commit d60da717c4206ed8b7149af50bd178dda6ec015f
parent 70c41e8a5142da0e060da46ce21a59fb17e24ad6
Author: Dan Stillman <dstillman@zotero.org>
Date: Thu, 31 Mar 2016 21:43:33 -0400
Merge pull request #937 from Juris-M/z4.0-bib-edit-conversion
HTML/RTF and RTF/HTML conversion for styled textbox
Diffstat:
1 file changed, 308 insertions(+), 97 deletions(-)
diff --git a/chrome/content/zotero/bindings/styled-textbox.xml b/chrome/content/zotero/bindings/styled-textbox.xml
@@ -49,57 +49,317 @@
this._iframe = document.getAnonymousElementByAttribute(this, "anonid", "rt-view");
- this._rtfMap = {
- "\\":"\\\\",
- "<em>":"\\i ",
- "</em>":"\\i0{}",
- "<i>":"\\i ",
- "</i>":"\\i0{}",
- "<strong>":"\\b ",
- "</strong>":"\\b0{}",
- "<b>":"\\b ",
- "</b>":"\\b0{}",
- "<br />":"\x0B",
- // there's no way to mimic a tab stop in CSS without
- // tables, which wouldn't work here.
- '<span class="tab"> </span>':"\\tab{}"
- };
+ this._htmlRTFmap = [
+ // Atomic units, HTML -> RTF (cleanup)
+ [/<br \/>/g, "\x0B"],
+ [/<span class=\"tab\"> <\/span>/g, "\\tab{}"],
+ [/‘/g, "‘"],
+ [/’/g, "’"],
+ [/“/g, "“"],
+ [/”/g, "”"],
+ [/ /g, "\u00A0"],
+ [/"(\w)/g, "“$1"],
+ [/([\w,.?!])"/g, "$1”"],
+ [/<p>/g, ""],
+ //[/<\/p>(?!\s*$)/g, "\\par{}"],
+ [/<\/?div[^>]*>/g, ""],
+ //[/ /g, " "],
+ //[/\u00A0/g, " "],
+ [/[\x7F-\uFFFF]/g, function(aChar) { return "\\uc0\\u"+aChar.charCodeAt(0).toString()+"{}"}]
+ ];
- this._rtfToHtmlMap = [
+ this._rtfHTMLmap = [
+ // Atomic units, RTF -> HTML (cleanup)
[/\\uc0\{?\\u([0-9]+)\}?(?:{}| )?/g, function(wholeStr, aCode) { return String.fromCharCode(aCode) }],
[/\\tab(?:\{\}| )/g, '<span class="tab"> </span>'],
- [/(?:\\par{}|\\\r?\n)/g, "</p><p>"],
- [/\\super (.*?)\\nosupersub{}/g, "<sup>$1</sup>"],
- [/\\sub (.*?)\\nosupersub{}/g, "<sub>$1</sub>"],
- // for backwards compatibility with Zotero < 2.1
- ["\\i0 ", "</em>"],
- ["\\b0 ", "</strong>"]
+ [/(?:\\par{}|\\\r?\n)/g, "</p><p>"]
];
- this._htmlToRtfMap = [
- [/"(\w)/, "“$1"],
- [/([\w,.?!])"/, "$1”"],
- ["<p>", ""],
- //[/<\/p>(?!\s*$)/g, "\\par{}"],
- [/<\/?div[^>]*>/g, ""],
- ["<sup>", "\\super "],
- ["</sup>", "\\nosupersub{}"],
- ["<sub>", "\\sub "],
- ["</sub>", "\\nosupersub{}"]
- ];
+ this.init = function() {
+ if (this.initialized) return;
+ // Tag data
+ var _rexData = [
+ [
+ [
+ ["<span style=\"font-variant:small-caps;\">"],
+ ["{\\scaps ", "{\\scaps{}"]
+ ],
+ [
+ ["<\/span>"],
+ ["}"]
+ ]
+ ],
+ [
+ [
+ ["<span style=\"text-decoration:underline;\">"],
+ ["{\\ul{}", "{\\ul "]
+ ],
+ [
+ ["<\/span>"],
+ ["}"]
+ ]
+ ],
+ [
+ [
+ ["<sup>"],
+ ["\\super ", "\\super{}"]
+ ],
+ [
+ ["</sup>"],
+ ["\\nosupersub{}", "\\nosupersub "]
+ ]
+ ],
+ [
+ [
+ ["<sub>"],
+ ["\\sub ", "\\sub{}"]
+ ],
+ [
+ ["</sub>"],
+ ["\\nosupersub{}", "\\nosupersub "]
+ ]
+ ],
+ [
+ [
+ ["<em>"],
+ ["{\\i{}", "{\\i "]
+ ],
+ [
+ ["</em>"],
+ ["}"]
+ ]
+ ],
+ [
+ [
+ ["<i>"],
+ ["{\\i{}", "{\\i "]
+ ],
+ [
+ ["</i>"],
+ ["}"]
+ ]
+ ],
+ [
+ [
+ ["<b>"],
+ ["{\\b{}", "{\\b "]
+ ],
+ [
+ ["</b>"],
+ ["}"]
+ ]
+ ],
+ [
+ [
+ ["<strong>"],
+ ["{\\b{}", "{\\b "]
+ ],
+ [
+ ["</strong>"],
+ ["}"]
+ ]
+ ],
+ [
+ [
+ ["<span style=\"font-variant:normal;\">"],
+ ["{\\scaps0{}", "{\\scaps0 "]
+ ],
+ [
+ ["</span>"],
+ ["}"]
+ ]
+ ],
+ [
+ [
+ ["<span style=\"font-style:normal;\">"],
+ ["{\\i0{}", "{\\i0 "]
+ ],
+ [
+ ["</span>"],
+ ["}"]
+ ]
+ ],
+ [
+ [
+ ["<span style=\"font-weight:normal;\">"],
+ ["{\\b0{}", "{\\b0 "]
+ ],
+ [
+ ["</span>"],
+ ["}"]
+ ]
+ ]
+ ];
+
+ function longestFirst(a, b) {
+ if (a.length < b.length) {
+ return 1;
+ } else if (a.length > b.length) {
+ return -1;
+ } else {
+ return 0;
+ }
+ }
+
+ function composeRex(rexes, noGlobal) {
+ var lst = [];
+ for (var rex in rexes) {
+ lst.push(rex);
+ }
+ lst.sort(longestFirst);
+ var rexStr = "(?:" + lst.join("|") + ")";
+ return new RegExp(rexStr, "g");
+ }
+
+ // Create splitting regexps
+ function splitRexMaker(segment) {
+ var rexes = {};
+ for (var i=0,ilen=_rexData.length; i < ilen; i++) {
+ for (var j=0,jlen=_rexData[i].length; j < jlen; j++) {
+ for (var k=0,klen=_rexData[i][j][segment].length; k < klen; k++) {
+ rexes[_rexData[i][j][segment][k].replace("\\", "\\\\")] = true;
+ }
+ }
+ }
+ var ret = composeRex(rexes, true);
+ return ret;
+ }
+ this.rtfHTMLsplitRex = splitRexMaker(1);
+ this.htmlRTFsplitRex = splitRexMaker(0);
+
+ // Create open-tag sniffing regexp
+ function openSniffRexMaker(segment) {
+ var rexes = {};
+ for (var i=0,ilen=_rexData.length; i < ilen; i++) {
+ for (var j=0,jlen=_rexData[i][0][segment].length; j < jlen; j++) {
+ rexes[_rexData[i][0][segment][j].replace("\\", "\\\\")] = true;
+ }
+ }
+ return composeRex(rexes);
+ }
+ this.rtfHTMLopenSniffRex = openSniffRexMaker(1);
+ this.htmlRTFopenSniffRex = openSniffRexMaker(0);
+
+ // Create open-tag remapper
+ function openTagRemapMaker(segment) {
+ var ret = {};
+ for (var i=0,ilen=_rexData.length; i < ilen; i++) {
+ var master = _rexData[i][0][segment][0];
+ for (var j=0,jlen=_rexData[i][0][segment].length; j < jlen; j++) {
+ ret[_rexData[i][0][segment][j]] = master;
+ }
+ }
+ return ret;
+ }
+ this.rtfHTMLopenTagRemap = openTagRemapMaker(1);
+ this.htmlRTFopenTagRemap = openTagRemapMaker(0);
+
+ // Create open-tag-keyed close-tag sniffing regexps
+ function closeTagRexMaker(segment) {
+ var ret = {};
+ var rexes = {};
+ for (var i=0,ilen=_rexData.length; i < ilen; i++) {
+ var master = _rexData[i][0][segment][0];
+ for (var j=0,jlen=_rexData[i][1][segment].length; j < jlen; j++) {
+ rexes[_rexData[i][1][segment][j]] = true;
+ }
+ ret[master] = composeRex(rexes);
+ }
+ return ret;
+ }
+ this.rtfHTMLcloseTagRex = closeTagRexMaker(1);
+ this.htmlRTFcloseTagRex = closeTagRexMaker(0);
+
+ // Create open-tag-keyed open/close tag registry
+ function tagRegistryMaker(segment) {
+ var antisegment = 1;
+ if (segment == 1) {
+ antisegment = 0;
+ }
+ var ret = {};
+ for (var i=0,ilen=_rexData.length; i < ilen; i++) {
+ var master = _rexData[i][0][segment][0];
+ ret[master] = {
+ open: _rexData[i][0][antisegment][0],
+ close: _rexData[i][1][antisegment][0]
+ }
+ }
+ return ret;
+ }
+ this.rtfHTMLtagRegistry = tagRegistryMaker(1);
+ this.htmlRTFtagRegistry = tagRegistryMaker(0);
+
+ this.initialized = true;
+ }
+ this.init();
- this._rtfRexMap = [
- ["<span style=\"font-variant:small-caps;\">",
- /small-caps/,
- "\\scaps ",
- "\\scaps0{}"
- ],
- ["<span style=\"text-decoration:underline;\">",
- /underline/,
- "\\ul ",
- "\\ul0{}"
- ]
- ]
+ this.getSplit = function(mode, txt) {
+ if (!txt) return [];
+ var splt = txt.split(this[mode + "splitRex"]);
+ var mtch = txt.match(this[mode + "splitRex"]);
+ var lst = [splt[0]];
+ for (var i=1,ilen=splt.length; i < ilen; i++) {
+ lst.push(mtch[i-1]);
+ lst.push(splt[i]);
+ }
+ return lst;
+ }
+
+ this.getOpenTag = function(mode, str) {
+ var m = str.match(this[mode + "openSniffRex"]);
+ if (m) {
+ m = this[mode + "openTagRemap"][m[0]];
+ }
+ return m;
+ }
+
+ this.convert = function(mode, txt) {
+ var lst = this.getSplit(mode, txt);
+ var sdepth = 0;
+ var depth = 0;
+ for (var i=1,ilen=lst.length; i < ilen; i += 2) {
+ var openTag = this.getOpenTag(mode, lst[i]);
+ if (openTag) {
+ sdepth++;
+ depth = sdepth;
+ for (var j=(i+2),jlen=lst.length; j < jlen; j += 2) {
+ var closeTag = !this.getOpenTag(mode, lst[j]);
+ if (closeTag) {
+ if (depth === sdepth && lst[j].match(this[mode + "closeTagRex"][openTag])) {
+ lst[i] = this[mode + "tagRegistry"][openTag].open;
+ lst[j] = this[mode + "tagRegistry"][openTag].close;
+ break;
+ }
+ depth--;
+ } else {
+ depth++;
+ }
+ }
+ } else {
+ sdepth--;
+ }
+ }
+ return lst.join("");
+ }
+
+ this.htmlToRTF = function(txt) {
+ txt = this.convert("htmlRTF", txt);
+ for (var i=0,ilen=this._htmlRTFmap.length; i < ilen; i++) {
+ var entry = this._htmlRTFmap[i];
+ txt = txt.replace(entry[0], entry[1]);
+ }
+ txt = Zotero.Utilities.unescapeHTML(txt);
+ return txt.trim();
+ }
+
+ this.rtfToHTML = function(txt) {
+ for (var i=0,ilen=this._rtfHTMLmap.length; i < ilen; i++) {
+ var entry = this._rtfHTMLmap[i];
+ txt = txt.replace(entry[0], entry[1]);
+ }
+ txt = this.convert("rtfHTML", txt);
+ return txt;
+ }
this._constructed = true;
@@ -213,7 +473,6 @@
<!-- Sets or returns contents of rich text box -->
<property name="value">
<getter><![CDATA[
- const highcharRe = /[\x7F-\uFFFF]/g;
var output = this._editor.getContent();
if(this._format == "RTF") {
@@ -221,44 +480,7 @@
if(output.substr(0, 5) == "<div>" && output.substr(-6) == "</div>") {
output = output.substr(5, output.length-6);
}
-
- // do appropriate replacement operations
- for(var needle in this._rtfMap) {
- output = output.replace(needle, this._rtfMap[needle], "g");
- }
-
- // Preserve small caps and underlining
- for each (var tagspec in this._rtfRexMap){
- var l = output.split(/(<\/?span[^>]*>)/);
- var current_level = 0;
- var tag_level = [];
- for (var pos=1; pos<l.length; pos+=2) {
- var tag = l[pos];
- if (tag[1] == "/") {
- current_level--;
- if (current_level == tag_level[tag_level.length-1]) {
- tag_level.pop();
- l[pos] = tagspec[3];
- }
- } else {
- if (l[pos].match(tagspec[1])) {
- l[pos] = tagspec[2];
- tag_level.push(current_level);
- }
- current_level++;
- };
- };
- output = l.join("");
- }
-
- for each(var entry in this._htmlToRtfMap) {
- output = output.replace(entry[0], entry[1], "g");
- }
-
- output = Zotero.Utilities.unescapeHTML(output.replace(" ", " ", "g"))
- .replace("\u00A0", " ", "g")
- .replace(highcharRe, function(aChar) { return "\\uc0\\u"+aChar.charCodeAt(0).toString()+"{}" });
- output = Zotero.Utilities.trim(output);
+ output = this.htmlToRTF(output)
}
return output;
@@ -313,19 +535,8 @@
bodyStyle = "margin-left:"+(li/20+6)+"pt;text-indent:"+(fi/20)+"pt;";
}
- // do appropriate replacement operations
- for(var needle in this._rtfMap) {
- if(this._rtfMap[needle]) {
- html = html.replace(this._rtfMap[needle], needle, "g");
- }
- }
- for each (var tagspec in this._rtfRexMap){
- html = html.replace(tagspec[2], tagspec[0], "g");
- html = html.replace(tagspec[3], "</span>", "g");
- }
- for each(var entry in this._rtfToHtmlMap) {
- html = html.replace(entry[0], entry[1], "g");
- }
+ html = this.rtfToHTML(html);
+
html = '<div style="'+bodyStyle+'"><p>'+html+"</p></div>";
}
