Create Jira using Wiki content using REST API with text formatting

[Mahesh Mannanikat]

Hello All,

It's my pleasure to be part of this community and it's my first time working with Jira REST API (v2) and it's my first post to the group.

I have a requirement of creating Jira tickets from the content present in a confluence page. The data presented in a table with columns (mapped to Jira fields) and I need to read it and create Jira ticket using rest API.

I have wrote a Python script now, in that I used BeautifulSoup to extract the content from the table. And using that I am calling the jira_create from Python JIRA package. The challenge for me is to preserve the text formatting in the description field from wiki to the ticket. I see that V3 API has a built in support for ADF, but unfortunately I have only option of using the V2 now. I tried html to markdown conversion using markdownify, but it is not helping me much.

Can you please suggest me a preferred way to do this in ATLASSIAN world? 

Thank you

Mahesh

[Evgenii · Accepted solution]

Hi, @Mahesh Mannanikat 

Welcome to Altassian Community

You're very creative person. And it looks like you're trying to recreate the wheel :)

Atlassian already has python library to work with Confluence, and you don't need to use BSoup.

Take a look here: https://atlassian-python-api.readthedocs.io/

As for Markdown Conversion: although you mentioned that markdownify didn't work for you, you could explore other Python libraries specifically designed for converting HTML to Markdown. Some popular ones include html2text, mistune, pandoc, and turndown. These libraries may provide better results in preserving text formatting during the conversion process.

[Mahesh Mannanikat]

Thank you @Evgenii for your response and so excited to see how quick I am getting the response in this community. I am already using the same library to read the page, sorry that I didn't include that info in my post as I was more focusing on the text formatting/markdown.

Here is how I read the content from page

# Fetch Confluence page content
confluence = Confluence(url=confluence_url, username=jira_username, password=jira_password)
page = confluence.get_page_by_title(space='MySpace', title='Test Page-Delete', expand="body.storage")
confluence_page = page["body"]["storage"]["value"]

BSoup is used only to extract info from the table

# Parse Confluence page HTML
soup = BeautifulSoup(confluence_page, "html.parser")

# Find the table in the Confluence page
table = soup.find("table")

issues = []

# Iterate over table rows (skipping the header row)
for row in table.find_all("tr")[1:]:
   # Initialize issue fields
   issue_fields = {}

# Table column mapping
column_ref = ['', 'Task Name', '', 'Task Details', '', '', '', '', '', '']

# Extract data from each column and map it to Jira issue fields
for idx, cell in enumerate(row.find_all("td")):
   column_name = None
   column_value = None

   if column_ref[idx]: # skipping blank columns
      column_name = field_mappings.get(column_ref[idx])
      column_value = cell.get_text()
      if column_name == "Description":
         # For description, maintaining HTML to retain the text format info
         column_value = MarkdownConverter().convert_soup(cell)
      issue_fields[column_name] = column_value
   issues.append(issue_fields)

I have tried out all of the alternative converters you have mentioned with few others, but the result is same. My main challenge is the indents in the ordered list items. While I print the content locally, I see it properly indented but while the text displays in Jira web UI, it misses all of them. For example

The original text (also displaying with exact correct indents while printing the conversion result from HTML to Markdown)

1. Item 1
   1. Sub-item 1
   2. Sub-item 2
2. Item 2
   1. Sub-item 1
   2. Sub-item 2

But while Jira displays, the content is getting messed-up

1. Item 1
1. Sub-item 1
2. Sub-item 2
2. Item 2
1. Sub-item 1
2. Sub-item 2

 Not sure where the issue is, may be in API level or HTML rendering stage of Jira. Or may be due to my very limited knowledge in HTML and web development :) 

Thank you

Mahesh

[Evgenii]

Understood. 

In Jira, it's achieved with such construction:

# Line 1
## Line 1 - 1
## Line 1 - 2
# Line 2
## Line 2 - 1
## Line 2 - 2

I think, you'll have to add custom formatter, to generate such structure, and it'll work.

[Awarelogy]

@Mahesh Mannanikat  Hey can you please help me with the formatting as i am also stuggling with quite a similiar issue. 

https://community.atlassian.com/t5/Confluence-discussions/Unable-to-Create-formatted-Markdown-with-quot-representation/m-p/2427736#M11319