wehub-resource-sync
e768098d0e
tools_continuous_delivery / Private PyPI non-main branch release (push) Has been skipped
tools_continuous_delivery / Private PyPI main branch release (push) Failing after 2m42s
Publish Promptflow Doc / Build (push) Has been cancelled
Publish Promptflow Doc / Deploy (push) Has been cancelled
Flake8 Lint / flake8 (push) Has been cancelled
Spell check CI / Spell_Check (push) Has been cancelled
73 lines
2.0 KiB
Python
73 lines
2.0 KiB
Python
from enum import Enum
|
|
from promptflow.core import tool
|
|
|
|
MAX_CONV_ITEM_LEN = 1000
|
|
|
|
|
|
class ConversationModality(str, Enum):
|
|
TEXT = "text"
|
|
TRANSCRIPT = "transcript"
|
|
|
|
|
|
def create_conversation_item(name: str, text: str) -> dict:
|
|
return {
|
|
"participantId": name,
|
|
"role": name if name.lower() in {"customer", "agent"} else "generic",
|
|
"text": text
|
|
}
|
|
|
|
|
|
def parse_conversation_line(line: str) -> list[dict]:
|
|
name_and_text = line.split(":", maxsplit=1)
|
|
name = name_and_text[0].strip()
|
|
text = name_and_text[1].strip()
|
|
conv_items = []
|
|
sentences = [s.strip() for s in text.split(".")]
|
|
buffer = ""
|
|
|
|
for sentence in sentences:
|
|
if len(buffer.strip()) + len(sentence) + 2 >= MAX_CONV_ITEM_LEN:
|
|
conv_items.append(create_conversation_item(name, buffer.strip()))
|
|
buffer = ""
|
|
buffer += " " + sentence + "."
|
|
|
|
conv_items.append(create_conversation_item(name, buffer.strip()))
|
|
return conv_items
|
|
|
|
|
|
@tool
|
|
def create_conversation(text: str,
|
|
modality: ConversationModality,
|
|
language: str,
|
|
id: int) -> dict:
|
|
"""
|
|
This tool creates a conversation input for conversation-based
|
|
language skills.
|
|
|
|
Conversation text is assumed to be of the following form:
|
|
<speaker id>: <speaker text>
|
|
<speaker id>: <speaker text>
|
|
...
|
|
|
|
:param text: conversation text.
|
|
:param modality: conversation modality.
|
|
:param language: conversation language.
|
|
:param id: conversation id.
|
|
"""
|
|
conv_items = []
|
|
id = 0
|
|
lines = text.replace(" ", "\n").split("\n")
|
|
lines = filter(lambda line: len(line.strip()) != 0, lines)
|
|
for line in lines:
|
|
for conv_item in parse_conversation_line(line):
|
|
id += 1
|
|
conv_item["id"] = id
|
|
conv_items.append(conv_item)
|
|
|
|
return {
|
|
"conversationItems": conv_items,
|
|
"language": language,
|
|
"modality": modality,
|
|
"id": str(id)
|
|
}
|