ThaiTextSummarization/visualizer.py at master · suchanun/ThaiTextSummarization · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
import plotly.graph_objects as go
import streamlit as st
import plotly
import numpy as np
class Displayer:
    @staticmethod
    def show_summary_and_text(info):
        st.header(info['title'])
        text = info['text']
        paragraphs = info['paragraphs']
        sentences_with_scores = sorted(info['sentences_with_scores'],key=lambda tup: tup[0])
        paragraph_numbers = sorted(list(paragraphs.keys()))
        st.subheader('My Summary')
        last_phrase_index = -1
        prev_phrases = ''
        selected_sentences = []
        selected_indices = []
        print(paragraphs)
        for p_num in paragraph_numbers:
            paragraph = paragraphs[p_num]
            # st.text('p_num == {}'.format(p_num))
            for phrases_group in paragraph:
                sentences_tokens = [tup[1] for tup in phrases_group]
                selected_indices.extend([tup[0] for tup in phrases_group])
                sentences = ' '.join(sentences_tokens)
                selected_sentences.extend(sentences_tokens)
                phrase_index = phrases_group[0][0]
                # st.text('phrase_index == {}, sentences == '.format(phrase_index, sentences))
                #print(repr(sentences))
                # print('last_phrase_index == {}'.format(last_phrase_index))
                if phrase_index == last_phrase_index + 1:
                    prev_phrases += ' '+ sentences
                else:
                    if prev_phrases != '':# start new phrases
                        st.markdown(prev_phrases)
                    prev_phrases = sentences
                last_phrase_index = phrases_group[-1][0]

        if prev_phrases != '':
            st.markdown(prev_phrases)
        sentences_group = dict()
        sentences_group[0] = selected_sentences
        highlighted_text = Highlighter.get_highlighted_html(text, sentences_group).replace('\n','\n\n')

        y = [tup[2] for tup in sentences_with_scores]
        Displayer.display_selected_in_graph(x=list(range(len(sentences_with_scores))),y=y,selected_indices=selected_indices)
        st.subheader('Highlights Displayed')
        st.markdown(highlighted_text, unsafe_allow_html=True)
        print(repr(highlighted_text))
        # for tup in sentences_with_scores:
        #     st.markdown(tup)
        # st.markdown(text)


    @staticmethod
    def display_selected_in_graph(x,y,selected_indices,mode='lines+markers'):#color_selected='tomato',
        # n = len(selected_indices)
        # y_not_selected = [y[i] for i in range(n) if i not in selected_indices]
        # y_selected = [y[i] for i in selected_indices]
        marker_color = np.zeros(len(x))#['tomato']
        # for idx in selected_indices:
        #     marker_color[idx]='ocean'
        np.put(marker_color,selected_indices,1)
        fig = go.Figure(data=go.Scatter(x=x, y=y, mode=mode,marker=dict(color=marker_color,\
                                                            colorscale=plotly.colors.sequential.Bluered)))#,colorscale='Hot'
        st.plotly_chart(fig)
    @staticmethod
    def display_figure(x,y,title,xaxis_title,yaxis_title,mode='lines+markers'):

        fig = go.Figure(data=go.Scatter(x=x, y=y, mode=mode ))
        fig.update_layout(title=title,
                          xaxis_title=xaxis_title,
                          yaxis_title=yaxis_title)
        st.plotly_chart(fig)

class Highlighter:

    @staticmethod
    def get_highlighted_html(text,sentences_group, intersect_group_id=2):
        highlighted_indices = Highlighter.get_highlight_indices(text, sentences_group, intersect_group_id)
        return Highlighter.compute_highlighted_text(text, highlighted_indices)

    @staticmethod
    def get_highlight_indices( text, sentences_group, intersect_group_id):
        def has_duplicate( start_end_indices, start_idx):
            for index in start_end_indices:
                if index[0] == start_idx:
                    return True
            return False

        text = text.lower()
        indices_with_group_id = []
        indices_by_groupID = dict()
        for group_id in sentences_group:
            indices_by_groupID[group_id] = set()
            for sentence in sentences_group[group_id]:
                sentence = sentence.lower().strip()
                start_idx = text.find(sentence)
                while has_duplicate(indices_by_groupID[group_id], start_idx):
                    start_idx = text.find(sentence, start_idx + len(sentence))
                if start_idx == -1:
                    continue
                index = (start_idx, start_idx + len(sentence))
                indices_by_groupID[group_id].add(index)
        intersect_indices = set.intersection(*[indices_by_groupID[gid] for gid in sentences_group])
        for group_id in sentences_group:
            indices = indices_by_groupID[group_id]
            for index in indices:
                if index not in intersect_indices:
                    indices_with_group_id.append(index + (group_id,))
        for idx in intersect_indices:
            indices_with_group_id.append(idx + (intersect_group_id,))
        ans =  sorted(indices_with_group_id, key=lambda my_tuple: my_tuple[0])

        return ans

    @staticmethod
    def compute_highlighted_text(text, indices, colors={0:'#D0F15F',1:'#90F9E3',2:'#E9B2ED'}):
        highlighted_text = ''
        last_pos = 0

        for index in indices:
            start, end, color_id = index
            color_code = colors[color_id]

            highlighted_text += text[last_pos:start] + '<span style="background-color: {}">'.format(color_code) + text[
                                                                                                                  start:end] + '</span>'
            last_pos = end
        highlighted_text += text[last_pos:]
        return highlighted_text