This is an automated email from the ASF dual-hosted git repository.
tustvold pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/arrow-rs.git
The following commit(s) were added to refs/heads/master by this push:
new c26bb8107 Add Fields::contains (#4026)
c26bb8107 is described below
commit c26bb810700149e912b020048ed378e09e5267df
Author: Raphael Taylor-Davies <1781103+tustv...@users.noreply.github.com>
AuthorDate: Thu Apr 6 20:10:32 2023 +0100
Add Fields::contains (#4026)
* Add Fields::contains
* Fix typo
---
arrow-schema/src/field.rs | 15 +++------------
arrow-schema/src/fields.rs | 17 +++++++++++++++++
arrow-schema/src/schema.rs | 20 ++++++++------------
3 files changed, 28 insertions(+), 24 deletions(-)
diff --git a/arrow-schema/src/field.rs b/arrow-schema/src/field.rs
index 1af157e4d..5edd5be7a 100644
--- a/arrow-schema/src/field.rs
+++ b/arrow-schema/src/field.rs
@@ -522,18 +522,9 @@ impl Field {
// self need to be nullable or both of them are not nullable
&& (self.nullable || !other.nullable)
// make sure self.metadata is a superset of other.metadata
- && match (&self.metadata.is_empty(), &other.metadata.is_empty()) {
- (_, true) => true,
- (true, false) => false,
- (false, false) => {
- other.metadata().iter().all(|(k, v)| {
- match self.metadata().get(k) {
- Some(s) => s == v,
- None => false
- }
- })
- }
- }
+ && other.metadata.iter().all(|(k, v1)| {
+ self.metadata.get(k).map(|v2| v1 == v2).unwrap_or_default()
+ })
}
/// Return size of this instance in bytes.
diff --git a/arrow-schema/src/fields.rs b/arrow-schema/src/fields.rs
index 07dff2aae..b93735328 100644
--- a/arrow-schema/src/fields.rs
+++ b/arrow-schema/src/fields.rs
@@ -66,6 +66,23 @@ impl Fields {
pub fn find(&self, name: &str) -> Option<(usize, &FieldRef)> {
self.0.iter().enumerate().find(|(_, b)| b.name() == name)
}
+
+ /// Check to see if `self` is a superset of `other`
+ ///
+ /// In particular returns true if both have the same number of fields, and
[`Field::contains`]
+ /// for each field across self and other
+ ///
+ /// In other words, any record that conforms to `other` should also
conform to `self`
+ pub fn contains(&self, other: &Fields) -> bool {
+ if Arc::ptr_eq(&self.0, &other.0) {
+ return true;
+ }
+ self.len() == other.len()
+ && self
+ .iter()
+ .zip(other.iter())
+ .all(|(a, b)| Arc::ptr_eq(a, b) || a.contains(b))
+ }
}
impl Default for Fields {
diff --git a/arrow-schema/src/schema.rs b/arrow-schema/src/schema.rs
index 2cc892f5a..a00e8a588 100644
--- a/arrow-schema/src/schema.rs
+++ b/arrow-schema/src/schema.rs
@@ -314,22 +314,18 @@ impl Schema {
Some((idx, field.as_ref()))
}
- /// Check to see if `self` is a superset of `other` schema. Here are the
comparison rules:
+ /// Check to see if `self` is a superset of `other` schema.
///
- /// * `self` and `other` should contain the same number of fields
- /// * for every field `f` in `other`, the field in `self` with
corresponding index should be a
- /// superset of `f`.
- /// * self.metadata is a superset of other.metadata
+ /// In particular returns true if `self.metadata` is a superset of
`other.metadata`
+ /// and [`Fields::contains`] for `self.fields` and `other.fields`
///
- /// In other words, any record conforms to `other` should also conform to
`self`.
+ /// In other words, any record that conforms to `other` should also
conform to `self`.
pub fn contains(&self, other: &Schema) -> bool {
- self.fields.len() == other.fields.len()
- && self.fields.iter().zip(other.fields.iter()).all(|(f1, f2)|
f1.contains(f2))
// make sure self.metadata is a superset of other.metadata
- && other.metadata.iter().all(|(k, v1)| match self.metadata.get(k) {
- Some(v2) => v1 == v2,
- _ => false,
- })
+ self.fields.contains(&other.fields)
+ && other.metadata.iter().all(|(k, v1)| {
+ self.metadata.get(k).map(|v2| v1 == v2).unwrap_or_default()
+ })
}
}
