Document how to use CDC #2871
Document how to use CDC #2871
Conversation
There was a problem hiding this comment.
Great contribution @Pavan-Nambi!
@sivukhin and I were just discussing updating the docs to include this, so amazing you contributed just in time! 🤩
I've left a few suggestions, @sivukhin and I can go through these, and once applied, we can get this merged.
Co-authored-by: Jamie Barton <jamie@notrab.dev>
Co-authored-by: Jamie Barton <jamie@notrab.dev>
Co-authored-by: Jamie Barton <jamie@notrab.dev>
Co-authored-by: Jamie Barton <jamie@notrab.dev>
Co-authored-by: Jamie Barton <jamie@notrab.dev>
Co-authored-by: Jamie Barton <jamie@notrab.dev>
|
Hey, thanks, i accepted commit suggestions , btw can u provide any input at #1993 - it's related to cdc - i m not exactly sure if that's what we want. thanks. |
|
cc @notrab |
Co-authored-by: Jamie Barton <jamie@notrab.dev>
Co-authored-by: Jamie Barton <jamie@notrab.dev>
|
|
||
| ``` | ||
|
|
||
| If you modify your table schema (adding/dropping columns), the `table_columns_json_array()` function returns the current schema, not the historical one. This can lead to incorrect results when decoding older CDC records. Manually track schema versions by storing the output of `table_columns_json_array()` before making schema changes. |
There was a problem hiding this comment.
This is first mention of that function. Let's create small section which will describe "helper" function to work with CDC (table_columns_json_array and bin_record_json_object)
| ├───────────┼─────────────┼─────────────┼───────────────┼────┼──────────┼──────────────────────────────────────────────────────────────────────────────┼───────────────┤ | ||
| │ 2 │ 1756713176 │ 1 │ users │ 1 │ │ John │ │ | ||
| ├───────────┼─────────────┼─────────────┼───────────────┼────┼──────────┼──────────────────────────────────────────────────────────────────────────────┼───────────────┤ | ||
| │ 3 │ 1756713176 │ 1 │ users │ 2 │ │ Jane │ │ |
There was a problem hiding this comment.
Should we use helper functions in the example in order to have human-readable before/after columns?
There was a problem hiding this comment.
@Pavan-Nambi, thanks a lot for the contribution!
I left few comments and some of them can be bit more complex than other NITs. So feel free to delay comments fixes if they are unclear to you.
Generally, I think this is great start and we can merge this PR and improve docs on top of this work in separate branches.
Co-authored-by: Jamie Barton <jamie@notrab.dev>
Co-authored-by: Jamie Barton <jamie@notrab.dev>
Co-authored-by: Jamie Barton <jamie@notrab.dev>
Co-authored-by: Jamie Barton <jamie@notrab.dev>
|
|
||
|
|
||
|
|
||
| - **`change_id` (INTEGER)** |
There was a problem hiding this comment.
I got most of this from test_cdc , if i wrote anything wrong lmk, thanks
| Granular details about the change. | ||
| - For UPDATE: shows specific column modifications. | ||
|
|
||
|
|
There was a problem hiding this comment.
I am assuming this is intended behaviour, when i tried cdc with transactions, this is the behaviour i experienced
|
I did not resolved this comment - this.. i m not sure either , tbh that table looks readable to me but perhaps we can make it more clear with just displaying before and after? idk. this comment - not really sure how these functions work - sorry i am new to codebase. |
Fixes #2221