QCAD Bugtracker

When compiling QCAD 3.4.2.0 on Linux, the transactionlistener plugin gets installed to the wrong folder. The attached patch provides the correct directory path that trasactionlistener.so file gets installed too.

g++ -c -pipe -O2 -Wall -W -pthread -D_THREAD_SAFE -fPIC -DQCAD_DLL -DPTHREADS 
  -DHAVE_PTHREAD_H -DQT_NO_DEBUG -DQT_WEBKIT_LIB -DQT_SCRIPT_LIB
  -DQT_SCRIPTTOOLS_LIB -DQT_SVG_LIB -DQT_SQL_LIB -DQT_XMLPATTERNS_LIB -DQT_XML_LIB
  -DQT_OPENGL_LIB -DQT_GUI_LIB -DQT_NETWORK_LIB -DQT_CORE_LIB -DQT_SHARED
  -I/usr/local/share/qt4/mkspecs/freebsd-g++ -I. -I/usr/local/include/qt4/QtHelp
  -I/usr/local/include/qt4/QtDesigner -I/usr/local/include/qt4/QtUiTools
  -I/usr/local/include/qt4/QtCore -I/usr/local/include/qt4/QtNetwork
  -I/usr/local/include/qt4/QtGui -I/usr/local/include/qt4/QtOpenGL
  -I/usr/local/include/qt4/QtXml -I/usr/local/include/qt4/QtXmlPatterns
  -I/usr/local/include/qt4/QtSql -I/usr/local/include/qt4/QtSvg
  -I/usr/local/include/qt4/QtScriptTools -I/usr/local/include/qt4/QtScript
  -I/usr/local/include/qt4/QtWebKit -I/usr/local/include/qt4 -I../../../../core
  -I../../../../core/math -I../../../../grid -I../../../../snap -I../../../../gui
  -I../../../../entity -I../../../../operations -I../../../../scripting
  -I../../../../stemmer -I../../../../scripting/ecmaapi
  -I../../../../scripting/ecmaapi/generated
  -I../../../../scripting/ecmaapi/adapters -I../../../../io/dxf
  -I../../../../spatialindex -I../../../../3rdparty -I. -I/usr/local/include
  -Irelease -I/usr/local/include/qt4 -I/usr/local/include -o
  release/qtscript_QPinchGesture.o
  ../../generated_cpp/com_trolltech_qt_gui/qtscript_QPinchGesture.cpp
../../generated_cpp/com_trolltech_qt_gui/qtscript_QPinchGesture.cpp:53: error: 
  redefinition of 'struct QMetaTypeId<QFlags<QPinchGesture::ChangeFlag> >'
/usr/local/include/qt4/QtGui/qgesture.h:202: error: previous definition of 
  'struct QMetaTypeId<QFlags<QPinchGesture::ChangeFlag> >'
../../generated_cpp/com_trolltech_qt_gui/qtscript_QPinchGesture.cpp:240: 
  warning: 'QScriptValue qtscript_QPinchGesture_prototype_call(QScriptContext*,
  QScriptEngine*)' defined but not used
*** [release/qtscript_QPinchGesture.o] Error code 1

$ sed '202!d' /usr/local/include/qt4/QtGui/qgesture.h
Q_DECLARE_METATYPE(QPinchGesture::ChangeFlags)

$ sed '53!d' <...>/qt-labs-qtscriptgenerator-4.8.5/<...>/qtscript_QPinchGesture.cpp
Q_DECLARE_METATYPE(QFlags<QPinchGesture::ChangeFlag>)

Using Qt version 4.8.5 in /usr/local/lib/qt4.

If I “Trim Both” a pair of lines which already meet, then neither line changes appearance on screen (as expected). However, after saving and reloading the file, both lines are now missing.

To reproduce:

1. Create a new drawing
2. Draw a 10×10 square
3. Save the drawing
4. Close and reopen. The square shows up as expected.
5. Use the Trim Both tool on a pair of lines meeting at one of the corners. No change in the display.
6. Save the drawing
7. Close and reopen. The lines used in step 5. have disappeared.

If one or both of the lines changes length, either lengthened or shortened, then the changed line is preserved.

Build information:

Version: 3.4.2.0 (3.4.2)
Build Date: Oct 25 2013
Qt Version: 4.7.4
Compiler: MSVC++ 10.0 (2010)

Running on Windows 7 64-bit.

Points are not being exported.

Properties with title ‘Z’ are not displayed in the property editor. This is correct but should not apply to custom, user defined properties.

Screen based linetypes not working for splines, texts (CAD fonts) and dimension arrows.

Arcs and circles are exported with too many segments if the drawing scale is below 1.0 (e.g. 1:100).

Entering a coordinate (e.g. 20,20) results in direct distance entry instead (angle from relative zero to mouse pointer, length 20).

Line pattern factor for zero weight lines defaults to 0 which can lead to division by zero.

See discussion at:
http://www.qcad.org/rsforum/viewtopic.php?f=31&t=2614

Yes, I had the ‘look for upgrades selected’ and ticked in the dialog. I only acknowledged the available update through the forum announcement made!

6a5c66 on both FreeBSD and Linux (particular example on FreeBSD). The following diff is present, but it should be a no-op in this regard:

--- a/shared.pri
+++ b/shared.pri
@@ -11,6 +11,7 @@ DEFINES += QCAD_DLL
 # output directory:
 CONFIG(debug, debug|release) {
     ROUTDIR = debug
+    QMAKE_CXXFLAGS_DEBUG += -ggdb3
 } else {
     ROUTDIR = release
 }

qmake-qt4 qcad.pro CONFIG=debug

Goes on a while, then bombs off at

g++ -c -pipe -g -ggdb3 -Wall -W -fPIC -DQCAD_DLL -DPTHREADS -DHAVE_PTHREAD_H
    -I/usr/local/share/qt4/mkspecs/freebsd-g++
    -I.
    -I/usr/local/include/qt4/QtHelp
    -I/usr/local/include/qt4/QtDesigner
    -I/usr/local/include/qt4/QtUiTools
    -I../../../../core
    -I../../../../core/math
    -I../../../../grid
    -I../../../../snap
    -I../../../../gui
    -I../../../../entity
    -I../../../../operations
    -I../../../../scripting
    -I../../../../stemmer
    -I../../../../scripting/ecmaapi
    -I../../../../scripting/ecmaapi/generated
    -I../../../../scripting/ecmaapi/adapters
    -I../../../../io/dxf
    -I../../../../spatialindex
    -I../../../../3rdparty
    -I.
    -I/usr/local/include/qt4
    -I/usr/local/include
    -o debug/qtscript_Global.o
    ../../generated_cpp/com_trolltech_qt_core/qtscript_Global.cpp

../../generated_cpp/com_trolltech_qt_core/qtscript_Global.cpp:6:25: error: qmetaobject.h: No such file or directory
../../generated_cpp/com_trolltech_qt_core/qtscript_Global.cpp:8:20: error: QVariant: No such file or directory
../../generated_cpp/com_trolltech_qt_core/qtscript_Global.cpp:10:21: error: qglobal.h: No such file or directory

The corresponding good release compile line is:

g++ -c -pipe -O2 -Wall -W -pthread -D_THREAD_SAFE -fPIC -DQCAD_DLL
    -DPTHREADS -DHAVE_PTHREAD_H -DQT_NO_DEBUG -DQT_WEBKIT_LIB
    -DQT_SCRIPT_LIB -DQT_SCRIPTTOOLS_LIB -DQT_SVG_LIB -DQT_SQL_LIB
    -DQT_XMLPATTERNS_LIB -DQT_XML_LIB -DQT_OPENGL_LIB -DQT_GUI_LIB
    -DQT_NETWORK_LIB -DQT_CORE_LIB -DQT_SHARED
    -I/usr/local/share/qt4/mkspecs/freebsd-g++
    -I.
    -I/usr/local/include/qt4/QtHelp
    -I/usr/local/include/qt4/QtDesigner
    -I/usr/local/include/qt4/QtUiTools
    -I/usr/local/include/qt4/QtCore
    -I/usr/local/include/qt4/QtNetwork
    -I/usr/local/include/qt4/QtGui
    -I/usr/local/include/qt4/QtOpenGL
    -I/usr/local/include/qt4/QtXml
    -I/usr/local/include/qt4/QtXmlPatterns
    -I/usr/local/include/qt4/QtSql
    -I/usr/local/include/qt4/QtSvg
    -I/usr/local/include/qt4/QtScriptTools
    -I/usr/local/include/qt4/QtScript
    -I/usr/local/include/qt4/QtWebKit
    -I/usr/local/include/qt4
    -I../../../../core
    -I../../../../core/math
    -I../../../../grid
    -I../../../../snap
    -I../../../../gui
    -I../../../../entity
    -I../../../../operations
    -I../../../../scripting
    -I../../../../stemmer
    -I../../../../scripting/ecmaapi
    -I../../../../scripting/ecmaapi/generated
    -I../../../../scripting/ecmaapi/adapters
    -I../../../../io/dxf
    -I../../../../spatialindex
    -I../../../../3rdparty
    -I.
    -I/usr/local/include
    -Irelease
    -I/usr/local/include/qt4
    -I/usr/local/include
    -o release/qtscript_Global.o
    ../../generated_cpp/com_trolltech_qt_core/qtscript_Global.cpp

The diff between them is suspiciously big (unique items only):

includes-wise:

--- incbad      2013-10-20 11:45:37.302711863 +0200
+++ incgood     2013-10-20 11:45:31.899711566 +0200
@@ -16,7 +16,19 @@
 -I../../../../stemmer
 -I/usr/local/include
 -I/usr/local/include/qt4
+-I/usr/local/include/qt4/QtCore
 -I/usr/local/include/qt4/QtDesigner
+-I/usr/local/include/qt4/QtGui
 -I/usr/local/include/qt4/QtHelp
+-I/usr/local/include/qt4/QtNetwork
+-I/usr/local/include/qt4/QtOpenGL
+-I/usr/local/include/qt4/QtScript
+-I/usr/local/include/qt4/QtScriptTools
+-I/usr/local/include/qt4/QtSql
+-I/usr/local/include/qt4/QtSvg
 -I/usr/local/include/qt4/QtUiTools
+-I/usr/local/include/qt4/QtWebKit
+-I/usr/local/include/qt4/QtXml
+-I/usr/local/include/qt4/QtXmlPatterns
 -I/usr/local/share/qt4/mkspecs/freebsd-g++
+-Irelease

Defines-wise:

--- defbad      2013-10-20 11:51:22.405716122 +0200
+++ defgood     2013-10-20 11:52:05.362693170 +0200
@@ -1,3 +1,17 @@
 -DHAVE_PTHREAD_H
 -DPTHREADS
 -DQCAD_DLL
+-DQT_CORE_LIB
+-DQT_GUI_LIB
+-DQT_NETWORK_LIB
+-DQT_NO_DEBUG
+-DQT_OPENGL_LIB
+-DQT_SCRIPTTOOLS_LIB
+-DQT_SCRIPT_LIB
+-DQT_SHARED
+-DQT_SQL_LIB
+-DQT_SVG_LIB
+-DQT_WEBKIT_LIB
+-DQT_XMLPATTERNS_LIB
+-DQT_XML_LIB
+-D_THREAD_SAFE

qmetaobject.h is on Qt/ and QtCore/, both of which is missing from the debug include path.

On Linux, it’s complaining about QtScript/QScriptExtensionPlugin (which is in /usr/include/qt4), diffs between includes:

--- incbad      2013-10-20 12:04:32.612710645 +0200
+++ incgood     2013-10-20 12:04:58.129714235 +0200
@@ -14,7 +14,21 @@
 -I../../../../snap
 -I../../../../spatialindex
 -I../../../../stemmer
+-I/usr/X11R6/include
+-I/usr/include/qt4
+-I/usr/include/qt4/QtCore
 -I/usr/include/qt4/QtDesigner
+-I/usr/include/qt4/QtGui
 -I/usr/include/qt4/QtHelp
+-I/usr/include/qt4/QtNetwork
+-I/usr/include/qt4/QtOpenGL
+-I/usr/include/qt4/QtScript
+-I/usr/include/qt4/QtScriptTools
+-I/usr/include/qt4/QtSql
+-I/usr/include/qt4/QtSvg
 -I/usr/include/qt4/QtUiTools
+-I/usr/include/qt4/QtWebKit
+-I/usr/include/qt4/QtXml
+-I/usr/include/qt4/QtXmlPatterns
 -I/usr/share/qt4/mkspecs/linux-g++-64
+-Irelease

Defines:

--- defbad      2013-10-20 12:07:16.961712166 +0200
+++ defgood     2013-10-20 12:07:38.831670098 +0200
@@ -1,4 +1,18 @@
 -DHAVE_PTHREAD_H
 -DPTHREADS
 -DQCAD_DLL
+-DQT_CORE_LIB
+-DQT_GUI_LIB
+-DQT_NETWORK_LIB
+-DQT_NO_DEBUG
+-DQT_OPENGL_LIB
+-DQT_SCRIPTTOOLS_LIB
+-DQT_SCRIPT_LIB
+-DQT_SHARED
+-DQT_SQL_LIB
+-DQT_SVG_LIB
 -DQT_WEBKIT
+-DQT_WEBKIT_LIB
+-DQT_XMLPATTERNS_LIB
+-DQT_XML_LIB
+-D_REENTRANT

I suspect it was a79f70, that’s when setCustomProperty() and friends changed.

(slightly reformatted so it doesn’t completely break preformatted display)

g++ -c -m64 -pipe -O2 -Wall -W -D_REENTRANT -DQT_WEBKIT -DQCAD_DLL -DPTHREADS
    -DHAVE_PTHREAD_H -DQT_NO_DEBUG -DQT_WEBKIT_LIB -DQT_SCRIPT_LIB
    -DQT_SCRIPTTOOLS_LIB -DQT_SVG_LIB -DQT_SQL_LIB -DQT_XMLPATTERNS_LIB -DQT_XML_LIB
    -DQT_OPENGL_LIB -DQT_GUI_LIB -DQT_NETWORK_LIB -DQT_CORE_LIB -DQT_SHARED
    -I/usr/share/qt4/mkspecs/linux-g++-64 -I. -I/usr/include/qt4/QtHelp
    -I/usr/include/qt4/QtDesigner -I/usr/include/qt4/QtUiTools
    -I/usr/include/qt4/QtCore -I/usr/include/qt4/QtNetwork -I/usr/include/qt4/QtGui
    -I/usr/include/qt4/QtOpenGL -I/usr/include/qt4/QtXml
    -I/usr/include/qt4/QtXmlPatterns -I/usr/include/qt4/QtSql
    -I/usr/include/qt4/QtSvg -I/usr/include/qt4/QtScriptTools
    -I/usr/include/qt4/QtScript -I/usr/include/qt4/QtWebKit -I/usr/include/qt4
    -I../../../src/core -I../../../src/core/math -I../../../src/grid
    -I../../../src/snap -I../../../src/gui -I../../../src/entity
    -I../../../src/operations -I../../../src/scripting -I../../../src/stemmer
    -I../../../src/scripting/ecmaapi -I../../../src/scripting/ecmaapi/generated
    -I../../../src/scripting/ecmaapi/adapters -I../../../src/io/dxf
    -I../../../src/spatialindex -I../../../src/3rdparty -I/usr/X11R6/include
    -Irelease -o release/MainWindow.o MainWindow.cpp
In file included from MainWindow.cpp:13:0:
../../../src/spatialindex/RSpatialIndexNavel.h: In member function ‘virtual void
    RSpatialIndexNavel::Visitor::visitData(std::vector<const SpatialIndex::IData*>&)’:
../../../src/spatialindex/RSpatialIndexNavel.h:191:77: warning: format ‘%Ld’
    expects argument of type ‘long long int’, but argument 2 has type
    ‘SpatialIndex::id_type {aka long int}’ [-Wformat]
MainWindow.cpp: In constructor ‘MainWindow::MainWindow(QWidget*)’:
MainWindow.cpp:79:54: error: no matching function for call to
    ‘RLineEntity::setCustomProperty(const char [14], int)’
MainWindow.cpp:79:54: note: candidate is:
../../../src/core/RObject.h:165:10: note:
    void RObject::setCustomProperty(const QString&, const QString&, const QVariant&)
../../../src/core/RObject.h:165:10: note:   candidate expects 3 arguments, 2 provided
MainWindow.cpp:80:66: error: no matching function for call to
    ‘RLineEntity::setCustomProperty(const char [17], const char [10])’
MainWindow.cpp:80:66: note: candidate is:
../../../src/core/RObject.h:165:10: note:
    void RObject::setCustomProperty(const QString&, const QString&, const QVariant&)
../../../src/core/RObject.h:165:10: note:   candidate expects 3 arguments, 2 provided
make[4]: *** [release/MainWindow.o] Error 1

Linux, Qt 4.8.1, GCC 4.6.4

It would be nice if we could use the Stretch tool also on blocks.

There are so many possibility’s to build a really good (time-saving) block-list /library with preassembled parts but if I have always to explode / ruin the block to use the stretch tool doesn’t helps the productivity.

As came up towards the end of  FS#892 .

What happens is a change in src/3rdparty/dxflib/src/dl_dxf.cpp does trigger a rebuild of release/libdxflib.a, but this does not, in turn, trigger a re-link of plugins/libqcaddxf.so, a (the?) consumer of libdxflib.a.

This causes partial rebuilds to be broken (I think I got puzzled by this earlier too).

How to reproduce: on a fully built tree, `touch src/3rdparty/dxflib/src/dl_dxf.cpp’ then `make’. Nothing but libdxflib.a gets rebuilt.

When we create a draft in version 2.2.2 we put dashed lines and the length is 0. We see a dash lines. (see attach file).

When we open the same draft in the version 3.3.2 we don’t see the dashed lines we see a continous lines.

If I changed in the dxf this lines DASHED 370 0 per DASHED 370 5 and changes the preference of the drawing it’s work.

But we don’t want to open all of 180000 drawings to change the length of the dashed lines from 0 to 5. We want that the version 3 print the dashed lines as the version 2.

Thanks

Eric

A window with a drawing that has unsafed changes gets an asterisk appended to its name in the tab title. When there too many tabs the title gets shortened and the star cannot be seen then.

It was nice, when the star remained visible, after or in front of the ellipses.

Also it was nice, when the “windows” menu would mark files with changes in the same way.

This seems to have somehow surfaced once again, this time on FreeBSD (that is, Linux builds are unaffected).

This diff, based on the suggestion from the forum post, seems to fix it:

--- a/src/3rdparty/qt-labs-qtscriptgenerator-4.8.5/qtbindings/qtbindings.pro
+++ b/src/3rdparty/qt-labs-qtscriptgenerator-4.8.5/qtbindings/qtbindings.pro
@@ -13,3 +13,7 @@ SUBDIRS = qtscript_core \
          qtscript_uitools
 
          #qs_eval
+
+freebsd* {
+       SUBDIRS -= qtscript_phonon
+}

Somehow (especially in light of the forum post) it doesn’t seem a very robust solution, though.

QCAD community, Git rev 3ff487, FreeBSD 9.2, Qt 4.8.5.

Apply a font name replacement table to a drawing. E.g. change font ‘romanc’ to ‘Arial’, etc.

The block hierarchy is lost after saving. Only one item appears in the block after reload, and it is doubled outside, too.

attached file:

Tried to put the four patches and the bar in a “Pzeile” block, this works, but after save and open, only the bar is in the block and is doubled to layer “0”.

The attached file is the destroyed version with removed doublette.

See:
http://www.qcad.org/rsforum/viewtopic.php?f=33&t=2595

XP/32

Info Tools - Area: “Add measurement to drawing” doesn’t work as expected

Projecting text into an isometric plane crashes QCAD - unless the text is first exploded, which is not always an acceptable appearance.

When opening or closing QCAD the command line isn’t staying active.

If the segment of an ellipse is limited by another ellipse with the same ratio and rotation, the break out tool cannot find the correct intersection to break out to.

See also:
http://www.qcad.org/rsforum/viewtopic.php?f=33&t=2586&view=unread#unread

Inserting part library items with attributes definitions can result in attributes being created with invalid layer IDs (from the source document but invalid in the target document).

Inserting a block with attributes into a drawing with a different unit results in wrongly scaled attribute entities.

Workaround:
Sync block attributes after insert.

Dear Sir:
I am an user in Taiwan. I need to use Chinese. While editing, the font appearance can be display in view port, but after saving-closing-re-opening, the font type is missing. For example, while I type Chinese with font type “B” or “C” and save the file, the font all become font type “A” with reopening and the “Main Font” selector inside “Text” dialog window is blank. I need to re-open “Text” dialogue window and re-select font type, then click the working view port, the display of text changes. Its inconvient. In previous version it’s ok. I can’t remember the exact version of well-saving-Chinese-font function because my computer is not in my house. It might be version 3.1.x. I don’t think it’s a major problem, but could you please deal with this problem while having time.

If the keep proportions button is checked in the scale dialog, no scaling is performed.

Port CXF font editor to QCAD 3.
- Load CXF fonts (load glyphs into blocks)
- Save back to CXF format

Break up spline at given point, break out segment, etc.

Title says it all. Cannot get QCAD to synchronise attributes on Fedora 19 box with QCAD 3.2.2 installed as the machine hangs (a single core of CPU maxed out). This is not present on my Windows machine (Windwows 7).

I can provide a sample file privately if needed.

Kurt

Once the print dialog is displayed, it will print even if I “cancel” or “x” the page.

This affects only certain KDE installations:
- Fedora 12 with KDE
- Kubuntu 13.04

Add matrix transformations to projection tools.

See also:
http://www.qcad.org/rsforum/viewtopic.php?f=32&t=2569

Insert block with attributes: third attribute input not working

Add an option to measuring tools to add results permanently to the drawing (on current layer, with current layer attributes).

Help - Reference Manual does not have any effect on Windows.

Somewhere in the 3.1 series the SVG exporter (both precise and PG) started to make lines that are zero (0) width in CAD 0.01 (@scale 1:1) points wide in exported files (0.1 @scale 1:10). I use such lines to constrain exported files to a common size and rather not have them show. In my view, this is a regression, as there is no reason for changing line weight, isnt it? If I wanted the lines to have weight, I’d draw them with a weight.

Texts in CAD fonts with characters for which no glyph is defined report a wrong bounding box which always includes the absolute zero point. This makes the spatial index ineffective for drawings containing many such texts (typically complex drawings from international users).

Inverting the current selection is significantly slower than select all, deselect all or other selection tools.

SVG import fails if no resolution is specified for SVGImporter.