Querying jira tickets using from langchain_community.agent_toolkits.jira.toolkit import JiraToolkit

[shalinibnair]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

from langchain.agents import AgentType, initialize_agent
from langchain_community.agent_toolkits.jira.toolkit import JiraToolkit

_prompt = PromptTemplate(input_variables=["input"], template=template)
prompt = _prompt.format(
    input="get parent issues and title from all tasks and stories from DHP project in jira"
)

agent = initialize_agent(
    tools=tools, llm=llm, agent=AgentType.ZERO_SHOT_REACT_DESCRIPTION, verbose=True
)
response = agent.run(prompt)

Description

I am trying to use JiraToolkit to get the records from jira based on a question asked by the user.
The issue I am facing is that the answer returned by the agent return only first 50. How to get the entire result.
Langchain documentation gives sample of deprecated usage

System Info

latest stable version of langchain

[dosubot[bot]]

Hey @shalinibnair! How can I assist you today? I'm here to help with any bugs, questions, or contributions you have in mind.

To handle pagination and retrieve more than the first 50 results when querying Jira using the JiraToolkit, you will need to adjust the pagination settings in your query. Unfortunately, the specific details on how to modify the pagination limit or fetch additional pages using the JiraToolkit are not provided in the available documentation.

However, you can try the following general approach to handle pagination:

1. Modify the JQL Query: Adjust your JQL query to include pagination parameters such as startAt and maxResults.

2. Iterate Over Pages: Implement a loop to fetch results page by page until all results are retrieved.

Here's an example of how you might modify your code to handle pagination:

from langchain.agents import AgentType, initialize_agent
from langchain_community.agent_toolkits.jira.toolkit import JiraToolkit
from langchain_community.utilities.jira import JiraAPIWrapper

# Initialize the Jira API wrapper
jira_api_wrapper = JiraAPIWrapper(api_key="your_api_key", server="your_jira_server")

# Create the JiraToolkit from the JiraAPIWrapper
jira_toolkit = JiraToolkit.from_jira_api_wrapper(jira_api_wrapper)

# Initialize the agent with the JiraToolkit tools
agent = initialize_agent(tools=jira_toolkit.get_tools(), llm=llm, agent=AgentType.ZERO_SHOT_REACT_DESCRIPTION, verbose=True)

# Define the JQL query
jql_query = "project = DHP AND (issuetype = Task OR issuetype = Story)"

# Function to fetch all results with pagination
def fetch_all_results(jql_query, max_results=50):
    start_at = 0
    all_results = []
    while True:
        paginated_query = f"{jql_query} ORDER BY created ASC&startAt={start_at}&maxResults={max_results}"
        prompt = _prompt.format(input=paginated_query)
        response = agent.run(prompt)
        results = response['issues']
        if not results:
            break
        all_results.extend(results)
        start_at += max_results
    return all_results

# Fetch all results
all_issues = fetch_all_results(jql_query)

This code snippet demonstrates how to handle pagination by iterating over pages of results and combining them into a single list. Adjust the max_results parameter as needed to control the number of results per page.

If you need more specific details or examples, you may need to refer to the Jira API documentation or the implementation details of the JiraAPIWrapper and JiraToolkit in the LangChain repository.

To continue talking to Dosu, mention @dosu.