Move method GooString::hasUnicodeMarkerLE to UTF.h

... and rename it to hasUnicodeByteOrderMarkLE.

This allows to replace GooString by std::string in a few places.
(In a future commit)

7 files changed, 15 insertions, 9 deletions

diff --git a/cpp/poppler-private.cpp b/cpp/poppler-private.cpp
index 6953724f..9e89c7b8 100644
--- a/cpp/poppler-private.cpp
+++ b/cpp/poppler-private.cpp
@@ -65,7 +65,7 @@ ustring detail::unicode_GooString_to_ustring(const GooString *str)
     const char *data = str->c_str();
     const int len = str->getLength();
 
-    const bool is_unicodeLE = str->hasUnicodeMarkerLE();
+    const bool is_unicodeLE = hasUnicodeByteOrderMarkLE(str->toStr());
     const bool is_unicode = hasUnicodeByteOrderMark(str->toStr()) || is_unicodeLE;
     int i = is_unicode ? 2 : 0;
     ustring::size_type ret_len = len - i;
diff --git a/glib/poppler-document.cc b/glib/poppler-document.cc
index 9c516146..5dc034c5 100644
--- a/glib/poppler-document.cc
+++ b/glib/poppler-document.cc
@@ -1101,7 +1101,7 @@ char *_poppler_goo_string_to_utf8(const GooString *s)
 
     if (hasUnicodeByteOrderMark(s->toStr())) {
         result = g_convert(s->c_str() + 2, s->getLength() - 2, "UTF-8", "UTF-16BE", nullptr, nullptr, nullptr);
-    } else if (s->hasUnicodeMarkerLE()) {
+    } else if (hasUnicodeByteOrderMarkLE(s->toStr())) {
         result = g_convert(s->c_str() + 2, s->getLength() - 2, "UTF-8", "UTF-16LE", nullptr, nullptr, nullptr);
     } else {
         int len;
diff --git a/goo/GooString.h b/goo/GooString.h
index 33f037e0..05aa630f 100644
--- a/goo/GooString.h
+++ b/goo/GooString.h
@@ -243,9 +243,6 @@ public:
     // Return true if string ends with suffix
     using std::string::ends_with;
 
-    bool hasUnicodeMarkerLE() const { return hasUnicodeMarkerLE(*this); }
-    static bool hasUnicodeMarkerLE(const std::string &s) { return s.size() >= 2 && s[0] == '\xff' && s[1] == '\xfe'; }
-
     POPPLER_PRIVATE_EXPORT void prependUnicodeMarker();
 };
 
diff --git a/poppler/UTF.cc b/poppler/UTF.cc
index 48645780..74c5a113 100644
--- a/poppler/UTF.cc
+++ b/poppler/UTF.cc
@@ -91,7 +91,7 @@ std::vector<Unicode> TextStringToUCS4(const std::string &textStr)
     if (hasUnicodeByteOrderMark(textStr)) {
         isUnicode = true;
         isUnicodeLE = false;
-    } else if (GooString::hasUnicodeMarkerLE(textStr)) {
+    } else if (hasUnicodeByteOrderMarkLE(textStr)) {
         isUnicode = false;
         isUnicodeLE = true;
     } else {
diff --git a/poppler/UTF.h b/poppler/UTF.h
index 9f0290db..812506eb 100644
--- a/poppler/UTF.h
+++ b/poppler/UTF.h
@@ -27,9 +27,12 @@
 #include "CharTypes.h"
 #include "poppler_private_export.h"
 
-// Magic bytes that mark the byte order in a UTF-16 unicode string
+// Magic bytes that mark the byte order in a UTF-16 unicode string (big-endian case)
 constexpr std::string_view unicodeByteOrderMark = "\xFE\xFF";
 
+// Magic bytes that mark the byte order in a UTF-16 unicode string (little-endian case)
+constexpr std::string_view unicodeByteOrderMarkLE = "\xFF\xFE";
+
 // Convert a UTF-16 string to a UCS-4
 //   utf16      - utf16 bytes
 //   utf16_len  - number of UTF-16 characters
@@ -54,6 +57,12 @@ inline bool hasUnicodeByteOrderMark(const std::string &s)
     return s.starts_with(unicodeByteOrderMark);
 }
 
+// check whether string starts with Little-Endian byte order mark
+inline bool hasUnicodeByteOrderMarkLE(const std::string &s)
+{
+    return s.starts_with(unicodeByteOrderMarkLE);
+}
+
 // is a unicode whitespace character
 bool UnicodeIsWhitespace(Unicode ucs4);
 
diff --git a/qt5/src/poppler-private.cc b/qt5/src/poppler-private.cc
index 3f4d286f..3e0f0f57 100644
--- a/qt5/src/poppler-private.cc
+++ b/qt5/src/poppler-private.cc
@@ -122,7 +122,7 @@ QString UnicodeParsedString(const std::string &s1)
         return QString();
     }
 
-    if (hasUnicodeByteOrderMark(s1) || GooString::hasUnicodeMarkerLE(s1)) {
+    if (hasUnicodeByteOrderMark(s1) || hasUnicodeByteOrderMarkLE(s1)) {
         return QString::fromUtf16(reinterpret_cast<const ushort *>(s1.c_str()), s1.size() / 2);
     } else {
         int stringLength;
diff --git a/qt6/src/poppler-private.cc b/qt6/src/poppler-private.cc
index 052360bd..480bb5ec 100644
--- a/qt6/src/poppler-private.cc
+++ b/qt6/src/poppler-private.cc
@@ -122,7 +122,7 @@ QString UnicodeParsedString(const std::string &s1)
         return QString();
     }
 
-    if (hasUnicodeByteOrderMark(s1) || GooString::hasUnicodeMarkerLE(s1)) {
+    if (hasUnicodeByteOrderMark(s1) || hasUnicodeByteOrderMarkLE(s1)) {
         return QString::fromUtf16(reinterpret_cast<const char16_t *>(s1.c_str()), s1.size() / 2);
     } else {
         int stringLength;