summaryrefslogtreecommitdiff
path: root/converters/libvisio
AgeCommit message (Collapse)AuthorFilesLines
2017-04-22Revbump after icu updateadam2-4/+4
2017-01-01Revbump after boost updateadam2-4/+4
2016-12-04Recursive revbump from textproc/icu 58.1ryoon2-4/+4
2016-10-07Revbump post boost updateadam2-4/+4
2016-07-09Bump PKGREVISION for perl-5.24.0 for everything mentioning perl.wiz1-2/+2
2016-04-11Recursive revbump from textproc/icu 57.1ryoon2-3/+4
2016-01-03Update libvisio to 0.1.5:wiz2-7/+7
libvisio 0.1.5 - Rewrite text output to avoid empty spans and to make the function a bit more readable for common mortal (wo)man. - Initial emulation of line rounding property (tdf#90603). - Make --help output of all command line tools more help2man-friendly.
2015-12-30Update libvisio to 0.1.4:wiz2-8/+7
libvisio 0.1.4 - Implement overriding of colour of lines and of text from layer properties (tdf#50309, tdf#68392) and the visibility and/or printability of a layer. - Fix incorrect handling of stencil text (tdf#90154). - Basic initial implementation of hatch fill (tdf#44552, tdf#76835). - Implement support of bullets/lists (tdf#92349) - Implement support of default tab-stops and custom tab-sets in paragraph properties. - Fix for reading of names of pages if present. - Fix build with boost 1.59. - Instead of line-break, spit out a new paragraph when a paragraph break is found
2015-11-03Add SHA512 digests for distfiles for converters categoryagc1-1/+2
Problems found with existing distfile: distfiles/libiconv-1.13-cp932.patch.gz No changes made to the libiconv distinfo file. Otherwise, existing SHA1 digests verified and found to be the same on the machine holding the existing distfiles (morden). All existing SHA1 digests retained for now as an audit trail.
2015-10-10Recursive revbump from textproc/icuryoon1-2/+2
2015-08-15Bump PKGREVISION for librevenge boost fix.wiz2-3/+4
2015-07-26Update to 0.1.3:wiz3-23/+6
libvisio 0.1.3 - Fix a build failure with some compilers. - Do not use C++11 syntax. libvisio 0.1.2 - Fix several build problems with different compilers. - Import more metadata properties. - Fix theme color import (tdf#92270). - Fix a bunch of crashes and hangs when reading broken documents, found with the help of american-fuzzy-lop. - Fix several memory leaks. - Ignore errors when parsing metadata. This makes some documents readable again. - Fix handling of character background (tdf#92271).
2015-06-12Recursive PKGREVISION bump for all packages mentioning 'perl',wiz1-2/+2
having a PKGNAME of p5-*, or depending such a package, for perl-5.22.0.
2015-04-17Revbump after updating devel/boost-libsadam1-2/+2
2015-04-06Revbump after updating textproc/icuadam2-4/+4
2015-02-02cppunit is required afer alladam1-2/+2
2015-02-02Fixed buildlinking; disabled documentation building as in other libraries.adam3-481/+15
2015-01-28Fix PLIST for doxygen-1.8.9 and depend on that version.wiz2-16/+10
Bump PKGREVISION.
2015-01-04Update to 0.1.1:wiz4-8/+40
libvisio 0.1.1 - Fix a minor issue reported by Coverity. - Check arguments of public functions. Passing NULL no longer causes a crash. - Set a minimal thickness for arrow markers to ensure they are visible. - Fix a crash when reading a broken file, found with the help of american-fuzzy-lop. - Use symbol visibility on Linux. The library only exports the two public functions now. - Import document metadata (fdo#86664).
2014-11-07Revbump after updating boostadam1-2/+2
2014-10-07Revbump after updating libwebp and icuadam1-2/+2
2014-08-13Revbump after boost-libs updateadam1-1/+2
2014-07-22Update to 0.1.0. Ride libwpd bump.wiz4-55/+39
libvisio 0.1.0 - switch to librevenge
2014-07-22Recursive bump for libwpd shlib name change.wiz1-2/+2
2014-05-29Bump for perl-5.20.0.wiz1-2/+2
Do it for all packages that * mention perl, or * have a directory name starting with p5-*, or * depend on a package starting with p5- like last time, for 5.18, where this didn't lead to complaints. Let me know if you have any this time.
2014-04-28Dance the doxygen!joerg2-7/+4
2014-04-09recursive bump from icu shlib major bump.obache1-2/+2
2014-01-05Fix build with doxygen-1.8.6 and depend on it. PKGREVISION++wiz2-87/+87
2013-11-20Revbump after updating devel/boost-libsadam1-2/+2
2013-11-20recursive bump from boost-lib shlib major bump.obache1-1/+2
2013-10-22Update to 0.0.31: changes undocumented.wiz3-9/+23
2013-05-31Bump all packages for perl-5.18, thatwiz1-2/+2
a) refer 'perl' in their Makefile, or b) have a directory name of p5-*, or c) have any dependency on any p5-* package Like last time, where this caused no complaints.
2013-02-15recursive bump from boost-lib shlib update.obache1-1/+2
2013-01-06Update to 0.0.24:wiz2-6/+6
Bump version in view of a possible release When cached fg and bg colours are 0, trust index more coverity: copying uninitialized value coverity: do not let VisioDocument::parse throw coverity: drop duplicated code coverity: consistently check ret. value of read*Data Trying to support initially Text Fields Fixed line height is in inches :(
2012-12-16Update to 0.0.23:wiz2-6/+6
For VSD6 and VSD11 trust more the cached hex then the colour index
2012-12-08Update to 0.0.22:wiz3-209/+308
Bump version Trying to unify the stencil extraction and the other parsing function Those ones don't need to be virtual anymore Unify all VSD2 - VSD5 in one parser Unify blob parsing over binary format versions Unsigned signed comparison Little fixup handle blog in the same way for all <= 5 Bump version, because I can Field List is a list too Fix background page detection for versions < 6 Simplify the inheritance a little bit Fix foreign in VSD 5- Some fixes for line and fill styles Some colour reading fixes for VSD <= 2 Fix stencil extraction for vsd <= 5 Miscellaneous fixes + text for VSD <= 5 Consider correctly the font encoding information Vertical alignment in center is default in Visio Trying to read the blob in the same way in all 16-bit versions Spit actual images in VSD1 Some more work on VSD1 and VSD2 parsing + some stream fixes Some more progress on version 5 Read VSD5 fill and line information Remove the generated files only on distclean Typedefs for MSVC Abstraction of reading of integers Some progress in VSD5 parsing Some more work on vsd1/2/3/4/5 Some more VSD2 - VSD5 infrastructure - work in progress Some more chunk work, but slow Setting framework for different parsing of v5 and v2 Set groundwork for vsd5 and lower chunk parsing Extend the parsing capacity to versions 1 - 5 Don't crash on empty or almost empty pointer order list Detect non-OLE binary Visio documents Fix warning C4244 ´initializing´ : conversion from ´double´ to ´unsigned char´, possible loss of data Some more work on VSD5 parser Allow the parser to parse VSD4 and VSD5 + read pointers in < 6 correctly Fix underlinking with gold. Fix a memory leak Fix extraction of stencils from xml file-formats Fix scope of the variable 'X' can be reduced Missing include for older compilers Silence some warnings Allow specifying boost include directory too Add libxml2 include and library directories Add the Visual Studio 2010 files Those don't need to be executable Add Visual Studio 2008 files Like in libcdr, detect each pkg-config dependency separately Fill missing char/para style only with legitimate default styles Default para/char style is first para/char style if no style 0 is present + fill inexisting styles from default Collect default para/char styles Don't overwrite charCount on style override Functions to reset charCount in char/para styles and a function to get a char/para list level Make charCount non-optional in paragraph styles Make charCount non-optional in character styles Fix some typos in the VSDXParser Actually the relative beziers give control points Update alternative build systems Some infrastructure for overriding of character and paragraph styles Let parser recover on errors Whitespace is significant in this context Clean page shape list at the beginning of a page A horrible typo that takes hours to discover :) Consolidate the way of reading font information A little work on text block styles VDX: actually read the text block properties Only one charlist and paralist can be in a shape Override paragraph and character lists Read TextXForm data too Calm valgrind Don't handle utf-8 as ansi + try to not run out of bounds Don't prefill the stencil text Actually output some text Trying to parse text in XML parsers Extract IX reading to a function Stub ParaIX and CharIX reading in VDX and VSDX parsers Read the stylesheet in second pass Initialize the default colour palette even if no colour section exists in file In a generic way skip unhandled sections inheritance of the styles is handled in the collectors Skip useless sections instead of diving into them default excentricity cannot be 0 Some more boost::optional foo Simplify the reading of data to reuse as much as one can NURBS and Polyline support for VDX and VSDX Stub the parsing of polyline and nurbs functions Move more stuff to boost::optional unify some stuff between the VSDX and VDX parsers boost::optional::is_initialized is deprecated Fix a typo If there are both Master and MasterShape attributes missing, it is simply master-less shape Removing unused fill-related variables Refactoring the fill style output Remove unused member variables Simplify the function handling line styles A little change in style outputting, still more to come when time allows Fix showing of foregins + write styles only if needed All styles pushed using optional styles Allow getting optional styles from the stylesheets Don't default-construct black backgrounds Don't append double Z segment and assure that ellipse is always closed Don't draw markers on closed path segments No point to dupplicate code in two different functions Consider also styles ids 0 A little refactoring of styles in VSDContentCollector add override for optional styles too readUnsignedCharData -> readByteData + move a bit Move the getColourString to a better place No need for a whole template for 2 sorts Actually disables -Weffc++ when requested Pass more properties as boost::optional pass the noLine, noFill, noShow to boost::optional too Some more path elements using boost::optional Pass some of the attribute reading to boost::optional VDX overrides geometries the same way as VSD, VSDX overrides only what needs to be overridden The common class can have those two data members Try to fill the first 24 colour values Implement empty geometry element to handle deleted segments like a big boy Shadow offset for Y-axis must be inverted Assure that the geometry elements will be always output sorted by keys + fix leak Fix some little problems in the VDX and VSDX parsers Trying to fix the extended colour reading Try to detect when the shape is not started Determine the first stencil shape correctly Clean up reading of VSDX shape properties. Remove uninitialised variable compiler warning. Read byte from VSDX XML string. For use with VSDX values of type PtgByte (unsigned byte). Some fixes in colour Some infrastructure to parse the relative geometry rows Simplify the styles a bit Simplify the styles code a bit Some work on styles Some wireframe output for VDX and VSDX parsers Parsing line, fill and text block information in VSDXParser Convenience pointer to access the current geometry list Prefill shape's m_geometries from stencil m_geometries being a map Make the test documents in VSDX and VDX parsers pass Fix one hang in VSDX parser Basic support of raster images in VDX and VSDX parser Try not to crash in XML based parsers _flushShape for VDXParser too _flushShape uses member variable anyway Add Foreign reading functions Add some readability Read the xform directly to the shape structure Some work on the xml-based parsers Slight modification of the base 64 decoder addShapeId when shapes document order is their rendering order Append WPXBinaryData from a base64 encoded string Flush all shape information to the collectors in one run/function Use different shapeLists for page shapes and group shapes send shapesOrder only if it is not empty Flush shapes order whenever we know it Add m_shapeList member to VSDShape First check for non-empty and then look for top() Adapt collectors to the information that will be available from VDX and VSDX parsers too Remove some leaks VSDParser: use more of the VSDShape internals to keep information Implement VSDShape::clear() Get rid of m_fieldList Revert a part that creates more problems then it solves If a key is not found in a map, operator[] default-constructs an element and returns its reference Trying to improve the field-related code Some fixes of fields and line-breaks Try to pass to the collectors the shape at once Some little improvements A bit more of simplification Avoid dynamic_cast Don't add empty geometries Trying to refactor a bit more Fix debug build Diminish pointer use a bit Fix possible uninitialized variables More VSDParser refactoring Try to finalize the _flushShape function Some constness changes Store some more stuff in the VSDShape object Some more styles passing simplification Simplify the passing of line and fill information Simplify the collectXFormData and collectTxtXform functions a bit Some more simplification of the VSDParser Simplyfy handling of shapes a bit m_stencilShape -> m_shape Access directly the stencil shape using master page and shape IDs VSDStencilShape -> VSDShape Some cppcheck cleaning Some more configurability about warnings Some more readability an error-proof changes This is a bit more readable Add stencil xform and txtxform Try to start to parse shapes Don't pass geomFlags to the collectors Don't crash on invalid namespace uris in SolutionXML elements Don't abort on themed values Try to unify stuff into a common XML parser base Simplify a bit Output pages in VSDXParser too Handle normal and background pages differently Read the information about background page in VDX files too Pass to collectors information about background page Detect a background page early Include VSDXMLTokenMap.h in the VSDXParser and move the later into the libtokenmap helper library Some framework for VSDX parser And also fix some bugs, so that regression suite is happy Pass the transparency to collectors as double VDX: Output now finally empty pages with the right sizes Make the VSD parser a bit more robust - Some documents can have empty groups/shapes, so handle that graciously. - Only wrap by try block the reading part, not the whole nested function calls in order not to handle exceptions thrown by other functions. Our assumptions about stream formats are a bit fragile, so don't fail if we are not right Simplifying a bit the colour stuff Rewriting the collectors to be more useful for VDX too VDX: parse xform data Parse VDX TextBlock properties Add a token id Parse VDX fill and line properties Some more VDX parser stuff Start to parse the font and colour list in VDX parser Add all VDX elements For security, check for sed Export only useful functions + speed-up rebuild when tokens change Setting up a hash framework for intelligent work with xml tokens requires gperf >= 3.0.0 at build time Try to avoid memory management of libxml2 strings as much as possible Allow building against older libwpd-stream, but without performance boost Set a custom error handler for xmlTextReader Speed-up zip central directory search, but depend on git version of libwpd Add BIPU VDX parser Extract binary data streams pointed by <Rel .../> tags too Flatten the xml that reader functions have to read by following the <Rel .../> tags Adapting alternative build systems Add VDX typedetection Spin off the XML stuff in view of VDX support Simplify a bit the VSDXParser Clean up the paths pushed to the API Suck in the collectors to the VSDX picture Find a way to rebase the targets just after reading of the relationships Some more relationship and stream parsing boilerplate Add function to get relationships stream for a target Visio 2013: Parse OPC relationships and make the typedetection a bit more serious Some BIPU typedetection for vsdx + some parser boilerplate Stub the VSDX parser pointless whitespace cleanup to increase my commit count Updating the alternative build systems too Prepare for the new VSDX file-format (ooxml version) fix reassignment a value before the old one has been used Don't output the last line break in a paragraph Prevent automake warnings Trying to output line breaks that are not paragraph breaks Some initial handling of 0x0* type of streams Bump version so that one can detect the presence of the new stencil parser BIPU vss2* support Initial stencils extraction set WPXParser::m_extractStencils to true in order to get stencil pages instead of normal pages Trying to unversalize the parser Extract chunk reader logic in order to reuse with blobs Trully implement 'level continuum' Simplify a bit the code Revert "Implement 'level continuum'" This reverts commit e44063e93b8f0ff9bcd62a26e91045192030773a. Implement 'level continuum' Little tiny cleanup Those streams are blobs, and we will not descend them, so return -> break for consistency Getting rid of handleForeignOle Get rid of handleStencilForeign function Move handleStencilPage to handleStream Unify a bit the logic of different stream parser functions
2012-10-03Drop superfluous PKG_DESTDIR_SUPPORT, "user-destdir" is default these days.asau1-3/+1
2012-08-31Update PLIST for doxygen-1.8.2 and depend on it. Bump PKGREVISION.wiz2-4/+8
2012-08-29Recursive bump from boost-libs update.obache1-1/+2
2012-07-29Update to 0.0.19:wiz2-7/+6
Bump version to 0.0.19 Fixing fdo#50990 reuse m_currentShapeLevel for stylesheets too Another tiny incremental change A little refactoring of the level change handling Order of pages is their order of arrival Revert "Drawing pages in right order" This modification is not needed anymore, since we approach the reordering of pointers differently This reverts commit 9371595a11a0858fc55ea9fce67dba12b39a6e3e. Reorder pointers to streams according to the list Handle 0x8* and 0xd* streams in an equivalent way Push information to collector even from streams if that is where it is to be found Keep pointer numbers around Get rid of handleStencils + some fixes in styles Clear font name before appending to it Get rid of handlePages function Removing unused constructor Recursive parsing of streams implemented Not giving any output as of yet though. Please, use STABLE-0-0-0 branch if you want to see pictures. I will merge this one there only after regression test calls me "good boy". Little correction Unify handling of chunks in stencils and normal pages Some more refactoring Extracting individual stream handling into a separate function Extract handling of streams into a separate function astyle Fix endless loop with text fields in VSD6 Appease automake 1.12 !=, not == coverity: protect against self-assignment coverity: protect against self-assignment coverity: protect against self-assignment coverity: memory leak coverity: bin superfluous copy ctors coverity: bin superfluous copy ctor coverity: bin superfluous copy ctors
2012-07-07New doxygen uses different files.prlw12-6/+22
2012-07-02Revbump after updating boostadam1-1/+2
2012-07-01Update to 0.0.18:wiz3-7/+11
Disable optimization in debug build Fix prefix ++/-- operators for non-primitive types This langid does not make sense and we don't even use it Trying to make encoding work in version 6 Sync this with libcdr Fix utf16 parsing Some more rework of text encoding conversions Some character conversion rework + some debug messages Some belts and braces Related: fdo#48601 FILEOPEN: Open Visio's file with bitmap's blocks This fdo contains 2 files with which I had a different crash for each of them. This patch fixes one of the 2 crashes. For the moment no more cppcheck reports Resolves: fdo#50988 bumping version Fix another crash in exotic circumstances And make it even compile Don't push_back(.back()) Fix some crashes on windows And produce a valid xml too Avoid SVG crash for missing svg:id Make the thing even build Break line only at the end of the line Add vsd2text tool to dump text from Visio diagrams More warnings to be safer later Unused variable Resolves: fdo#50178 Reading Visio file causes crash include stdint.h and inttypes.h inconditionally in no-autofoo no-windows build Group the filled and stroked paths that are part of the same geometry
2012-06-03Update to 0.0.17, changes undocument (AFAICT)wiz3-7/+13
2012-04-22Update to 0.0.16:wiz2-6/+6
A better fix that does not alter Bruno's work Adapt the end and start line markers to recent changes of svg:d in LO Read longer integers in one run A bit more of sanity tabs -> spaces Consider svg:x, svg:y, svg:width, svg:height separately Generate SVG for bitmap fill if asked for emtpy, don't use size
2012-03-16Initial import of libvisio-0.0.15, packaged for wip by Francois Tigeot.wiz5-0/+399
Libvisio is library providing ability to interpret and import visio diagrams into various applications.
generated by cgit v1.2.3 (git 2.39.1) at 2024-05-18 21:09:13 +0000