geruh commented on code in PR #3009:
URL: https://github.com/apache/iceberg-python/pull/3009#discussion_r2985267794
##########
pyiceberg/manifest.py:
##########
@@ -1184,6 +1184,92 @@ def existing(self, entry: ManifestEntry) ->
ManifestWriter:
return self
+class RollingManifestWriter:
+ """As opposed to ManifestWriter, a rolling writer could produce multiple
manifest files."""
+
+ _ROWS_DIVISOR = 250
Review Comment:
Yeah this is a direct port of the java implementations
[`RollingmanifestWriter.ROW_DIVISOR`](https://github.com/apache/iceberg/blob/a7a09d43b10e4584faf5598c40bc8fc424c402e1/core/src/main/java/org/apache/iceberg/RollingManifestWriter.java#L31).
this added divisor is basically a interval where we check the file size
against target-size-bytes every 250 entries rather than on every write. But
after some digging it looks like in PyIceberg `tell()` on both the PyArrow and
fsspec `OutputStream` implementations is essentially free and use their own
counters so maybe we can simplify this check and ignore the `ROW_DIVISOR`, but
still doesn't hurt to have wdyt?
--
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.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
