加入ratio調整及滑鼠事件

generalText.py

@@ -2,6 +2,7 @@ import jieba
 import csv
 import nltk
 import re
+import json
 from jieba import posseg
 from nltk import tokenize
 from langdetect import detect
@@ -31,38 +32,37 @@ def processText(randId, text, stopwords):
         return ''
     lang = detect(text)
     sentenses = []
+    sentenses_raw = []
     print(lang)
     if (lang == 'zh-cn' or lang == 'zh-tw' or lang == 'ko'):
         splitted = re.split('。|[\n]+', text)
         print(splitted)
         cutted = []
-        for i in splitted:
-            cutted.append(filterPOS(i))
+        for spl in splitted:
+            cutted.append(filterPOS(spl))
         print(cutted)
-        for i in cutted:
-            result = []
-            for j in i:
-                if (j in stopwords):
-                    continue
-                result.append(j)
-                if (len(result) >= 20):
-                    sentenses.append(' '.join(result.copy()))
-                    result = []
-            if (result != []):
-                sentenses.append(' '.join(result))
+        for spl, raw in zip(cutted, splitted):
+            sentenses.append(' '.join(spl))
+            sentenses_raw.append(raw)
     else:
         sentenses = []
-        for sentence in tokenize.sent_tokenize(text):
-            words = sentence.lower().split(' ')
+        sentenses_raw = []
+        for sentence_raw in tokenize.sent_tokenize(text):
+            words = sentence_raw.lower().split(' ')
             print([''.join([a for a in w1 if a.isalpha()]) for w1 in words])
             sentence = ' '.join([w for w in [''.join([a for a in w1 if a.isalpha()]) for w1 in words] if w not in [sw.lower() for sw in stopwords]])
             sentenses.append(sentence)
+            sentenses_raw.append(sentence_raw)
     result = []
-    result.append(['id', 'text', 'count'])
-    for index, sentence in enumerate(sentenses):
-        result.append([index, sentence, 1000])
-    with open('data/' + randId + '.tsv', 'w', newline='', encoding="utf-8") as f:
-        writer = csv.writer(f, delimiter='\t')
-        writer.writerows(result)
-        f.close()
-    return ('data/' + randId + '.tsv')
+    for index, raw_pair in enumerate(zip(sentenses, sentenses_raw)):
+        sentence, sentence_raw = raw_pair
+        result.append({
+            'id': index,
+            'text': sentence,
+            'count': 10,
+            'rawtxt': sentence_raw,
+        })
+
+    with open('data/' + randId + '.json', 'w', newline='', encoding="utf-8") as fp:
+        json.dump(result, fp, ensure_ascii=False, indent=4)
+    return ('data/' + randId + '.json')

static/js/generalText.js

@@ -1,6 +1,15 @@
 var tsvPath
 var stopwords = []
 
+const init = () => {
+    $(window).on('mousemove', function (e) {
+        $('#nodeTitle').css({
+            left: e.pageX,
+            top: e.pageY
+        })
+    })
+}
+
 function clearStopWord() {
     stopwords = []
     $('#sweContainer').html('')
@@ -86,22 +95,34 @@ function submit() {
 
 function buildSentetree() {
     console.log("Build.")
-    var model;
-    var tree;
-    var data;
-    const graph = d3.tsv(tsvPath, buildTree);
+    let model;
+    let tree;
+    let data;
+    const graph = d3.json(tsvPath, buildTree);
 
     function buildTree(error, rawdata) {
+        console.log(rawdata)
         const data = rawdata.map(d => Object.assign({}, d, { count: +d.count }));
+        console.log({ data })
+        let minRatio = $('#minRatio').val()
+        let maxRatio = $('#maxRatio').val()
+        console.log({ minRatio, maxRatio })
         model = new SentenTree.SentenTreeBuilder()
             .tokenize(SentenTree.tokenizer.tokenizeBySpace)
             .transformToken(token => (/score(d|s)?/.test(token) ? 'score' : token))
-            .buildModel(data, { maxSupportRatio: 1 });
+            .buildModel(data, { maxSupportRatio: maxRatio, minSupportRatio: minRatio });
         tree = new SentenTree.SentenTreeVis('#vis', {
             fontSize: [15, 40],
             gapBetweenGraph: 10
         });
         tree.data(model.getRenderedGraphs(2))
+            .on('nodeMouseenter', (node) => {
+                $('#nodeTitle').removeClass('hidden')
+                $('#nodeTitleContent').html(node.data.topEntries.map((n) => data[n.id].rawtxt).join('<br>'))
+            })
+            .on('nodeMouseleave', () => {
+                $('#nodeTitle').addClass('hidden')
+            })
         new ResizeSensor(jQuery('#d3kitRoot'), function () {
             var scale, origin;
             scale = Math.min(2, ($('#graph').outerWidth()) / ($('#d3kitRoot').outerWidth() + 60))
@@ -133,3 +154,5 @@ function countWords() {
     console.log(wordCount)
     $("#wordcount").html('字數：' + wordCount)
 }
+
+init()

templates/generalTxt.html

@@ -9,6 +9,9 @@
 </head>
 
 <body>
+    <div id="nodeTitle" class="nodeTitle hidden">
+        <div id="nodeTitleContent">test</div>
+    </div>
     <div id="stopWordEditorLayer" class="info hidden">
         <div id="stopWordEditor">
             <h4 id="sweTitle" style="margin:10px; display: inline;">編輯停用詞</h4>
@@ -39,13 +42,17 @@
     <div id='heading'>
         <h2>泛用文字視覺化工具</h2>
         <p>SentenTree <a href="https://github.com/twitter/SentenTree">https://github.com/twitter/SentenTree</a></p>
-        <p id='comment'>這是泛用.txt檔視覺化工具，能夠簡單處理文字檔的視覺化。</p>
+        <p id='comment'>這是泛用文字視覺化工具，能夠簡單處理文字檔的視覺化。</p>
         <p id='comment'>支援的語言：繁體中文、英文以及所有使用空格分詞的語言。</p>
         <p id='comment'>使用繁體中文Jieba斷詞器，不保證簡體中文能夠正常使用。</p>
     </div>
     <div style="margin:10px;">
         <button class="general-button" type="button" id="editSWButton" style="margin:10px 0px;"
             onclick="showStopwordEditor()">編輯停用詞</button>
+        <label>Min Ratio</label>
+        <input type="number" step="0.0001" id="minRatio" value="0.001" min="0.0001" max="1">
+        <label>Max Ratio</label>
+        <input type="number" step="0.0001" id="maxRatio" value="1" min="0.0001" max="1">
     </div>
     <div id='rawText' class=''>
         <div id="wordcount">字數：0</div>