Rebased ref, commits from common ancestor:
commit 55bd739f985758b044e1705713de1fc80f82f8af
Author: Kevin Suo <[email protected]>
AuthorDate: Sat Jul 10 11:47:39 2021 +0800
Commit: Thorsten Behrens <[email protected]>
CommitDate: Sun Oct 30 13:46:11 2022 +0100
tdf#78427 sdext.pdfimport: No need to read a font file for the purpose of...
...determining the bold/italic/underline etc.
The purpose for reading a font file is that in case the font attributes
determined by the xpdfimport process is not enough, then we use the lo core
font classes which read in the font file and then determine whether it is bold,
italic etc.
However, while this works in some cases, it does not work in many cases
when the font file was actually a subset and a toUnicode map is followed in the
PDF, see tdf#78427.
In addition, in case the information collected from the xpdfimport process
is enough, there is no need to read the font file.
This commit removes the read of font file part. Also, this commit uses
gfxFont->getNameWithoutSubsetTag() to get the font name with the subset tags
removed, thus simplified the code in wrapper.cxx while also improves performace
as the remove of subset tags is only run when the font is a subset (the
previous code did this for all the font names).
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/118733
Reviewed-by: Noel Grandin <[email protected]>
Tested-by: Jenkins
(cherry picked from commit da59686672fd2bc98f8cb28d5f04dc978b50ac13)
restore compatibility with older popplers
with poppler 20.09:
/home/rene/LibreOffice/git/libreoffice-7-2/sdext/source/pdfimport/xpdfwrapper/pdfioutdev_gpl.cxx:
In member function 'int
pdfi::PDFOutDev::parseFont(long long int, GfxFont*,
GfxState*) const':
/home/rene/LibreOffice/git/libreoffice-7-2/sdext/source/pdfimport/xpdfwrapper/pdfioutdev_gpl.cxx:438:39:
error: 'class
GfxFont' has no member named 'getNameWithoutSubsetTag'
https://www.google.com/search?q=getNameWithoutSubsetTag&oq=getNameWithoutSubsetTag&aqs=chrome..69i57.784j0j7&sourceid=chrome&ie=UTF-8
suggests it was added in 20.12
Change-Id: I4eacd2d740cb689ff9b3c6cab59376e01b1ba162
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/118977
Tested-by: René Engelhard <[email protected]>
Tested-by: Jenkins
Reviewed-by: Noel Grandin <[email protected]>
(cherry picked from commit 07087041610ca8351d764c838ae07fa58f3bdf9e)
Conflicts:
sdext/source/pdfimport/xpdfwrapper/pdfioutdev_gpl.cxx
diff --git a/sdext/source/pdfimport/wrapper/wrapper.cxx
b/sdext/source/pdfimport/wrapper/wrapper.cxx
index f91612a5a201..b4a10780741d 100644
--- a/sdext/source/pdfimport/wrapper/wrapper.cxx
+++ b/sdext/source/pdfimport/wrapper/wrapper.cxx
@@ -499,12 +499,6 @@ void Parser::parseFontFamilyName( FontAttributes& rResult )
const sal_Unicode* pCopy = rResult.familyName.getStr();
sal_Int32 nLen = rResult.familyName.getLength();
- // parse out truetype subsets (e.g. BAAAAA+Thorndale)
- if( nLen > 8 && pCopy[6] == '+' )
- {
- pCopy += 7;
- nLen -= 7;
- }
// TODO: Looks like this block needs to be refactored
while( nLen )
@@ -626,52 +620,6 @@ void Parser::readFont()
// extract textual attributes (bold, italic in the name, etc.)
parseFontFamilyName(aResult);
- // need to read font file?
- if( nFileLen )
- {
- uno::Sequence<sal_Int8> aFontFile(nFileLen);
- readBinaryData( aFontFile );
-
- awt::FontDescriptor aFD;
- uno::Sequence< uno::Any > aArgs(1);
- aArgs[0] <<= aFontFile;
-
- try
- {
- uno::Reference< beans::XMaterialHolder > xMat(
-
m_xContext->getServiceManager()->createInstanceWithArgumentsAndContext(
- "com.sun.star.awt.FontIdentificator", aArgs, m_xContext ),
- uno::UNO_QUERY );
- if( xMat.is() )
- {
- uno::Any aRes( xMat->getMaterial() );
- if( aRes >>= aFD )
- {
- if (!aFD.Name.isEmpty())
- {
- aResult.familyName = aFD.Name;
- parseFontFamilyName(aResult);
- }
- aResult.isBold = (aFD.Weight > 100.0);
- aResult.isItalic = (aFD.Slant == awt::FontSlant_OBLIQUE
||
- aFD.Slant == awt::FontSlant_ITALIC
);
- aResult.isUnderline = false;
- aResult.size = 0;
- }
- }
- }
- catch( uno::Exception& )
- {
- }
-
- if( aResult.familyName.isEmpty() )
- {
- // last fallback
- aResult.familyName = "Arial";
- aResult.isUnderline = false;
- }
-
- }
if (!m_xDev)
m_xDev.disposeAndReset(VclPtr<VirtualDevice>::Create());
diff --git a/sdext/source/pdfimport/xpdfwrapper/pdfioutdev_gpl.cxx
b/sdext/source/pdfimport/xpdfwrapper/pdfioutdev_gpl.cxx
index 0aabffe41488..427a6c58109e 100644
--- a/sdext/source/pdfimport/xpdfwrapper/pdfioutdev_gpl.cxx
+++ b/sdext/source/pdfimport/xpdfwrapper/pdfioutdev_gpl.cxx
@@ -432,14 +432,19 @@ int PDFOutDev::parseFont( long long nNewId, GfxFont*
gfxFont, GfxState* state )
FontAttributes aNewFont;
int nSize = 0;
-#if POPPLER_CHECK_VERSION(0, 64, 0)
- const
+#if POPPLER_CHECK_VERSION(20, 12, 0)
+ std::string familyName = gfxFont->getNameWithoutSubsetTag();
+#else
+ std::string familyName = gfxFont->getName()->toStr();
+ if (familyName.length() > 7 && familyName.at(6) == '+')
+ {
+ familyName = familyName.substr(7);
+ }
#endif
- GooString* pFamily = gfxFont->getName();
- if( pFamily )
+ if( familyName != "" )
{
aNewFont.familyName.clear();
- aNewFont.familyName.append( gfxFont->getName() );
+ aNewFont.familyName.append( familyName );
}
else
{
commit a4ef98773062acdc5808a5df12f68652549fe98f
Author: Sam James <[email protected]>
AuthorDate: Fri Sep 2 04:31:18 2022 +0100
Commit: Thorsten Behrens <[email protected]>
CommitDate: Sun Oct 30 13:46:11 2022 +0100
Fix build with Poppler 22.09.0
With Poppler 22.09.0, LO fails to build with:
```
/var/tmp/portage/app-office/libreoffice-7.3.5.2/work/libreoffice-7.3.5.2/sdext/source/pdfimport/xpdfwrapper/pdfioutdev_gpl.cxx:682:36:
error: too many arguments to function call, expected single argument 'start',
have 3 arguments
state->getLineDash(&dashArray, &arrayLen, &startOffset);
~~~~~~~~~~~~~~~~~~ ^~~~~~~~~~~~~~~~~~~~~~~
/usr/include/poppler/GfxState.h:1506:32: note: 'getLineDash' declared here
const std::vector<double> &getLineDash(double *start)
^
1 error generated.
```
Poppler changed the getLineDash interface:
```
- void getLineDash(double **dash, int *length, double *start)
+ const std::vector<double> &getLineDash(double *start)
```
Signed-off-by: Sam James <[email protected]>
Change-Id: I29e18f20d7650a7fcac1bc8ab4aaa04aaa2ab8fb
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/139249
Tested-by: Jenkins
Reviewed-by: Michael Stahl <[email protected]>
(cherry picked from commit b7d63694985bbb1cf86eb71769feadb28ce68c17)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/139835
(cherry picked from commit 89d2ecd6bc71fc6e581cff595b18ae67a13d8b11)
diff --git a/sdext/source/pdfimport/xpdfwrapper/pdfioutdev_gpl.cxx
b/sdext/source/pdfimport/xpdfwrapper/pdfioutdev_gpl.cxx
index fa3d22f14c5e..0aabffe41488 100644
--- a/sdext/source/pdfimport/xpdfwrapper/pdfioutdev_gpl.cxx
+++ b/sdext/source/pdfimport/xpdfwrapper/pdfioutdev_gpl.cxx
@@ -661,8 +661,15 @@ void PDFOutDev::updateLineDash(GfxState *state)
return;
assert(state);
- double* dashArray; int arrayLen; double startOffset;
+ int arrayLen; double startOffset;
+#if POPPLER_CHECK_VERSION(22, 9, 0)
+ const std::vector<double> &dash = state->getLineDash(&startOffset);
+ const double* dashArray = dash.data();
+ arrayLen = dash.size();
+#else
+ double* dashArray;
state->getLineDash(&dashArray, &arrayLen, &startOffset);
+#endif
printf( "updateLineDash" );
if( arrayLen && dashArray )
commit 5a941f163c2bb1ae0bb831261acb35848c9035f7
Author: Nathan Pratta Teodosio <[email protected]>
AuthorDate: Tue Jun 21 08:47:14 2022 -0300
Commit: Thorsten Behrens <[email protected]>
CommitDate: Sun Oct 30 13:46:11 2022 +0100
Update for Poppler 22.06
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/136261
Tested-by: Jenkins
Reviewed-by: Michael Stahl <[email protected]>
(cherry picked from commit 0d0469b4302dfe95b016a6f04b145834b79d5ed3)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/136319
Reviewed-by: Adolfo Jayme Barrientos <[email protected]>
(cherry picked from commit 56c7e666ebf47d6cdf29adf85fc5de56246be86d)
Follow-up for Poppler 22.06 update
Change-Id: I8ee9f1a53cc4389e6a4d44e9765b478b5edfffd4
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/136342
Tested-by: Jenkins
Reviewed-by: Michael Stahl <[email protected]>
(cherry picked from commit ad1ffc62e40c2409b610dfff25a8483b1f2556ad)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/136382
Reviewed-by: Adolfo Jayme Barrientos <[email protected]>
(cherry picked from commit 71fb0151f985924af60c383331b214a5a27790a5)
Conflicts:
sdext/source/pdfimport/xpdfwrapper/pdfioutdev_gpl.cxx
diff --git a/sdext/source/pdfimport/xpdfwrapper/pdfioutdev_gpl.cxx
b/sdext/source/pdfimport/xpdfwrapper/pdfioutdev_gpl.cxx
index 94e771b9e8dc..fa3d22f14c5e 100644
--- a/sdext/source/pdfimport/xpdfwrapper/pdfioutdev_gpl.cxx
+++ b/sdext/source/pdfimport/xpdfwrapper/pdfioutdev_gpl.cxx
@@ -456,12 +456,21 @@ int PDFOutDev::parseFont( long long nNewId, GfxFont*
gfxFont, GfxState* state )
{
// TODO(P3): Unfortunately, need to read stream twice, since
// we must write byte count to stdout before
+#if POPPLER_CHECK_VERSION(22, 6, 0)
+ std::optional<std::vector<unsigned char>> pBuf =
gfxFont->readEmbFontFile( m_pDoc->getXRef() );
+ if ( pBuf )
+ {
+ aNewFont.isEmbedded = true;
+ nSize = pBuf->size();
+ }
+#else
char* pBuf = gfxFont->readEmbFontFile( m_pDoc->getXRef(), &nSize );
if( pBuf )
{
aNewFont.isEmbedded = true;
gfree(pBuf);
}
+#endif
}
m_aFontMap[ nNewId ] = aNewFont;
@@ -474,13 +483,29 @@ void PDFOutDev::writeFontFile( GfxFont* gfxFont ) const
return;
int nSize = 0;
+#if POPPLER_CHECK_VERSION(22, 6, 0)
+ std::optional<std::vector<unsigned char>> pBuf = gfxFont->readEmbFontFile(
m_pDoc->getXRef() );
+ if ( pBuf )
+ nSize = pBuf->size();
+ if ( nSize == 0 )
+ return;
+#else
char* pBuf = gfxFont->readEmbFontFile( m_pDoc->getXRef(), &nSize );
if( !pBuf )
return;
+#endif
// ---sync point--- see SYNC STREAMS above
fflush(stdout);
+#if POPPLER_CHECK_VERSION(22, 6, 0)
+ if( fwrite(pBuf->data(), sizeof(*pBuf->data()), nSize, g_binary_out) !=
static_cast<size_t>(nSize) )
+ {
+ exit(1); // error
+ }
+ // ---sync point--- see SYNC STREAMS above
+ fflush(g_binary_out);
+#else
if( fwrite(pBuf, sizeof(char), nSize, g_binary_out) !=
static_cast<size_t>(nSize) )
{
gfree(pBuf);
@@ -489,6 +514,7 @@ void PDFOutDev::writeFontFile( GfxFont* gfxFont ) const
// ---sync point--- see SYNC STREAMS above
fflush(g_binary_out);
gfree(pBuf);
+#endif
}
#if POPPLER_CHECK_VERSION(0, 83, 0)
@@ -742,7 +768,11 @@ void PDFOutDev::updateFont(GfxState *state)
{
assert(state);
+#if POPPLER_CHECK_VERSION(22, 6, 0)
+ GfxFont *gfxFont = state->getFont().get();
+#else
GfxFont *gfxFont = state->getFont();
+#endif
if( !gfxFont )
return;
diff --git a/sdext/source/pdfimport/xpdfwrapper/wrapper_gpl.cxx
b/sdext/source/pdfimport/xpdfwrapper/wrapper_gpl.cxx
index 8666ab15cfa6..521a1f91df24 100644
--- a/sdext/source/pdfimport/xpdfwrapper/wrapper_gpl.cxx
+++ b/sdext/source/pdfimport/xpdfwrapper/wrapper_gpl.cxx
@@ -119,6 +119,15 @@ int main(int argc, char **argv)
_setmode( _fileno( g_binary_out ), _O_BINARY );
#endif
+#if POPPLER_CHECK_VERSION(22, 6, 0)
+ PDFDoc aDoc( std::make_unique<GooString>(pFileName),
+ std::optional<GooString>(pOwnerPasswordStr),
+ std::optional<GooString>(pUserPasswordStr) );
+
+ PDFDoc aErrDoc( std::make_unique<GooString>(pErrFileName),
+ std::optional<GooString>(pOwnerPasswordStr),
+ std::optional<GooString>(pUserPasswordStr) );
+#else
PDFDoc aDoc( pFileName,
pOwnerPasswordStr,
pUserPasswordStr );
@@ -126,6 +135,7 @@ int main(int argc, char **argv)
PDFDoc aErrDoc( pErrFileName,
pOwnerPasswordStr,
pUserPasswordStr );
+#endif
// Check various permissions for aDoc.
PDFDoc &rDoc = aDoc.isOk()? aDoc: aErrDoc;
