From af8db02078175e02abffaa4664d3aee0f0cc400d Mon Sep 17 00:00:00 2001
From: Hiroyuki Sato <hiroysato@gmail.com>
Date: Mon, 31 Aug 2026 10:45:50 +0900
Subject: [PATCH 01/13] GH-51090: [CI][C++] Fix shellcheck errors in
 cpp/build-support/fuzzing/generate_corpuses.sh

---
 .pre-commit-config.yaml                       |  8 +---
 .../fuzzing/generate_corpuses.sh              | 48 +++++++++----------
 2 files changed, 25 insertions(+), 31 deletions(-)

diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
index 9ddf332c57ca..eec8492f8264 100644
--- a/.pre-commit-config.yaml
+++ b/.pre-commit-config.yaml
@@ -288,13 +288,7 @@ repos:
           (
           ?^c_glib/.*\.sh$|
           ?^ci/.*\.sh$|
-          ?^cpp/build-support/build-lz4-lib\.sh$|
-          ?^cpp/build-support/build-zstd-lib\.sh$|
-          ?^cpp/build-support/get-upstream-commit\.sh$|
-          ?^cpp/build-support/run-test\.sh$|
-          ?^cpp/build-support/update-flatbuffers\.sh$|
-          ?^cpp/build-support/update-thrift\.sh$|
-          ?^cpp/build-support/vendor-flatbuffers\.sh$|
+          ?^cpp/build-support/.*\.sh$|
           ?^cpp/examples/minimal_build/run\.sh$|
           ?^cpp/examples/tutorial_examples/run\.sh$|
           ?^cpp/src/arrow/flight/sql/odbc/install/mac/postinstall$|
diff --git a/cpp/build-support/fuzzing/generate_corpuses.sh b/cpp/build-support/fuzzing/generate_corpuses.sh
index 07afa793dc6b..81e4b9827b81 100755
--- a/cpp/build-support/fuzzing/generate_corpuses.sh
+++ b/cpp/build-support/fuzzing/generate_corpuses.sh
@@ -29,7 +29,7 @@ set -ex
 CORPUS_DIR=/tmp/corpus
 PANDAS_DIR=/tmp/pandas
 
-ARROW_ROOT=$(cd $(dirname "$BASH_SOURCE")/../../..; pwd)
+ARROW_ROOT=$(cd "$(dirname "${BASH_SOURCE[0]}")/../../.." && pwd)
 ARROW_CPP=$ARROW_ROOT/cpp
 OUT=$1
 
@@ -42,48 +42,48 @@ OUT=$1
 rm -rf ${CORPUS_DIR}
 ${OUT}/arrow-ipc-generate-fuzz-corpus -stream ${CORPUS_DIR}
 # Add "golden" IPC integration files
-IPC_INTEGRATION_FILES=$(find ${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration -name "*.stream")
+IPC_INTEGRATION_FILES=$(find "${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration" -name "*.stream")
 [ -z "${IPC_INTEGRATION_FILES}" ] && exit 1
 # Several IPC integration files can have the same name, make sure
 # they all appear in the corpus by numbering the duplicates.
-cp --backup=numbered ${IPC_INTEGRATION_FILES} ${CORPUS_DIR}
-${ARROW_CPP}/build-support/fuzzing/pack_corpus.py ${CORPUS_DIR} ${OUT}/arrow-ipc-stream-fuzz_seed_corpus.zip
+cp --backup=numbered "${IPC_INTEGRATION_FILES}" ${CORPUS_DIR}
+"${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" ${CORPUS_DIR} "${OUT}"/arrow-ipc-stream-fuzz_seed_corpus.zip
 
 rm -rf ${CORPUS_DIR}
-${OUT}/arrow-ipc-generate-fuzz-corpus -file ${CORPUS_DIR}
-IPC_INTEGRATION_FILES=$(find ${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration -name "*.arrow_file")
+"${OUT}/arrow-ipc-generate-fuzz-corpus" -file "${CORPUS_DIR}"
+IPC_INTEGRATION_FILES=$(find "${ARROW_ROOT}"/testing/data/arrow-ipc-stream/integration -name "*.arrow_file")
 [ -z "${IPC_INTEGRATION_FILES}" ] && exit 1
-cp --backup=numbered ${IPC_INTEGRATION_FILES} ${CORPUS_DIR}
-${ARROW_CPP}/build-support/fuzzing/pack_corpus.py ${CORPUS_DIR} ${OUT}/arrow-ipc-file-fuzz_seed_corpus.zip
+cp --backup=numbered "${IPC_INTEGRATION_FILES}" "${CORPUS_DIR}"
+"${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}"/arrow-ipc-file-fuzz_seed_corpus.zip
 
-rm -rf ${CORPUS_DIR}
-${OUT}/arrow-ipc-generate-tensor-fuzz-corpus -stream ${CORPUS_DIR}
-${ARROW_CPP}/build-support/fuzzing/pack_corpus.py ${CORPUS_DIR} ${OUT}/arrow-ipc-tensor-stream-fuzz_seed_corpus.zip
+rm -rf "${CORPUS_DIR}"
+"${OUT}/arrow-ipc-generate-tensor-fuzz-corpus -stream" "${CORPUS_DIR}"
+"${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}/arrow-ipc-tensor-stream-fuzz_seed_corpus.zip"
 
 # Parquet file-level fuzzer
 
 rm -rf ${CORPUS_DIR}
-${OUT}/parquet-arrow-generate-fuzz-corpus ${CORPUS_DIR}
+"${OUT}/parquet-arrow-generate-fuzz-corpus" "${CORPUS_DIR}"
 # Add Parquet testing examples
-cp ${ARROW_CPP}/submodules/parquet-testing/data/*.parquet ${CORPUS_DIR}
-cp ${ARROW_CPP}/submodules/parquet-testing/bad_data/*.parquet ${CORPUS_DIR}
-${ARROW_CPP}/build-support/fuzzing/pack_corpus.py ${CORPUS_DIR} ${OUT}/parquet-arrow-fuzz_seed_corpus.zip
+cp "${ARROW_CPP}/submodules/parquet-testing/data/*.parquet" "${CORPUS_DIR}"
+cp "${ARROW_CPP}/submodules/parquet-testing/bad_data/*.parquet" "${CORPUS_DIR}"
+"${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}/parquet-arrow-fuzz_seed_corpus.zip"
 
 # Parquet encoding fuzzer
 
 rm -rf ${CORPUS_DIR}
-${OUT}/parquet-generate-encoding-fuzz-corpus ${CORPUS_DIR}
-${ARROW_CPP}/build-support/fuzzing/pack_corpus.py ${CORPUS_DIR} ${OUT}/parquet-encoding-fuzz_seed_corpus.zip
+"${OUT}/parquet-generate-encoding-fuzz-corpus" "${CORPUS_DIR}"
+"${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}/parquet-encoding-fuzz_seed_corpus.zip"
 
 # CSV
 
-rm -rf ${PANDAS_DIR}
-git clone --depth=1 https://github.com/pandas-dev/pandas ${PANDAS_DIR}
+rm -rf "${PANDAS_DIR}"
+git clone --depth=1 https://github.com/pandas-dev/pandas "${PANDAS_DIR}"
 
-rm -rf ${CORPUS_DIR}
-${OUT}/arrow-csv-generate-fuzz-corpus ${CORPUS_DIR}
+rm -rf "${CORPUS_DIR}"
+"${OUT}/arrow-csv-generate-fuzz-corpus" "${CORPUS_DIR}"
 # Add examples from arrow-testing repo
-cp ${ARROW_ROOT}/testing/data/csv/*.csv ${CORPUS_DIR}
+cp "${ARROW_ROOT}"/testing/data/csv/*.csv "${CORPUS_DIR}"
 # Add examples from Pandas test suite
-find ${PANDAS_DIR}/ -name "*.csv" -exec cp --backup=numbered '{}' ${CORPUS_DIR} \;
-${ARROW_CPP}/build-support/fuzzing/pack_corpus.py ${CORPUS_DIR} ${OUT}/arrow-csv-fuzz_seed_corpus.zip
+find "${PANDAS_DIR}/" -name "*.csv" -exec cp --backup=numbered '{}' "${CORPUS_DIR}" \;
+"${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}/arrow-csv-fuzz_seed_corpus.zip"

From 5de572f69a403facc95792d62f3cf58885bfa459 Mon Sep 17 00:00:00 2001
From: Hiroyuki Sato <hiroysato@gmail.com>
Date: Mon, 31 Aug 2026 10:57:53 +0900
Subject: [PATCH 02/13] GH-51090: [CI][C++] Fix shellcheck errors in
 cpp/build-support/fuzzing/generate_corpuses.sh

---
 cpp/build-support/fuzzing/generate_corpuses.sh | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/cpp/build-support/fuzzing/generate_corpuses.sh b/cpp/build-support/fuzzing/generate_corpuses.sh
index 81e4b9827b81..9fe28f3912ad 100755
--- a/cpp/build-support/fuzzing/generate_corpuses.sh
+++ b/cpp/build-support/fuzzing/generate_corpuses.sh
@@ -39,17 +39,17 @@ OUT=$1
 
 # Arrow IPC
 
-rm -rf ${CORPUS_DIR}
-${OUT}/arrow-ipc-generate-fuzz-corpus -stream ${CORPUS_DIR}
+rm -rf "${CORPUS_DIR}"
+"${OUT}/arrow-ipc-generate-fuzz-corpus" -stream "${CORPUS_DIR}"
 # Add "golden" IPC integration files
 IPC_INTEGRATION_FILES=$(find "${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration" -name "*.stream")
 [ -z "${IPC_INTEGRATION_FILES}" ] && exit 1
 # Several IPC integration files can have the same name, make sure
 # they all appear in the corpus by numbering the duplicates.
-cp --backup=numbered "${IPC_INTEGRATION_FILES}" ${CORPUS_DIR}
-"${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" ${CORPUS_DIR} "${OUT}"/arrow-ipc-stream-fuzz_seed_corpus.zip
+cp --backup=numbered "${IPC_INTEGRATION_FILES}" "${CORPUS_DIR}"
+"${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}"/arrow-ipc-stream-fuzz_seed_corpus.zip
 
-rm -rf ${CORPUS_DIR}
+rm -rf "${CORPUS_DIR}"
 "${OUT}/arrow-ipc-generate-fuzz-corpus" -file "${CORPUS_DIR}"
 IPC_INTEGRATION_FILES=$(find "${ARROW_ROOT}"/testing/data/arrow-ipc-stream/integration -name "*.arrow_file")
 [ -z "${IPC_INTEGRATION_FILES}" ] && exit 1
@@ -62,11 +62,11 @@ rm -rf "${CORPUS_DIR}"
 
 # Parquet file-level fuzzer
 
-rm -rf ${CORPUS_DIR}
+rm -rf "${CORPUS_DIR}"
 "${OUT}/parquet-arrow-generate-fuzz-corpus" "${CORPUS_DIR}"
 # Add Parquet testing examples
-cp "${ARROW_CPP}/submodules/parquet-testing/data/*.parquet" "${CORPUS_DIR}"
-cp "${ARROW_CPP}/submodules/parquet-testing/bad_data/*.parquet" "${CORPUS_DIR}"
+cp "${ARROW_CPP}"/submodules/parquet-testing/data/*.parquet "${CORPUS_DIR}"
+cp "${ARROW_CPP}"/submodules/parquet-testing/bad_data/*.parquet "${CORPUS_DIR}"
 "${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}/parquet-arrow-fuzz_seed_corpus.zip"
 
 # Parquet encoding fuzzer

From f6caadac6eaa599c3fef53fd828ed3ccce89a210 Mon Sep 17 00:00:00 2001
From: Hiroyuki Sato <hiroysato@gmail.com>
Date: Tue, 1 Sep 2026 14:02:01 +0900
Subject: [PATCH 03/13] Use array

---
 cpp/build-support/fuzzing/generate_corpuses.sh | 18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)

diff --git a/cpp/build-support/fuzzing/generate_corpuses.sh b/cpp/build-support/fuzzing/generate_corpuses.sh
index 9fe28f3912ad..d6d1db1d8e83 100755
--- a/cpp/build-support/fuzzing/generate_corpuses.sh
+++ b/cpp/build-support/fuzzing/generate_corpuses.sh
@@ -42,18 +42,24 @@ OUT=$1
 rm -rf "${CORPUS_DIR}"
 "${OUT}/arrow-ipc-generate-fuzz-corpus" -stream "${CORPUS_DIR}"
 # Add "golden" IPC integration files
-IPC_INTEGRATION_FILES=$(find "${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration" -name "*.stream")
-[ -z "${IPC_INTEGRATION_FILES}" ] && exit 1
+mapfile -d '' IPC_INTEGRATION_FILES < <(
+  find "${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration" \
+    -name "*.stream" -print0
+)
+[ -z "${IPC_INTEGRATION_FILES[@]}" ] && exit 1
 # Several IPC integration files can have the same name, make sure
 # they all appear in the corpus by numbering the duplicates.
-cp --backup=numbered "${IPC_INTEGRATION_FILES}" "${CORPUS_DIR}"
+cp --backup=numbered "${IPC_INTEGRATION_FILES[@]}" "${CORPUS_DIR}"
 "${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}"/arrow-ipc-stream-fuzz_seed_corpus.zip
 
 rm -rf "${CORPUS_DIR}"
 "${OUT}/arrow-ipc-generate-fuzz-corpus" -file "${CORPUS_DIR}"
-IPC_INTEGRATION_FILES=$(find "${ARROW_ROOT}"/testing/data/arrow-ipc-stream/integration -name "*.arrow_file")
-[ -z "${IPC_INTEGRATION_FILES}" ] && exit 1
-cp --backup=numbered "${IPC_INTEGRATION_FILES}" "${CORPUS_DIR}"
+mapfile -d '' IPC_INTEGRATION_FILES < <(
+  find "${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration" \
+    -name "*.arrow_file" -print0
+)
+[ -z "${IPC_INTEGRATION_FILES[@]}" ] && exit 1
+cp --backup=numbered "${IPC_INTEGRATION_FILES[@]}" "${CORPUS_DIR}"
 "${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}"/arrow-ipc-file-fuzz_seed_corpus.zip
 
 rm -rf "${CORPUS_DIR}"

From b3f458c0bdc6d5ff5d04ed01738b4e5236807180 Mon Sep 17 00:00:00 2001
From: Hiroyuki Sato <hiroysato@gmail.com>
Date: Tue, 1 Sep 2026 14:16:55 +0900
Subject: [PATCH 04/13] Fix SC2198

---
 cpp/build-support/fuzzing/generate_corpuses.sh | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/cpp/build-support/fuzzing/generate_corpuses.sh b/cpp/build-support/fuzzing/generate_corpuses.sh
index d6d1db1d8e83..1d7a79d402d5 100755
--- a/cpp/build-support/fuzzing/generate_corpuses.sh
+++ b/cpp/build-support/fuzzing/generate_corpuses.sh
@@ -46,7 +46,7 @@ mapfile -d '' IPC_INTEGRATION_FILES < <(
   find "${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration" \
     -name "*.stream" -print0
 )
-[ -z "${IPC_INTEGRATION_FILES[@]}" ] && exit 1
+[ "${#IPC_INTEGRATION_FILES[@]}" -eq 0 ] && exit 1
 # Several IPC integration files can have the same name, make sure
 # they all appear in the corpus by numbering the duplicates.
 cp --backup=numbered "${IPC_INTEGRATION_FILES[@]}" "${CORPUS_DIR}"
@@ -58,7 +58,7 @@ mapfile -d '' IPC_INTEGRATION_FILES < <(
   find "${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration" \
     -name "*.arrow_file" -print0
 )
-[ -z "${IPC_INTEGRATION_FILES[@]}" ] && exit 1
+[ "${#IPC_INTEGRATION_FILES[@]}" -eq 0 ] && exit 1
 cp --backup=numbered "${IPC_INTEGRATION_FILES[@]}" "${CORPUS_DIR}"
 "${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}"/arrow-ipc-file-fuzz_seed_corpus.zip
 

From a9f358f06cf89edf392623f8c9040f2cfe91ab13 Mon Sep 17 00:00:00 2001
From: Hiroyuki Sato <hiroysato@gmail.com>
Date: Tue, 1 Sep 2026 14:40:29 +0900
Subject: [PATCH 05/13] fix invalid quote

---
 cpp/build-support/fuzzing/generate_corpuses.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/cpp/build-support/fuzzing/generate_corpuses.sh b/cpp/build-support/fuzzing/generate_corpuses.sh
index 1d7a79d402d5..e18443051257 100755
--- a/cpp/build-support/fuzzing/generate_corpuses.sh
+++ b/cpp/build-support/fuzzing/generate_corpuses.sh
@@ -63,7 +63,7 @@ cp --backup=numbered "${IPC_INTEGRATION_FILES[@]}" "${CORPUS_DIR}"
 "${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}"/arrow-ipc-file-fuzz_seed_corpus.zip
 
 rm -rf "${CORPUS_DIR}"
-"${OUT}/arrow-ipc-generate-tensor-fuzz-corpus -stream" "${CORPUS_DIR}"
+"${OUT}/arrow-ipc-generate-tensor-fuzz-corpus" -stream "${CORPUS_DIR}"
 "${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}/arrow-ipc-tensor-stream-fuzz_seed_corpus.zip"
 
 # Parquet file-level fuzzer

From 47b45b59bfe0c2467a272b48539800929d907049 Mon Sep 17 00:00:00 2001
From: Hiroyuki Sato <hiroysato@gmail.com>
Date: Tue, 1 Sep 2026 15:32:52 +0900
Subject: [PATCH 06/13] Add comment

---
 cpp/build-support/fuzzing/generate_corpuses.sh | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/cpp/build-support/fuzzing/generate_corpuses.sh b/cpp/build-support/fuzzing/generate_corpuses.sh
index e18443051257..2bc1f74626e4 100755
--- a/cpp/build-support/fuzzing/generate_corpuses.sh
+++ b/cpp/build-support/fuzzing/generate_corpuses.sh
@@ -42,10 +42,12 @@ OUT=$1
 rm -rf "${CORPUS_DIR}"
 "${OUT}/arrow-ipc-generate-fuzz-corpus" -stream "${CORPUS_DIR}"
 # Add "golden" IPC integration files
+# Store the files found by the find command in an array.
 mapfile -d '' IPC_INTEGRATION_FILES < <(
   find "${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration" \
     -name "*.stream" -print0
 )
+# Exit with an error if find returns no files.
 [ "${#IPC_INTEGRATION_FILES[@]}" -eq 0 ] && exit 1
 # Several IPC integration files can have the same name, make sure
 # they all appear in the corpus by numbering the duplicates.
@@ -54,10 +56,12 @@ cp --backup=numbered "${IPC_INTEGRATION_FILES[@]}" "${CORPUS_DIR}"
 
 rm -rf "${CORPUS_DIR}"
 "${OUT}/arrow-ipc-generate-fuzz-corpus" -file "${CORPUS_DIR}"
+# Store the files found by the find command in an array.
 mapfile -d '' IPC_INTEGRATION_FILES < <(
   find "${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration" \
     -name "*.arrow_file" -print0
 )
+# Exit with an error if find returns no files.
 [ "${#IPC_INTEGRATION_FILES[@]}" -eq 0 ] && exit 1
 cp --backup=numbered "${IPC_INTEGRATION_FILES[@]}" "${CORPUS_DIR}"
 "${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}"/arrow-ipc-file-fuzz_seed_corpus.zip

From c22d5464f8e26bc316e519205bc517c703fede27 Mon Sep 17 00:00:00 2001
From: Hiroyuki Sato <hiroysato@gmail.com>
Date: Tue, 1 Sep 2026 16:08:06 +0900
Subject: [PATCH 07/13] fix quote

---
 cpp/build-support/fuzzing/generate_corpuses.sh | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/cpp/build-support/fuzzing/generate_corpuses.sh b/cpp/build-support/fuzzing/generate_corpuses.sh
index 2bc1f74626e4..38f40a713dfc 100755
--- a/cpp/build-support/fuzzing/generate_corpuses.sh
+++ b/cpp/build-support/fuzzing/generate_corpuses.sh
@@ -52,7 +52,7 @@ mapfile -d '' IPC_INTEGRATION_FILES < <(
 # Several IPC integration files can have the same name, make sure
 # they all appear in the corpus by numbering the duplicates.
 cp --backup=numbered "${IPC_INTEGRATION_FILES[@]}" "${CORPUS_DIR}"
-"${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}"/arrow-ipc-stream-fuzz_seed_corpus.zip
+"${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}/arrow-ipc-stream-fuzz_seed_corpus.zip"
 
 rm -rf "${CORPUS_DIR}"
 "${OUT}/arrow-ipc-generate-fuzz-corpus" -file "${CORPUS_DIR}"
@@ -64,7 +64,7 @@ mapfile -d '' IPC_INTEGRATION_FILES < <(
 # Exit with an error if find returns no files.
 [ "${#IPC_INTEGRATION_FILES[@]}" -eq 0 ] && exit 1
 cp --backup=numbered "${IPC_INTEGRATION_FILES[@]}" "${CORPUS_DIR}"
-"${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}"/arrow-ipc-file-fuzz_seed_corpus.zip
+"${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}/arrow-ipc-file-fuzz_seed_corpus.zip"
 
 rm -rf "${CORPUS_DIR}"
 "${OUT}/arrow-ipc-generate-tensor-fuzz-corpus" -stream "${CORPUS_DIR}"

From 3a17d27cfc87ff4da9df91c5cbf2fa669e2f317e Mon Sep 17 00:00:00 2001
From: Hiroyuki Sato <hiroysato@gmail.com>
Date: Sun, 6 Sep 2026 21:58:43 +0900
Subject: [PATCH 08/13] Use shell glob instead of find command

---
 .../fuzzing/generate_corpuses.sh              | 30 +++++++++----------
 1 file changed, 14 insertions(+), 16 deletions(-)

diff --git a/cpp/build-support/fuzzing/generate_corpuses.sh b/cpp/build-support/fuzzing/generate_corpuses.sh
index 38f40a713dfc..2f14f8af964c 100755
--- a/cpp/build-support/fuzzing/generate_corpuses.sh
+++ b/cpp/build-support/fuzzing/generate_corpuses.sh
@@ -41,29 +41,27 @@ OUT=$1
 
 rm -rf "${CORPUS_DIR}"
 "${OUT}/arrow-ipc-generate-fuzz-corpus" -stream "${CORPUS_DIR}"
+
+IPC_INTEGRATION_DIR="${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration"
+
 # Add "golden" IPC integration files
-# Store the files found by the find command in an array.
-mapfile -d '' IPC_INTEGRATION_FILES < <(
-  find "${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration" \
-    -name "*.stream" -print0
-)
-# Exit with an error if find returns no files.
-[ "${#IPC_INTEGRATION_FILES[@]}" -eq 0 ] && exit 1
 # Several IPC integration files can have the same name, make sure
 # they all appear in the corpus by numbering the duplicates.
-cp --backup=numbered "${IPC_INTEGRATION_FILES[@]}" "${CORPUS_DIR}"
+# Use shell globbing to keep this ShellCheck-safe and compatible with Bash 3.2 on macOS.
+# IPC integration files are expected to be directly under each subdirectory.
+cp --backup=numbered \
+  "${IPC_INTEGRATION_DIR}"/*/*.stream \
+  "${CORPUS_DIR}"
+
 "${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}/arrow-ipc-stream-fuzz_seed_corpus.zip"
 
 rm -rf "${CORPUS_DIR}"
 "${OUT}/arrow-ipc-generate-fuzz-corpus" -file "${CORPUS_DIR}"
-# Store the files found by the find command in an array.
-mapfile -d '' IPC_INTEGRATION_FILES < <(
-  find "${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration" \
-    -name "*.arrow_file" -print0
-)
-# Exit with an error if find returns no files.
-[ "${#IPC_INTEGRATION_FILES[@]}" -eq 0 ] && exit 1
-cp --backup=numbered "${IPC_INTEGRATION_FILES[@]}" "${CORPUS_DIR}"
+
+cp --backup=numbered \
+  "${IPC_INTEGRATION_DIR}"/*/*.arrow_file \
+  "${CORPUS_DIR}"
+
 "${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}/arrow-ipc-file-fuzz_seed_corpus.zip"
 
 rm -rf "${CORPUS_DIR}"

From 7cec9c64b3e584d0fa755c0ff6eb012aebab0cba Mon Sep 17 00:00:00 2001
From: Hiroyuki Sato <hiroysato@gmail.com>
Date: Mon, 7 Sep 2026 09:20:29 +0900
Subject: [PATCH 09/13] Revert "Use shell glob instead of find command"

This reverts commit 3a17d27cfc87ff4da9df91c5cbf2fa669e2f317e.
---
 .../fuzzing/generate_corpuses.sh              | 30 ++++++++++---------
 1 file changed, 16 insertions(+), 14 deletions(-)

diff --git a/cpp/build-support/fuzzing/generate_corpuses.sh b/cpp/build-support/fuzzing/generate_corpuses.sh
index 2f14f8af964c..38f40a713dfc 100755
--- a/cpp/build-support/fuzzing/generate_corpuses.sh
+++ b/cpp/build-support/fuzzing/generate_corpuses.sh
@@ -41,27 +41,29 @@ OUT=$1
 
 rm -rf "${CORPUS_DIR}"
 "${OUT}/arrow-ipc-generate-fuzz-corpus" -stream "${CORPUS_DIR}"
-
-IPC_INTEGRATION_DIR="${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration"
-
 # Add "golden" IPC integration files
+# Store the files found by the find command in an array.
+mapfile -d '' IPC_INTEGRATION_FILES < <(
+  find "${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration" \
+    -name "*.stream" -print0
+)
+# Exit with an error if find returns no files.
+[ "${#IPC_INTEGRATION_FILES[@]}" -eq 0 ] && exit 1
 # Several IPC integration files can have the same name, make sure
 # they all appear in the corpus by numbering the duplicates.
-# Use shell globbing to keep this ShellCheck-safe and compatible with Bash 3.2 on macOS.
-# IPC integration files are expected to be directly under each subdirectory.
-cp --backup=numbered \
-  "${IPC_INTEGRATION_DIR}"/*/*.stream \
-  "${CORPUS_DIR}"
-
+cp --backup=numbered "${IPC_INTEGRATION_FILES[@]}" "${CORPUS_DIR}"
 "${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}/arrow-ipc-stream-fuzz_seed_corpus.zip"
 
 rm -rf "${CORPUS_DIR}"
 "${OUT}/arrow-ipc-generate-fuzz-corpus" -file "${CORPUS_DIR}"
-
-cp --backup=numbered \
-  "${IPC_INTEGRATION_DIR}"/*/*.arrow_file \
-  "${CORPUS_DIR}"
-
+# Store the files found by the find command in an array.
+mapfile -d '' IPC_INTEGRATION_FILES < <(
+  find "${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration" \
+    -name "*.arrow_file" -print0
+)
+# Exit with an error if find returns no files.
+[ "${#IPC_INTEGRATION_FILES[@]}" -eq 0 ] && exit 1
+cp --backup=numbered "${IPC_INTEGRATION_FILES[@]}" "${CORPUS_DIR}"
 "${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}/arrow-ipc-file-fuzz_seed_corpus.zip"
 
 rm -rf "${CORPUS_DIR}"

From baf318541beae63a7c24fdac4c9d3054cc5536b2 Mon Sep 17 00:00:00 2001
From: Hiroyuki Sato <hiroysato@gmail.com>
Date: Mon, 7 Sep 2026 09:27:05 +0900
Subject: [PATCH 10/13] Use while .. read

---
 cpp/build-support/fuzzing/generate_corpuses.sh | 18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)

diff --git a/cpp/build-support/fuzzing/generate_corpuses.sh b/cpp/build-support/fuzzing/generate_corpuses.sh
index 38f40a713dfc..23da9fedeede 100755
--- a/cpp/build-support/fuzzing/generate_corpuses.sh
+++ b/cpp/build-support/fuzzing/generate_corpuses.sh
@@ -42,9 +42,12 @@ OUT=$1
 rm -rf "${CORPUS_DIR}"
 "${OUT}/arrow-ipc-generate-fuzz-corpus" -stream "${CORPUS_DIR}"
 # Add "golden" IPC integration files
-# Store the files found by the find command in an array.
-mapfile -d '' IPC_INTEGRATION_FILES < <(
-  find "${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration" \
+# Use read instead of mapfile to keep this compatible with Bash 3.2 on macOS.
+IPC_INTEGRATION_FILES=()
+while IFS= read -r -d '' IPC_INTEGRATION_FILE; do
+  IPC_INTEGRATION_FILES+=("${IPC_INTEGRATION_FILE}")
+done < <(
+  find "${IPC_INTEGRATION_DIR}" \
     -name "*.stream" -print0
 )
 # Exit with an error if find returns no files.
@@ -56,9 +59,12 @@ cp --backup=numbered "${IPC_INTEGRATION_FILES[@]}" "${CORPUS_DIR}"
 
 rm -rf "${CORPUS_DIR}"
 "${OUT}/arrow-ipc-generate-fuzz-corpus" -file "${CORPUS_DIR}"
-# Store the files found by the find command in an array.
-mapfile -d '' IPC_INTEGRATION_FILES < <(
-  find "${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration" \
+
+IPC_INTEGRATION_FILES=()
+while IFS= read -r -d '' IPC_INTEGRATION_FILE; do
+  IPC_INTEGRATION_FILES+=("${IPC_INTEGRATION_FILE}")
+done < <(
+  find "${IPC_INTEGRATION_DIR}" \
     -name "*.arrow_file" -print0
 )
 # Exit with an error if find returns no files.

From ddb09076d1bb121ef1111692bcaf4b33004633de Mon Sep 17 00:00:00 2001
From: Hiroyuki Sato <hiroysato@gmail.com>
Date: Mon, 7 Sep 2026 09:32:12 +0900
Subject: [PATCH 11/13] Fix missing variable

---
 cpp/build-support/fuzzing/generate_corpuses.sh | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/cpp/build-support/fuzzing/generate_corpuses.sh b/cpp/build-support/fuzzing/generate_corpuses.sh
index 23da9fedeede..844f12fdaf8e 100755
--- a/cpp/build-support/fuzzing/generate_corpuses.sh
+++ b/cpp/build-support/fuzzing/generate_corpuses.sh
@@ -47,7 +47,7 @@ IPC_INTEGRATION_FILES=()
 while IFS= read -r -d '' IPC_INTEGRATION_FILE; do
   IPC_INTEGRATION_FILES+=("${IPC_INTEGRATION_FILE}")
 done < <(
-  find "${IPC_INTEGRATION_DIR}" \
+  find "${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration" \
     -name "*.stream" -print0
 )
 # Exit with an error if find returns no files.
@@ -64,7 +64,7 @@ IPC_INTEGRATION_FILES=()
 while IFS= read -r -d '' IPC_INTEGRATION_FILE; do
   IPC_INTEGRATION_FILES+=("${IPC_INTEGRATION_FILE}")
 done < <(
-  find "${IPC_INTEGRATION_DIR}" \
+  find "${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration" \
     -name "*.arrow_file" -print0
 )
 # Exit with an error if find returns no files.

From 354d28a886c5d04ce08ad62e370e276ee142c449 Mon Sep 17 00:00:00 2001
From: Hiroyuki Sato <hiroysato@gmail.com>
Date: Mon, 7 Sep 2026 10:46:06 +0900
Subject: [PATCH 12/13] Remove find result check

---
 cpp/build-support/fuzzing/generate_corpuses.sh | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/cpp/build-support/fuzzing/generate_corpuses.sh b/cpp/build-support/fuzzing/generate_corpuses.sh
index 844f12fdaf8e..fdcfbdacc894 100755
--- a/cpp/build-support/fuzzing/generate_corpuses.sh
+++ b/cpp/build-support/fuzzing/generate_corpuses.sh
@@ -50,8 +50,6 @@ done < <(
   find "${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration" \
     -name "*.stream" -print0
 )
-# Exit with an error if find returns no files.
-[ "${#IPC_INTEGRATION_FILES[@]}" -eq 0 ] && exit 1
 # Several IPC integration files can have the same name, make sure
 # they all appear in the corpus by numbering the duplicates.
 cp --backup=numbered "${IPC_INTEGRATION_FILES[@]}" "${CORPUS_DIR}"
@@ -67,8 +65,6 @@ done < <(
   find "${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration" \
     -name "*.arrow_file" -print0
 )
-# Exit with an error if find returns no files.
-[ "${#IPC_INTEGRATION_FILES[@]}" -eq 0 ] && exit 1
 cp --backup=numbered "${IPC_INTEGRATION_FILES[@]}" "${CORPUS_DIR}"
 "${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}/arrow-ipc-file-fuzz_seed_corpus.zip"
 

From 700bf788d9a917aecc6ec2af4079197ec772929d Mon Sep 17 00:00:00 2001
From: Hiroyuki Sato <hiroysato@gmail.com>
Date: Mon, 7 Sep 2026 13:35:58 +0900
Subject: [PATCH 13/13] Remove while..read

---
 .../fuzzing/generate_corpuses.sh              | 25 ++++++-------------
 1 file changed, 8 insertions(+), 17 deletions(-)

diff --git a/cpp/build-support/fuzzing/generate_corpuses.sh b/cpp/build-support/fuzzing/generate_corpuses.sh
index fdcfbdacc894..2d95e110b516 100755
--- a/cpp/build-support/fuzzing/generate_corpuses.sh
+++ b/cpp/build-support/fuzzing/generate_corpuses.sh
@@ -42,30 +42,21 @@ OUT=$1
 rm -rf "${CORPUS_DIR}"
 "${OUT}/arrow-ipc-generate-fuzz-corpus" -stream "${CORPUS_DIR}"
 # Add "golden" IPC integration files
-# Use read instead of mapfile to keep this compatible with Bash 3.2 on macOS.
-IPC_INTEGRATION_FILES=()
-while IFS= read -r -d '' IPC_INTEGRATION_FILE; do
-  IPC_INTEGRATION_FILES+=("${IPC_INTEGRATION_FILE}")
-done < <(
-  find "${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration" \
-    -name "*.stream" -print0
-)
 # Several IPC integration files can have the same name, make sure
 # they all appear in the corpus by numbering the duplicates.
-cp --backup=numbered "${IPC_INTEGRATION_FILES[@]}" "${CORPUS_DIR}"
+find "${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration" \
+  -name "*.stream" \
+  -exec cp --backup=numbered '{}' "${CORPUS_DIR}" \;
+
 "${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}/arrow-ipc-stream-fuzz_seed_corpus.zip"
 
 rm -rf "${CORPUS_DIR}"
 "${OUT}/arrow-ipc-generate-fuzz-corpus" -file "${CORPUS_DIR}"
 
-IPC_INTEGRATION_FILES=()
-while IFS= read -r -d '' IPC_INTEGRATION_FILE; do
-  IPC_INTEGRATION_FILES+=("${IPC_INTEGRATION_FILE}")
-done < <(
-  find "${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration" \
-    -name "*.arrow_file" -print0
-)
-cp --backup=numbered "${IPC_INTEGRATION_FILES[@]}" "${CORPUS_DIR}"
+find "${ARROW_ROOT}/testing/data/arrow-ipc-stream/integration" \
+  -name "*.arrow_file" \
+  -exec cp --backup=numbered '{}' "${CORPUS_DIR}" \;
+
 "${ARROW_CPP}/build-support/fuzzing/pack_corpus.py" "${CORPUS_DIR}" "${OUT}/arrow-ipc-file-fuzz_seed_corpus.zip"
 
 rm -rf "${CORPUS_DIR}"