File size: 9,050 Bytes
c398ab5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
from utlis.helper import *

initialize_session_state()

with st.sidebar:
    st.image("logo.png", width=170)
    st.title("AGDC")
    # Get List of models
    llms = ['gpt-3.5-turbo', 'gemini']
    st.session_state.llm = st.selectbox("Choose LLM",llms)
    st.session_state.genre = st.radio(
    "Choose option",
    ["Select document", "Add document(s)","Delete service(s)", "Delete document(s)"])
    
    if st.session_state.genre=="Add document(s)":
        st.title('Add Document(s)')
        # Check service status
        # Get all available services
        add_new_service = st.checkbox("Add new service")
        if add_new_service:
            new_service = st.text_input("Enter service name")
            # Get list of Embedding models
            
            if  new_service and st.button('Add'):
                add_service(st.session_state.token,new_service)
        data = {"token": st.session_state.token}
        json_data = json.dumps(data)
        headers = {'Content-Type': 'application/json'}
        services  = requests.get(SERVICES_API,data=json_data, headers=headers)
        services =json.loads(services.text)
        if len(services)>0:
           st.session_state.service = st.selectbox("Choose Service",services)


        if len(services)>0:
            st.session_state.uploaded_files = st.file_uploader("Upload PDF file",  type=["pdf"], accept_multiple_files=False)
            if st.session_state.uploaded_files:
                st.session_state.process = st.button('Process')
                if st.session_state.process:
                    add_document(st.session_state.token,st.session_state.service)

    elif st.session_state.genre=="Select document":
        st.title('Scrape Document')
        data = {"token": st.session_state.token}
        json_data = json.dumps(data)
        headers = {'Content-Type': 'application/json'}
        services  = requests.get(SERVICES_API,data=json_data, headers=headers)
        services =json.loads(services.text)

        if len(services)>0:
            st.session_state.service_slected_to_chat = st.selectbox("Choose Service",services)
            data = {"token": st.session_state.token, "servicename": st.session_state.service_slected_to_chat}
            json_data = json.dumps(data)
            headers = {'Content-Type': 'application/json'}
            history_document  = requests.get(DOCUMENT_API,data=json_data, headers=headers)
            history_document =json.loads(history_document.text).get("documents",[])
            history_document = [doc["documentname"] for doc in history_document]
            st.session_state.doument_slected_to_chat = st.selectbox("Choose Documnet",history_document)
            data = {"token": st.session_state.token, "service_name": st.session_state.service_slected_to_chat,"document_name":st.session_state.doument_slected_to_chat}
            json_data = json.dumps(data)
            headers = {'Content-Type': 'application/json'}
            number_pages = requests.get(GET_NUM_PAGES,data=json_data, headers=headers)
            number_pages =json.loads(number_pages.text).get("num_pages")
            page_options = list(range(1, int(number_pages) + 1))

            st.session_state.start_page = st.selectbox("Start Page",page_options)
            st.session_state.end_page = st.selectbox("End Page", page_options, index=len(page_options) - 1)
            st.session_state.method = st.selectbox("Chunking Method", ["chunk_per_page", "personalize_chunking"])
            if st.session_state.method=="personalize_chunking":
               st.session_state.split_token = st.text_area("Split Token")
        else:
            st.session_state.service_slected_to_chat = None

            
    elif st.session_state.genre == "Delete service(s)":
        st.title('Delete Service(s)')
        data = {"token": st.session_state.token}
        json_data = json.dumps(data)
        headers = {'Content-Type': 'application/json'}
        services  = requests.get(SERVICES_API,data=json_data, headers=headers)
        services =json.loads(services.text)
        if len(services)>=2:
            services.append("ALL")
            # Get list of documents from histrory
        if "ALL" in services:
            service_slected = st.multiselect(
                    "",services ,default="ALL"
                    )
        elif len(services)==1:
            service_slected = st.multiselect(
                    "",services,default=services[0]
                    )
        else:
            service_slected = st.multiselect(
                    "",services
                    )
        if "ALL" in service_slected:
            service_slected = services
            service_slected.remove("ALL")
        st.write("You selected:", service_slected)

        if len(service_slected) > 0:
            st.session_state.delete = st.button('Delete')
            if st.session_state.delete:
                delete_service(st.session_state.token ,service_slected)
        
    elif st.session_state.genre == "Delete document(s)":
        st.title('Delete Document(s)')
        data = {"token": st.session_state.token}
        json_data = json.dumps(data)
        headers = {'Content-Type': 'application/json'}
        services  = requests.get(SERVICES_API,data=json_data, headers=headers)
        services =json.loads(services.text)
        if len(services)>0:
            service = st.selectbox("Choose Service",services)
            data = {"token": st.session_state.token, "servicename": service}
            json_data = json.dumps(data)
            headers = {'Content-Type': 'application/json'}
            history_document  = requests.get(DOCUMENT_API,data=json_data, headers=headers)
            history_document =json.loads(history_document.text).get("documents",[])
            history_document = [doc["documentname"] for doc in history_document]
            if len(history_document)>=2:
                history_document.append("ALL")
            # Get list of documents from histrory
            if "ALL" in history_document:
                document_slected_to_delete = st.multiselect(
                    "",history_document ,default="ALL"
                    )
            elif len(history_document)==1:
                document_slected_to_delete = st.multiselect(
                    "",history_document,default=history_document[0]
                    )
            else:
                document_slected_to_delete = st.multiselect(
                    "",history_document
                    )
            if "ALL" in document_slected_to_delete:
                document_slected_to_delete = history_document
                document_slected_to_delete.remove("ALL")

            st.write("You selected:", document_slected_to_delete)
            if len(document_slected_to_delete) > 0:
                st.session_state.delete = st.button('Delete')
                if st.session_state.delete:
                    delete_document(st.session_state.token,st.session_state.service ,document_slected_to_delete)

css_style = """
<style>
.title {
    white-space: nowrap;
}
</style>
"""

st.markdown(css_style, unsafe_allow_html=True)

with st.container():
    st.markdown('<h1 class="title">Augmented Generative Document Scraper</h1>', unsafe_allow_html=True)
    if st.session_state.genre=="Select document" and st.session_state.service_slected_to_chat:
        schema = display_and_validate_schema()
        comments = None
        if schema and st.checkbox("Add comments")  :
            keys = get_all_keys(schema)
            comments = handle_comments(keys)
        if schema and st.button('Process') :
            data = {"token": st.session_state.token,
            "service_name": st.session_state.service_slected_to_chat,
            "document_name": st.session_state.doument_slected_to_chat,
            "method": st.session_state.method,
            "model": st.session_state.llm,
            "schema": schema,
            "comment": comments,
            "split_token": st.session_state.split_token if st.session_state.method == "personalize_chunking" else "",
            "start_page": st.session_state.start_page,
            "end_page": st.session_state.end_page}
            json_data = json.dumps(data)
            headers = {'Content-Type': 'application/json'}
            response  = requests.get(RESPONSE_API,data=json_data, headers=headers)
            response_data = json.loads(response.text)
            if response_data.get('status')=='success':
                json_str =response_data.get("json")
                json_str_formatted = json.dumps(json_str)
    
                # Encode this JSON string to bytes, which is required for the download
                json_bytes = json_str_formatted.encode('utf-8')
                st.download_button(
                    label="Download JSON",
                    data=json_bytes,
                    file_name="results.json",
                    mime="application/json"
                )
            else:
                st.error("Error in processing document")