fdo#77996: writerfilter: RTF import: re-work destination text buffering

The problem in the bugdoc is that the ';' style terminator is encoded as
{\uc1 \u59 ?}, i.e. inside a group, so when reading the ';' the
aDestinationText of the top group is empty and the style name is lost.
Or since the style name characters are encoded in the same way, every
character is lost once the group closes.

The same problem affects some of the document properties.

Introduce an abstraction of RTFParserState::aDestinationText so that for
every destination only one of these buffers is used, regardless of
nested group structures; the aDestinationText buffer is only switched
when entering a new destination.

Also, the \revtbl and \stylesheet destinations do not contain entries
directly, i.e., every entry must be in a sub-group, so remove some
special-casing for these; however, for \fonttbl the entries may be in
groups or not.

(cherry picked from commit 82e17dbb2a16c7653a163139f0eea51faa4d46b8)

Conflicts:
	writerfilter/source/rtftok/rtfdocumentimpl.cxx

Change-Id: Ica276a8b730e4a707530471ba27bfdd1582b8890
Reviewed-on: https://gerrit.libreoffice.org/10464
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Miklos Vajna <vmiklos@collabora.co.uk>

2 files changed, 57 insertions, 0 deletions

diff --git a/sw/qa/extras/rtfimport/data/fdo77996.rtf b/sw/qa/extras/rtfimport/data/fdo77996.rtf
new file mode 100644
index 000000000000..b5b4fdec7905
--- /dev/null
+++ b/sw/qa/extras/rtfimport/data/fdo77996.rtf
@@ -0,0 +1,35 @@
+{\rtf \ansi \ansicpg0 \deff0 \stshfdbch1 \stshfloch0 \stshfhich0 \deflang1033 \deflangfe1033
+
+{\fonttbl
+{\f0 \froman \fcharset0 \fprq2 Times New Roman{\*\falt Times New Roman};}
+{\f1 \fnil \fcharset134 \fprq0 {\uc1 \u23435 ?}{\uc1 \u20307 ?}{\*\falt {\uc1 \u23435 ?}{\uc1 \u20307 ?}};}
+{\f6 \fnil \fcharset0 \fprq0 Wingdings{\*\falt Wingdings};}
+{\f10 \fnil \fcharset0 \fprq0 Calibri{\*\falt Calibri};}
+{\f11 \fnil \fcharset134 \fprq0 {\uc1 \u24494 ?}{\uc1 \u36719 ?}{\uc1 \u38597 ?}{\uc1 \u40657 ?}{\*\falt {\uc1 \u24494 ?}{\uc1 \u36719 ?}{\uc1 \u38597 ?}{\uc1 \u40657 ?}};}
+}
+
+{\stylesheet
+{\qj \li0 \ri0 \nowidctlpar \aspalpha \aspnum \adjustright \lin0 \rin0 \itap0 \fs21 \kerning2 \dbch \af1 \hich \af10 \loch \f10 \lang1033 \langnp1033 \langfe2052 \langfenp2052 \snext0 Normal{\uc1 \u59 ?}}
+{\s1 \ql \li0 \ri0 \sb100 \lisb0 \sa100 \lisa0 \sbauto1 \saauto1 \widctlpar \aspalpha \aspnum \adjustright \lin0 \rin0 \itap0 \outlinelevel0 \b1 \fs48 \kerning36 \dbch \af1 \hich \af1 \loch \f1 \lang1033 \langnp1033 \langfe2052 \langfenp2052 \sbasedon0 \snext1 \slink18 heading 1{\uc1 \u59 ?}}
+{\s2 \ql \li0 \ri0 \sb100 \lisb0 \sa100 \lisa0 \sbauto1 \saauto1 \widctlpar \aspalpha \aspnum \adjustright \lin0 \rin0 \itap0 \outlinelevel1 \b1 \fs36 \kerning0 \dbch \af1 \hich \af1 \loch \f1 \lang1033 \langnp1033 \langfe2052 \langfenp2052 \sbasedon0 \snext2 \slink16 heading 2{\uc1 \u59 ?}}
+{\*\cs10 \snext10 Default Paragraph Font{\uc1 \u59 ?}}
+{\*\cs15 \fs18 \v0 \sbasedon10 \snext15 {\uc1 \u97 ?}{\uc1 \u117 ?}{\uc1 \u116 ?}{\uc1 \u104 ?}{\uc1 \u111 ?}{\uc1 \u114 ?}{\uc1 \u59 ?}}
+{\*\cs16 \b1 \fs36 \kerning0 \dbch \af1 \hich \af1 \loch \f1 \sbasedon10 \snext16 \slink2 {\uc1 \u26631 ?}{\uc1 \u39064 ?}{\uc1 \u32 ?}{\uc1 \u50 ?}{\uc1 \u32 ?}{\uc1 \u67 ?}{\uc1 \u104 ?}{\uc1 \u97 ?}{\uc1 \u114 ?}{\uc1 \u59 ?}}
+{\*\cs17 \b1 \sbasedon10 \snext17 strong{\uc1 \u59 ?}}
+{\*\cs18 \b1 \fs48 \kerning36 \dbch \af1 \hich \af1 \loch \f1 \sbasedon10 \snext18 \slink1 {\uc1 \u26631 ?}{\uc1 \u39064 ?}{\uc1 \u32 ?}{\uc1 \u49 ?}{\uc1 \u32 ?}{\uc1 \u67 ?}{\uc1 \u104 ?}{\uc1 \u97 ?}{\uc1 \u114 ?}{\uc1 \u59 ?}}
+{\s19 \ql \li0 \ri0 \sb375 \lisb0 \sa100 \lisa0 \saauto1 \widctlpar \aspalpha \aspnum \adjustright \lin0 \rin0 \itap0 \fs18 \kerning0 \dbch \af1 \hich \af1 \loch \f1 \lang1033 \langnp1033 \langfe2052 \langfenp2052 \sbasedon0 \snext19 {\uc1 \u101 ?}{\uc1 \u120 ?}{\uc1 \u116 ?}{\uc1 \u114 ?}{\uc1 \u97 ?}{\uc1 \u99 ?}{\uc1 \u116 ?}{\uc1 \u50 ?}{\uc1 \u59 ?}}
+}
+
+{\info
+{\title {\uc1 \u21414 ?}{\uc1 \u38376 ?}{\uc1 \u38056 ?}{\uc1 \u19994 ?}{\uc1 \u32929 ?}{\uc1 \u20221 ?}{\uc1 \u26377 ?}{\uc1 \u38480 ?}{\uc1 \u20844 ?}{\uc1 \u21496 ?}}
+{\author {\uc1 \u65 ?}{\uc1 \u108 ?}{\uc1 \u110 ?}{\uc1 \u32 ?}{\uc1 \u76 ?}{\uc1 \u105 ?}{\uc1 \u110 ?}{\uc1 \u32 ?}{\uc1 \u40 ?}{\uc1 \u66 ?}{\uc1 \u101 ?}{\uc1 \u105 ?}{\uc1 \u32 ?}{\uc1 \u74 ?}{\uc1 \u105 ?}{\uc1 \u110 ?}{\uc1 \u103 ?}{\uc1 \u41 ?}}
+{\operator {\uc1 \u106 ?}{\uc1 \u97 ?}{\uc1 \u121 ?}}
+}
+
+
+{\loch \af11 \hich \af11 \dbch \f11
+{\uc1 \u21414 ?}{\uc1 \u38376 ?}{\uc1 \u38056 ?}{\uc1 \u19994 ?}{\uc1 \u32929 ?}{\uc1 \u20221 ?}{\uc1 \u26377 ?}{\uc1 \u38480 ?}{\uc1 \u20844 ?}{\uc1 \u21496 ?}
+}
+\par
+
+}
diff --git a/sw/qa/extras/rtfimport/rtfimport.cxx b/sw/qa/extras/rtfimport/rtfimport.cxx
index 128e2a3b1069..2a956dbe2487 100644
--- a/sw/qa/extras/rtfimport/rtfimport.cxx
+++ b/sw/qa/extras/rtfimport/rtfimport.cxx
@@ -1267,6 +1267,28 @@ DECLARE_RTFIMPORT_TEST(testN823675, "n823675.rtf")
     CPPUNIT_ASSERT_EQUAL(OUString("Symbol"), aFont.Name);
 }
 
+DECLARE_RTFIMPORT_TEST(testFdo77996, "fdo77996.rtf")
+{
+    // all styles were imported as name "0"
+    uno::Reference<container::XNameAccess> xChars(getStyles("CharacterStyles"));
+    CPPUNIT_ASSERT(!xChars->hasByName("0"));
+    CPPUNIT_ASSERT(xChars->hasByName("strong"));
+    CPPUNIT_ASSERT(xChars->hasByName("author"));
+    uno::Reference<container::XNameAccess> xParas(getStyles("ParagraphStyles"));
+    CPPUNIT_ASSERT(!xParas->hasByName("0"));
+    CPPUNIT_ASSERT(xParas->hasByName("extract2"));
+    // some document properties were lost
+    uno::Reference<document::XDocumentPropertiesSupplier> xDocumentPropertiesSupplier(mxComponent, uno::UNO_QUERY);
+    uno::Reference<document::XDocumentProperties> xProps(xDocumentPropertiesSupplier->getDocumentProperties());
+    CPPUNIT_ASSERT_EQUAL(OUString("Aln Lin (Bei Jing)"), xProps->getAuthor());
+    CPPUNIT_ASSERT_EQUAL(
+        OUString("\xe5\x8e\xa6\xe9\x97\xa8\xe9\x92\xa8\xe4\xb8\x9a\xe8\x82\xa1\xe4\xbb\xbd\xe6\x9c\x89\xe9\x99\x90\xe5\x85\xac\xe5\x8f\xb8", 30,
+            RTL_TEXTENCODING_UTF8),
+        xProps->getTitle());
+    uno::Reference<beans::XPropertySet> xUDProps(xProps->getUserDefinedProperties(), uno::UNO_QUERY);
+    CPPUNIT_ASSERT_EQUAL(OUString("jay"), getProperty<OUString>(xUDProps, "Operator"));
+}
+
 DECLARE_RTFIMPORT_TEST(testFdo47802, "fdo47802.rtf")
 {
     uno::Reference<drawing::XDrawPageSupplier> xDrawPageSupplier(mxComponent, uno::UNO_QUERY);