Fokko commented on code in PR #5116:
URL: https://github.com/apache/iceberg/pull/5116#discussion_r908085208
##########
python/src/iceberg/avro/reader.py:
##########
@@ -211,26 +282,39 @@ class MapReader(Reader):
def read(self, decoder: BinaryDecoder) -> dict:
read_items = {}
- block_count = decoder.read_long()
+ block_count = decoder.read_int()
while block_count != 0:
if block_count < 0:
block_count = -block_count
# We ignore the block size for now
- _ = decoder.read_long()
+ _ = decoder.read_int()
for _ in range(block_count):
key = self.key.read(decoder)
read_items[key] = self.value.read(decoder)
- block_count = decoder.read_long()
+ block_count = decoder.read_int()
return read_items
+ def skip(self, decoder: BinaryDecoder) -> None:
+ block_count = decoder.read_int()
+ while block_count != 0:
+ if block_count < 0:
+ block_count = -block_count
+ block_size = decoder.read_int()
+ decoder.skip(block_size)
+ else:
+ for _ in range(block_count):
+ self.key.skip(decoder)
+ self.value.skip(decoder)
+ block_count = decoder.read_int()
Review Comment:
I've refactored this to a function:
```python
def _skip_map_array(decoder: BinaryDecoder, skip_entry: Callable) -> None:
"""Skips over an array or map
Both the array and map are encoded similar, and we can re-use
the logic of skipping in an efficient way.
From the Avro spec:
Maps (and arrays) are encoded as a series of blocks.
Each block consists of a long count value, followed by that many
key/value pairs in the case of a map,
and followed by that many array items in the case of an array. A block
with count zero indicates the
end of the map. Each item is encoded per the map's value schema.
If a block's count is negative, its absolute value is used, and the
count is followed immediately by a
long block size indicating the number of bytes in the block. This block
size permits fast skipping
through data, e.g., when projecting a record to a subset of its fields.
Args:
decoder:
The decoder that reads the types from the underlying data
skip_entry:
Function to skip over the underlying data, element in case of an
array, and the
key/value in the case of a map
"""
block_count = decoder.read_int()
while block_count != 0:
if block_count < 0:
# The length in bytes in encoded, so we can skip over it right
away
block_size = decoder.read_int()
decoder.skip(block_size)
else:
[skip_entry() for _ in range(block_count)]
block_count = decoder.read_int()
```
--
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]
