[drill] branch master updated: DRILL-8064: Cannot query text files with names that contain a backslash (#2392)

[dzamo]

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

dzamo pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/drill.git


The following commit(s) were added to refs/heads/master by this push:
     new 1758029  DRILL-8064: Cannot query text files with names that contain a 
backslash (#2392)
1758029 is described below

commit 1758029dc36980b06c0ef21cee6c20d07dc578d9
Author: James Turton <9107319+dz...@users.noreply.github.com>
AuthorDate: Sat Dec 4 06:23:26 2021 +0200

    DRILL-8064: Cannot query text files with names that contain a backslash 
(#2392)
    
    * Unescape root path in dfs plugin.
    
    * Fix a spelling mistake in a comment.
---
 .../org/apache/drill/exec/store/dfs/FileSelection.java   | 11 ++++++++---
 .../java/org/apache/drill/exec/store/dfs/TestGlob.java   | 16 ++++++++++++++++
 2 files changed, 24 insertions(+), 3 deletions(-)

diff --git 
a/exec/java-exec/src/main/java/org/apache/drill/exec/store/dfs/FileSelection.java
 
b/exec/java-exec/src/main/java/org/apache/drill/exec/store/dfs/FileSelection.java
index 6700114..ebe1a43 100644
--- 
a/exec/java-exec/src/main/java/org/apache/drill/exec/store/dfs/FileSelection.java
+++ 
b/exec/java-exec/src/main/java/org/apache/drill/exec/store/dfs/FileSelection.java
@@ -270,15 +270,20 @@ public class FileSelection {
     Stopwatch timer = logger.isDebugEnabled() ? Stopwatch.createStarted() : 
null;
     boolean hasWildcard = path.contains(WILD_CARD);
 
-    Path combined = new Path(parent, 
DrillStringUtils.removeLeadingSlash(path));
+    String child = DrillStringUtils.removeLeadingSlash(path);
+    Path combined = new Path(parent, child);
+    // Unescape chars escaped with '\' for our root path to be consistent with 
what
+    // fs.globStatus(...) below will do with them, c.f. DRILL-8064
+    Path root = new Path(parent, DrillStringUtils.unescapeJava(child));
+
     if (!allowAccessOutsideWorkspace) {
       checkBackPaths(new Path(parent).toUri().getPath(), 
combined.toUri().getPath(), path);
     }
-    FileStatus[] statuses = fs.globStatus(combined); // note: this would 
expand wildcards
+    FileStatus[] statuses = fs.globStatus(combined); // note: this will expand 
wildcards
     if (statuses == null) {
       return null;
     }
-    FileSelection fileSel = create(Arrays.asList(statuses), null, combined);
+    FileSelection fileSel = create(Arrays.asList(statuses), null, root);
     if (timer != null) {
       logger.debug("FileSelection.create() took {} ms ", 
timer.elapsed(TimeUnit.MILLISECONDS));
       timer.stop();
diff --git 
a/exec/java-exec/src/test/java/org/apache/drill/exec/store/dfs/TestGlob.java 
b/exec/java-exec/src/test/java/org/apache/drill/exec/store/dfs/TestGlob.java
index 2ca2120..03fba37 100644
--- a/exec/java-exec/src/test/java/org/apache/drill/exec/store/dfs/TestGlob.java
+++ b/exec/java-exec/src/test/java/org/apache/drill/exec/store/dfs/TestGlob.java
@@ -31,6 +31,10 @@ public class TestGlob extends BaseTestQuery {
   @BeforeClass
   public static void setupTestFiles() {
     dirTestWatcher.copyResourceToRoot(Paths.get("multilevel"));
+    dirTestWatcher.copyResourceToRoot(
+      Paths.get("emptyStrings.csv"),
+      Paths.get("globEscapeCharIsA\\Backslash.csv")
+    );
   }
 
   @Test
@@ -75,4 +79,16 @@ public class TestGlob extends BaseTestQuery {
       .build()
       .run();
   }
+
+  @Test
+  // DRILL-8064
+  public void testGlobEscapeCharRootTextFile() throws Exception {
+    testBuilder()
+      .sqlQuery("select count(*) from 
dfs.`globEscapeCharIsA\\\\Backslash.csv`")
+      .unOrdered()
+      .baselineColumns("EXPR$0")
+      .baselineValues(3L)
+      .build()
+      .run();
+  }
 }