[programs] Fix infinite loop when empty input is passed to trainer

When an empty input file was passed to the dictionary trainer, it would infinite loop.
The added test case exposes the bug, and is fixed with this PR.

diff --git a/programs/dibio.c b/programs/dibio.c
index fddbc9e5769849e42742c5a866293a5668be3711..f6757dd34880fa943cf99a7f6d4aa7364b44a872 100644 (file)
--- a/programs/dibio.c
+++ b/programs/dibio.c
@@ -128,8 +128,11 @@ static int DiB_loadFiles(
     while ( nbSamplesLoaded < sstSize && fileIndex < nbFiles ) {
         size_t fileDataLoaded;
         S64 const fileSize = DiB_getFileSize(fileNamesTable[fileIndex]);
-        if (fileSize <= 0) /* skip if zero-size or file error */
+        if (fileSize <= 0) {
+            /* skip if zero-size or file error */
+            ++fileIndex;
             continue;
+        }
 
         f = fopen( fileNamesTable[fileIndex], "rb");
         if (f == NULL)

diff --git a/tests/cli-tests/dict-builder/empty-input.sh b/tests/cli-tests/dict-builder/empty-input.sh
new file mode 100755 (executable)
index 0000000..b500bfe
--- /dev/null
+++ b/tests/cli-tests/dict-builder/empty-input.sh
@@ -0,0 +1,9 @@
+#!/bin/sh
+set -e
+for i in $(seq 50); do
+    datagen -s$i > file$i
+done
+touch empty
+
+set -v
+zstd -q --train empty file*

diff --git a/tests/cli-tests/dict-builder/empty-input.sh.stderr.exact b/tests/cli-tests/dict-builder/empty-input.sh.stderr.exact
new file mode 100644 (file)
index 0000000..2747e76
--- /dev/null
+++ b/tests/cli-tests/dict-builder/empty-input.sh.stderr.exact
@@ -0,0 +1 @@
+zstd -q --train empty file*