312 lines
9.9 KiB
JavaScript
312 lines
9.9 KiB
JavaScript
|
/*!
|
||
|
|
* Snowball JavaScript Library v0.3
|
||
|
|
* http://code.google.com/p/urim/
|
||
|
|
* http://snowball.tartarus.org/
|
||
|
|
*
|
||
|
|
* Copyright 2010, Oleg Mazko
|
||
|
|
* http://www.mozilla.org/MPL/
|
||
|
|
*/
|
||
|
|
|
||
|
|
/**
|
||
|
|
* export the module via AMD, CommonJS or as a browser global
|
||
|
|
* Export code from https://github.com/umdjs/umd/blob/master/returnExports.js
|
||
|
|
*/
|
||
|
|
(function (root, factory) {
|
||
|
|
if (typeof define === "function" && define.amd) {
|
||
|
|
// AMD. Register as an anonymous module.
|
||
|
|
define(factory);
|
||
|
|
} else if (typeof exports === "object") {
|
||
|
|
/**
|
||
|
|
* Node. Does not work with strict CommonJS, but
|
||
|
|
* only CommonJS-like environments that support module.exports,
|
||
|
|
* like Node.
|
||
|
|
*/
|
||
|
|
module.exports = factory();
|
||
|
|
} else {
|
||
|
|
// Browser globals (root is window)
|
||
|
|
factory()(root.lunr);
|
||
|
|
}
|
||
|
|
})(this, function () {
|
||
|
|
/**
|
||
|
|
* Just return a value to define the module export.
|
||
|
|
* This example returns an object, but the module
|
||
|
|
* can return a function as the exported value.
|
||
|
|
*/
|
||
|
|
return function (lunr) {
|
||
|
|
/* provides utilities for the included stemmers */
|
||
|
|
lunr.stemmerSupport = {
|
||
|
|
Among: function (s, substring_i, result, method) {
|
||
|
|
this.toCharArray = function (s) {
|
||
|
|
var sLength = s.length,
|
||
|
|
charArr = new Array(sLength);
|
||
|
|
for (var i = 0; i < sLength; i++) charArr[i] = s.charCodeAt(i);
|
||
|
|
return charArr;
|
||
|
|
};
|
||
|
|
|
||
|
|
if ((!s && s != "") || (!substring_i && substring_i != 0) || !result)
|
||
|
|
throw (
|
||
|
|
"Bad Among initialisation: s:" +
|
||
|
|
s +
|
||
|
|
", substring_i: " +
|
||
|
|
substring_i +
|
||
|
|
", result: " +
|
||
|
|
result
|
||
|
|
);
|
||
|
|
this.s_size = s.length;
|
||
|
|
this.s = this.toCharArray(s);
|
||
|
|
this.substring_i = substring_i;
|
||
|
|
this.result = result;
|
||
|
|
this.method = method;
|
||
|
|
},
|
||
|
|
SnowballProgram: function () {
|
||
|
|
var current;
|
||
|
|
return {
|
||
|
|
bra: 0,
|
||
|
|
ket: 0,
|
||
|
|
limit: 0,
|
||
|
|
cursor: 0,
|
||
|
|
limit_backward: 0,
|
||
|
|
setCurrent: function (word) {
|
||
|
|
current = word;
|
||
|
|
this.cursor = 0;
|
||
|
|
this.limit = word.length;
|
||
|
|
this.limit_backward = 0;
|
||
|
|
this.bra = this.cursor;
|
||
|
|
this.ket = this.limit;
|
||
|
|
},
|
||
|
|
getCurrent: function () {
|
||
|
|
var result = current;
|
||
|
|
current = null;
|
||
|
|
return result;
|
||
|
|
},
|
||
|
|
in_grouping: function (s, min, max) {
|
||
|
|
if (this.cursor < this.limit) {
|
||
|
|
var ch = current.charCodeAt(this.cursor);
|
||
|
|
if (ch <= max && ch >= min) {
|
||
|
|
ch -= min;
|
||
|
|
if (s[ch >> 3] & (0x1 << (ch & 0x7))) {
|
||
|
|
this.cursor++;
|
||
|
|
return true;
|
||
|
|
}
|
||
|
|
}
|
||
|
|
}
|
||
|
|
return false;
|
||
|
|
},
|
||
|
|
in_grouping_b: function (s, min, max) {
|
||
|
|
if (this.cursor > this.limit_backward) {
|
||
|
|
var ch = current.charCodeAt(this.cursor - 1);
|
||
|
|
if (ch <= max && ch >= min) {
|
||
|
|
ch -= min;
|
||
|
|
if (s[ch >> 3] & (0x1 << (ch & 0x7))) {
|
||
|
|
this.cursor--;
|
||
|
|
return true;
|
||
|
|
}
|
||
|
|
}
|
||
|
|
}
|
||
|
|
return false;
|
||
|
|
},
|
||
|
|
out_grouping: function (s, min, max) {
|
||
|
|
if (this.cursor < this.limit) {
|
||
|
|
var ch = current.charCodeAt(this.cursor);
|
||
|
|
if (ch > max || ch < min) {
|
||
|
|
this.cursor++;
|
||
|
|
return true;
|
||
|
|
}
|
||
|
|
ch -= min;
|
||
|
|
if (!(s[ch >> 3] & (0x1 << (ch & 0x7)))) {
|
||
|
|
this.cursor++;
|
||
|
|
return true;
|
||
|
|
}
|
||
|
|
}
|
||
|
|
return false;
|
||
|
|
},
|
||
|
|
out_grouping_b: function (s, min, max) {
|
||
|
|
if (this.cursor > this.limit_backward) {
|
||
|
|
var ch = current.charCodeAt(this.cursor - 1);
|
||
|
|
if (ch > max || ch < min) {
|
||
|
|
this.cursor--;
|
||
|
|
return true;
|
||
|
|
}
|
||
|
|
ch -= min;
|
||
|
|
if (!(s[ch >> 3] & (0x1 << (ch & 0x7)))) {
|
||
|
|
this.cursor--;
|
||
|
|
return true;
|
||
|
|
}
|
||
|
|
}
|
||
|
|
return false;
|
||
|
|
},
|
||
|
|
eq_s: function (s_size, s) {
|
||
|
|
if (this.limit - this.cursor < s_size) return false;
|
||
|
|
for (var i = 0; i < s_size; i++)
|
||
|
|
if (current.charCodeAt(this.cursor + i) != s.charCodeAt(i))
|
||
|
|
return false;
|
||
|
|
this.cursor += s_size;
|
||
|
|
return true;
|
||
|
|
},
|
||
|
|
eq_s_b: function (s_size, s) {
|
||
|
|
if (this.cursor - this.limit_backward < s_size) return false;
|
||
|
|
for (var i = 0; i < s_size; i++)
|
||
|
|
if (
|
||
|
|
current.charCodeAt(this.cursor - s_size + i) != s.charCodeAt(i)
|
||
|
|
)
|
||
|
|
return false;
|
||
|
|
this.cursor -= s_size;
|
||
|
|
return true;
|
||
|
|
},
|
||
|
|
find_among: function (v, v_size) {
|
||
|
|
var i = 0,
|
||
|
|
j = v_size,
|
||
|
|
c = this.cursor,
|
||
|
|
l = this.limit,
|
||
|
|
common_i = 0,
|
||
|
|
common_j = 0,
|
||
|
|
first_key_inspected = false;
|
||
|
|
while (true) {
|
||
|
|
var k = i + ((j - i) >> 1),
|
||
|
|
diff = 0,
|
||
|
|
common = common_i < common_j ? common_i : common_j,
|
||
|
|
w = v[k];
|
||
|
|
for (var i2 = common; i2 < w.s_size; i2++) {
|
||
|
|
if (c + common == l) {
|
||
|
|
diff = -1;
|
||
|
|
break;
|
||
|
|
}
|
||
|
|
diff = current.charCodeAt(c + common) - w.s[i2];
|
||
|
|
if (diff) break;
|
||
|
|
common++;
|
||
|
|
}
|
||
|
|
if (diff < 0) {
|
||
|
|
j = k;
|
||
|
|
common_j = common;
|
||
|
|
} else {
|
||
|
|
i = k;
|
||
|
|
common_i = common;
|
||
|
|
}
|
||
|
|
if (j - i <= 1) {
|
||
|
|
if (i > 0 || j == i || first_key_inspected) break;
|
||
|
|
first_key_inspected = true;
|
||
|
|
}
|
||
|
|
}
|
||
|
|
while (true) {
|
||
|
|
var w = v[i];
|
||
|
|
if (common_i >= w.s_size) {
|
||
|
|
this.cursor = c + w.s_size;
|
||
|
|
if (!w.method) return w.result;
|
||
|
|
var res = w.method();
|
||
|
|
this.cursor = c + w.s_size;
|
||
|
|
if (res) return w.result;
|
||
|
|
}
|
||
|
|
i = w.substring_i;
|
||
|
|
if (i < 0) return 0;
|
||
|
|
}
|
||
|
|
},
|
||
|
|
find_among_b: function (v, v_size) {
|
||
|
|
var i = 0,
|
||
|
|
j = v_size,
|
||
|
|
c = this.cursor,
|
||
|
|
lb = this.limit_backward,
|
||
|
|
common_i = 0,
|
||
|
|
common_j = 0,
|
||
|
|
first_key_inspected = false;
|
||
|
|
while (true) {
|
||
|
|
var k = i + ((j - i) >> 1),
|
||
|
|
diff = 0,
|
||
|
|
common = common_i < common_j ? common_i : common_j,
|
||
|
|
w = v[k];
|
||
|
|
for (var i2 = w.s_size - 1 - common; i2 >= 0; i2--) {
|
||
|
|
if (c - common == lb) {
|
||
|
|
diff = -1;
|
||
|
|
break;
|
||
|
|
}
|
||
|
|
diff = current.charCodeAt(c - 1 - common) - w.s[i2];
|
||
|
|
if (diff) break;
|
||
|
|
common++;
|
||
|
|
}
|
||
|
|
if (diff < 0) {
|
||
|
|
j = k;
|
||
|
|
common_j = common;
|
||
|
|
} else {
|
||
|
|
i = k;
|
||
|
|
common_i = common;
|
||
|
|
}
|
||
|
|
if (j - i <= 1) {
|
||
|
|
if (i > 0 || j == i || first_key_inspected) break;
|
||
|
|
first_key_inspected = true;
|
||
|
|
}
|
||
|
|
}
|
||
|
|
while (true) {
|
||
|
|
var w = v[i];
|
||
|
|
if (common_i >= w.s_size) {
|
||
|
|
this.cursor = c - w.s_size;
|
||
|
|
if (!w.method) return w.result;
|
||
|
|
var res = w.method();
|
||
|
|
this.cursor = c - w.s_size;
|
||
|
|
if (res) return w.result;
|
||
|
|
}
|
||
|
|
i = w.substring_i;
|
||
|
|
if (i < 0) return 0;
|
||
|
|
}
|
||
|
|
},
|
||
|
|
replace_s: function (c_bra, c_ket, s) {
|
||
|
|
var adjustment = s.length - (c_ket - c_bra),
|
||
|
|
left = current.substring(0, c_bra),
|
||
|
|
right = current.substring(c_ket);
|
||
|
|
current = left + s + right;
|
||
|
|
this.limit += adjustment;
|
||
|
|
if (this.cursor >= c_ket) this.cursor += adjustment;
|
||
|
|
else if (this.cursor > c_bra) this.cursor = c_bra;
|
||
|
|
return adjustment;
|
||
|
|
},
|
||
|
|
slice_check: function () {
|
||
|
|
if (
|
||
|
|
this.bra < 0 ||
|
||
|
|
this.bra > this.ket ||
|
||
|
|
this.ket > this.limit ||
|
||
|
|
this.limit > current.length
|
||
|
|
)
|
||
|
|
throw "faulty slice operation";
|
||
|
|
},
|
||
|
|
slice_from: function (s) {
|
||
|
|
this.slice_check();
|
||
|
|
this.replace_s(this.bra, this.ket, s);
|
||
|
|
},
|
||
|
|
slice_del: function () {
|
||
|
|
this.slice_from("");
|
||
|
|
},
|
||
|
|
insert: function (c_bra, c_ket, s) {
|
||
|
|
var adjustment = this.replace_s(c_bra, c_ket, s);
|
||
|
|
if (c_bra <= this.bra) this.bra += adjustment;
|
||
|
|
if (c_bra <= this.ket) this.ket += adjustment;
|
||
|
|
},
|
||
|
|
slice_to: function () {
|
||
|
|
this.slice_check();
|
||
|
|
return current.substring(this.bra, this.ket);
|
||
|
|
},
|
||
|
|
eq_v_b: function (s) {
|
||
|
|
return this.eq_s_b(s.length, s);
|
||
|
|
},
|
||
|
|
};
|
||
|
|
},
|
||
|
|
};
|
||
|
|
|
||
|
|
lunr.trimmerSupport = {
|
||
|
|
generateTrimmer: function (wordCharacters) {
|
||
|
|
var startRegex = new RegExp("^[^" + wordCharacters + "]+");
|
||
|
|
var endRegex = new RegExp("[^" + wordCharacters + "]+$");
|
||
|
|
|
||
|
|
return function (token) {
|
||
|
|
// for lunr version 2
|
||
|
|
if (typeof token.update === "function") {
|
||
|
|
return token.update(function (s) {
|
||
|
|
return s.replace(startRegex, "").replace(endRegex, "");
|
||
|
|
});
|
||
|
|
} else {
|
||
|
|
// for lunr version 1
|
||
|
|
return token.replace(startRegex, "").replace(endRegex, "");
|
||
|
|
}
|
||
|
|
};
|
||
|
|
},
|
||
|
|
};
|
||
|
|
};
|
||
|
|
});
|