App/Gui: Refactor expression completer and unit expressions #3062

realthunder · 2020-02-16T10:21:27Z

Forum thread https://forum.freecadweb.org/viewtopic.php?f=17&t=43412

The refactor are mostly in Gui/ExpressionCompleter.cpp file. The misleadingly huge change set is due to some minor change in expression scanner and parser source file, which resulted huge changes in generated code.

vocx-fc · 2020-02-17T05:53:21Z

It seems the Travis tests fail because of the Spreadsheet unit tests. Maybe you can re-write these as well, or comment them temporarily, so that Travis finishes testing the rest of the code.

realthunder · 2020-02-17T07:17:40Z

That's strange. I ran the test several times to make sure it'll pass unchanged. I'll investigate.

realthunder · 2020-02-17T10:22:42Z

@vocx-fc Can you try build my branch and see if you have problem with the Spreadsheet test?

@wwmayer Can you shed some light on this? Anything special about the running test in Travis? I have built Python2 and 3 on both Linux and Window in three computer setup, all of them can pass the spreadsheet test.

wwmayer · 2020-02-17T12:24:51Z

Have you tried to run the command line version? When only testing with the GUI version then additional modules are loaded and things apparently seem to work while in cmd line version they fail.

realthunder · 2020-02-17T12:30:38Z

Yes, I tried command line version on all of my computers.

vocx-fc · 2020-02-17T15:47:03Z

@realthunder

I fail to compile your branch. It errors very soon.

[  5%] Building CXX object src/Base/CMakeFiles/FreeCADBase.dir/QuantityPyImp.cpp.o
In file included from /opt/freecad-source-vocx/src/Base/Quantity.h:27:0,
                 from /opt/freecad-source-vocx/src/Base/QuantityPyImp.cpp:25:
/opt/freecad-source-vocx/src/Base/Unit.h:98:23: error: ‘vector’ in namespace ‘std’ does not name a template type
     static const std::vector<std::pair<Unit,const char*> > &unitTypes();
                       ^~~~~~
In file included from /opt/freecad-source-vocx/src/Base/QuantityPyImp.cpp:25:0:
/opt/freecad-source-vocx/src/Base/Quantity.h:160:23: error: ‘vector’ in namespace ‘std’ does not name a template type
     static const std::vector<UnitInfo> &unitInfo();
                       ^~~~~~

I see your branch is a bit old, based on 16b2b9a. I tried to rebase it onto master but I get conflicts.

CONFLICT (content): Merge conflict in src/Gui/ExpressionCompleter.h
Auto-merging src/Gui/ExpressionCompleter.cpp
error: Failed to merge in the changes.
Patch failed at 0004 Gui: refactor ExpressionCompleter

realthunder · 2020-02-17T21:53:19Z

I fail to compile your branch. It errors very soon.

Another strange issue, maybe they are related. How did you clone my branch? My initial submit of this PR had this build error. I force pushed the fix along with an upstream rebase. Maybe travis got stalled cache or something because of the force push? But then again, how did you get the non-existent revision?

Please check the header file src/Base/Unit.h. The fixed version shall include additional header .

vocx-fc · 2020-02-17T22:15:10Z

I git add remote your branch as a remote in my system, and then git checkouted your branch. I have now git fetched again your repository, and it now seems the branch is updated (forced push). Now it seems to be based on a76bc63. I will try compiling again.

realthunder · 2020-02-17T22:32:31Z

Hmm... maybe it's github getting some cloud syncing problem. I don't know what did you see, but on my computer viewing this very PR, it shows my force push at 22 hours ago, and your first comment 17 hours ago. So, by right, you shouldn't be able to get the old revision, unless you have cloned my branch very soon after my PR submission?

vocx-fc · 2020-02-17T22:36:55Z

I think what happened is that I added your repository earlier, around the time you made the pull request, but I had not git fetched the latest changes. So, I had fetched the reference to an old version of your branch.

vocx-fc · 2020-02-18T00:09:49Z

I compiled successfully, opened FreeCAD, tested the Spreadsheet, and everything worked fine.

I ran the unit tests of Spreadsheet and these are some of the messages that I get. They appear on the terminal, but they don't make the tests fail.

bin/FreeCAD -t TestSpreadsheet

testAggregates (TestSpreadsheet.SpreadsheetCases)
<Spreadsheet> Cell.cpp(592): Unnamed#Spreadsheet001.H6: Quantity::operator +=(): Unit mismatch in plus operation
<Spreadsheet> Cell.cpp(592): Unnamed#Spreadsheet001.H5: Quantity::operator -(): Unit mismatch in minus operation
<Spreadsheet> Cell.cpp(592): Unnamed#Spreadsheet001.H3: Quantity::operator >(): quantities need to have same unit to compare
<Spreadsheet> Cell.cpp(592): Unnamed#Spreadsheet001.H2: Quantity::operator <(): quantities need to have same unit to compare
<Spreadsheet> Cell.cpp(592): Unnamed#Spreadsheet001.H1: Quantity::operator +=(): Unit mismatch in plus operation
<Spreadsheet> Cell.cpp(592): Unnamed#Spreadsheet001.D1: Invalid number of entries: at least two required.
1e-07 <App> Document.cpp(3464): Failed to recompute Unnamed#Spreadsheet001: One or more cells failed contains errors.
Recompute failed! Please check report view.
Test all aggregate functions ... ok
testAlias (TestSpreadsheet.SpreadsheetCases)
Playing with aliases ... ok
testAmbiguousAlias (TestSpreadsheet.SpreadsheetCases)
Try to set the same alias twice (bug #2402) ... ok
testClearAlias (TestSpreadsheet.SpreadsheetCases)
This was causing a crash ... ok
testCrossDocumentLinks (TestSpreadsheet.SpreadsheetCases)
Expressions across files are not saved (bug #2442) ... ok
testFunctions (TestSpreadsheet.SpreadsheetCases)
<Spreadsheet> Cell.cpp(592): Unnamed#Spreadsheet.C27: Units must be equal.
cath(7.0710678118654764mm; 5mm; 4)
<Spreadsheet> Cell.cpp(592): Unnamed#Spreadsheet.C26: Units must be equal.
cath(5mm; 3)
<Spreadsheet> Cell.cpp(592): Unnamed#Spreadsheet.C24: Units must be equal.
hypot(3mm; 4; 5)
<Spreadsheet> Cell.cpp(592): Unnamed#Spreadsheet.C23: Units must be equal.
hypot(3mm; 4)
<Spreadsheet> Cell.cpp(592): Unnamed#Spreadsheet.D21: Exponent is not allowed to have a unit.
pow(7mm; 4mm)
<Spreadsheet> Cell.cpp(592): Unnamed#Spreadsheet.C20: Units must be equal.
atan2(3mm; 3)
<Spreadsheet> Cell.cpp(592): Unnamed#Spreadsheet.B17: Unit must be empty.
tanh(0.5mm)
<Spreadsheet> Cell.cpp(592): Unnamed#Spreadsheet.B16: Unit must be empty.
cosh(0.5mm)
<Spreadsheet> Cell.cpp(592): Unnamed#Spreadsheet.B15: Unit must be empty.
sinh(0.5mm)
<Spreadsheet> Cell.cpp(592): Unnamed#Spreadsheet.B14: Unit must be empty.
atan(0.5mm)
<Spreadsheet> Cell.cpp(592): Unnamed#Spreadsheet.B13: Unit must be empty.
acos(0.5mm)
<Spreadsheet> Cell.cpp(592): Unnamed#Spreadsheet.B12: Unit must be empty.
asin(0.5mm)
<Spreadsheet> Cell.cpp(592): Unnamed#Spreadsheet.C7: Unit must be empty.
log10(-3mm)
<Spreadsheet> Cell.cpp(592): Unnamed#Spreadsheet.C6: Unit must be empty.
log(-3mm)
<Spreadsheet> Cell.cpp(592): Unnamed#Spreadsheet.C5: Unit must be empty.
exp(-3mm)
0.431293 <App> Document.cpp(3464): Failed to recompute Unnamed#Spreadsheet: One or more cells failed contains errors.
Recompute failed! Please check report view.
Test all built-in simple functions ... ok
testInsertRows (TestSpreadsheet.SpreadsheetCases)
Inserting rows -- check renaming of internal cells ... ok
testInvoluteGear (TestSpreadsheet.SpreadsheetCases)
Support of boolean or integer values ... ok
testIssue3128 (TestSpreadsheet.SpreadsheetCases)
Regression test for issue 3128; mod should work with arbitrary units for both arguments ... ok
testIssue3225 (TestSpreadsheet.SpreadsheetCases)
Inserting rows -- check renaming of internal cells ... ok
testIssue3363 (TestSpreadsheet.SpreadsheetCases)
Regression test for issue 3363; Nested conditionals statement fails with additional conditional statement in false-branch ... ok
testIssue3432 (TestSpreadsheet.SpreadsheetCases)
Regression test for issue 3432; numbers with units are ignored from aggregates ... ok
testMatrix (TestSpreadsheet.SpreadsheetCases)
<Spreadsheet> Cell.cpp(592): Unnamed#Spreadsheet.H2: Cannot invert singular matrix
G2 ^ -1
0.829919 <App> Document.cpp(3464): Failed to recompute Unnamed#Spreadsheet: One or more cells failed contains errors.
Recompute failed! Please check report view.
Test Matrix/Vector/Placement/Rotation operations ... ok
testNumbers (TestSpreadsheet.SpreadsheetCases)
Test different numbers ... ok
testPlacementName (TestSpreadsheet.SpreadsheetCases)
Object name is equal to property name (bug #2389) ... ok
testPrecedence (TestSpreadsheet.SpreadsheetCases)
Precedence -- test precedence for relational operators and conditional operator. ... ok
testRelationalOperators (TestSpreadsheet.SpreadsheetCases)
Test relational operators ... ok
testRemoveRows (TestSpreadsheet.SpreadsheetCases)
Removing rows -- check renaming of internal cells ... ok
testRenameAlias (TestSpreadsheet.SpreadsheetCases)
Test renaming of alias1 to alias2 in a spreadsheet ... ok
testRenameAlias2 (TestSpreadsheet.SpreadsheetCases)
Test renaming of alias1 to alias2 in a spreadsheet, when referenced from another object ... ok
testRenameAlias3 (TestSpreadsheet.SpreadsheetCases)
Test renaming of document object referenced from another object ... ok
testSetInvalidAlias (TestSpreadsheet.SpreadsheetCases)
Try to use a cell address as alias name ... ok
testSetInvalidAlias2 (TestSpreadsheet.SpreadsheetCases)
Try to use a unit (reserved word) as alias name ... ok
testSketcher (TestSpreadsheet.SpreadsheetCases)
Mixup of Label and Name (bug #2407) ... ok
testUnits (TestSpreadsheet.SpreadsheetCases)
Units -- test unit calculations. ... ok

----------------------------------------------------------------------
Ran 25 tests in 1.545s

OK
System exit

chennes

First, thanks and kudos to @realthunder for writing this -- I love the functionality, it makes expressions much faster to write and much less error prone. I also appreciate the "discoverability" that this feature adds to FreeCAD, allowing me to easily see what properties are available where. I've added a few very minor comments within the code. I cannot pretend to understand all of what I am reading here, I am still learning the FreeCAD codebase, and this is a pretty daunting change to look through! So my review is mostly that it compiles and runs, and works as intended, and I don't see anything too horrifying in the code. Well, seeing a union in modern C++ scares me a little... :) I know, that was there before...

chennes · 2020-11-13T23:52:10Z

src/App/Property.cpp

@@ -130,7 +130,8 @@ const boost::any Property::getPathValue(const ObjectIdentifier &path) const

 void Property::getPaths(std::vector<ObjectIdentifier> &paths) const
 {
-    paths.emplace_back(getContainer(), getName());
+    (void)paths;


A comment about why getPaths does nothing would be nice - I don't understand why it's been disabled.

The reason is mentioned in 3b5388d commit message.

Hah, I was about to repeat myself here! Can you please add a short comment in the source code, so readers don't have to dig through git blame?

src/Mod/Spreadsheet/App/PropertySheet.cpp

chennes · 2020-11-14T00:19:55Z

src/Gui/ExpressionCompleter.cpp

+                if(role == Qt::EditRole)
+                    return QString::fromLatin1(".%1.").arg(
+                        QString::fromLatin1(sobj->getNameInDocument()));
+                else if(role == Qt::UserRole && obj->getPropertyByName(sobj->getNameInDocument())) {


I think it might be a good idea to give Qt::UserRole (and Qt::UserRole+1, which I think is also used later on) more descriptive names within this file, either via a locally-defined enumeration, or just some const ints.

Fair point. I'll change that later

ezzieyguywuf

Yikes! I started this review but never hit submit (months and months ago).

I'm going to go ahead and submit this - this review is incomplete but hopefully it can begin some good conversations

ezzieyguywuf · 2020-08-27T21:28:51Z

src/App/Property.cpp

 void Property::getPaths(std::vector<ObjectIdentifier> &paths) const
 {
-    paths.emplace_back(getContainer(), getName());
+    (void)paths;
+    // paths.emplace_back(getContainer(), getName());
 }


It looks like this function is not being used any more - why not just remove altogether?

This function is still used by some property to expose its sub-value to be accessed by the expression only. The property can choose to expose sub-value either through Python, which is also accessible by expression.

ezzieyguywuf · 2020-08-27T21:34:42Z

src/App/PropertyGeo.cpp

+
+    // Unlike the above path (which provides units that are not available
+    // through python, the following paths provides the same value. They are no
+    // longer needed, because the expression completer will now dig into all
+    // python attributes.
+
+    // paths.push_back(ObjectIdentifier(*this)
+    //                 << ObjectIdentifier::SimpleComponent(ObjectIdentifier::String("Rotation"))
+    //                 << ObjectIdentifier::SimpleComponent(ObjectIdentifier::String("Axis"))
+    //                 << ObjectIdentifier::SimpleComponent(ObjectIdentifier::String("x")));
+    // paths.push_back(ObjectIdentifier(*this)
+    //                 << ObjectIdentifier::SimpleComponent(ObjectIdentifier::String("Rotation"))
+    //                 << ObjectIdentifier::SimpleComponent(ObjectIdentifier::String("Axis"))
+    //                 << ObjectIdentifier::SimpleComponent(ObjectIdentifier::String("y")));
+    // paths.push_back(ObjectIdentifier(*this)
+    //                 << ObjectIdentifier::SimpleComponent(ObjectIdentifier::String("Rotation"))
+    //                 << ObjectIdentifier::SimpleComponent(ObjectIdentifier::String("Axis"))
+    //                 << ObjectIdentifier::SimpleComponent(ObjectIdentifier::String("z")));


Rather than commenting this out, it should be deleted altogether.

Then, this could be a standalone commit, explaining in the commit message what you have here in your comments.

The commit 3b5388d that did the change mention the reason in its commit message. The modification here just provides some more detail with context.

ezzieyguywuf · 2020-08-27T21:40:56Z

src/App/ObjectIdentifier.h

@@ -69,6 +69,7 @@ AppExport std::string quote(const std::string &input, bool toPython=false);
    _t(_t &&other) { *this = std::move(other); }\
    _t &operator=(_t &&other)

+


Why is this blank line added?

ezzieyguywuf · 2020-08-27T22:02:05Z

src/App/ObjectIdentifier.h

@@ -174,10 +183,13 @@ class AppExport ObjectIdentifier {
                int begin=INT_MAX, int end=INT_MAX, int step=1);

        static Component SimpleComponent(const char * _component);
-
-        static Component SimpleComponent(const String & _component);
+        static Component SimpleComponent(const String &_component); 


There is a trailing space on this line - it should be deleted.

static Component SimpleComponent(const String &_component); //<- the trailing space is here after the ;

ezzieyguywuf

I'll keep submitting my reviews as I finish them. Please let me know if you'd prefer for me to wait until I'm done with everything before submitting.

ezzieyguywuf · 2020-11-18T21:51:10Z

src/App/DynamicProperty.h

@@ -75,6 +75,8 @@ class AppExport DynamicProperty
    //@{
    /// Get all properties of the class (including parent)
    void getPropertyList(std::vector<Property*> &List) const;
+    /// get all properties with their names
+    void getPropertyNamedList(std::vector<std::pair<const char*,Property*> > &List) const;


Doesn't the getPropertyMap contain all this same information?

Why is it necessary to have this info in a std::vector<std::pair> rather than the std::map that is already provided?

For example, I see this is used in ExpressionParser.cpp but couldn't that range-based for-loop loop over the key-value pairs in the std::map that is already provided?

Commit bdc1c835 introduced this API and mentioned the usage in its commit message. It is for property aliases. Originally PropertySheet just add a real duplicated property with a different name as alias, which is a waste of resources.

ezzieyguywuf · 2020-11-18T23:54:08Z

src/App/ExpressionParser.h

@@ -71,7 +71,10 @@ struct AppExport Expression::Component {
 class  AppExport UnitExpression : public Expression {
    TYPESYSTEM_HEADER_WITH_OVERRIDE();
 public:
-    UnitExpression(const App::DocumentObject *_owner = 0, const Base::Quantity & _quantity = Base::Quantity(), const std::string & _unitStr = std::string());
+
+    static UnitExpression *create(const App::DocumentObject *_owner, const char *unit);


This static method seems unnecessary. I can see that it is being used in the parser (yacc), but it is simply a workaround due to a misuse of Flex/Yacc.

There is a very nice overview here, but to summarize, the lexer should parse an arbitrary string into tokens.

In this case, the lexer should determine whether or not a given string is a valid UNIT.

Next, the lexer can assume that any UNIT it recieves is valid, since the lexer took care of the grammar. The job of the lexer is to ensure that the tokens created by the lexer are put together in a valid order.

Commit d1e6fe6 introduced this change and explained the reason behind it. When new units are introduced (which happens quite often), the name may clash with existing property names or aliases. It is possible to disambiguate the meaning of the token under the usage context. However, the lexer has no concept of context, which is why it is not suitable for interpreting unit here.

ezzieyguywuf · 2020-11-19T00:01:59Z

src/App/ExpressionParser.h

@@ -101,7 +106,7 @@ class  AppExport UnitExpression : public Expression {

 private:
    Base::Quantity quantity;
-    std::string unitStr; /**< The unit string from the original parsed string */
+    const char *unitStr; /**< The unit string from the original parsed string */


Why has this type been changed?

I can see that the UnitExpression constructor has been updated such that it takes 2 rather than 3 arguments, with unitstr defaulting to 0 in a protected constructor when needed but...why?

What is the purpose of this change? It seems to obfuscate and make things more difficult than they need to be.

The purpose of commit d1e6fe6 is to concentrate all unit string in one place as constant string, for both parsing and printing. Any unknown unit will be interpreted as a VariableExpression. So there is little reason to use std::string here.

realthunder · 2020-11-19T05:20:39Z

@ezzieyguywuf Please continue. I'll reply your comments later.

ezzieyguywuf · 2020-12-14T12:03:53Z

@realthunder saw your replies to my comments, thank you for taking the time to read through them. I will find some time to review your replies and provide some responses of my own, as well as to continue the code review (i had stopped earlier to give you a chance to catch up)

berndhahnebach · 2022-07-05T19:40:30Z

fixed ... FreeCAD/FreeCAD-CI@a11fb14

and improved .... FreeCAD/FreeCAD-CI@fb0f2a9

berndhahnebach · 2022-07-05T20:22:31Z

for feature branch PR_3062. Pipeline 574892909 was triggered at ad9cf27. All CI branches and pipelines.

gabrielgrant · 2022-09-13T00:15:12Z

this has been pending for quite a while. what's needed to move this forward (other than fixing some merge conflicts that have accumulated during the delay, of course)?

realthunder · 2022-11-03T22:28:52Z

PR suspended for now, and will resume after topo naming merge.

adrianinsaval · 2022-11-23T14:13:10Z

i know this is on hold but for the record I get an error when compiling:

FAILED: src/Mod/Part/App/CMakeFiles/Part.dir/ShapeUpgrade/UnifySameDomainPyImp.cpp.o 
/usr/bin/ccache /usr/lib/ccache/bin/c++ -DBOOST_ATOMIC_DYN_LINK -DBOOST_ATOMIC_NO_LIB -DBOOST_DATE_TIME_DYN_LINK -DBOOST_DATE_TIME_NO_LIB -DBOOST_FILESYSTEM_DYN_LINK -DBOOST_FILESYSTEM_NO_LIB -DBOOST_PROGRAM_OPTIONS_DYN_LINK -DBOOST_PROGRAM_OPTIONS_NO_LIB -DBOOST_REGEX_DYN_LINK -DBOOST_REGEX_NO_LIB -DBOOST_SYSTEM_DYN_LINK -DBOOST_SYSTEM_NO_LIB -DBOOST_THREAD_DYN_LINK -DBOOST_THREAD_NO_LIB -DFCUseFreeType -DHAVE_CONFIG_H -DHAVE_LIMITS_H -DPart_EXPORTS -DQT_CORE_LIB -DQT_NO_DEBUG -DQT_XML_LIB -D_OCC64 -I/fakepath/src/FreeCAD/build/src/Mod/Part/App/Part_autogen/include -I/fakepath/src/FreeCAD/build -I/fakepath/src/FreeCAD/build/src -I/fakepath/src/FreeCAD/src -I/fakepath/src/FreeCAD/build/src/Mod/Part/App -I/fakepath/src/FreeCAD/src/Mod/Part/App -I/usr/include/opencascade -I/usr/include/python3.10 -I/usr/include/freetype2 -isystem /usr/include/qt -isystem /usr/include/qt/QtCore -isystem /usr/lib/qt/mkspecs/linux-g++ -isystem /usr/include/qt/QtXml -Wall -Wextra -Wno-write-strings -march=x86-64 -mtune=generic -O2 -pipe -fno-plt -fexceptions         -Wp,-D_FORTIFY_SOURCE=2 -Wformat -Werror=format-security         -fstack-clash-protection -fcf-protection -Wp,-D_GLIBCXX_ASSERTIONS -fPIC -w -fPIC -fPIC -std=gnu++17 -MD -MT src/Mod/Part/App/CMakeFiles/Part.dir/ShapeUpgrade/UnifySameDomainPyImp.cpp.o -MF src/Mod/Part/App/CMakeFiles/Part.dir/ShapeUpgrade/UnifySameDomainPyImp.cpp.o.d -o src/Mod/Part/App/CMakeFiles/Part.dir/ShapeUpgrade/UnifySameDomainPyImp.cpp.o -c /fakepath/src/FreeCAD/src/Mod/Part/App/ShapeUpgrade/UnifySameDomainPyImp.cpp
/fakepath/src/FreeCAD/src/Mod/Part/App/ShapeUpgrade/UnifySameDomainPyImp.cpp: In member function ‘virtual int Part::UnifySameDomainPy::PyInit(PyObject*, PyObject*)’:
/fakepath/src/FreeCAD/src/Mod/Part/App/ShapeUpgrade/UnifySameDomainPyImp.cpp:64:14: error: ‘unique_ptr’ is not a member of ‘std’
   64 |         std::unique_ptr<ShapeUpgrade_UnifySameDomain> ptr(new ShapeUpgrade_UnifySameDomain(shp,
      |              ^~~~~~~~~~
/fakepath/src/FreeCAD/src/Mod/Part/App/ShapeUpgrade/UnifySameDomainPyImp.cpp:37:1: note: ‘std::unique_ptr’ is defined in header ‘<memory>’; did you forget to ‘#include <memory>’?
   36 | #include "PartPyCXX.h"
  +++ |+#include <memory>
   37 | 
/fakepath/src/FreeCAD/src/Mod/Part/App/ShapeUpgrade/UnifySameDomainPyImp.cpp:64:53: error: expected primary-expression before ‘>’ token
   64 |         std::unique_ptr<ShapeUpgrade_UnifySameDomain> ptr(new ShapeUpgrade_UnifySameDomain(shp,
      |                                                     ^
/fakepath/src/FreeCAD/src/Mod/Part/App/ShapeUpgrade/UnifySameDomainPyImp.cpp:64:55: error: ‘ptr’ was not declared in this scope
   64 |         std::unique_ptr<ShapeUpgrade_UnifySameDomain> ptr(new ShapeUpgrade_UnifySameDomain(shp,
      |                                                       ^~~

Adding #include <memory> as suggested fixes it.

adrianinsaval · 2022-12-03T00:16:14Z

@realthunder whenever you come back to this please take into account the feature introduced with #7688 and adapt it to use the new syntax to reference constraints that have a name that isn't a valid identifier, if I'm not mistaken it is something like .Constraints[<<name string>>]

maxwxyz · 2024-04-27T06:32:26Z

Could this also be resumed now, after the recent toponaming merges?

realthunder force-pushed the ExprCompleter branch from 996ed50 to 64fdd50 Compare February 17, 2020 00:13

realthunder force-pushed the ExprCompleter branch 3 times, most recently from 4689686 to f7f84c0 Compare February 27, 2020 03:44

realthunder force-pushed the ExprCompleter branch from e51749f to 648f1b4 Compare April 14, 2020 23:35

realthunder force-pushed the ExprCompleter branch from 648f1b4 to 04bd777 Compare May 22, 2020 10:09

realthunder force-pushed the ExprCompleter branch 3 times, most recently from 2886179 to 39cd5c1 Compare June 16, 2020 00:58

chennes previously approved these changes Nov 14, 2020

View reviewed changes

ezzieyguywuf suggested changes Nov 18, 2020

View reviewed changes

ezzieyguywuf reviewed Nov 19, 2020

View reviewed changes

wwmayer added the For 0.20 label Feb 9, 2021

berndhahnebach added Core Issue or PR touches core sections (App, Gui, Base) of FreeCAD and removed For 0.20 labels Sep 17, 2021

realthunder force-pushed the ExprCompleter branch from a1c4e0c to de1ef61 Compare October 4, 2021 23:00

FreeCAD deleted a comment from freecadci Oct 5, 2021

FreeCAD deleted a comment from freecadci Jul 5, 2022

realthunder mentioned this pull request Jul 13, 2022

Remove deprecated href #7172

Closed

7 tasks

luzpaz added ✋ On hold This PR must not be merged before some condition is met WB Spreadsheet Related to the Spreadsheet Workbench Expressions labels Nov 8, 2022

luzpaz marked this pull request as draft November 8, 2022 14:15

adrianinsaval mentioned this pull request Nov 23, 2022

CarbonCopy should use names of constraints if available. #5965

Closed

adrianinsaval mentioned this pull request Dec 14, 2022

[GUI] Add missing include #8045

Merged

This comment was marked as off-topic.

Sign in to view

		@@ -69,6 +69,7 @@ AppExport std::string quote(const std::string &input, bool toPython=false);
		_t(_t &&other) { *this = std::move(other); }\
		_t &operator=(_t &&other)

App/Gui: Refactor expression completer and unit expressions #3062

Are you sure you want to change the base?

App/Gui: Refactor expression completer and unit expressions #3062

Conversation

realthunder commented Feb 16, 2020 • edited

vocx-fc commented Feb 17, 2020

realthunder commented Feb 17, 2020

realthunder commented Feb 17, 2020

wwmayer commented Feb 17, 2020

realthunder commented Feb 17, 2020

vocx-fc commented Feb 17, 2020

realthunder commented Feb 17, 2020

vocx-fc commented Feb 17, 2020

realthunder commented Feb 17, 2020 • edited

vocx-fc commented Feb 17, 2020

vocx-fc commented Feb 18, 2020

chennes left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

ezzieyguywuf left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

ezzieyguywuf left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

realthunder Dec 14, 2020 • edited

Choose a reason for hiding this comment

Choose a reason for hiding this comment

realthunder Dec 14, 2020 • edited

Choose a reason for hiding this comment

realthunder commented Nov 19, 2020

ezzieyguywuf commented Dec 14, 2020

berndhahnebach commented Jul 5, 2022 • edited

berndhahnebach commented Jul 5, 2022

gabrielgrant commented Sep 13, 2022 • edited

realthunder commented Nov 3, 2022

adrianinsaval commented Nov 23, 2022 • edited

adrianinsaval commented Dec 3, 2022

This comment was marked as off-topic.

This comment was marked as off-topic.

maxwxyz commented Apr 27, 2024

realthunder commented Feb 16, 2020 •

edited

realthunder commented Feb 17, 2020 •

edited

realthunder Dec 14, 2020 •

edited

realthunder Dec 14, 2020 •

edited

berndhahnebach commented Jul 5, 2022 •

edited

gabrielgrant commented Sep 13, 2022 •

edited

adrianinsaval commented Nov 23, 2022 •

edited