Dev.to Markdown Tables/Bold Broken? Fixing the `markdownify` Double Processing Issue

I had a bit of trouble with Markdown tables and bold text getting messed up on Dev.to. It was pretty annoying to see my posts come out all jumbled every time I published. I wanted to share how I solved this issue.

Attempts and Pitfalls

When I tried to publish my posts on Dev.to, the tables and bold text I wrote in Markdown weren't displaying correctly and were showing up broken. At first, I thought I had written the Markdown syntax wrong and double-checked it multiple times.

| Header 1 | Header 2 |
|---|---|
| **Bold Text** | Some Data |

Even with a simple table and bold text like the above, it looked strange on Dev.to. It was definitely rendering fine locally.

It seemed like there was some kind of post-processing happening on the server that was handling the Markdown again. So, I started suspecting that if I was using a library like markdownify, maybe it was being processed twice.

The Cause

In the end, the problem was with the content_en field. This field already contained the content in Markdown format, and I was processing it again with markdownify, which caused a double Markdown conversion. Since it was already Markdown, trying to convert it to Markdown again messed up the syntax.

The Solution

The fix was simple. I modified the logic to use the content_en field directly if it was already in Markdown format, instead of reprocessing it with markdownify.

# Original code (assumed)
# content_en = "..." # Already in Markdown format
# processed_content = markdownify(content_en)
Modified code
content_en = "..." # Already in Markdown format
if not is_already_markdown(content_en): # In reality, you'd need logic to check the format of content_en
processed_content = markdownify(content_en)
else:
processed_content = content_en
When applying this in practice, it might vary depending on what values the content_en field holds
and how a function like is_already_markdown is implemented.
The important thing is to avoid extra conversion when it's already Markdown.

By skipping the markdownify processing for data that was already in Markdown format, the issue was resolved.

Results

• Markdown tables in posts published on Dev.to are now displayed correctly.
• Text formatting like bold and italics is no longer broken and appears as intended.
• I no longer spend time worrying about my content breaking after publishing.

In Summary — To Avoid the Same Pitfall

• [ ] When using Markdown conversion libraries, make it a habit to check if the input data is already in Markdown format.
• [ ] Clearly understand the format of data stored in fields like content_en and design your processing logic accordingly.
• [ ] Double processing is a common culprit for unexpected bugs. Always pay close attention to your data flow.