From 5fd38fd7a0f56b784e1dcf7830925ab07ad9e9ac Mon Sep 17 00:00:00 2001
From: Marc Cenac <marc.cenac@datadoghq.com>
Date: Fri, 25 Oct 2024 07:17:35 -0500
Subject: [PATCH 1/3] Azure: Fix ADLSLocation file parsing

---
 .../org/apache/iceberg/azure/adlsv2/ADLSLocation.java     | 3 +--
 .../org/apache/iceberg/azure/adlsv2/ADLSLocationTest.java | 8 ++++++++
 2 files changed, 9 insertions(+), 2 deletions(-)

diff --git a/azure/src/main/java/org/apache/iceberg/azure/adlsv2/ADLSLocation.java b/azure/src/main/java/org/apache/iceberg/azure/adlsv2/ADLSLocation.java
index e73093512b82..876384e44f5f 100644
--- a/azure/src/main/java/org/apache/iceberg/azure/adlsv2/ADLSLocation.java
+++ b/azure/src/main/java/org/apache/iceberg/azure/adlsv2/ADLSLocation.java
@@ -64,8 +64,7 @@ class ADLSLocation {
     }
 
     String uriPath = matcher.group(2);
-    uriPath = uriPath == null ? "" : uriPath.startsWith("/") ? uriPath.substring(1) : uriPath;
-    this.path = uriPath.split("\\?", -1)[0].split("#", -1)[0];
+    this.path = uriPath == null ? "" : uriPath.startsWith("/") ? uriPath.substring(1) : uriPath;
   }
 
   /** Returns Azure storage account. */
diff --git a/azure/src/test/java/org/apache/iceberg/azure/adlsv2/ADLSLocationTest.java b/azure/src/test/java/org/apache/iceberg/azure/adlsv2/ADLSLocationTest.java
index 867b54b4c7e3..037799ed6b02 100644
--- a/azure/src/test/java/org/apache/iceberg/azure/adlsv2/ADLSLocationTest.java
+++ b/azure/src/test/java/org/apache/iceberg/azure/adlsv2/ADLSLocationTest.java
@@ -101,4 +101,12 @@ public void testQueryAndFragmentNoPath() {
     assertThat(location.container().get()).isEqualTo("container");
     assertThat(location.path()).isEqualTo("");
   }
+
+  @ParameterizedTest
+  @ValueSource(strings = {"file?.txt", "file%3F.txt"})
+  public void testQuestionMarkInFileName(String path) {
+    String fullPath = String.format("abfs://container@account.dfs.core.windows.net/%s", path);
+    ADLSLocation location = new ADLSLocation(fullPath);
+    assertThat(location.path()).contains(path);
+  }
 }

From e9691c9ccfc8ffb538ceaed9f2455421cdf5dfb8 Mon Sep 17 00:00:00 2001
From: Marc Cenac <marc.cenac@datadoghq.com>
Date: Fri, 25 Oct 2024 07:19:17 -0500
Subject: [PATCH 2/3] Azure: Remove invalid test cases from ADLSLocationTest

---
 .../azure/adlsv2/ADLSLocationTest.java        | 20 -------------------
 1 file changed, 20 deletions(-)

diff --git a/azure/src/test/java/org/apache/iceberg/azure/adlsv2/ADLSLocationTest.java b/azure/src/test/java/org/apache/iceberg/azure/adlsv2/ADLSLocationTest.java
index 037799ed6b02..403886f4b28e 100644
--- a/azure/src/test/java/org/apache/iceberg/azure/adlsv2/ADLSLocationTest.java
+++ b/azure/src/test/java/org/apache/iceberg/azure/adlsv2/ADLSLocationTest.java
@@ -82,26 +82,6 @@ public void testNoPath() {
     assertThat(location.path()).isEqualTo("");
   }
 
-  @Test
-  public void testQueryAndFragment() {
-    String p1 = "abfs://container@account.dfs.core.windows.net/path/to/file?query=foo#123";
-    ADLSLocation location = new ADLSLocation(p1);
-
-    assertThat(location.storageAccount()).isEqualTo("account.dfs.core.windows.net");
-    assertThat(location.container().get()).isEqualTo("container");
-    assertThat(location.path()).isEqualTo("path/to/file");
-  }
-
-  @Test
-  public void testQueryAndFragmentNoPath() {
-    String p1 = "abfs://container@account.dfs.core.windows.net?query=foo#123";
-    ADLSLocation location = new ADLSLocation(p1);
-
-    assertThat(location.storageAccount()).isEqualTo("account.dfs.core.windows.net");
-    assertThat(location.container().get()).isEqualTo("container");
-    assertThat(location.path()).isEqualTo("");
-  }
-
   @ParameterizedTest
   @ValueSource(strings = {"file?.txt", "file%3F.txt"})
   public void testQuestionMarkInFileName(String path) {

From b39772b3c1c024e93e68f082aa38573bd829c3b3 Mon Sep 17 00:00:00 2001
From: Marc Cenac <marc.cenac@datadoghq.com>
Date: Mon, 28 Oct 2024 19:00:43 -0500
Subject: [PATCH 3/3] Update Javadocs with reference to ADLS URI

---
 .../org/apache/iceberg/azure/adlsv2/ADLSLocation.java  | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/azure/src/main/java/org/apache/iceberg/azure/adlsv2/ADLSLocation.java b/azure/src/main/java/org/apache/iceberg/azure/adlsv2/ADLSLocation.java
index 876384e44f5f..5af590628fe8 100644
--- a/azure/src/main/java/org/apache/iceberg/azure/adlsv2/ADLSLocation.java
+++ b/azure/src/main/java/org/apache/iceberg/azure/adlsv2/ADLSLocation.java
@@ -25,14 +25,16 @@
 import org.apache.iceberg.relocated.com.google.common.base.Preconditions;
 
 /**
- * This class represents a fully qualified location in Azure expressed as a URI.
+ * This class represents a fully qualified location to a file or directory in Azure Data Lake
+ * Storage Gen2 storage.
  *
- * <p>Locations follow the conventions used by Hadoop's Azure support, i.e.
+ * <p>Locations follow a URI like structure to identify resources
  *
  * <pre>{@code abfs[s]://[<container>@]<storage account host>/<file path>}</pre>
  *
- * <p>See <a href="https://hadoop.apache.org/docs/stable/hadoop-azure/abfs.html">Hadoop Azure
- * Support</a>
+ * <p>See <a
+ * href="https://learn.microsoft.com/en-us/azure/storage/blobs/data-lake-storage-introduction-abfs-uri#uri-syntax">Azure
+ * Data Lake Storage URI</a>
  */
 class ADLSLocation {
   private static final Pattern URI_PATTERN = Pattern.compile("^abfss?://([^/?#]+)(.*)?$");