commit ed56301abd8ba11463b340b4c5c6f7af57ed6623
author Łukasz Wojniłowicz <lukasz.wojnilowicz@gmail.com> Sat May 16 12:53:50 2020 +0200
committer Jeff Genovy <29107334+jefgen@users.noreply.github.com> Wed Jun 24 11:38:41 2020 -0700
tree 5ebf5f7ddfe0abd1bb32e58a9cd5849033422273
parent ef12882fdbbfc581f0b55fc423467356057da064

ICU-20545 Ensure that path ends with detected file separator

CharString, when asked, appends U_FILE_SEP_CHAR at the end of the string
it holds, if it won't find U_FILE_SEP_CHAR or U_FILE_ALT_SEP_CHAR there.
The problem starts if the dir variable uses
U_FILE_ALT_SEP_CHAR which is not equal to U_FILE_SEP_CHAR. Then the
resulting path could look like this
../data\
instead of this
../data/

This patch uses U_FILE_SEP_CHAR unless it detects that the dir variable
doesn't use it, and uses U_FILE_ALT_SEP_CHAR instead.

icu4c/source/common/charstr.cpp

diff --git a/icu4c/source/common/charstr.cpp b/icu4c/source/common/charstr.cpp
index dda29da..4878c73 100644
--- a/icu4c/source/common/charstr.cpp
+++ b/icu4c/source/common/charstr.cpp
@@ -197,7 +197,7 @@ CharString &CharString::appendPathPart(StringPiece s, UErrorCode &errorCode) {
     }
     char c;
     if(len>0 && (c=buffer[len-1])!=U_FILE_SEP_CHAR && c!=U_FILE_ALT_SEP_CHAR) {
-        append(U_FILE_SEP_CHAR, errorCode);
+        append(getDirSepChar(), errorCode);
     }
     append(s, errorCode);
     return *this;
@@ -207,9 +207,19 @@ CharString &CharString::ensureEndsWithFileSeparator(UErrorCode &errorCode) {
     char c;
     if(U_SUCCESS(errorCode) && len>0 &&
             (c=buffer[len-1])!=U_FILE_SEP_CHAR && c!=U_FILE_ALT_SEP_CHAR) {
-        append(U_FILE_SEP_CHAR, errorCode);
+        append(getDirSepChar(), errorCode);
     }
     return *this;
 }
 
+char CharString::getDirSepChar() const {
+    char dirSepChar = U_FILE_SEP_CHAR;
+#if (U_FILE_SEP_CHAR != U_FILE_ALT_SEP_CHAR)
+    // We may need to return a different directory separator when building for Cygwin or MSYS2.
+    if(len>0 && !uprv_strchr(data(), U_FILE_SEP_CHAR) && uprv_strchr(data(), U_FILE_ALT_SEP_CHAR))
+        dirSepChar = U_FILE_ALT_SEP_CHAR;
+#endif
+    return dirSepChar;
+}
+
 U_NAMESPACE_END

icu4c/source/common/charstr.h

diff --git a/icu4c/source/common/charstr.h b/icu4c/source/common/charstr.h
index 23b950e..9400fbb 100644
--- a/icu4c/source/common/charstr.h
+++ b/icu4c/source/common/charstr.h
@@ -141,13 +141,13 @@ class U_COMMON_API CharString : public UMemory {
 
     /**
      * Appends a filename/path part, e.g., a directory name.
-     * First appends a U_FILE_SEP_CHAR if necessary.
+     * First appends a U_FILE_SEP_CHAR or U_FILE_ALT_SEP_CHAR if necessary.
      * Does nothing if s is empty.
      */
     CharString &appendPathPart(StringPiece s, UErrorCode &errorCode);
 
     /**
-     * Appends a U_FILE_SEP_CHAR if this string is not empty
+     * Appends a U_FILE_SEP_CHAR or U_FILE_ALT_SEP_CHAR if this string is not empty
      * and does not already end with a U_FILE_SEP_CHAR or U_FILE_ALT_SEP_CHAR.
      */
     CharString &ensureEndsWithFileSeparator(UErrorCode &errorCode);
@@ -160,6 +160,12 @@ class U_COMMON_API CharString : public UMemory {
 
     CharString(const CharString &other); // forbid copying of this class
     CharString &operator=(const CharString &other); // forbid copying of this class
+
+    /**
+     * Returns U_FILE_ALT_SEP_CHAR if found in string, and U_FILE_SEP_CHAR is not found.
+     * Otherwise returns U_FILE_SEP_CHAR.
+     */
+    char getDirSepChar() const;
 };
 
 U_NAMESPACE_END