liurenjie1024 commented on issue #246:
URL: https://github.com/apache/iceberg-rust/issues/246#issuecomment-2003753550
Close by #259 Feel free to open when if necessary.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use t
liurenjie1024 closed issue #246: Loading table from metadata file directly.
URL: https://github.com/apache/iceberg-rust/issues/246
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment
a-agmon commented on issue #246:
URL: https://github.com/apache/iceberg-rust/issues/246#issuecomment-1998907464
> > Maybe we can also incorporate something like this at this stage. like
adding a _readonly_ property that will default to false, while the StaticTable
sets it as true.
> > ``
liurenjie1024 commented on issue #246:
URL: https://github.com/apache/iceberg-rust/issues/246#issuecomment-1998709854
> Maybe we can also incorporate something like this at this stage. like
adding a _readonly_ property that will default to false, while the StaticTable
sets it as true.
>
a-agmon commented on issue #246:
URL: https://github.com/apache/iceberg-rust/issues/246#issuecomment-1998312408
Maybe we can also incorporate something like this at this stage. like adding
a _readonly_ property that will default to false, while the StaticTable sets it
as true.
```rus
liurenjie1024 commented on issue #246:
URL: https://github.com/apache/iceberg-rust/issues/246#issuecomment-1997160521
> Alought I still stick with my idea that Table should be able to build
without catalog and perform runtime checks againest read_only status, returning
errors if user tried
Xuanwo commented on issue #246:
URL: https://github.com/apache/iceberg-rust/issues/246#issuecomment-1996773784
> Good point, how about allowing access inner Table as unsafe?
Providing a `StaticTable::into_table()` should work fine to me.
Alought I still stick with my idea that `
liurenjie1024 commented on issue #246:
URL: https://github.com/apache/iceberg-rust/issues/246#issuecomment-1996770933
If that still not works for you, I agree that we provide both apis.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on t
liurenjie1024 commented on issue #246:
URL: https://github.com/apache/iceberg-rust/issues/246#issuecomment-1996767094
> Perhaps we can also implement `from_metadata_file()` on `Table`? or
provide an accessor to `Table` from `StaticTable` with some warning on usage?
Good point, how ab
Xuanwo commented on issue #246:
URL: https://github.com/apache/iceberg-rust/issues/246#issuecomment-1996763599
> Will following code work for you?
The issue I encountered is that the `Table`'s read-only status is determined
at runtime, not compile time. Introducing a `const READ_ONLY:
liurenjie1024 commented on issue #246:
URL: https://github.com/apache/iceberg-rust/issues/246#issuecomment-1996758139
> > It's kind of weird to me to have both of them. Is there any concrete use
case?
>
>
>
> Database like databend supports loading iceberg table from both cata
a-agmon commented on issue #246:
URL: https://github.com/apache/iceberg-rust/issues/246#issuecomment-1996749250
Perhaps we can also implement `from_metadata_file()` on `Table`? or provide
an accessor to `Table` from `StaticTable` with some warning on usage?
--
This is an automated messag
Xuanwo commented on issue #246:
URL: https://github.com/apache/iceberg-rust/issues/246#issuecomment-1996714186
> It's kind of weird to me to have both of them. Is there any concrete use
case?
Database like databend supports loading iceberg table from both catalog and
paths of a stora
liurenjie1024 commented on issue #246:
URL: https://github.com/apache/iceberg-rust/issues/246#issuecomment-1996709656
> > But The motivation behind Is that we want to do it in a type safe way
without runtime check.
>
>
>
> Do you think it's a good idea to provide both ways?
Xuanwo commented on issue #246:
URL: https://github.com/apache/iceberg-rust/issues/246#issuecomment-1996678394
> But The motivation behind Is that we want to do it in a type safe way
without runtime check.
Do you think it's a good idea to provide both ways?
--
This is an automated
liurenjie1024 commented on issue #246:
URL: https://github.com/apache/iceberg-rust/issues/246#issuecomment-1996611762
> Should simply returning an error for modification operations be acceptable?
Yeah, that's one approach. But The motivation behind Is that we want to do
it in a type s
Xuanwo commented on issue #246:
URL: https://github.com/apache/iceberg-rust/issues/246#issuecomment-1996603012
From the discussion:
https://apache-iceberg.slack.com/archives/C05HTENMJG4/p1710392414486259
> The reason we need StaticTable Is that we don't want to allow modidication
on
liurenjie1024 commented on issue #246:
URL: https://github.com/apache/iceberg-rust/issues/246#issuecomment-1993843083
Hi, @a-agmon I've reviewed the pr and it looks great! I've left small
comments to improve it.
--
This is an automated message from the Apache Git Service.
To respond to th
a-agmon commented on issue #246:
URL: https://github.com/apache/iceberg-rust/issues/246#issuecomment-1993467063
Hi @liurenjie1024 ,
Please take a look at the
[PR](https://github.com/apache/iceberg-rust/pull/259)
How does this look?
--
This is an automated message from the Apache G
a-agmon commented on issue #246:
URL: https://github.com/apache/iceberg-rust/issues/246#issuecomment-1991294429
Thanks @liurenjie1024
Sounds good to me.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above
liurenjie1024 commented on issue #246:
URL: https://github.com/apache/iceberg-rust/issues/246#issuecomment-1991256797
Hi, @a-agmon
I think both method could be provided and the second version could be
implemented as a simple wrapper as the first one.
What I want to add is that
a-agmon commented on issue #246:
URL: https://github.com/apache/iceberg-rust/issues/246#issuecomment-1991011883
Hi @liurenjie1024 ,
Thanks for adding this in response to my question on Slack :-)
I would be happy to work on this as good-first-issue if you think its a good
idea.
I h
22 matches
Mail list logo