Fix surrogate pairs handling when tweaking script info

Followup to: commit 3af30ba Author: Khaled Hosny <[email protected]> Date: Thu Jul 27 19:03:28 2023 +0300 sw: Handle surrogate pairs when tweaking script info and: commit d6efe8c Author: Khaled Hosny <[email protected]> Date: Thu Jul 27 20:39:22 2023 +0300 tdf#112594: Group NNBSP with the Mongolian characters after it Change-Id: Ie273c457e4f3ed31a3372bc8eb0eb0055c1b97b1 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/155053 Tested-by: Jenkins Reviewed-by: Michael Stahl <[email protected]>
LibreOffice · Jul 31, 2023 · 5526c52 · 5526c52
1 parent 74963f6
commit 5526c52
Show file tree

Hide file tree

Showing 4 changed files with 14 additions and 5 deletions.
diff --git a/editeng/source/editeng/impedit2.cxx b/editeng/source/editeng/impedit2.cxx
@@ -1742,8 +1742,7 @@ void ImpEditEngine::InitScriptTypes( sal_Int32 nPara )
                     (nPrevChar == 0x202F /* NNBSP, tdf#112594 */ &&
                      u_getIntPropertyValue(nChar, UCHAR_SCRIPT) == USCRIPT_MONGOLIAN))
                 {
-                    --nPos;
-                    rTypes.back().nEndPos--;
+                    rTypes.back().nEndPos = nPos = nPrevPos;
                     break;
                 }
             }

diff --git a/sw/qa/core/text/data/scriptinfo-surrogate-pairs.fodt b/sw/qa/core/text/data/scriptinfo-surrogate-pairs.fodt
@@ -287,6 +287,7 @@
     <text:sequence-decl text:display-outline-level="0" text:name="Figure"/>
    </text:sequence-decls>
    <text:p text:style-name="Standard">11◌𐻽</text:p>
+   <text:p text:style-name="Standard">11𝐀َ</text:p>
   </office:text>
  </office:body>
 </office:document>
diff --git a/sw/qa/core/text/text.cxx b/sw/qa/core/text/text.cxx
@@ -1475,9 +1475,18 @@ CPPUNIT_TEST_FIXTURE(SwCoreTextTest, testScriptinfosurrogatePairs)
     // Without the fix it fails with:
     // - Expected: 11
     // - Actual  : 11◌
-    assertXPath(pXmlDoc, "//SwParaPortion/SwLineLayout/SwLinePortion[1]", "portion", u"11");
-    assertXPath(pXmlDoc, "//SwParaPortion/SwLineLayout/SwLinePortion[2]", "portion",
+    assertXPath(pXmlDoc, "//txt[1]/SwParaPortion/SwLineLayout/SwLinePortion[1]", "portion", u"11");
+    assertXPath(pXmlDoc, "//txt[1]/SwParaPortion/SwLineLayout/SwLinePortion[2]", "portion",
                 u"\u25CC\U00010EFD");
+
+    // Without the fix this would crash because we got a lone surrogate that
+    // can’t be converted to UTF-8, but if it were not for that it might fail
+    // with something like:
+    // - Expected: 11
+    // - Actual  : 11𝐀
+    assertXPath(pXmlDoc, "//txt[2]/SwParaPortion/SwLineLayout/SwLinePortion[1]", "portion", u"11");
+    assertXPath(pXmlDoc, "//txt[2]/SwParaPortion/SwLineLayout/SwLinePortion[2]", "portion",
+                u"\U0001D400\u064E");
 }
 
 CPPUNIT_TEST_FIXTURE(SwCoreTextTest, testTdf112594)

diff --git a/sw/source/core/text/porlay.cxx b/sw/source/core/text/porlay.cxx
@@ -1457,7 +1457,7 @@ void SwScriptInfo::InitScriptInfo(const SwTextNode& rNode,
                 (nPrevChar == CHAR_NNBSP &&
                  u_getIntPropertyValue(nChar, UCHAR_SCRIPT) == USCRIPT_MONGOLIAN))
             {
-                --nPos;
+                nPos = nPrevPos;
             }
         }
         m_ScriptChanges.emplace_back(TextFrameIndex(nPos), nScript);