Created
February 18, 2012 11:26
-
-
Save hippietrail/1858881 to your computer and use it in GitHub Desktop.
Sort and normalize Georgian requested entries page of English Wiktionary
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
<html> | |
<head> | |
<!--Load the AJAX API--> | |
<style type="text/css"> | |
</style> | |
<script type="text/javascript" src="http://code.jquery.com/jquery-latest.min.js"></script> | |
<script type="text/javascript"> | |
$(document).ready(function(){ | |
// initialize stuff | |
// ... | |
$.getJSON("http://en.wiktionary.org/w/api.php?callback=?", | |
{ | |
action: "query", | |
format: "json", | |
titles: "Wiktionary:Requested_entries_(Georgian)", | |
prop: "revisions", | |
rvprop: "content" | |
}, | |
function(data) { | |
var src = data.query.pages[Object.keys(data.query.pages)[0]].revisions[0]["*"]; | |
var heads = src.match(/^==.*==$/mg); | |
var bodies = src.split(/^==.*==$/mg); | |
for (var i = 0; i < heads.length; ++i) { | |
var x = heads[i].match(/^==([^ ]*)\s+\[(.*)\]==$/); | |
if (x) { | |
var g = x[1], ipa = x[2], b = bodies[i+1]; | |
var b0 = b; | |
// match [[kaka]] (enen) if "foo" is georgian (and "bar" is latin) | |
b = b.replace(/\[\[(.*?[\u10a0-\u10fb].*?)]]\s*\((.*?[a-zA-Z].*?)\)/g, "{{l|ka|$1|tr=$2}}"); | |
// match [[kaka]] as long as at least one character is georgian | |
b = b.replace(/\[\[(.*?[\u10a0-\u10fb].*?)]]/g, "{{l|ka|$1}}"); | |
// match [[ruru]] as long as at least one character is russian / cyrillic | |
b = b.replace(/\[\[(.*?[\u0400-\u04ff].*?)]]/g, "{{l|ru|$1}}"); | |
// match [[hyhy]] as long as at least one character is armenian | |
b = b.replace(/\[\[(.*?[\u0531-\u058a].*?)]]/g, "{{l|hy|$1}}"); | |
// {{l|ka|kaka}} -> {{l|ka|kaka|tr=trtr}} | |
b = b.replace(/{{l\|ka\|([^|]*?[\u10a0-\u10fb][|^}]*?)}}/g, replacetr); | |
var unchanged = (b0 == b); | |
var lines = b.split(/\r?\n/); | |
var obj = {}; | |
var arr = []; | |
var oldkey; | |
$.each(lines, function(i, l) { | |
var m; | |
var k = ""; | |
// old style links [[foo]] (bar) | |
if (m = l.match(/^\*\s*\[\[(.*?)\]\]/)) { | |
k = m[1]; | |
k = k.replace(/[- ]/g, ""); | |
obj[k] = l; | |
arr.push(k); | |
// template style links {{l|ka|foo|tr=bar}} | |
} else if (m = l.match(/^\*\s*{{l\|ka\|([^\|}]*?)[\|}]/)) { | |
k = m[1]; | |
k = k.replace(/[- ]/g, ""); | |
obj[k] = l; | |
arr.push(k); | |
} else if (l.match(/^\*:/)) { | |
obj[oldkey] += "\n" + l; | |
} | |
if (k != "") { | |
oldkey = k; | |
} | |
}); | |
var sarr = arr.slice(0); | |
sarr.sort(); | |
var sorted = arraycompare(sarr, arr); | |
var after = ""; | |
if (!sorted || !unchanged) { | |
after = "<pre>"; | |
for (var j = 0; j < sarr.length; ++j) { | |
if (j != 0) { | |
after += "\n"; | |
} | |
after += obj[sarr[j]]; | |
} | |
after += "</pre>"; | |
} | |
var html = "<div id=\"" + g + "\"><h2>" + g + " <tt>[" + ipa + "]</tt></h2><dl>" | |
+ "</dl>" + (sorted ? "sorted" : "NOT SORTED") + "\n" + (unchanged ? "unchanged" : "CHANGED") + "</div>" | |
+ after; | |
$(html).appendTo("#app"); | |
} | |
} | |
} | |
); | |
}); | |
function arraycompare(a, b) { | |
for (var i = 0; i < a.length; ++i) { | |
if (a[i] != b[i]) { | |
return false; | |
} | |
} | |
return true; | |
} | |
function replacetr(m, c) { | |
var tr = transliterate('ka', 'tr', c); | |
r = "{{l|ka|" + c + "|tr=" + tr + "}}"; | |
return r; | |
} | |
var ka_tr = { | |
'ა': 'a', | |
'ბ': 'b', | |
'გ': 'g', | |
'დ': 'd', | |
'ე': 'e', | |
'ვ': 'v', | |
'ზ': 'z', | |
'თ': [ 't\'', 't', 'T' ], | |
'ი': 'i', | |
'კ': [ 'k', 'k\'' ], | |
'ლ': 'l', | |
'მ': 'm', | |
'ნ': 'n', | |
'ო': 'o', | |
'პ': [ 'p', 'p\'' ], | |
'ჟ': [ 'ž', 'zh', 'j', 'J' ], | |
'რ': 'r', | |
'ს': 's', | |
'ტ': [ 't', 't\'' ], | |
'უ': 'u', | |
'ფ': [ 'p\'', 'p' ], | |
'ქ': [ 'k\'', 'k', 'q' ], | |
'ღ': [ 'ḡ', 'gh', 'R' ], | |
'ყ': [ 'q', 'y', 'q' ], | |
'შ': [ 'š', 'sh', 'S' ], | |
'ჩ': [ 'č\'', 'ch', 'C' ], | |
'ც': [ 'c\'', 'c', 'ts' ], | |
'ძ': [ 'j', 'dz', 'Z' ], | |
'წ': [ 'c', 'c\'', 'ts', 'w', 'ts\'' ], | |
'ჭ': [ 'č', 'ch', 'W', 'ch\'' ], | |
'ხ': [ 'x', 'kh' ], | |
'ჯ': [ 'ǰ', 'j' ], | |
'ჰ': 'h', | |
}; | |
var tr_ka = { | |
'a': 'ა', | |
'b': 'ბ', | |
'g': 'გ', | |
'd': 'დ', | |
'e': 'ე', | |
'v': 'ვ', | |
'z': 'ზ', | |
't\'': 'თ', // ტ | |
'i': 'ი', | |
'k': 'კ', // ქ | |
'l': 'ლ', | |
'm': 'მ', | |
'n': 'ნ', | |
'o': 'ო', | |
'p': 'პ', // ფ | |
'ž': 'ჟ', | |
'r': 'რ', | |
's': 'ს', | |
't': 'ტ', // თ | |
'u': 'უ', | |
'p\'': 'ფ', // პ | |
'k\'': 'ქ', // კ | |
'ḡ': 'ღ', | |
'q': 'ყ', // ქ | |
'š': 'შ', | |
'č\'': 'ჩ', | |
'c\'': 'ც', | |
'j': 'ძ', | |
'c': 'წ', | |
'č': 'ჭ', | |
'x': 'ხ', | |
'ǰ': 'ჯ', | |
'h': 'ჰ', | |
}; | |
// J̌ ǰ | |
function transliterate(src, dst, srctxt) { | |
var l = srctxt.length; | |
var i = 0; | |
var dsttxt = ''; | |
var hash = src === 'ka' ? ka_tr : tr_ka; | |
for (i = 0; i < l; ++i) { | |
var a = srctxt.charAt(i); | |
var b = i+1 < l ? srctxt.charAt(i+1) : ''; | |
var c = a + b; | |
var x; | |
if (c in hash) { | |
x = hash[c]; | |
++i; | |
} else if (a in hash) { | |
x = hash[a]; | |
} else { | |
x = a; | |
} | |
if ($.isArray(x)) | |
x = x[0]; | |
dsttxt += x; | |
} | |
return dsttxt; | |
} | |
</script> | |
</head> | |
<body> | |
<div id="app"> | |
</div> | |
</body> | |
</html> |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment