) {
 		html.push('<table>' + (trs[a].concat(trs[a + 1])).join('\n') + '</table>');
 	}
+	html.push('<table id="tq_raw"><tr><td>raw</td><td><input id="q_raw" type="text" name="q" /></td></tr></table>');
 	ebi('srch_form').innerHTML = html.join('\n');
 
 	var o = QSA('#op_search input');
@@ -1050,33 +1051,80 @@ document.onkeydown = function (e) {
 			var chk = ebi(id.slice(0, -1) + 'c');
 			chk.checked = ((v + '').length > 0);
 		}
+
+		if (id != "q_raw")
+			encode_query();
+
 		clearTimeout(search_timeout);
 		if (Date.now() - search_in_progress > 30 * 1000)
 			search_timeout = setTimeout(do_search, 200);
 	}
 
+	function encode_query() {
+		var q = '';
+		for (var a = 0; a < sconf.length; a++) {
+			for (var b = 1; b < sconf[a].length; b++) {
+				var k = sconf[a][b][0],
+					chk = 'srch_' + k + 'c',
+					tvs = ebi('srch_' + k + 'v').value.split(/ /g);
+
+				if (!ebi(chk).checked)
+					continue;
+
+				for (var c = 0; c < tvs.length; c++) {
+					var tv = tvs[c];
+					q += ' and ';
+
+					if (k == 'adv') {
+						q += tv.replace(/ /g, " and ");
+						continue;
+					}
+
+					if (k.length == 3) {
+						q += k.replace(/sz/, 'size').replace(/dt/, 'date').replace(/l$/, ' >= ').replace(/u$/, ' <= ') + tv;
+						continue;
+					}
+
+					if (k == 'path' || k == 'name' || k == 'tags') {
+						var not = ' ';
+						if (tv.slice(0, 1) == '-') {
+							tv = tv.slice(1);
+							not = ' not ';
+						}
+
+						if (tv.slice(0, 1) == '^') {
+							tv = tv.slice(1);
+						}
+						else {
+							tv = '*' + tv;
+						}
+
+						if (tv.slice(-1) == '$') {
+							tv = tv.slice(0, -1);
+						}
+						else {
+							tv += '*';
+						}
+
+						q += k + not + 'like ' + tv;
+					}
+				}
+			}
+		}
+		ebi('q_raw').value = q.slice(5);
+	}
+
 	function do_search() {
 		search_in_progress = Date.now();
 		srch_msg(false, "searching...");
 		clearTimeout(search_timeout);
 
-		var params = {},
-			o = QSA('#op_search input[type="text"]');
-
-		for (var a = 0; a < o.length; a++) {
-			var chk = ebi(o[a].getAttribute('id').slice(0, -1) + 'c');
-			if (!chk.checked)
-				continue;
-
-			params[o[a].getAttribute('name')] = o[a].value;
-		}
-		// ebi('srch_q').textContent = JSON.stringify(params, null, 4);
 		var xhr = new XMLHttpRequest();
 		xhr.open('POST', '/?srch', true);
 		xhr.setRequestHeader('Content-Type', 'text/plain');
 		xhr.onreadystatechange = xhr_search_results;
 		xhr.ts = Date.now();
-		xhr.send(JSON.stringify(params));
+		xhr.send(JSON.stringify({"q": ebi('q_raw').value}));
 	}
 
 	function xhr_search_results() {

docs/notes.sh

@@ -80,6 +80,12 @@ command -v gdate && date() { gdate "$@"; }; while true; do t=$(date +%s.%N); (ti
 var t=[]; var b=document.location.href.split('#')[0].slice(0, -1); document.querySelectorAll('#u2tab .prog a').forEach((x) => {t.push(b+encodeURI(x.getAttribute("href")))}); console.log(t.join("\n"));
 
 
+##
+## bash oneliners
+
+# get the size and video-id of all youtube vids in folder, assuming filename ends with -id.ext, and create a copyparty search query
+find -maxdepth 1 -printf '%s %p\n' | sort -n | awk '!/-([0-9a-zA-Z_-]{11})\.(mkv|mp4|webm)$/{next} {sub(/\.[^\.]+$/,"");n=length($0);v=substr($0,n-10);print $1, v}' | tee /dev/stderr | awk 'BEGIN {p="("} {printf("%s name like -%s.* ",p,$2);p="or"} END {print ")\n"}' | cat >&2
+
 ##
 ## sqlite3 stuff