Roger Cigol

...Also: which compiler are you using. C++ "classic" is the 32 bit compiler developed by Borland but basically frozen quite some number of years ago. The recommended compilers are Clang32 for 32 bit targets and Clang64 for 64bit targets. 12.1 also introduced a new tool chain for clang 64bit which is expected to replace the earlier 64bit system (this is currently called clang64 modern) but in 12.1 it has some minor limitations which may make it not your first choice if you are still finding your way around. These limitations are understood by Embarcadero and are expected to be removed in next version or two with the goal being to replace the clang64 approach with this "clang 64 modern" approach. You select the compiler used for 32 bit applications using the menu Project | Options and then under Builiding | C++ compiler there is an entry "Use classic borland compiler" (I suggest you set this to "false"). You select the compiler used for 64 bit applications using the "target platform" in the project tree.

Hi Newbee, I wonder what version you are using here? Embarcadero have tried various experiments with the auto completion type editing functionality in recent editions. It is only in 12.1 - with reasonable integration of the visual assists components from Whole Tomato that things are working well. I find 12.1 auto completion fast and reliable.

Me also - and I use the C++ side a lot and that too is also good.

This is true for design time connections. But you can compile to 64 bit application and then connect at run time (but if you do this you CANNOT connect at design time because of the limitation I mention above).

You can only open MS Access files on a PC using either 64 bit apps (in which case you need to install the 64 bit MS Access drivers into Windows) or 32 bit apps (in which case you need MS Access on the machine (or at least the drivers). You cannot have both on the machine at the same time.

Everyone accepts that it is correct that you need elevated privileges to install. "There is no difference between them" means for me either route has always led to being able to run the IDE from a "standard user" account without elevated privileges.

I've used ISO and web based installers for different installations - no difference between them.

I've been using RAD Studio since 2009 (ie pre "XE" days") and never had to use elevated privileges to run the IDE. 12.1 exactly the same..... I second this.

I have a short sample XML file which is as follows <?xml version="1.0"?> <authors> <author> <name>abc</name> <posts> <items> <title lang="eng1">abc item 1</title> <content lang="eng2">abc item 1</content> </items> <items> <title lang="eng3">abc item 2</title> <content lang="eng4">abc item 2</content> </items> <items> <title lang="eng5">abc item 3</title> <content>abc item 3</content> </items> <items> <title>abc item 4</title> <content lang="eng6">abc item 4</content> </items> </posts> </author> </authors> I need a schema (ie a file *.xsd format) for this xml format. Schema creation tools I have tried don't seem to be able to handle the fact that nodes such as <title>...</title> contain both an attribute and text data. If I use a tool to create a schema *.xsd and then recreate a sample xml from the created schema I always find that the resulting xml created is missing either the attribute or the data, meaning the schema I have generated does NOT match my XML file. Can anyone create a schema that does correctly map to an XML file with nodes with both attribute and text data? I know it's not a delphi question (or a C++ Builder question) but I also know there are some XML gurus on Delphi Praxis that might be able to solve my problem very easily ! Thanks in anticipation

Wow! super fast response. I wasn't aware of that useful site - so thanks for the link. I will try this xsd and see how I get on.... Thank you Lars.

Exactly, I've checked my tool box, I don't have a BMW ZGW device or a way of simulating one........ I would also need someone to tell me what it was supposed to do in the event of different failures when communicating with the BMW ZGW device.....

I have said before: You can tell a good software engineer if the first question they ask is "how am I going to test this?". If I were to translate your code into Delphi how would I test it to know that it works ?

I agree - the key thing is that a text file (including a unicode text file) does not contain information about the display of the characters so any attempt at alignment can never be guaranteed to have success. Thinking out of the box - what about the plain old TAB character? using this as a separator between your columns might result in any reasonable file viewer lining them up for you. Always good to try a simple solution first.....

just add a bool variable to keep track of if mp1MyType is valid or not ? and when not valid (equivalent to your old NULL value) simply ignore the value of mp1MyType ? seems simple enough to fix the code snippet you posted....

One way is to add two dummy series, one with Y values that are at the smallest maximum value you want to allow and one with Y values at the largest minimum value you want to allow. Then set the colour of the two dummy lines equal to the background colour (so they aren't visible to the user). Not the most elegant solution but a devious workable one.....

How about an example of how you have to pick your "epsilon" value with knowledge of what's going on: If I am working out shortest route to walk from London to Birmingham I might use floating point values to store route distance in km (roughly 175 km). I might consider any route that is within 1km of another route as being "equal" from the perspective of route assessment. So I can set a large epsilon when making my route comparisons. But if I am launching a module and trying to land it on a flat bit of the moon my route is roughly 400,000km but I want the landing point to be accurate to a couple of meters so I would need to choose a much tighter epsilon. David is right - offering a "hidden" arbitrary epsilon value that works for "most cases" is a route that is often convenient but it is a potentially dangerous approach. Floating point values are the right approach for many coding problems but they do need using with an understanding of how they work and the inherent "limitations" that result from how they work.

Your file refers to a custom schema with the line <GetCAPRangeResponse xmlns="http://www.capnetwork.co.uk/CAPVehicles/"> If you type this URL into your web browser you get a 404 error - ie this schema has been removed or the website is down.

This normally works fine (of course). Questions: What Delphi/RAD Studio version are you using. Are you designing VCL or FMX? Do you get a problem if you start with a completely new form ?

...and don't forget the built in Delphi Rest debugger tool (under the tools menu). Can be handy for checking what you should be sending...

I like 12.0 patch 1 for C++ 64bit - the improved visual assist stuff works pretty well and the debugger is improved over 11.x

"Which is the best way?" questions in C++ are always tricky! The language has evolved so much (and continues to do so). We C++ Builder users have the added complication of the use of the VCL or FireMonkey functions (that come from the Delphi RTL). How important it is to keep your code portable is certainly one question to consider. Re: text files : also take into account encoding possiblities - if you need to anticipate different encodings then the VCL support is often the easiest route. I tend to use TStringList as the store and then the LoadFromFile() and SaveToFile() functions.

Find Symbol works for me - with both clang64 or with classic 32 bit (not tried it with clang32). But I have to actually select the text (typically by dragging over it with the mouse) before using the search ! find symbol

I have a Win64 C++ Builder VCL FireDAC application. I have a TFDQuery that is used as the source for a TBindSourceDB component which in turn is used for livebindings for conveniently displaying the query results on a string grid. With the change from RAD Studio 11 to RAD Studio 12 (patch 1) came the introduction of better support for fields of TimeStamp with offset (timezone) types. With RAD Studio 11 it seems that these were just treated as TimeStamp fields (ie the time zone offset was ignored). Code that worked in RAD Studio 11 stopped working in RAD Studio 12 (patch 1) and I traced the problem to the fact that one of the field definitions for this particular TFDQuery now needed to be changed from type TSQLTimeStamp (which worked in 11.) to TSQLTimeStampOffset (which works in 12-patch 1). The underlying type in the postgreSQL database for this field has always been "time stamp with time zone". I have this all understood and it is all now working following this change of type. MY QUESTION Is there a convenient way to change the TYPE of a field definition through the IDE? I couldn't find a way to do what you would think should be a simple task. In the end I "cut" the query component to the clip board, pasted it into a text editor, changed the type as described above, copied the text editor contents to the clip board and then pasted it back into the VCL data module. Not actually too difficult but hardly what you would expect to have to do in a RAD Studio IDE! I suspect I am missing a trick, hence this appeal for advice.... Attachments: file "Structure.jpg" shows the field (after my change of data type) selected in the structure pane. In the properties pane there are lots of properties, but not (or I couldn't find it!) the field type. "NewField.jpg" shows the "wiard" form that appears if you add a new field to the field definition - here there is a drop down box in which all the possible field types appear and you can pick the one you need. But I can't find a way of getting this box up again for an existing field... (Not sure if this should have been posted under category: IDE, VCL or Database - so I went for "Database").

I don't actually see that you have to do anything here - just don't take any action for the case(s) you want to ignore. If you do want to skip an extra step you can always deliberately increment the counting value - ie use this line inside the loop.... i = i + 1;

Interested in the future? Embarcadero - the future of clang - webinar