add text area option
Browse files- app.py +38 -20
- utlis/__pycache__/constant.cpython-39.pyc +0 -0
- utlis/__pycache__/helper.cpython-39.pyc +0 -0
- utlis/helper.py +27 -0
app.py
CHANGED
|
@@ -33,11 +33,21 @@ with st.sidebar:
|
|
| 33 |
|
| 34 |
|
| 35 |
if len(services)>0:
|
| 36 |
-
st.session_state.
|
| 37 |
-
if st.session_state.
|
| 38 |
-
st.session_state.
|
| 39 |
-
if st.session_state.
|
| 40 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 41 |
|
| 42 |
elif st.session_state.genre=="Select document":
|
| 43 |
st.title('Scrape Document')
|
|
@@ -69,7 +79,8 @@ with st.sidebar:
|
|
| 69 |
st.session_state.method = st.selectbox("Chunking Method", ["chunk_per_page", "personalize_chunking"])
|
| 70 |
if st.session_state.method=="personalize_chunking":
|
| 71 |
st.session_state.split_token = st.text_area("Split Token")
|
| 72 |
-
elif st.session_state.doument_slected_to_chat.split("_")[-1]=="txt":
|
|
|
|
| 73 |
st.session_state.method = st.selectbox("Chunking Method", ["personalize_chunking"])
|
| 74 |
st.session_state.split_token = st.text_area("Split Token")
|
| 75 |
else:
|
|
@@ -160,6 +171,12 @@ st.markdown(css_style, unsafe_allow_html=True)
|
|
| 160 |
|
| 161 |
with st.container():
|
| 162 |
st.markdown('<h1 class="title">Augmented Generative Document Scraper</h1>', unsafe_allow_html=True)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 163 |
if st.session_state.genre=="Select document" and st.session_state.service_slected_to_chat:
|
| 164 |
schema = display_and_validate_schema()
|
| 165 |
comments = None
|
|
@@ -182,7 +199,8 @@ with st.container():
|
|
| 182 |
headers = {'Content-Type': 'application/json'}
|
| 183 |
response = requests.get(RESPONSE_API,data=json_data, headers=headers)
|
| 184 |
response_data = json.loads(response.text)
|
| 185 |
-
elif st.session_state.doument_slected_to_chat.split("_")[-1]=="txt":
|
|
|
|
| 186 |
data = {"token": st.session_state.token,
|
| 187 |
"service_name": st.session_state.service_slected_to_chat,
|
| 188 |
"document_name": st.session_state.doument_slected_to_chat,
|
|
@@ -195,18 +213,18 @@ with st.container():
|
|
| 195 |
headers = {'Content-Type': 'application/json'}
|
| 196 |
response = requests.get(RESPONSE_TXT_API,data=json_data, headers=headers)
|
| 197 |
response_data = json.loads(response.text)
|
| 198 |
-
|
| 199 |
-
|
| 200 |
-
|
| 201 |
|
| 202 |
-
|
| 203 |
-
|
| 204 |
-
|
| 205 |
-
|
| 206 |
-
|
| 207 |
-
|
| 208 |
-
|
| 209 |
-
|
| 210 |
-
|
| 211 |
-
|
| 212 |
|
|
|
|
| 33 |
|
| 34 |
|
| 35 |
if len(services)>0:
|
| 36 |
+
st.session_state.doc_ortext = st.radio("Choose option",["Documnt", "Text area"])
|
| 37 |
+
if st.session_state.doc_ortext=="Documnt":
|
| 38 |
+
st.session_state.uploaded_files = st.file_uploader("Upload PDF file", type=["pdf","txt"], accept_multiple_files=False)
|
| 39 |
+
if st.session_state.uploaded_files:
|
| 40 |
+
st.session_state.process = st.button('Process')
|
| 41 |
+
if st.session_state.process:
|
| 42 |
+
add_document(st.session_state.token,st.session_state.service)
|
| 43 |
+
# elif st.session_state.doc_ortext=="Text area":
|
| 44 |
+
# st.session_state.name_text_area = st.container().text_area("Enter name of the text area")
|
| 45 |
+
# st.session_state.text_area = st.container().text_area("Enter text")
|
| 46 |
+
|
| 47 |
+
# if st.session_state.text_area:
|
| 48 |
+
# st.session_state.process = st.container().button('Process')
|
| 49 |
+
# if st.session_state.process:
|
| 50 |
+
# add_text_document(st.session_state.token,st.session_state.service)
|
| 51 |
|
| 52 |
elif st.session_state.genre=="Select document":
|
| 53 |
st.title('Scrape Document')
|
|
|
|
| 79 |
st.session_state.method = st.selectbox("Chunking Method", ["chunk_per_page", "personalize_chunking"])
|
| 80 |
if st.session_state.method=="personalize_chunking":
|
| 81 |
st.session_state.split_token = st.text_area("Split Token")
|
| 82 |
+
#elif st.session_state.doument_slected_to_chat.split("_")[-1]=="txt":
|
| 83 |
+
else:
|
| 84 |
st.session_state.method = st.selectbox("Chunking Method", ["personalize_chunking"])
|
| 85 |
st.session_state.split_token = st.text_area("Split Token")
|
| 86 |
else:
|
|
|
|
| 171 |
|
| 172 |
with st.container():
|
| 173 |
st.markdown('<h1 class="title">Augmented Generative Document Scraper</h1>', unsafe_allow_html=True)
|
| 174 |
+
if st.session_state.genre=="Add document(s)" and st.session_state.doc_ortext == "Text area":
|
| 175 |
+
st.session_state.name_text_area = st.text_input("Enter name of the text area:")
|
| 176 |
+
st.session_state.text_area = st.text_area("Enter text:")
|
| 177 |
+
if st.session_state.text_area:
|
| 178 |
+
if st.button('Process Text'):
|
| 179 |
+
add_text_document(st.session_state.token, st.session_state.service)
|
| 180 |
if st.session_state.genre=="Select document" and st.session_state.service_slected_to_chat:
|
| 181 |
schema = display_and_validate_schema()
|
| 182 |
comments = None
|
|
|
|
| 199 |
headers = {'Content-Type': 'application/json'}
|
| 200 |
response = requests.get(RESPONSE_API,data=json_data, headers=headers)
|
| 201 |
response_data = json.loads(response.text)
|
| 202 |
+
#elif st.session_state.doument_slected_to_chat.split("_")[-1]=="txt":
|
| 203 |
+
else:
|
| 204 |
data = {"token": st.session_state.token,
|
| 205 |
"service_name": st.session_state.service_slected_to_chat,
|
| 206 |
"document_name": st.session_state.doument_slected_to_chat,
|
|
|
|
| 213 |
headers = {'Content-Type': 'application/json'}
|
| 214 |
response = requests.get(RESPONSE_TXT_API,data=json_data, headers=headers)
|
| 215 |
response_data = json.loads(response.text)
|
| 216 |
+
if response_data.get('status')=='success':
|
| 217 |
+
json_str =response_data.get("json")
|
| 218 |
+
json_str_formatted = json.dumps(json_str)
|
| 219 |
|
| 220 |
+
# Encode this JSON string to bytes, which is required for the download
|
| 221 |
+
json_bytes = json_str_formatted.encode('utf-8')
|
| 222 |
+
st.download_button(
|
| 223 |
+
label="Download JSON",
|
| 224 |
+
data=json_bytes,
|
| 225 |
+
file_name="results.json",
|
| 226 |
+
mime="application/json"
|
| 227 |
+
)
|
| 228 |
+
else:
|
| 229 |
+
st.error("Error in processing document")
|
| 230 |
|
utlis/__pycache__/constant.cpython-39.pyc
CHANGED
|
Binary files a/utlis/__pycache__/constant.cpython-39.pyc and b/utlis/__pycache__/constant.cpython-39.pyc differ
|
|
|
utlis/__pycache__/helper.cpython-39.pyc
CHANGED
|
Binary files a/utlis/__pycache__/helper.cpython-39.pyc and b/utlis/__pycache__/helper.cpython-39.pyc differ
|
|
|
utlis/helper.py
CHANGED
|
@@ -117,6 +117,33 @@ def add_service(token,servicename):
|
|
| 117 |
st.success(f"{servicename} added successfully")
|
| 118 |
else:
|
| 119 |
st.error(response.text)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 120 |
def add_document(token,servicename):
|
| 121 |
|
| 122 |
|
|
|
|
| 117 |
st.success(f"{servicename} added successfully")
|
| 118 |
else:
|
| 119 |
st.error(response.text)
|
| 120 |
+
def add_text_document(token, servicename):
|
| 121 |
+
# Retrieve text and document name from session state
|
| 122 |
+
document_text = st.session_state.text_area
|
| 123 |
+
document_name = st.session_state.name_text_area.replace(" ", "_").replace("(", "_").replace(")", "_").replace("-", "_").replace(".", "_")
|
| 124 |
+
|
| 125 |
+
# Encode the document text as Base64
|
| 126 |
+
encoded_text = base64.b64encode(document_text.encode('utf-8')).decode('utf-8')
|
| 127 |
+
|
| 128 |
+
url = ADD_STORE_DOCUMENT
|
| 129 |
+
# Prepare the JSON payload
|
| 130 |
+
data = {
|
| 131 |
+
"token": token,
|
| 132 |
+
"service_name": servicename,
|
| 133 |
+
"document_name": document_name,
|
| 134 |
+
"file": encoded_text # Assuming the API can handle Base64 encoded text under the 'file' key
|
| 135 |
+
}
|
| 136 |
+
|
| 137 |
+
# Convert the dictionary to a JSON formatted string and send the POST request
|
| 138 |
+
headers = {'Content-Type': 'application/json'}
|
| 139 |
+
response = requests.post(url, data=json.dumps(data), headers=headers)
|
| 140 |
+
status = json.loads(response.text).get("status")
|
| 141 |
+
|
| 142 |
+
if status == "success":
|
| 143 |
+
st.success(f"{document_name} uploaded successfully as text")
|
| 144 |
+
else:
|
| 145 |
+
st.error(f"{document_name} not uploaded successfully")
|
| 146 |
+
|
| 147 |
def add_document(token,servicename):
|
| 148 |
|
| 149 |
|