[Feature] CSV Support

[RoyiAvital]

Problem Description

Just like there is a specialized view for JSON / TOML / etc.
It should have a good view of the columns with colors (Colors like RainbowCSV).
The columns should be aligned (Fixed Width Format, See UltraEdit).

Proposed Solution

Build something comparable to the CSV support in EmEditor.

Use Cases

1. Viewing large CSV's.
2. Doing basic row / column operations (Delete / Move).
3. Easy to navigate between columns / rows.

Mockups / Examples

See EmEditor.

Additional Context

It could be a great differentable

Priority

How important is this feature to your workflow?

• Critical - blocking my work
• High - significant improvement
• Medium - nice to have
• Low - minor convenience

[sr79368142]

Hi! Protect your privacy, buddy!😸😸😸

[RoyiAvital]

Hi! Protect your privacy, buddy!😸😸😸

You mean the image of the CSV data?
This is an image from the UltraEdit site (See https://www.ultraedit.com/wp-content/uploads/2022/11/working-with-csv4.png).
I assume it is completely made up list.

[OlaProeis]

Hey, thanks for the suggestion!

CSV support is an interesting idea. I could see it working similar to how JSON/YAML/TOML files get the tree viewer - basically a table view where you can see and edit cells.

A few questions to understand what you're looking for:

Do you mainly want to view CSVs or edit them too? Are you working with large files that might need special handling for performance? And do you need just comma-separated, or also tabs and semicolons?

Let me know more about your use case and I'll add this to the backlog. Thanks!

[RoyiAvital]

I'd say the priorities would be:

1. Viewing large files in a table view. Performance is surely the main motivation. Otherwise one can use Excel.
2. Searching (Search and Replace with Regular Expression, etc...) within the file.
3. Edit at the cell level.
4. Moving / Deleting Rows and Columns.

The important is to keep the table view.
So the user see the Table abstraction of the file while the engine keep syncing it with the RAW file data.
The EmEditor does it perfectly with amazing speed and efficiency.

I am not sure how hard is to support TSV (Tabs) and SSV (Semicolon, is there such thing?).
I guess most CSV parsers have support for all of them so from your point of view it will be "free".

[OlaProeis]

Thanks for the detailed writeup! This is a well-scoped request.

The references help a lot - RainbowCSV style coloring and EmEditor's fixed-width alignment are good targets to aim for. I can see how this fits with Ferrite's existing structured data support (JSON/YAML/TOML tree viewer).

A few thoughts on scope:

Basic version could include:

• Column-aware syntax highlighting with alternating colors
• Fixed-width column alignment in rendered view
• Click to select column/row

Fuller version would add:

• Virtual scrolling for large files (important for CSVs which can get huge)
• Basic operations like delete row, move column
• Multiple delimiter support (comma, tab, semicolon, pipe)
• Header row detection

Honestly this is a bigger feature than it might seem - proper CSV handling with large file performance, editing operations, and column alignment is a decent chunk of work. But I agree it would be a nice differentiator since most markdown editors ignore CSV entirely.

I'll add this to the backlog. Can't promise a timeline but it's on my radar now.

Quick question - what priority would you give this for your workflow? And roughly how large are the CSVs you typically work with?

[RoyiAvital]

I do a lot of ML so CSV's are my daily work.
I'd say the bigger ones are 0.1-1.5 [GB].

[OlaProeis]

0.1-1.5 GB - that's serious scale! That changes the approach quite a bit.

For files that size we'd need memory-mapped file access and streaming parsing rather than loading into memory. EmEditor is impressive precisely because it handles this well.

Honest assessment: doing this right is a bigger project than I initially scoped. For v0.2.5 I'm planning to add CSV support that works well for "normal" sized files (up to maybe 50-100MB) with:

• Table view with rainbow column coloring
• Fixed-width alignment
• Virtual scrolling
• Delimiter detection

For GB-scale files, that would need to be a dedicated effort - probably v0.3.0 or later with proper memory-mapped I/O. I don't want to ship something that chokes on your real workload.

Would a "works great up to 100MB, handles larger files gracefully but slower" version be useful as a first step? Or is sub-GB basically unusable for your workflow?

[RoyiAvital]

It will be highly useful and handling it in iterative manner is the way to go.
I gave the range of the large files. Most of them are in the range of 10-100 [MB].

If you improve each iteration it would be great.
By the way, JSON files can also be quite large. So aiming to support memory mapped files should be on the roadmap. Even if farther down the road.

[OlaProeis]

10-100MB is very doable for v0.2.5.

Iterative approach it is:

• v0.2.5: Table view, rainbow colors, virtual scrolling - solid up to ~100MB
• Later: Memory-mapped I/O for GB-scale files

Good call on JSON too - large JSON files have the same issue. I'll add memory-mapped file support to the long-term roadmap as a general optimization that benefits CSV, JSON, and other structured formats.

Thanks for being flexible on scope. Will ping this issue when CSV lands.

[OlaProeis]

Hi @RoyiAvital,

CSV support has been implemented in v0.2.5 (just released)!

What's included:

• Native table view for CSV and TSV files with fixed-width column alignment
• Rainbow column coloring for improved readability (like RainbowCSV)
• Auto-detection of delimiters (comma, tab, semicolon, pipe)
• Intelligent header row detection and highlighting

Coming in v0.2.6:

• Virtual scrolling for large files (thousands of rows)
• Basic row/column operations

Please try it out and let us know if it meets your needs!

Closing the core feature request as completed.