version bump 0.20.2

see https://regexide.com for more details
2024-04-04 21:20:28 -04:00 · 2024-04-04 21:20:28 -04:00 · a0bed2a97d
commit a0bed2a97d
parent 0941ff97a3
44 changed files with 9720 additions and 8896 deletions
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@ -6,6 +6,7 @@ changes may not be included if they are not expected to break existing code.

 ## v0.20.2

+* Reworked parsing methods to avoid slow regexes (CVE-2024-22363)
 * HTML properly encode data-v attribute
 * SYLK read and write error cells

--- a/bits/01_version.js
+++ b/bits/01_version.js
@ -1 +1 @@
-XLSX.version = '0.20.1';
+XLSX.version = '0.20.2';
--- a/bits/10_ssf.js
+++ b/bits/10_ssf.js
@ -408,7 +408,7 @@ function write_num_f2(r/*:Array<string>*/, aval/*:number*/, sign/*:string*/)/*:s
 	return sign + (aval === 0 ? "" : ""+aval) + fill(" ", r[1].length + 2 + r[4].length);
 }
 var dec1 = /^#*0*\.([0#]+)/;
-var closeparen = /\).*[0#]/;
+var closeparen = /\)[^)]*[0#]/;
 var phone = /\(###\) ###\\?-####/;
 function hashq(str/*:string*/)/*:string*/ {
 	var o = "", cc;
--- a/bits/18_cfb.js
+++ b/bits/18_cfb.js
@ -1683,7 +1683,7 @@ function parse_mime(cfb/*:CFBContainer*/, data/*:Array<string>*/, root/*:string*
 	for(;di < 10; ++di) {
 		var line = data[di];
 		if(!line || line.match(/^\s*$/)) break;
-		var m = line.match(/^(.*?):\s*([^\s].*)$/);
+		var m = line.match(/^([^:]*?):\s*([^\s].*)$/);
 		if(m) switch(m[1].toLowerCase()) {
 			case "content-location": fname = m[2].trim(); break;
 			case "content-type": ctype = m[2].trim(); break;
--- a/bits/20_jsutils.js
+++ b/bits/20_jsutils.js
@ -87,7 +87,7 @@ function parseDate(str/*:string*/, date1904/*:boolean*/)/*:Date*/ {
 	if(m) return new Date(Date.UTC(+m[1], +m[2]-1, +m[3], 0, 0, 0, 0));
 	/* TODO: 1900-02-29T00:00:00.000 should return a flag to treat as a date code (affects xlml) */
 	m = str.match(pdre3);
-	if(m) return new Date(Date.UTC(+m[1], +m[2]-1, +m[3], +m[4], +m[5], ((m[6] && parseInt(m[6].slice(1), 10))|| 0), ((m[7] && parseInt(m[7].slice(1), 10))||0)));
+	if(m) return new Date(Date.UTC(+m[1], +m[2]-1, +m[3], +m[4], +m[5], ((m[6] && parseInt(m[6].slice(1), 10))|| 0), ((m[7] && parseInt((m[7] + "0000").slice(1,4), 10))||0)));
 	var d = new Date(str);
 	return d;
 }
@ -154,7 +154,7 @@ function fuzzynum(s/*:string*/)/*:number*/ {
 	var wt = 1;
 	var ss = s.replace(/([\d]),([\d])/g,"$1$2").replace(/[$]/g,"").replace(/[%]/g, function() { wt *= 100; return "";});
 	if(!isNaN(v = Number(ss))) return v / wt;
-	ss = ss.replace(/[(](.*)[)]/,function($$, $1) { wt = -wt; return $1;});
+	ss = ss.replace(/[(]([^()]*)[)]/,function($$, $1) { wt = -wt; return $1;});
 	if(!isNaN(v = Number(ss))) return v / wt;
 	return v;
 }
@ -224,6 +224,15 @@ function local_to_utc(local) {
 	return new Date(Date.UTC(local.getFullYear(), local.getMonth(), local.getDate(), local.getHours(), local.getMinutes(), local.getSeconds(), local.getMilliseconds()));
 }

+function remove_doctype(str) {
+	var preamble = str.slice(0, 1024);
+	var si = preamble.indexOf("<!DOCTYPE");
+	if(si == -1) return str;
+	var m = str.match(/<[\w]/);
+	if(!m) return str;
+	return str.slice(0, si) + str.slice(m.index);
+}
+
 /* str.match(/<!--[\s\S]*?-->/g) --> str_match_ng(str, "<!--", "-->") */
 function str_match_ng(str, s, e) {
  var out = [];
@ -257,11 +266,13 @@ function str_remove_ng(str, s, e) {
  return out.join("");
 }

-/* str.match(/<tag[^>]*?>([\s\S]*?)</tag>/) --> str_match_xml(str, "tag") */
+/* str.match(/<tag\b[^>]*?>([\s\S]*?)</tag>/) --> str_match_xml(str, "tag") */
+var xml_boundary = { " ": 1, "\t": 1, "\r": 1, "\n": 1, ">": 1 };
 function str_match_xml(str, tag) {
-	var si = str.indexOf('<' + tag);
+	var si = str.indexOf('<' + tag), w = tag.length + 1, L = str.length;
+	while(si >= 0 && si <= L - w && !xml_boundary[str.charAt(si + w)]) si = str.indexOf('<' + tag, si+1);
 	if(si === -1) return null;
-	var sf = str.indexOf(">", si);
+	var sf = str.indexOf(">", si + tag.length);
 	if(sf === -1) return null;
 	var et = "</" + tag + ">";
 	var ei = str.indexOf(et, sf);
@ -269,12 +280,13 @@ function str_match_xml(str, tag) {
 	return [str.slice(si, ei + et.length), str.slice(sf + 1, ei)];
 }

+/* str.match(/<(?:\w+:)?tag\b[^<>]*?>([\s\S]*?)<\/(?:\w+:)?tag>/) --> str_match_xml(str, "tag") */
 var str_match_xml_ns = /*#__PURE__*/(function() {
 	var str_match_xml_ns_cache = {};
 	return function str_match_xml_ns(str, tag) {
 		var res = str_match_xml_ns_cache[tag];
 		if(!res) str_match_xml_ns_cache[tag] = res = [
-			new RegExp('<(?:\\w+:)?'+tag+'[^>]*>', "g"),
+			new RegExp('<(?:\\w+:)?'+tag+'\\b[^<>]*>', "g"),
 			new RegExp('</(?:\\w+:)?'+tag+'>', "g")
 		];
 		res[0].lastIndex = res[1].lastIndex = 0;
@ -291,24 +303,74 @@ var str_match_xml_ns = /*#__PURE__*/(function() {
 	};
 })();

+/* str.match(/<(?:\w+:)?tag\b[^<>]*?>([\s\S]*?)<\/(?:\w+:)?tag>/g) --> str_match_xml_ns_g(str, "tag") */
 var str_match_xml_ns_g = /*#__PURE__*/(function() {
 	var str_match_xml_ns_cache = {};
 	return function str_match_xml_ns(str, tag) {
 		var out = [];
 		var res = str_match_xml_ns_cache[tag];
 		if(!res) str_match_xml_ns_cache[tag] = res = [
-			new RegExp('<(?:\\w+:)?'+tag+'[^>]*>', "g"),
+			new RegExp('<(?:\\w+:)?'+tag+'\\b[^<>]*>', "g"),
 			new RegExp('</(?:\\w+:)?'+tag+'>', "g")
 		];
 		res[0].lastIndex = res[1].lastIndex = 0;
 		var m;
 		while((m = res[0].exec(str))) {
 			var si = m.index;
-			var sf = res[0].lastIndex;
 			res[1].lastIndex = res[0].lastIndex;
 			m = res[1].exec(str);
 			if(!m) return null;
-			var ei = m.index;
+			var ef = res[1].lastIndex;
+			out.push(str.slice(si, ef));
+			res[0].lastIndex = res[1].lastIndex;
+		}
+		return out.length == 0 ? null : out;
+	};
+})();
+var str_remove_xml_ns_g = /*#__PURE__*/(function() {
+	var str_remove_xml_ns_cache = {};
+	return function str_remove_xml_ns_g(str, tag) {
+		var out = [];
+		var res = str_remove_xml_ns_cache[tag];
+		if(!res) str_remove_xml_ns_cache[tag] = res = [
+			new RegExp('<(?:\\w+:)?'+tag+'\\b[^<>]*>', "g"),
+			new RegExp('</(?:\\w+:)?'+tag+'>', "g")
+		];
+		res[0].lastIndex = res[1].lastIndex = 0;
+		var m;
+		var si = 0, ef = 0;
+		while((m = res[0].exec(str))) {
+			si = m.index;
+			out.push(str.slice(ef, si));
+			ef = si;
+			res[1].lastIndex = res[0].lastIndex;
+			m = res[1].exec(str);
+			if(!m) return null;
+			ef = res[1].lastIndex;
+			res[0].lastIndex = res[1].lastIndex;
+		}
+		out.push(str.slice(ef));
+		return out.length == 0 ? "" : out.join("");
+	};
+})();
+
+/* str.match(/<(?:\w+:)?tag\b[^>]*?>([\s\S]*?)<\/(?:\w+:)?tag>/gi) --> str_match_xml_ns_ig(str, "tag") */
+var str_match_xml_ig = /*#__PURE__*/(function() {
+	var str_match_xml_ns_cache = {};
+	return function str_match_xml_ns(str, tag) {
+		var out = [];
+		var res = str_match_xml_ns_cache[tag];
+		if(!res) str_match_xml_ns_cache[tag] = res = [
+			new RegExp('<'+tag+'\\b[^<>]*>', "ig"),
+			new RegExp('</'+tag+'>', "ig")
+		];
+		res[0].lastIndex = res[1].lastIndex = 0;
+		var m;
+		while((m = res[0].exec(str))) {
+			var si = m.index;
+			res[1].lastIndex = res[0].lastIndex;
+			m = res[1].exec(str);
+			if(!m) return null;
 			var ef = res[1].lastIndex;
 			out.push(str.slice(si, ef));
 			res[0].lastIndex = res[1].lastIndex;
--- a/bits/22_xmlutils.js
+++ b/bits/22_xmlutils.js
@ -1,6 +1,6 @@
 var XML_HEADER = '<?xml version="1.0" encoding="UTF-8" standalone="yes"?>\r\n';
-var attregexg=/([^"\s?>\/]+)\s*=\s*((?:")([^"]*)(?:")|(?:')([^']*)(?:')|([^'">\s]+))/g;
-var tagregex1=/<[\/\?]?[a-zA-Z0-9:_-]+(?:\s+[^"\s?>\/]+\s*=\s*(?:"[^"]*"|'[^']*'|[^'">\s=]+))*\s*[\/\?]?>/mg, tagregex2 = /<[^>]*>/g;
+var attregexg=/\s([^"\s?>\/]+)\s*=\s*((?:")([^"]*)(?:")|(?:')([^']*)(?:')|([^'">\s]+))/g;
+var tagregex1=/<[\/\?]?[a-zA-Z0-9:_-]+(?:\s+[^"\s?<>\/]+\s*=\s*(?:"[^"]*"|'[^']*'|[^'"<>\s=]+))*\s*[\/\?]?>/mg, tagregex2 = /<[^<>]*>/g;
 var tagregex = /*#__PURE__*/XML_HEADER.match(tagregex1) ? tagregex1 : tagregex2;
 var nsregex=/<\w*:/, nsregex2 = /<(\/?)\w+:/;
 function parsexmltag(tag/*:string*/, skip_root/*:?boolean*/, skip_LC/*:?boolean*/)/*:any*/ {
@ -11,7 +11,7 @@ function parsexmltag(tag/*:string*/, skip_root/*:?boolean*/, skip_LC/*:?boolean*
 	if(eq === tag.length) return z;
 	var m = tag.match(attregexg), j=0, v="", i=0, q="", cc="", quot = 1;
 	if(m) for(i = 0; i != m.length; ++i) {
-		cc = m[i];
+		cc = m[i].slice(1);
 		for(c=0; c != cc.length; ++c) if(cc.charCodeAt(c) === 61) break;
 		q = cc.slice(0,c).trim();
 		while(cc.charCodeAt(c+1) == 32) ++c;
@ -171,21 +171,21 @@ var htmldecode/*:{(s:string):string}*/ = /*#__PURE__*/(function() {
 				// Remove new lines and spaces from start of content
 				.replace(/^[\t\n\r ]+/, "")
 				// Remove new lines and spaces from end of content
-				.replace(/[\t\n\r ]+$/,"")
+				.replace(/(^|[^\t\n\r ])[\t\n\r ]+$/,"$1")
 				// Added line which removes any white space characters after and before html tags
-				.replace(/>\s+/g,">").replace(/\s+</g,"<")
+				.replace(/>\s+/g,">").replace(/\b\s+</g,"<")
 				// Replace remaining new lines and spaces with space
 				.replace(/[\t\n\r ]+/g, " ")
 				// Replace <br> tags with new lines
 				.replace(/<\s*[bB][rR]\s*\/?>/g,"\n")
 				// Strip HTML elements
-				.replace(/<[^>]*>/g,"");
+				.replace(/<[^<>]*>/g,"");
 		for(var i = 0; i < entities.length; ++i) o = o.replace(entities[i][0], entities[i][1]);
 		return o;
 	};
 })();

-var vtvregex = /<\/?(?:vt:)?variant>/g, vtmregex = /<(?:vt:)([^>]*)>([\s\S]*)</;
+var vtvregex = /<\/?(?:vt:)?variant>/g, vtmregex = /<(?:vt:)([^<"'>]*)>([\s\S]*)</;
 function parseVector(data/*:string*/, opts)/*:Array<{v:string,t:string}>*/ {
 	var h = parsexmltag(data);

@ -230,9 +230,8 @@ function xlml_normalize(d)/*:string*/ {
 	if(typeof Uint8Array !== 'undefined' && d instanceof Uint8Array) return utf8read(a2s(ab2a(d)));
 	throw new Error("Bad input format: expected Buffer or string");
 }
-/* UOS uses CJK in tags */
-var xlmlregex = /<(\/?)([^\s?><!\/:]*:|)([^\s?<>:\/]+)(?:[\s?:\/](?:[^>=]|="[^"]*?")*)?>/mg;
-//var xlmlregex = /<(\/?)([a-z0-9]*:|)(\w+)[^>]*>/mg;
+/* UOS uses CJK in tags, ODS uses invalid XML */
+var xlmlregex = /<([\/]?)([^\s?><!\/:"]*:|)([^\s?<>:\/"]+)(?:\s+[^<>=?"'\s]+="[^"]*?")*\s*[\/]?>/mg;

 var XMLNS = ({
 	CORE_PROPS: 'http://schemas.openxmlformats.org/package/2006/metadata/core-properties',
--- a/bits/27_csfutils.js
+++ b/bits/27_csfutils.js
@ -128,22 +128,24 @@ function sheet_add_aoa(_ws/*:?Worksheet*/, data/*:AOA*/, opts/*:?any*/)/*:Worksh
 			var _origin/*:CellAddress*/ = typeof o.origin == "string" ? decode_cell(o.origin) : o.origin;
 			_R = _origin.r; _C = _origin.c;
 		}
-		if(!ws["!ref"]) ws["!ref"] = "A1:A1";
 	}
 	var range/*:Range*/ = ({s: {c:10000000, r:10000000}, e: {c:0, r:0}}/*:any*/);
-	if(ws['!ref']) {
+	if(ws["!ref"]){
 		var _range = safe_decode_range(ws['!ref']);
 		range.s.c = _range.s.c;
 		range.s.r = _range.s.r;
 		range.e.c = Math.max(range.e.c, _range.e.c);
 		range.e.r = Math.max(range.e.r, _range.e.r);
-		if(_R == -1) range.e.r = _R = _range.e.r + 1;
+		if(_R == -1) range.e.r = _R = (ws["!ref"] ? _range.e.r + 1 : 0);
+	} else {
+		range.s.c = range.e.c = range.s.r = range.e.r = 0;
 	}
-	var row = [];
+	var row = [], seen = false;
 	for(var R = 0; R != data.length; ++R) {
 		if(!data[R]) continue;
 		if(!Array.isArray(data[R])) throw new Error("aoa_to_sheet expects an array of arrays");
 		var __R = _R + R, __Rstr = "" + (__R + 1);
+		//console.log("!!", R, _R, __R);
 		if(dense) {
 			if(!ws["!data"][__R]) ws["!data"][__R] = [];
 			row = ws["!data"][__R];
@ -156,6 +158,7 @@ function sheet_add_aoa(_ws/*:?Worksheet*/, data/*:AOA*/, opts/*:?any*/)/*:Worksh
 			if(range.s.c > __C) range.s.c = __C;
 			if(range.e.r < __R) range.e.r = __R;
 			if(range.e.c < __C) range.e.c = __C;
+			seen = true;
 			if(data[R][C] && typeof data[R][C] === 'object' && !Array.isArray(data[R][C]) && !(data[R][C] instanceof Date)) cell = data[R][C];
 			else {
 				if(Array.isArray(cell.v)) { cell.f = data[R][C][1]; cell.v = cell.v[0]; }
@ -185,7 +188,7 @@ function sheet_add_aoa(_ws/*:?Worksheet*/, data/*:AOA*/, opts/*:?any*/)/*:Worksh
 			}
 		}
 	}
-	if(range.s.c < 10000000) ws['!ref'] = encode_range(range);
+	if(seen && range.s.c < 10400000) ws['!ref'] = encode_range(range);
 	return ws;
 }
 function aoa_to_sheet(data/*:AOA*/, opts/*:?any*/)/*:Worksheet*/ { return sheet_add_aoa(null, data, opts); }
--- a/bits/35_custprops.js
+++ b/bits/35_custprops.js
@ -1,5 +1,5 @@
 /* 15.2.12.2 Custom File Properties Part */
-var custregex = /<[^>]+>[^<]*/g;
+var custregex = /<[^<>]+>[^<]*/g;
 function parse_cust_props(data/*:string*/, opts) {
 	var p = {}, name = "";
 	var m = data.match(custregex);
--- a/bits/38_xlstypes.js
+++ b/bits/38_xlstypes.js
@ -233,8 +233,8 @@ function parse_PropertySet(blob, PIDSI) {
 				/* [MS-OSHARED] 2.3.3.2.3.1.2 + PROPVARIANT */
 				switch(blob[blob.l]) {
 					case 0x41 /*VT_BLOB*/: blob.l += 4; val = parse_BLOB(blob); break;
-					case 0x1E /*VT_LPSTR*/: blob.l += 4; val = parse_VtString(blob, blob[blob.l-4]).replace(/\u0000+$/,""); break;
-					case 0x1F /*VT_LPWSTR*/: blob.l += 4; val = parse_VtString(blob, blob[blob.l-4]).replace(/\u0000+$/,""); break;
+					case 0x1E /*VT_LPSTR*/: blob.l += 4; val = parse_VtString(blob, blob[blob.l-4]).replace(/(^|[^\u0000])\u0000+$/,"$1"); break;
+					case 0x1F /*VT_LPWSTR*/: blob.l += 4; val = parse_VtString(blob, blob[blob.l-4]).replace(/(^|[^\u0000])\u0000+$/,"$1"); break;
 					case 0x03 /*VT_I4*/: blob.l += 4; val = blob.read_shift(4, 'i'); break;
 					case 0x13 /*VT_UI4*/: blob.l += 4; val = blob.read_shift(4); break;
 					case 0x05 /*VT_R8*/: blob.l += 4; val = blob.read_shift(8, 'f'); break;
--- a/bits/40_harb.js
+++ b/bits/40_harb.js
@ -118,7 +118,7 @@ function dbf_to_aoa(buf, opts)/*:AOA*/ {
 	var ww = l7 ? 32 : 11;
 	while(d.l < hend && d[d.l] != 0x0d) {
 		field = ({}/*:any*/);
-		field.name = (typeof $cptable !== "undefined" ? $cptable.utils.decode(current_cp, d.slice(d.l, d.l+ww)) : a2s(d.slice(d.l, d.l + ww))).replace(/[\u0000\r\n].*$/g,"");
+		field.name = (typeof $cptable !== "undefined" ? $cptable.utils.decode(current_cp, d.slice(d.l, d.l+ww)) : a2s(d.slice(d.l, d.l + ww))).replace(/[\u0000\r\n][\S\s]*$/g,"");
 		d.l += ww;
 		field.type = String.fromCharCode(d.read_shift(1));
 		if(ft != 0x02 && !l7) field.offset = d.read_shift(4);
@ -176,7 +176,7 @@ function dbf_to_aoa(buf, opts)/*:AOA*/ {
 			switch(fields[C].type) {
 				case 'C':
 					// NOTE: it is conventional to write '  /  /  ' for empty dates
-					if(s.trim().length) out[R][C] = s.replace(/\s+$/,"");
+					if(s.trim().length) out[R][C] = s.replace(/([^\s])\s+$/,"$1");
 					break;
 				case 'D':
 					if(s.length === 8) {
@ -1121,7 +1121,7 @@ function read_wb_ID(d, opts) {
 		return out;
 	} catch(e) {
 		o.WTF = OLD_WTF;
-		if(!e.message.match(/SYLK bad record ID/) && OLD_WTF) throw e;
+		if((e.message.indexOf("SYLK bad record ID") == -1) && OLD_WTF) throw e;
 		return PRN.to_workbook(d, opts);
 	}
 }
--- a/bits/42_sstxml.js
+++ b/bits/42_sstxml.js
@ -4,7 +4,7 @@ function parse_rpr(rpr) {
 	var pass = false;
 	if(m) for(;i!=m.length; ++i) {
 		var y = parsexmltag(m[i]);
-		switch(y[0].replace(/\w*:/g,"")) {
+		switch(y[0].replace(/<\w*:/g,"<")) {
 			/* 18.8.12 condense CT_BooleanProperty */
 			/* ** not required . */
 			case '<condense': break;
@ -167,8 +167,7 @@ var rs_to_html = /*#__PURE__*/(function parse_rs_factory() {
 })();

 /* 18.4.8 si CT_Rst */
-var sitregex = /<(?:\w+:)?t[^>]*>([^<]*)<\/(?:\w+:)?t>/g, sirregex = /<(?:\w+:)?r\b[^>]*>/;
-var sirphregex = /<(?:\w+:)?rPh.*?>([\s\S]*?)<\/(?:\w+:)?rPh>/g;
+var sitregex = /<(?:\w+:)?t\b[^<>]*>([^<]*)<\/(?:\w+:)?t>/g, sirregex = /<(?:\w+:)?r\b[^<>]*>/;
 function parse_si(x, opts) {
 	var html = opts ? opts.cellHTML : true;
 	var z = {};
@ -184,7 +183,7 @@ function parse_si(x, opts) {
 	/* 18.4.4 r CT_RElt (Rich Text Run) */
 	else if((/*y = */x.match(sirregex))) {
 		z.r = utf8read(x);
-		z.t = unescapexml(utf8read((x.replace(sirphregex, '').match(sitregex)||[]).join("").replace(tagregex,"")), true);
+		z.t = unescapexml(utf8read((str_remove_xml_ns_g(x, "rPh").match(sitregex)||[]).join("").replace(tagregex,"")), true);
 		if(html) z.h = rs_to_html(parse_rs(z.r));
 	}
 	/* 18.4.3 phoneticPr CT_PhoneticPr (TODO: needed for Asian support) */
@ -193,21 +192,20 @@ function parse_si(x, opts) {
 }

 /* 18.4 Shared String Table */
-var sstr0 = /<(?:\w+:)?sst([^>]*)>([\s\S]*)<\/(?:\w+:)?sst>/;
 var sstr1 = /<(?:\w+:)?(?:si|sstItem)>/g;
 var sstr2 = /<\/(?:\w+:)?(?:si|sstItem)>/;
 function parse_sst_xml(data/*:string*/, opts)/*:SST*/ {
 	var s/*:SST*/ = ([]/*:any*/), ss = "";
 	if(!data) return s;
 	/* 18.4.9 sst CT_Sst */
-	var sst = data.match(sstr0);
+	var sst = str_match_xml_ns(data, "sst");
 	if(sst) {
-		ss = sst[2].replace(sstr1,"").split(sstr2);
+		ss = sst[1].replace(sstr1,"").split(sstr2);
 		for(var i = 0; i != ss.length; ++i) {
 			var o = parse_si(ss[i].trim(), opts);
 			if(o != null) s[s.length] = o;
 		}
-		sst = parsexmltag(sst[1]); s.Count = sst.count; s.Unique = sst.uniqueCount;
+		sst = parsexmltag(sst[0].slice(0, sst[0].indexOf(">"))); s.Count = sst.count; s.Unique = sst.uniqueCount;
 	}
 	return s;
 }
--- a/bits/47_styxml.js
+++ b/bits/47_styxml.js
@ -3,7 +3,7 @@ function parse_borders(t, styles, themes, opts) {
 	styles.Borders = [];
 	var border = {};
 	var pass = false;
-	(t[0].match(tagregex)||[]).forEach(function(x) {
+	(t.match(tagregex)||[]).forEach(function(x) {
 		var y = parsexmltag(x);
 		switch(strip_ns(y[0])) {
 			case '<borders': case '<borders>': case '</borders>': break;
@ -78,7 +78,7 @@ function parse_fills(t, styles, themes, opts) {
 	styles.Fills = [];
 	var fill = {};
 	var pass = false;
-	(t[0].match(tagregex)||[]).forEach(function(x) {
+	(t.match(tagregex)||[]).forEach(function(x) {
 		var y = parsexmltag(x);
 		switch(strip_ns(y[0])) {
 			case '<fills': case '<fills>': case '</fills>': break;
@ -144,7 +144,7 @@ function parse_fonts(t, styles, themes, opts) {
 	styles.Fonts = [];
 	var font = {};
 	var pass = false;
-	(t[0].match(tagregex)||[]).forEach(function(x) {
+	(t.match(tagregex)||[]).forEach(function(x) {
 		var y = parsexmltag(x);
 		switch(strip_ns(y[0])) {
 			case '<fonts': case '<fonts>': case '</fonts>': break;
@ -264,7 +264,7 @@ function parse_numFmts(t, styles, opts) {
 	styles.NumberFmt = [];
 	var k/*Array<number>*/ = (keys(table_fmt)/*:any*/);
 	for(var i=0; i < k.length; ++i) styles.NumberFmt[k[i]] = table_fmt[k[i]];
-	var m = t[0].match(tagregex);
+	var m = t.match(tagregex);
 	if(!m) return;
 	for(i=0; i < m.length; ++i) {
 		var y = parsexmltag(m[i]);
@ -305,7 +305,7 @@ function parse_cellXfs(t, styles, opts) {
 	styles.CellXf = [];
 	var xf;
 	var pass = false;
-	(t[0].match(tagregex)||[]).forEach(function(x) {
+	(t.match(tagregex)||[]).forEach(function(x) {
 		var y = parsexmltag(x), i = 0;
 		switch(strip_ns(y[0])) {
 			case '<cellXfs': case '<cellXfs>': case '<cellXfs/>': case '</cellXfs>': break;
@ -370,36 +370,31 @@ function write_cellXfs(cellXfs)/*:string*/ {

 /* 18.8 Styles CT_Stylesheet*/
 var parse_sty_xml= /*#__PURE__*/(function make_pstyx() {
-var numFmtRegex = /<(?:\w+:)?numFmts([^>]*)>[\S\s]*?<\/(?:\w+:)?numFmts>/;
-var cellXfRegex = /<(?:\w+:)?cellXfs([^>]*)>[\S\s]*?<\/(?:\w+:)?cellXfs>/;
-var fillsRegex = /<(?:\w+:)?fills([^>]*)>[\S\s]*?<\/(?:\w+:)?fills>/;
-var fontsRegex = /<(?:\w+:)?fonts([^>]*)>[\S\s]*?<\/(?:\w+:)?fonts>/;
-var bordersRegex = /<(?:\w+:)?borders([^>]*)>[\S\s]*?<\/(?:\w+:)?borders>/;

 return function parse_sty_xml(data, themes, opts) {
 	var styles = {};
 	if(!data) return styles;
-	data = str_remove_ng(data, "<!--", "-->").replace(/<!DOCTYPE[^\[]*\[[^\]]*\]>/gm,"");
+	data = remove_doctype(str_remove_ng(data, "<!--", "-->"));
 	/* 18.8.39 styleSheet CT_Stylesheet */
 	var t;

 	/* 18.8.31 numFmts CT_NumFmts ? */
-	if((t=data.match(numFmtRegex))) parse_numFmts(t, styles, opts);
+	if((t=str_match_xml_ns(data, "numFmts"))) parse_numFmts(t[0], styles, opts);

 	/* 18.8.23 fonts CT_Fonts ? */
-	if((t=data.match(fontsRegex))) parse_fonts(t, styles, themes, opts);
+	if((t=str_match_xml_ns(data, "fonts"))) parse_fonts(t[0], styles, themes, opts);

 	/* 18.8.21 fills CT_Fills ? */
-	if((t=data.match(fillsRegex))) parse_fills(t, styles, themes, opts);
+	if((t=str_match_xml_ns(data, "fills"))) parse_fills(t[0], styles, themes, opts);

 	/* 18.8.5  borders CT_Borders ? */
-	if((t=data.match(bordersRegex))) parse_borders(t, styles, themes, opts);
+	if((t=str_match_xml_ns(data, "borders"))) parse_borders(t[0], styles, themes, opts);

 	/* 18.8.9  cellStyleXfs CT_CellStyleXfs ? */
 	/* 18.8.8  cellStyles CT_CellStyles ? */

 	/* 18.8.10 cellXfs CT_CellXfs ? */
-	if((t=data.match(cellXfRegex))) parse_cellXfs(t, styles, opts);
+	if((t=str_match_xml_ns(data, "cellXfs"))) parse_cellXfs(t[0], styles, opts);

 	/* 18.8.15 dxfs CT_Dxfs ? */
 	/* 18.8.42 tableStyles CT_TableStyles ? */
--- a/bits/49_theme.js
+++ b/bits/49_theme.js
@ -70,30 +70,24 @@ function parse_fontScheme(/*::t, themes, opts*/) { }
 /* 20.1.4.1.15 fmtScheme CT_StyleMatrix */
 function parse_fmtScheme(/*::t, themes, opts*/) { }

-var clrsregex = /<a:clrScheme([^>]*)>[\s\S]*<\/a:clrScheme>/;
-var fntsregex = /<a:fontScheme([^>]*)>[\s\S]*<\/a:fontScheme>/;
-var fmtsregex = /<a:fmtScheme([^>]*)>[\s\S]*<\/a:fmtScheme>/;
-
 /* 20.1.6.10 themeElements CT_BaseStyles */
 function parse_themeElements(data, themes, opts) {
 	themes.themeElements = {};

 	var t;

-	[
-		/* clrScheme CT_ColorScheme */
-		['clrScheme', clrsregex, parse_clrScheme],
-		/* fontScheme CT_FontScheme */
-		['fontScheme', fntsregex, parse_fontScheme],
-		/* fmtScheme CT_StyleMatrix */
-		['fmtScheme', fmtsregex, parse_fmtScheme]
-	].forEach(function(m) {
-		if(!(t=data.match(m[1]))) throw new Error(m[0] + ' not found in themeElements');
-		m[2](t, themes, opts);
-	});
-}
+	/* clrScheme CT_ColorScheme */
+	if(!(t=str_match_xml(data, "a:clrScheme"))) throw new Error('clrScheme not found in themeElements');
+	parse_clrScheme(t, themes, opts);

-var themeltregex = /<a:themeElements([^>]*)>[\s\S]*<\/a:themeElements>/;
+	/* fontScheme CT_FontScheme */
+	if(!(t=str_match_xml(data, "a:fontScheme"))) throw new Error('fontScheme not found in themeElements');
+	parse_fontScheme(t, themes, opts);
+
+	/* fmtScheme CT_StyleMatrix */
+	if(!(t=str_match_xml(data, "a:fmtScheme"))) throw new Error('fmtScheme not found in themeElements');
+	parse_fmtScheme(t, themes, opts);
+}

 /* 14.2.7 Theme Part */
 function parse_theme_xml(data/*:string*/, opts) {
@ -104,7 +98,7 @@ function parse_theme_xml(data/*:string*/, opts) {
 	var themes = {};

 	/* themeElements CT_BaseStyles */
-	if(!(t=data.match(themeltregex))) throw new Error('themeElements not found in theme');
+	if(!(t=str_match_xml(data, "a:themeElements"))) throw new Error('themeElements not found in theme');
 	parse_themeElements(t[0], themes, opts);
 	themes.raw = data;
 	return themes;
--- a/bits/55_vml.js
+++ b/bits/55_vml.js
@ -1,8 +1,7 @@
 /* L.5.5.2 SpreadsheetML Comments + VML Schema */
-var shapevmlregex = /<(?:\w+:)?shape(?:[^\w][^>]*)?>([\s\S]*?)<\/(?:\w+:)?shape>/g;
 function parse_vml(data/*:string*/, sheet, comments) {
 	var cidx = 0;
-	(data.match(shapevmlregex)||[]).forEach(function(m) {
+	(str_match_xml_ns_g(data, "shape")||[]).forEach(function(m) {
 		var type = "";
 		var hidden = true;
 		var aidx = -1;
--- a/bits/57_cmntxml.js
+++ b/bits/57_cmntxml.js
@ -4,22 +4,22 @@ function parse_comments_xml(data/*:string*/, opts)/*:Array<RawComment>*/ {
 	if(data.match(/<(?:\w+:)?comments *\/>/)) return [];
 	var authors/*:Array<string>*/ = [];
 	var commentList/*:Array<RawComment>*/ = [];
-	var authtag = data.match(/<(?:\w+:)?authors>([\s\S]*)<\/(?:\w+:)?authors>/);
+	var authtag = str_match_xml_ns(data, "authors");
 	if(authtag && authtag[1]) authtag[1].split(/<\/\w*:?author>/).forEach(function(x) {
 		if(x === "" || x.trim() === "") return;
-		var a = x.match(/<(?:\w+:)?author[^>]*>(.*)/);
+		var a = x.match(/<(?:\w+:)?author[^<>]*>(.*)/);
 		if(a) authors.push(a[1]);
 	});
-	var cmnttag = data.match(/<(?:\w+:)?commentList>([\s\S]*)<\/(?:\w+:)?commentList>/);
+	var cmnttag = str_match_xml_ns(data, "commentList");
 	if(cmnttag && cmnttag[1]) cmnttag[1].split(/<\/\w*:?comment>/).forEach(function(x) {
 		if(x === "" || x.trim() === "") return;
-		var cm = x.match(/<(?:\w+:)?comment[^>]*>/);
+		var cm = x.match(/<(?:\w+:)?comment[^<>]*>/);
 		if(!cm) return;
 		var y = parsexmltag(cm[0]);
 		var comment/*:RawComment*/ = ({ author: y.authorId && authors[y.authorId] || "sheetjsghost", ref: y.ref, guid: y.guid }/*:any*/);
 		var cell = decode_cell(y.ref);
 		if(opts.sheetRows && opts.sheetRows <= cell.r) return;
-		var textMatch = x.match(/<(?:\w+:)?text>([\s\S]*)<\/(?:\w+:)?text>/);
+		var textMatch = str_match_xml_ns(x, "text");
 		var rt = !!textMatch && !!textMatch[1] && parse_si(textMatch[1]) || {r:"",t:"",h:""};
 		comment.r = rt.r;
 		if(rt.r == "<t></t>") rt.t = rt.h = "";
--- a/bits/59_vba.js
+++ b/bits/59_vba.js
@ -13,7 +13,7 @@ function fill_vba_xls(cfb, vba) {
  vba.FullPaths.forEach(function(p, i) {
    if (i == 0)
      return;
-    var newpath = p.replace(/[^\/]*[\/]/, "/_VBA_PROJECT_CUR/");
+    var newpath = p.replace(/^[\/]*[^\/]*[\/]/, "/_VBA_PROJECT_CUR/");
    if (newpath.slice(-1) !== "/")
      CFB.utils.cfb_add(cfb, newpath, vba.FileIndex[i].content);
  });
--- a/bits/67_wsxml.js
+++ b/bits/67_wsxml.js
@ -2,16 +2,13 @@ function parse_ws_xml_dim(ws/*:Worksheet*/, s/*:string*/) {
 	var d = safe_decode_range(s);
 	if(d.s.r<=d.e.r && d.s.c<=d.e.c && d.s.r>=0 && d.s.c>=0) ws["!ref"] = encode_range(d);
 }
-var mergecregex = /<(?:\w:)?mergeCell ref=["'][A-Z0-9:]+['"]\s*[\/]?>/g;
-var sheetdataregex = /<(?:\w+:)?sheetData[^>]*>([\s\S]*)<\/(?:\w+:)?sheetData>/;
-var hlinkregex = /<(?:\w:)?hyperlink [^>]*>/mg;
+var mergecregex = /<(?:\w+:)?mergeCell ref=["'][A-Z0-9:]+['"]\s*[\/]?>/g;
+var hlinkregex = /<(?:\w+:)?hyperlink [^<>]*>/mg;
 var dimregex = /"(\w*:\w*)"/;
-var colregex = /<(?:\w:)?col\b[^>]*[\/]?>/g;
-var afregex = /<(?:\w:)?autoFilter[^>]*([\/]|>([\s\S]*)<\/(?:\w:)?autoFilter)>/g;
-var marginregex= /<(?:\w:)?pageMargins[^>]*\/>/g;
-var sheetprregex = /<(?:\w:)?sheetPr\b(?:[^>a-z][^>]*)?\/>/;
-var sheetprregex2= /<(?:\w:)?sheetPr[^>]*(?:[\/]|>([\s\S]*)<\/(?:\w:)?sheetPr)>/;
-var svsregex = /<(?:\w:)?sheetViews[^>]*(?:[\/]|>([\s\S]*)<\/(?:\w:)?sheetViews)>/;
+var colregex = /<(?:\w+:)?col\b[^<>]*[\/]?>/g;
+var afregex = /<(?:\w+:)?autoFilter[^>]*/g;
+var marginregex= /<(?:\w+:)?pageMargins[^<>]*\/>/g;
+var sheetprregex = /<(?:\w+:)?sheetPr\b[^<>]*?\/>/;

 /* 18.3 Worksheets */
 function parse_ws_xml(data/*:?string*/, opts, idx/*:number*/, rels, wb/*:WBWBProps*/, themes, styles)/*:Worksheet*/ {
@ -24,7 +21,7 @@ function parse_ws_xml(data/*:?string*/, opts, idx/*:number*/, rels, wb/*:WBWBPro
 	var refguess/*:Range*/ = ({s: {r:2000000, c:2000000}, e: {r:0, c:0} }/*:any*/);

 	var data1 = "", data2 = "";
-	var mtch/*:?any*/ = data.match(sheetdataregex);
+	var mtch/*:?any*/ = str_match_xml_ns(data, "sheetData");
 	if(mtch) {
 		data1 = data.slice(0, mtch.index);
 		data2 = data.slice(mtch.index + mtch[0].length);
@ -33,7 +30,7 @@ function parse_ws_xml(data/*:?string*/, opts, idx/*:number*/, rels, wb/*:WBWBPro
 	/* 18.3.1.82 sheetPr CT_SheetPr */
 	var sheetPr = data1.match(sheetprregex);
 	if(sheetPr) parse_ws_xml_sheetpr(sheetPr[0], s, wb, idx);
-	else if((sheetPr = data1.match(sheetprregex2))) parse_ws_xml_sheetpr2(sheetPr[0], sheetPr[1]||"", s, wb, idx, styles, themes);
+	else if((sheetPr = str_match_xml_ns(data1, "sheetPr"))) parse_ws_xml_sheetpr2(sheetPr[0], sheetPr[1]||"", s, wb, idx, styles, themes);

 	/* 18.3.1.35 dimension CT_SheetDimension */
 	var ridx = (data1.match(/<(?:\w*:)?dimension/)||{index:-1}).index;
@ -43,7 +40,7 @@ function parse_ws_xml(data/*:?string*/, opts, idx/*:number*/, rels, wb/*:WBWBPro
 	}

 	/* 18.3.1.88 sheetViews CT_SheetViews */
-	var svs = data1.match(svsregex);
+	var svs = str_match_xml_ns(data1, "sheetViews");
 	if(svs && svs[1]) parse_ws_xml_sheetviews(svs[1], wb);

 	/* 18.3.1.17 cols CT_Cols */
@ -242,7 +239,7 @@ function write_ws_xml_autofilter(data, ws, wb, idx)/*:string*/ {

 /* 18.3.1.88 sheetViews CT_SheetViews */
 /* 18.3.1.87 sheetView CT_SheetView */
-var sviewregex = /<(?:\w:)?sheetView(?:[^>a-z][^>]*)?\/?>/g;
+var sviewregex = /<(?:\w:)?sheetView(?:[^<>a-z][^<>]*)?\/?>/g;
 function parse_ws_xml_sheetviews(data, wb/*:WBWBProps*/) {
 	if(!wb.Views) wb.Views = [{}];
 	(data.match(sviewregex)||[]).forEach(function(r/*:string*/, i/*:number*/) {
@ -321,7 +318,7 @@ function write_ws_xml_cell(cell/*:Cell*/, ref, ws, opts, idx, wb, date1904)/*:st

 var parse_ws_xml_data = /*#__PURE__*/(function() {
 	var cellregex = /<(?:\w+:)?c[ \/>]/, rowregex = /<\/(?:\w+:)?row>/;
-	var rregex = /r=["']([^"']*)["']/, isregex = /<(?:\w+:)?is>([\S\s]*?)<\/(?:\w+:)?is>/;
+	var rregex = /r=["']([^"']*)["']/;
 	var refregex = /ref=["']([^"']*)["']/;

 return function parse_ws_xml_data(sdata/*:string*/, s, opts, guess/*:Range*/, themes, styles, wb) {
@ -425,7 +422,7 @@ return function parse_ws_xml_data(sdata/*:string*/, s, opts, guess/*:Range*/, th
 							sharedf[parseInt(ftag.si, 10)] = [ftag, ___f, tag.r];
 						}
 					}
-				} else if((cref=d.match(/<f[^>]*\/>/))) {
+				} else if((cref=d.match(/<f[^<>]*\/>/))) {
 					ftag = parsexmltag(cref[0]);
 					if(sharedf[ftag.si]) p.f = shift_formula_xlsx(sharedf[ftag.si][1], sharedf[ftag.si][2]/*[0].ref*/, tag.r);
 				}
@ -471,7 +468,7 @@ return function parse_ws_xml_data(sdata/*:string*/, s, opts, guess/*:Range*/, th
 					if(opts.cellHTML) p.h = escapehtml(p.v);
 					break;
 				case 'inlineStr':
-					cref = d.match(isregex);
+					cref = str_match_xml_ns(d, "is");
 					p.t = 's';
 					if(cref != null && (sstr = parse_si(cref[1]))) {
 						p.v = sstr.t;
@ -636,7 +633,7 @@ function write_ws_xml(idx/*:number*/, opts, wb/*:Workbook*/, rels)/*:string*/ {
 			if(!l[1].Target) return;
 			rel = ({"ref":l[0]}/*:any*/);
 			if(l[1].Target.charAt(0) != "#") {
-				rId = add_rels(rels, -1, escapexml(l[1].Target).replace(/#.*$/, ""), RELS.HLINK);
+				rId = add_rels(rels, -1, escapexml(l[1].Target).replace(/#[\s\S]*$/, ""), RELS.HLINK);
 				rel["r:id"] = "rId"+rId;
 			}
 			if((relc = l[1].Target.indexOf("#")) > -1) rel.location = escapexml(l[1].Target.slice(relc+1));
--- a/bits/68_wsbin.js
+++ b/bits/68_wsbin.js
@ -924,7 +924,7 @@ function write_HLINKS(ba, ws/*:Worksheet*/, rels) {
 	/* *BrtHLink */
 	ws['!links'].forEach(function(l) {
 		if(!l[1].Target) return;
-		var rId = add_rels(rels, -1, l[1].Target.replace(/#.*$/, ""), RELS.HLINK);
+		var rId = add_rels(rels, -1, l[1].Target.replace(/#[\s\S]*$/, ""), RELS.HLINK);
 		write_record(ba, 0x01EE /* BrtHLink */, write_BrtHLink(l, rId));
 	});
 	delete ws['!links'];
--- a/bits/69_chartxml.js
+++ b/bits/69_chartxml.js
@ -4,16 +4,16 @@ function parse_Cache(data/*:string*/)/*:[Array<number|string>, string, ?string]*
 	var f;

 	/* 21.2.2.150 pt CT_NumVal */
-	(data.match(/<c:pt idx="(\d*)">(.*?)<\/c:pt>/mg)||[]).forEach(function(pt) {
-		var q = pt.match(/<c:pt idx="(\d*?)"><c:v>(.*)<\/c:v><\/c:pt>/);
+	(data.match(/<c:pt idx="(\d*)"[^<>\/]*><c:v>([^<])<\/c:v><\/c:pt>/mg)||[]).forEach(function(pt) {
+		var q = pt.match(/<c:pt idx="(\d*)"[^<>\/]*><c:v>([^<]*)<\/c:v><\/c:pt>/);
 		if(!q) return;
 		col[+q[1]] = num ? +q[2] : q[2];
 	});

 	/* 21.2.2.71 formatCode CT_Xstring */
-	var nf = unescapexml((data.match(/<c:formatCode>([\s\S]*?)<\/c:formatCode>/) || ["","General"])[1]);
+	var nf = unescapexml((str_match_xml(data, "c:formatCode") || ["","General"])[1]);

-	(data.match(/<c:f>(.*?)<\/c:f>/mg)||[]).forEach(function(F) { f = F.replace(/<.*?>/g,""); });
+	(str_match_ng(data, "<c:f>", "</c:f>")||[]).forEach(function(F) { f = F.replace(/<[^<>]*>/g,""); });

 	return [col, nf, f];
 }
@ -28,7 +28,7 @@ function parse_chart(data/*:?string*/, name/*:string*/, opts, rels, wb, csheet)
 	var refguess = {s: {r:2000000, c:2000000}, e: {r:0, c:0} };

 	/* 21.2.2.120 numCache CT_NumData */
-	(data.match(/<c:numCache>[\s\S]*?<\/c:numCache>/gm)||[]).forEach(function(nc) {
+	(str_match_ng(data, "<c:numCache>", "</c:numCache>")||[]).forEach(function(nc) {
 		var cache = parse_Cache(nc);
 		refguess.s.r = refguess.s.c = 0;
 		refguess.e.c = C;
--- a/bits/75_xlml.js
+++ b/bits/75_xlml.js
@ -1,5 +1,5 @@
-var attregexg2=/([\w:]+)=((?:")([^"]*)(?:")|(?:')([^']*)(?:'))/g;
-var attregex2=/([\w:]+)=((?:")(?:[^"]*)(?:")|(?:')(?:[^']*)(?:'))/;
+var attregexg2=/\b((?:\w+:)?[\w]+)=((?:")([^"]*)(?:")|(?:')([^']*)(?:'))/g;
+var attregex2=/\b((?:\w+:)?[\w]+)=((?:")(?:[^"]*)(?:")|(?:')(?:[^']*)(?:'))/;
 function xlml_parsexmltag(tag/*:string*/, skip_root/*:?boolean*/) {
 	var words = tag.split(/\s+/);
 	var z/*:any*/ = ([]/*:any*/); if(!skip_root) z[0] = words[0];
@ -112,7 +112,7 @@ function parse_xlml_data(xml, ss, data, cell/*:any*/, base, styles, csty, row, a
 			break;
 		case 'String':
 			cell.t = 's'; cell.r = xlml_fixstr(unescapexml(xml));
-			cell.v = (xml.indexOf("<") > -1 ? unescapexml(ss||xml).replace(/<.*?>/g, "") : cell.r); // todo: BR etc
+			cell.v = (xml.indexOf("<") > -1 ? unescapexml(ss||xml).replace(/<[^<>]*>/g, "") : cell.r); // todo: BR etc
 			break;
 		case 'DateTime':
 			if(xml.slice(-1) != "Z") xml += "Z";
--- a/bits/79_html.js
+++ b/bits/79_html.js
@ -8,7 +8,7 @@ function html_to_sheet(str/*:string*/, _opts)/*:Workbook*/ {
 	if(!mtch) throw new Error("Invalid HTML: could not find <table>");
 	var mtch2/*:any*/ = str.match(/<\/table/i);
 	var i/*:number*/ = mtch.index, j/*:number*/ = mtch2 && mtch2.index || str.length;
-	var rows = split_regex(str.slice(i, j), /(:?<tr[^>]*>)/i, "<tr>");
+	var rows = split_regex(str.slice(i, j), /(:?<tr[^<>]*>)/i, "<tr>");
 	var R = -1, C = 0, RS = 0, CS = 0;
 	var range/*:Range*/ = {s:{r:10000000, c:10000000},e:{r:0,c:0}};
 	var merges/*:Array<Range>*/ = [];
@ -99,7 +99,7 @@ var HTML_BEGIN = '<html><head><meta charset="utf-8"/><title>SheetJS Table Export
 var HTML_END = '</body></html>';

 function html_to_workbook(str/*:string*/, opts)/*:Workbook*/ {
-	var mtch = str.match(/<table[\s\S]*?>[\s\S]*?<\/table>/gi);
+	var mtch = str_match_xml_ig(str, "table");
 	if(!mtch || mtch.length == 0) throw new Error("Invalid HTML: could not find <table>");
 	if(mtch.length == 1) {
 		var w = sheet_to_workbook(html_to_sheet(mtch[0], opts), opts);
--- a/bits/80_parseods.js
+++ b/bits/80_parseods.js
@ -5,9 +5,9 @@ function parse_text_p(text/*:string*//*::, tag*/)/*:Array<any>*/ {
 		.replace(/[\t\r\n]/g, " ").trim().replace(/ +/g, " ")
 		.replace(/<text:s\/>/g," ")
 		.replace(/<text:s text:c="(\d+)"\/>/g, function($$,$1) { return Array(parseInt($1,10)+1).join(" "); })
-		.replace(/<text:tab[^>]*\/>/g,"\t")
+		.replace(/<text:tab[^<>]*\/>/g,"\t")
 		.replace(/<text:line-break\/>/g,"\n");
-	var v = unescapexml(fixed.replace(/<[^>]*>/g,""));
+	var v = unescapexml(fixed.replace(/<[^<>]*>/g,""));

 	return [v];
 }
@ -17,10 +17,10 @@ function parse_ods_styles(d/*:string*/, _opts, _nfm) {
 	var number_format_map = _nfm || {};
 	var str = xlml_normalize(d);
 	xlmlregex.lastIndex = 0;
-	str = str_remove_ng(str, "<!--", "-->").replace(/<!DOCTYPE[^\[]*\[[^\]]*\]>/gm,"");
+	str = remove_doctype(str_remove_ng(str, "<!--", "-->"));
 	var Rn, NFtag, NF = "", tNF = "", y, etpos = 0, tidx = -1, infmt = false, payload = "";
 	while((Rn = xlmlregex.exec(str))) {
-		switch((Rn[3]=Rn[3].replace(/_.*$/,""))) {
+		switch((Rn[3]=Rn[3].replace(/_[\s\S]*$/,""))) {
 		/* Number Format Definitions */
 		case 'number-style': // <number:number-style> 16.29.2
 		case 'currency-style': // <number:currency-style> 16.29.8
@ -264,8 +264,8 @@ function parse_content_xml(d/*:string*/, _opts, _nfm)/*:Workbook*/ {
 		var isstub = false, intable = false;
 		var i = 0;
 		xlmlregex.lastIndex = 0;
-		str = str_remove_ng(str, "<!--", "-->").replace(/<!DOCTYPE[^\[]*\[[^\]]*\]>/gm,"");
-		while((Rn = xlmlregex.exec(str))) switch((Rn[3]=Rn[3].replace(/_.*$/,""))) {
+		str = remove_doctype(str_remove_ng(str, "<!--", "-->"));
+		while((Rn = xlmlregex.exec(str))) switch((Rn[3]=Rn[3].replace(/_[\s\S]*$/,""))) {

 			case 'table': case '工作表': // 9.1.2 <table:table>
 				if(Rn[1]==='/') {
--- a/bits/81_writeods.js
+++ b/bits/81_writeods.js
@ -67,7 +67,7 @@ function write_number_format_ods(nf/*:string*/, nfidx/*:string*/)/*:string*/ {
 		if((t=nf.match(/# (\?+)\/(\d+)/))) { payload += writextag("number:fraction", null, {"number:min-integer-digits":0, "number:min-numerator-digits": t[1].length, "number:denominator-value": +t[2]}); break j; }

 		/* percentages */
-		if((t=nf.match(/(\d+)(|\.\d+)%/))) { type = "percentage"; payload += writextag("number:number", null, {"number:decimal-places": t[2] && t.length - 1 || 0, "number:min-decimal-places": t[2] && t.length - 1 || 0, "number:min-integer-digits": t[1].length }) + "<number:text>%</number:text>"; break j; }
+		if((t=nf.match(/\b(\d+)(|\.\d+)%/))) { type = "percentage"; payload += writextag("number:number", null, {"number:decimal-places": t[2] && t.length - 1 || 0, "number:min-decimal-places": t[2] && t.length - 1 || 0, "number:min-integer-digits": t[1].length }) + "<number:text>%</number:text>"; break j; }

 		/* datetime */
 		var has_time = false;
@ -196,7 +196,7 @@ function write_names_ods(Names, SheetNames, idx) {
 		return "        " + writextag("table:named-range", null, {
 			"table:name": name.Name,
 			"table:cell-range-address": odsref,
-			"table:base-cell-address": odsref.replace(/[\.]?[^\.]*$/, ".$A$1")
+			"table:base-cell-address": odsref.replace(/[\.][^\.]*$/, ".$A$1")
 		});
 	}).join("\n") + "\n      </table:named-expressions>\n";
 }
@ -436,7 +436,7 @@ var write_content_ods/*:{(wb:any, opts:any):string}*/ = /* @__PURE__ */(function

 		if(opts.bookType == "fods") {
 			o.push('<office:document' + attr + fods + '>\n');
-			o.push(write_meta_ods().replace(/<office:document-meta.*?>/, "").replace(/<\/office:document-meta>/, "") + "\n");
+			o.push(write_meta_ods().replace(/<office:document-meta[^<>]*?>/, "").replace(/<\/office:document-meta>/, "") + "\n");
 			// TODO: settings (equiv of settings.xml for ODS)
 		} else o.push('<office:document-content' + attr  + '>\n');
 		// o.push('  <office:scripts/>\n');
--- a/bits/83_numbers.js
+++ b/bits/83_numbers.js
@ -1512,7 +1512,7 @@ function numbers_add_meta(mlist, newid, newloc) {
  mlist[3].push({ type: 2, data: write_shallow([
    [],
    [{ type: 0, data: write_varint49(newid) }],
-    [{ type: 2, data: stru8(newloc.replace(/-.*$/, "")) }],
+    [{ type: 2, data: stru8(newloc.replace(/-[\s\S]*$/, "")) }],
    [{ type: 2, data: stru8(newloc) }],
    [{ type: 2, data: new Uint8Array([2, 0, 0]) }],
    [{ type: 2, data: new Uint8Array([2, 0, 0]) }],
--- a/bits/85_parsezip.js
+++ b/bits/85_parsezip.js
@ -204,7 +204,7 @@ function parse_zip(zip/*:ZIP*/, opts/*:?ParseOpts*/)/*:Workbook*/ {
 		if(wbrels && wbrels[i]) {
 			path = 'xl/' + (wbrels[i][1]).replace(/[\/]?xl\//, "");
 			if(!safegetzipfile(zip, path)) path = wbrels[i][1];
-			if(!safegetzipfile(zip, path)) path = wbrelsfile.replace(/_rels\/.*$/,"") + wbrels[i][1];
+			if(!safegetzipfile(zip, path)) path = wbrelsfile.replace(/_rels\/[\S\s]*$/,"") + wbrels[i][1];
 			stype = wbrels[i][2];
 		} else {
 			path = 'xl/worksheets/sheet'+(i+1-nmode)+"." + wbext;
--- a/bits/90_utils.js
+++ b/bits/90_utils.js
@ -309,9 +309,9 @@ function book_append_sheet(wb/*:Workbook*/, ws/*:Worksheet*/, name/*:?string*/,
 	if(!name) for(; i <= 0xFFFF; ++i, name = undefined) if(wb.SheetNames.indexOf(name = "Sheet" + i) == -1) break;
 	if(!name || wb.SheetNames.length >= 0xFFFF) throw new Error("Too many worksheets");
 	if(roll && wb.SheetNames.indexOf(name) >= 0 && name.length < 32) {
-		var m = name.match(/(^.*?)(\d+)$/);
-		i = m && +m[2] || 0;
-		var root = m && m[1] || name;
+		var m = name.match(/\d+$/); // at this point, name length is capped at 32
+		i = m && +m[0] || 0;
+		var root = m && name.slice(0, m.index) || name;
 		for(++i; i <= 0xFFFF; ++i) if(wb.SheetNames.indexOf(name = root + i) == -1) break;
 	}
 	check_ws_name(name);
--- a/dist/xlsx.core.min.js
+++ b/dist/xlsx.core.min.js
--- a/dist/xlsx.core.min.map
+++ b/dist/xlsx.core.min.map
--- a/dist/xlsx.extendscript.js
+++ b/dist/xlsx.extendscript.js
--- a/dist/xlsx.full.min.js
+++ b/dist/xlsx.full.min.js
--- a/dist/xlsx.full.min.map
+++ b/dist/xlsx.full.min.map
--- a/dist/xlsx.mini.min.js
+++ b/dist/xlsx.mini.min.js
--- a/dist/xlsx.mini.min.map
+++ b/dist/xlsx.mini.min.map
--- a/modules/04_base64.ts
+++ b/modules/04_base64.ts
@ -60,9 +60,9 @@ function Base64_decode(input: string): string {
 	var o = "";
 	var c1=0, c2=0, c3=0, e1=0, e2=0, e3=0, e4=0;
 	if(input.slice(0,5) == "data:") {
-    var i = input.slice(0, 1024).indexOf(";base64,");
-    if(i > -1) input = input.slice(i+8);
-  }
+		var i = input.slice(0, 1024).indexOf(";base64,");
+		if(i > -1) input = input.slice(i+8);
+	}
 	input = input.replace(/[^\w\+\/\=]/g, "");
 	for(var i = 0; i < input.length;) {
 		e1 = Base64_map.indexOf(input.charAt(i++));
--- a/modules/59_vba.js
+++ b/modules/59_vba.js
@ -13,7 +13,7 @@ function fill_vba_xls(cfb, vba) {
  vba.FullPaths.forEach(function(p, i) {
    if (i == 0)
      return;
-    var newpath = p.replace(/[^\/]*[\/]/, "/_VBA_PROJECT_CUR/");
+    var newpath = p.replace(/^[\/]*[^\/]*[\/]/, "/_VBA_PROJECT_CUR/");
    if (newpath.slice(-1) !== "/")
      CFB.utils.cfb_add(cfb, newpath, vba.FileIndex[i].content);
  });
--- a/modules/59_vba.ts
+++ b/modules/59_vba.ts
@ -16,7 +16,7 @@ function make_vba_xls(cfb: CFBModule.CFB$Container) {
 function fill_vba_xls(cfb: CFBModule.CFB$Container, vba: CFBModule.CFB$Container): void {
 	vba.FullPaths.forEach(function(p, i) {
 		if(i == 0) return;
-		var newpath = p.replace(/[^\/]*[\/]/, "/_VBA_PROJECT_CUR/");
+		var newpath = p.replace(/^[\/]*[^\/]*[\/]/, "/_VBA_PROJECT_CUR/");
 		if(newpath.slice(-1) !== "/") CFB.utils.cfb_add(cfb, newpath, vba.FileIndex[i].content);
 	});
 }
--- a/modules/83_numbers.js
+++ b/modules/83_numbers.js
@ -1512,7 +1512,7 @@ function numbers_add_meta(mlist, newid, newloc) {
  mlist[3].push({ type: 2, data: write_shallow([
    [],
    [{ type: 0, data: write_varint49(newid) }],
-    [{ type: 2, data: stru8(newloc.replace(/-.*$/, "")) }],
+    [{ type: 2, data: stru8(newloc.replace(/-[\s\S]*$/, "")) }],
    [{ type: 2, data: stru8(newloc) }],
    [{ type: 2, data: new Uint8Array([2, 0, 0]) }],
    [{ type: 2, data: new Uint8Array([2, 0, 0]) }],
--- a/modules/83_numbers.ts
+++ b/modules/83_numbers.ts
@ -1281,7 +1281,7 @@ function numbers_iwa_find(cfb: CFB$Container, deps: Dependents, id: number) {
 function numbers_add_meta(mlist: ProtoMessage, newid: number, newloc: string) {
 	mlist[3].push({type: 2, data: write_shallow([ [],
 		[{type: 0, data: write_varint49(newid)}],
-		[{type: 2, data: stru8(newloc.replace(/-.*$/, "")) }],
+		[{type: 2, data: stru8(newloc.replace(/-[\s\S]*$/, "")) }],
 		[{type: 2, data: stru8(newloc)}],
 		[{type: 2, data: new Uint8Array([2, 0, 0])}],
 		[{type: 2, data: new Uint8Array([2, 0, 0])}],
--- a/xlsx.flow.js
+++ b/xlsx.flow.js
--- a/xlsx.full.min.js
+++ b/xlsx.full.min.js
--- a/xlsx.js
+++ b/xlsx.js
--- a/xlsx.mini.flow.js
+++ b/xlsx.mini.flow.js
--- a/xlsx.mini.js
+++ b/xlsx.mini.js
--- a/xlsx.mjs
+++ b/xlsx.mjs