[arrow-nanoarrow] branch main updated: fix: Improve readability of `ArrowArrayAllocateChildren()` (#199)

[paleolimbot]

This is an automated email from the ASF dual-hosted git repository.

paleolimbot pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/arrow-nanoarrow.git


The following commit(s) were added to refs/heads/main by this push:
     new 6494471  fix: Improve readability of `ArrowArrayAllocateChildren()` 
(#199)
6494471 is described below

commit 64944714534d488593ee930d383eb5581fb440fe
Author: Dirk Eddelbuettel <e...@debian.org>
AuthorDate: Sun May 21 21:22:43 2023 -0500

    fix: Improve readability of `ArrowArrayAllocateChildren()` (#199)
    
    In `ArrowArrayAllocateChildren()` two loops over all children follow
    each other directly. The first assigns a NULL to each element, then
    second overwrites that with an allocation. That makes the first loop
    effectively deadweight. It may be there on purpose in which case you
    just disregard and close this PR. But if it is an oversight this will
    make it ever-so-marginally faster as we can skip one loop over
    `n_children`.
    
    (I noticed another microscopic style difference: in the corresponding
    `ArrowSchemaAllocateChildren()` an `if (n_children != 0) {...}` wraps
    around the core code of function; there it is a simpler `if (n_children
    == 0) return ...`. Of course they are also the same "in practice" and I
    don't have a style preference either -- just curious that they differ.
    No need for change, but while at it you could....)
    
    ---------
    
    Co-authored-by: Dewey Dunnington <de...@fishandwhistle.net>
---
 src/nanoarrow/array.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/src/nanoarrow/array.c b/src/nanoarrow/array.c
index cd6a639..ec321e6 100644
--- a/src/nanoarrow/array.c
+++ b/src/nanoarrow/array.c
@@ -228,9 +228,7 @@ ArrowErrorCode ArrowArrayAllocateChildren(struct 
ArrowArray* array, int64_t n_ch
     return ENOMEM;
   }
 
-  for (int64_t i = 0; i < n_children; i++) {
-    array->children[i] = NULL;
-  }
+  memset(array->children, 0, n_children * sizeof(struct ArrowArray*));
 
   for (int64_t i = 0; i < n_children; i++) {
     array->children[i] = (struct ArrowArray*)ArrowMalloc(sizeof(struct 
ArrowArray));