Re: [Lazarus] is it me or a bug that i ve been fighting?
compared to a textmode version that's only 2K ;) :lol: How can I create a 2K executable with fpc? Vincent ___ Lazarus mailing list Lazarus@lazarus.freepascal.org http://www.lazarus.freepascal.org/mailman/listinfo/lazarus
Re: [Lazarus] is it me or a bug that i ve been fighting?
Zitat von waldo kitty wkitt...@windstream.net: Martin Friebe wrote: Anyway, unless you are willing to spent time on debugging the issue your self, and go into the depth of the Connector code, this will probably be of little help. i've finally been able to work out the following... MySQLConnection: TMySQL50Connection; TMySQL50Connection = Class(TConnectionName); TConnectionName = class (TSQLConnection) TSQLConnection = class (TDatabase) TDatabase = class(TCustomConnection) TCustomConnection = class(TComponent) TComponent = class(TPersistent,IUnknown,IInterfaceComponentReference) TPersistent = class(TObject) it works out that MySQLConnection.Close is TCustomConnection.Close but if i hover the mouse over Close in Procedure Close; in the public section of the TCustomConnection definition stuffs, the editor wants to point to the textfile close routine Close(var t:Text); which definitely isn't the same thing :( The hint window shows the next declaration. For example: MySQL50Connection.Close; When the mouse is over 'Close' it will show TCustomConnection.Close, because TConnectionName and TSQLConnection don't have a 'Close'. When the mouse is already over a declaration, then it shows the next declaration with this name, so you can see what this declaration hides. For example when the mouse is over the procedure Close; of TCustomConnection then it will show the 'Close' procedure of the system unit. and with all of the searching i've done, i still don't see anything that indicates where i can actually lay eyes on the code for TCustomConnection.Close Move editor cursor over 'procedure Close;' of TCustomConnection and press Ctrl+Shift+Down. It will take you to procedure TCustomConnection.Close; begin Connected := False; end; Do Ctrl+Click on 'Connected' and it will jump to property Connected: Boolean read GetConnected write SetConnected; Do Ctrl+Click on 'SetConnected' and it will jump to procedure SetConnected (Value : boolean); virtual; Since this is a virtual method it might be overloaded. But in this case it isn't. So press Ctrl+Shift+Down to jump to the method body. And so forth ... nor do i see anything indicating that it is assigned to another routine going by another name :? now i know why i much prefer the old-style procedural coding methods :P If you get the stack backtrace you can mail it and hope some one knows about... i still don't know what this involves :( http://wiki.lazarus.freepascal.org/Creating_a_Backtrace_with_GDB Mattias ___ Lazarus mailing list Lazarus@lazarus.freepascal.org http://www.lazarus.freepascal.org/mailman/listinfo/lazarus
Re: [Lazarus] is it me or a bug that i ve been fighting?
Op maandag 19-01-2009 om 18:29 uur [tijdzone -0500], schreef waldo kitty: Martin Friebe wrote: Anyway, unless you are willing to spent time on debugging the issue your self, and go into the depth of the Connector code, this will probably be of little help. i've finally been able to work out the following... MySQLConnection: TMySQL50Connection; TMySQL50Connection = Class(TConnectionName); TConnectionName = class (TSQLConnection) TSQLConnection = class (TDatabase) TDatabase = class(TCustomConnection) TCustomConnection = class(TComponent) TComponent = class(TPersistent,IUnknown,IInterfaceComponentReference) TPersistent = class(TObject) it works out that MySQLConnection.Close is TCustomConnection.Close but if i hover the mouse over Close in Procedure Close; in the public section of the TCustomConnection definition stuffs, the editor wants to point to the textfile close routine Close(var t:Text); which definitely isn't the same thing :( You can find the code in the fpc-code-dir under packages/fcl-db/src/base/database.inc. What primarily happens is that doDisConnect is called. and with all of the searching i've done, i still don't see anything that indicates where i can actually lay eyes on the code for TCustomConnection.Close nor do i see anything indicating that it is assigned to another routine going by another name :? Take a look at TDatabase.DoDisconnect in the same file. It first closes all queries which are linked to this dataset, and thereafter it closes all transactions. So try to close those yourself before you close the database-connection. See where it goes wrong... now i know why i much prefer the old-style procedural coding methods :P If you get the stack backtrace you can mail it and hope some one knows about... i still don't know what this involves :( Run the application in gdb, and when the crash occurs, type 'bt' on the gdb-prompt. But you must have the fcl-db package compiled with debug-information for it to make any sense. If you really want to try, add the fcl-db/src/base fcl-db/src/sqldb fcl-db/src/dbase and fcl-db/src/sqldb/mysql paths to your compiler-options path and rebuild your application within Lazarus. That way you can 'step' into the fcl-db sources... Joost. ___ Lazarus mailing list Lazarus@lazarus.freepascal.org http://www.lazarus.freepascal.org/mailman/listinfo/lazarus
Re: [Lazarus] is it me or a bug that i ve been fighting?
Joost van der Sluis wrote:
Op maandag 19-01-2009 om 18:29 uur [tijdzone -0500], schreef waldo
kitty:
it works out that MySQLConnection.Close is TCustomConnection.Close but if i
hover the mouse over Close in Procedure Close; in the public section of
the
TCustomConnection definition stuffs, the editor wants to point to the
textfile
close routine Close(var t:Text); which definitely isn't the same thing :(
You can find the code in the fpc-code-dir under
packages/fcl-db/src/base/database.inc. What primarily happens is that
doDisConnect is called.
hummm... ok...
and with all of the searching i've done, i still don't see anything that
indicates where i can actually lay eyes on the code for
TCustomConnection.Close
nor do i see anything indicating that it is assigned to another routine
going by
another name :?
Take a look at TDatabase.DoDisconnect in the same file. It first closes
all queries which are linked to this dataset, and thereafter it closes
all transactions.
ok... well, the query is closed by my code after the variables are loaded with
the data... then we attempt to close the connection via a routine that all will
use to close the connection... this routine does two things...
transaction.active := false;
connection.close;
is it enough to set transaction.active:=false? i note there's no
transaction.close... in any case, the error pops up when the connection.close
is
called... so, i'd say that, yes, i've closed the query and the transaction
before attempting to close the connection...
also, could it be that because i'm not using a datasource that is causing the
problem?
So try to close those yourself before you close the database-connection.
See where it goes wrong...
by the above, i am, aren't i?... in fact, here's code...
MySQLConnection: TMySQL50Connection;
CompanyDataSQLQuery: TSQLQuery;
SQLTransaction: TSQLTransaction;
the CompanyEdit_co_* fields are all TEdits...
procedure TkimsForm.LoadCompanyData(Sender: TObject);
begin
if MySQLConnection.Connected then begin
ShowStatus('Connected #1');
CloseConnection(Sender);
end;
if not MySQLConnection.Connected then ShowStatus('NOT Connected #1');
MySQLConnection.DatabaseName := 'foo';
MySQLConnection.HostName := 'bar';
MySQLConnection.Password := 'fubar';
MySQLConnection.UserName := 'snafu';
ShowStatus('Connecting to database ' + MySQLConnection.DatabaseName + '...');
MySQLConnection.Open;
// If we're connected, retrieve our company info
if MySQLConnection.Connected then begin
ShowStatus('Connected');
ShowStatus('Opening query...');
CompanyDataSQLQuery.Open;
if CompanyDataSQLQuery.Active then begin
CompanyDataSQLQuery.First;
ShowStatus('Query ACTIVE and at first record...');
while not CompanyDataSQLQuery.EOF do begin
ShowStatus('Loading data from row...');
CompanyEdit_co_name.text :=
CompanyDataSQLQuery.FieldByName('co_name').AsString;
CompanyEdit_co_address1.text :=
CompanyDataSQLQuery.FieldByName('co_address1').AsString;
CompanyEdit_co_address2.text :=
CompanyDataSQLQuery.FieldByName('co_address2').AsString;
CompanyEdit_co_city.text :=
CompanyDataSQLQuery.FieldByName('co_city').AsString;
CompanyEdit_co_state.text :=
CompanyDataSQLQuery.FieldByName('co_state').AsString;
CompanyEdit_co_postal.text :=
CompanyDataSQLQuery.FieldByName('co_postal').AsString;
CompanyEdit_co_phone.text :=
CompanyDataSQLQuery.FieldByName('co_phone').AsString;
CompanyEdit_co_fax.text :=
CompanyDataSQLQuery.FieldByName('co_fax').AsString;
CompanyEdit_co_contact.text :=
CompanyDataSQLQuery.FieldByName('co_contact').AsString;
CompanyEdit_co_taxpercent.text :=
CompanyDataSQLQuery.FieldByName('co_taxpercent').AsString;
CompanyEdit_co_commission_rate1.text :=
CompanyDataSQLQuery.FieldByName('co_commission_rate1').AsString;
CompanyEdit_co_commission_rate2.text :=
CompanyDataSQLQuery.FieldByName('co_commission_rate2').AsString;
CompanyEdit_co_commission_percent1.text :=
CompanyDataSQLQuery.FieldByName('co_commission_percent1').AsString;
CompanyEdit_co_commission_percent2.text :=
CompanyDataSQLQuery.FieldByName('co_commission_percent2').AsString;
CompanyDataSQLQuery.Next;
end;
ShowStatus('Closing query...');
CompanyDataSQLQuery.Close;
end;
if not CompanyDataSQLQuery.Active then
ShowStatus('Query NOT ACTIVE');
end;
if MySQLConnection.Connected then begin
ShowStatus('Connected #2');
CloseConnection(Sender);
end;
if not MySQLConnection.Connected then ShowStatus('NOT Connected #2');
end;
procedure TkimsForm.CloseConnection(Sender: TObject);
begin
// The SQLTransaction gets activated automatically, but before we can close
// the connection we have to set the SQLTransaction.Active to false.
if SQLTransaction.Active
Re: [Lazarus] is it me or a bug that i ve been fighting?
waldo kitty wrote:
procedure TkimsForm.CloseConnection(Sender: TObject);
begin
// The SQLTransaction gets activated automatically, but before we can close
// the connection we have to set the SQLTransaction.Active to false.
if SQLTransaction.Active then begin
ShowStatus('Deactivating SQLTransaction...');
SQLTransaction.Active := False;
SQLTransaction.
end;
whoops... the SQLTransaction. line is not supposed to be there... i had
placed
it while looking to see if there was a SQLTransaction.Close option and forgot
to
remove it before copy'n'paste to the message :?
--
NOTE: NEW EMAIL ADDRESS!!
_\/
(@@) Waldo Kitty, Waldo's Place USA
__ooO_( )_Ooo_ telnet://bbs.wpusa.dynip.com
_|_|_|_|_|_|_ http://www.wpusa.dynip.com
|_|_|_|_|_| ftp://ftp.wpusa.dynip.com
_|_Eat_SPAM_to_email_me!_YUM!__|_ wkitty42 -at- windstream.net
---
avast! Antivirus: Outbound message clean.
Virus Database (VPS): 090120-0, 01/20/2009
Tested on: 1/20/09 15:42:58
avast! - copyright (c) 1988-2009 ALWIL Software.
http://www.avast.com
___
Lazarus mailing list
Lazarus@lazarus.freepascal.org
http://www.lazarus.freepascal.org/mailman/listinfo/lazarus
Re: [Lazarus] is it me or a bug that i ve been fighting?
Mattias Gärtner wrote: Move editor cursor over 'procedure Close;' of TCustomConnection and press Ctrl+Shift+Down. ahhh, that's one i hadn't discovered... i did figure out the CTRL-Click thing when i had the CTRL pressed inadvertently while hovering over an item and saw that it changed to a link :P so, following along a slightly different path (DoDisconnect pointed to by Joost), i've come to the following in database.inc... procedure TDatabase.DoDisconnect; begin Closedatasets; Closetransactions; DoInternalDisConnect; if csloading in ComponentState then FOpenAfterRead := false; FConnected := False; end; Closedatasets, Closetransactions, and DoInternalDisConnect are all pointed as being in db.pas and the first two, Closedatasets and Closetransactions both come to code blocks with CTRL-Shift-Down but DoInternalDisConnect does not :? it is defined in db.pas as Procedure DoInternalDisConnect; Virtual;Abstract; and the IDE doesn't want to show me anything about it when i hover over it so now i don't know where to look further :( on a whim, i went tromping about and have found an empty DoInternalDisConnect routine in packages/fcl-db/src/sqldb/sqldb.pp procedure TSQLConnection.DoInternalDisconnect; begin end; and another in packages/fcl-db/src/sqldb/mysql/mysqlconn.inc procedure TConnectionName.DoInternalDisconnect; begin inherited DoInternalDisconnect; mysql_close(FMySQL); FMySQL:=Nil; ReleaseMysql; end; which would seem to want to take me to ReleaseMysql from packages/mysql/src/mysql.inc... perhaps i have the wrong idea about database apps?? in all of my coding years, the only time i've ever maintained an open connection to the database was way back in my days as a dBaseIII/dBaseIV coder working with relational databases... this was before and just as SQL stuffs started coming along... heck, what we call tables today were called databases back then ;) with that in mind, should i be maintaining a connection to the mysql database the whole time my app is running and only close it when i close my app? what about for situations where the database server may not allow persistent connections and may close the connection after some minutes? FWIW: this will be a relational type database app in that there is projected to be 4 other tables beside the company data table... three of those tables will be related into the 4th by unique (to each of the 3 tables) id numbers in three different columns of the 4th table... yeah, for those who are quick on the draw, the 4th table is for sales made by each employee (one table) to each customer (another table) for each inventory item (the 3rd table)... and all this just to be able to get to the point of implementing the custom math to calculate the final sale price, tax and employee commission for the sale... call it an exercise in reinventing the wheel, eh? :P -- NOTE: NEW EMAIL ADDRESS!! _\/ (@@) Waldo Kitty, Waldo's Place USA __ooO_( )_Ooo_ telnet://bbs.wpusa.dynip.com _|_|_|_|_|_|_ http://www.wpusa.dynip.com |_|_|_|_|_| ftp://ftp.wpusa.dynip.com _|_Eat_SPAM_to_email_me!_YUM!__|_ wkitty42 -at- windstream.net --- avast! Antivirus: Outbound message clean. Virus Database (VPS): 090120-0, 01/20/2009 Tested on: 1/20/09 17:13:17 avast! - copyright (c) 1988-2009 ALWIL Software. http://www.avast.com ___ Lazarus mailing list Lazarus@lazarus.freepascal.org http://www.lazarus.freepascal.org/mailman/listinfo/lazarus
Re: [Lazarus] is it me or a bug that i ve been fighting?
waldo kitty wrote: Martin Friebe wrote: but a couple of thinks that come to mind: 1) What version of FPC are you using (you only said you updated Lazarus); you refer to FPC 2.2.2, have you tried the fixes branch (2.2.3)? or trunc(2.3.1)? i'm using whatever comes with lazarus... i started with laz 0.9.26 and then completely removed it to install 0.9.27 because i found that some config settings are evidently hardcoded or shared in a/the registry when a dual install of 0.9.27 still had some things pointing to the old 0.9.26 stuffs :? :( On the daily snapshot page, you get a windows build with fpc 2.2.2 and another one with 2.2.3 (eq fixes branch of fpc) lookup -pcp (--primar-config-path) Lazarus does what a windows app should do, it stores setting in you users home director (depends on windows version) Then if you installed Lazarus (incl fpc) in one directory = do not rename the directory; otherwise you must update you fpc.cfg file (windows search to locate) If you have multiply installations, you may have to go to environment-options, and adapt the path to fpc (once you use -pcp) and that might be another bug or something where one can't install laz 0.9.26 to one directory and 0.9.27 to another and have each come out of the box without sharing any config settings or something... i dunno, as i'm still trying to come to grips with the whole GUI interface and drop this here and that there to create a hello world! GUI app that's 10Meg (tongue in cheek!) in size compared to a textmode version that's only 2K ;) :lol: use strip to get rid of the symbols (or compiler options dialog, unclick all the debug info) But leave it in while looking for a bug 2) You say you do not know the exact error: because the popup box doesn't state it = Try running the exe in the debugger, make sure you compile with debug symbols for gdb (-g, maybe also -gl). i've not changed any settings from the defaults on install... i have, a few times, run some stuff directly from the GUI in the debugger but even with 256Meg of RAM in this box, it gets a bit on the tight side at times which is why i switched over to simply building the apps and then shutting down laz and running the built code outside on its own... maybe that wasn't/isn't such a good idea? why not, but it doesn't help debugging (unless you do gdb by hand) ___ Lazarus mailing list Lazarus@lazarus.freepascal.org http://www.lazarus.freepascal.org/mailman/listinfo/lazarus
Re: [Lazarus] is it me or a bug that i ve been fighting?
waldo kitty wrote: Martin Friebe wrote: 2) You say you do not know the exact error: because the popup box doesn't state it = Try running the exe in the debugger, make sure you compile with debug symbols for gdb (-g, maybe also -gl). just following up on this (as my other reply hasn't srrived back from the list yet) but i have just attempted to, with laz 0.9.27 fpc 2.2.3, run my app from the laz gui via F9... i got a box stating Project raised exception class 'External SIGSEGV' at which point i could only click on the [OK] button in that box... then it appeared that my box was locked up but it took a minute or so and another box appeared titled Assembly with what appeared to be a list of addresses but i farkled things up by attempting to scroll the box to see what may have been listed and i lost the starting point that it gave me... Yes the assembly window, ignore that (and yes if you close it to early, it does mind that sometimes) Open view-debug-stack window (alt-ctrl-s) That will show you where in the Lazarus code the exception happened. Problem: Likely it is in the fpc part, which will be useless numbers. Or even in the mysql library. Anyway, unless you are willing to spent time on debugging the issue your self, and go into the depth of the Connector code, this will probably be of little help. If you get the stack backtrace you can mail it and hope some one knows about... . ___ Lazarus mailing list Lazarus@lazarus.freepascal.org http://www.lazarus.freepascal.org/mailman/listinfo/lazarus
Re: [Lazarus] is it me or a bug that i ve been fighting?
On 1/19/09, waldo kitty wkitt...@windstream.net wrote: and that might be another bug or something where one can't install laz 0.9.26 to one directory and 0.9.27 to another and have each come out of the box without sharing any config settings or something... It can be done. Just start lazarus with --pcp=path where you want lazarus to save settings I have 0.9.26 and an SVN version up and running on Windows with that --pcp thing. AFAIK before 0.9.27 settings were saved in Lazarus directory, after that in users config directory, which I guess is supplied by OS (on Windows with GetSpecialFolder (?)). Bart ___ Lazarus mailing list Lazarus@lazarus.freepascal.org http://www.lazarus.freepascal.org/mailman/listinfo/lazarus
Re: [Lazarus] is it me or a bug that i ve been fighting?
Martin Friebe wrote: Anyway, unless you are willing to spent time on debugging the issue your self, and go into the depth of the Connector code, this will probably be of little help. i've finally been able to work out the following... MySQLConnection: TMySQL50Connection; TMySQL50Connection = Class(TConnectionName); TConnectionName = class (TSQLConnection) TSQLConnection = class (TDatabase) TDatabase = class(TCustomConnection) TCustomConnection = class(TComponent) TComponent = class(TPersistent,IUnknown,IInterfaceComponentReference) TPersistent = class(TObject) it works out that MySQLConnection.Close is TCustomConnection.Close but if i hover the mouse over Close in Procedure Close; in the public section of the TCustomConnection definition stuffs, the editor wants to point to the textfile close routine Close(var t:Text); which definitely isn't the same thing :( and with all of the searching i've done, i still don't see anything that indicates where i can actually lay eyes on the code for TCustomConnection.Close nor do i see anything indicating that it is assigned to another routine going by another name :? now i know why i much prefer the old-style procedural coding methods :P If you get the stack backtrace you can mail it and hope some one knows about... i still don't know what this involves :( -- NOTE: NEW EMAIL ADDRESS!! _\/ (@@) Waldo Kitty, Waldo's Place USA __ooO_( )_Ooo_ telnet://bbs.wpusa.dynip.com _|_|_|_|_|_|_ http://www.wpusa.dynip.com |_|_|_|_|_| ftp://ftp.wpusa.dynip.com _|_Eat_SPAM_to_email_me!_YUM!__|_ wkitty42 -at- windstream.net --- avast! Antivirus: Outbound message clean. Virus Database (VPS): 090119-0, 01/19/2009 Tested on: 1/19/09 18:29:11 avast! - copyright (c) 1988-2009 ALWIL Software. http://www.avast.com ___ Lazarus mailing list Lazarus@lazarus.freepascal.org http://www.lazarus.freepascal.org/mailman/listinfo/lazarus
Re: [Lazarus] is it me or a bug that i ve been fighting?
thanks for your response, reenen... it does tend to make sense on the one hand but on the other i must ask if you have looked at and/or run the demo code i pointed to? it does check if the connection is open ;) procedure TFormTryMySQL.ConnectButtonClick(Sender: TObject); begin // Check if we have an active connection. If so, let's close it. if MySQLConnection1.Connected then begin SQLTransaction1.Active := False; MySQLConnection1.Close; end; blahfooblah...; when i place logging lines in the above, the last logging line that executes is the one right above MySQLConnection1.Close; which indicates the error in is the MySQLConnection1.Close; library code... i can see where what you are saying may be possible but it doesn't make sense because the next time the above is executed, it works properly (ie: there's no error).. what it almost sounds like is that in closing the connection object, something was not created during the open which somehow does get created during the close... again, i dunno and that's why i'm here... one would think that demo code should work and work properly (ie: without error) out of the box... i won't point back to my comment about having problems with several other of the wiki demo apps not working for some reason or another :? Reenen Laurie wrote: I am very uninformed when it comes to databases, but access violations happens when you do stuff to stuff that's not there. My guess is that for some reason your close code gets fired twice, and thus it's already closed by the time you get there the second time. So if there's a way to check if it's open, before you close it, I'd suggest try to close it inside an if open... Regards, -Reenen On Sun, Jan 18, 2009 at 9:39 PM, waldo kitty wkitt...@windstream.net mailto:wkitt...@windstream.net wrote: history: i'm a long time pascal coder... since TP3 first appeared on the market... then to TP5 and most of my time with TP6/BP6 and numerous commercial apps under my belt... i've dabbled with OOP stuffs, TP7/BP7 and even have a copy of delphi6-personal that i've not been able to wrap my head around... since my delphi was a downloaded copy from Inprise, it is also a bit harder to work with since i've no real documentation that i can lay in my lap to look over while working on code :? :( in any case, i've been fighting a battle with mysql50conn for a few weeks, now... i started with Lazaruz 0.9.26 and last night installed Lazarus 0.9.27 to see if the problem i'm facing may have been fixed in there but it has not and i'm not even certain if the problem is in the library code or possibly something on this old win98SE box :? ok, enough of the history and story telling... the details: with the tutorial DB code on the wiki (yeah, i was /finally/ able to get two or three of those working after some starts, stops and fits :P), there's several that check to see if the connection to the database is active and if it is, they attempt to close it... my problem is that the first attempt to close the connection _always_ throws an error... i don't know what specific error code because the popup box doesn't state it... the box only says Access violation. Press Ok to ignore and risk data corruption. Press Cancel to kill the program. [OK] [Cancel] i'm triggering my initial connection to the database via a button. i click the button, the connection takes place, data is retrieved, and the connection is attempted to close... that's when the error is thrown... however, if i hit the [OK] button to ignore the error then i can open and close the connection to the database with no problems... at least it appears that way because other routines gather other data from other tables and they all connect, gather, and disconnect without problems... all routines use the very same disconnect routine and this is what leads be to believe that there's something in the library code that's the problem... currently, all the code i'm using simply uses the close procedure of the connection object (ie: MySQL50Connection.Close)... i've only one graphical demo app that doesn't exhibit this problem but it is using mysql4 in the uses and a lot of manual code instead of the connector, transaction, and sqlquery objects... i also have the text modedemo app working and it doesn't have the problems either... now, i'm not sure, at all, if any of the (demo) apps i'm working with are using the libmysql.dll file that i placed in my windows/system(32?) directory or not... i believe that the textmode demo would not run without that dll but that was several weeks ago when i was
Re: [Lazarus] is it me or a bug that i ve been fighting?
Hi, I have never done any database stuff either, but AVs cry for a debugger. Maybe other people on this list can help you, but it will do no harm to give gdb (or some such) a chance. hih -- A: Because it messes up the order in which people normally read text. Q: Why is top-posting such a bad thing? A: Top-posting. Q: What is the most annoying thing in e-mail? ___ Lazarus mailing list Lazarus@lazarus.freepascal.org http://www.lazarus.freepascal.org/mailman/listinfo/lazarus
Re: [Lazarus] is it me or a bug that i ve been fighting?
Martin Friebe wrote: let me start with No I have not executed the demo code; :) i can appreciate that ;) but a couple of thinks that come to mind: 1) What version of FPC are you using (you only said you updated Lazarus); you refer to FPC 2.2.2, have you tried the fixes branch (2.2.3)? or trunc(2.3.1)? i'm using whatever comes with lazarus... i started with laz 0.9.26 and then completely removed it to install 0.9.27 because i found that some config settings are evidently hardcoded or shared in a/the registry when a dual install of 0.9.27 still had some things pointing to the old 0.9.26 stuffs :? :( and that might be another bug or something where one can't install laz 0.9.26 to one directory and 0.9.27 to another and have each come out of the box without sharing any config settings or something... i dunno, as i'm still trying to come to grips with the whole GUI interface and drop this here and that there to create a hello world! GUI app that's 10Meg (tongue in cheek!) in size compared to a textmode version that's only 2K ;) :lol: 2) You say you do not know the exact error: because the popup box doesn't state it = Try running the exe in the debugger, make sure you compile with debug symbols for gdb (-g, maybe also -gl). i've not changed any settings from the defaults on install... i have, a few times, run some stuff directly from the GUI in the debugger but even with 256Meg of RAM in this box, it gets a bit on the tight side at times which is why i switched over to simply building the apps and then shutting down laz and running the built code outside on its own... maybe that wasn't/isn't such a good idea? Or try to compile not as a windows app (switch off -WG) then you get a console window, and a lot of output. (Both options are in Project-compiler options-tab: linking) hummm... that's something i haven't tried and actually wasn't aware of... since i'm on a winwhatever box, i did turn on the windows app option in the settings but other than that, i've not altered any settings from their defaults (that i know of)... It could be in the RTL, in which case it would be helpful to rebuild this with debug support interesting... how to do that, i dunno but it would also seem that others might attempt to run the same demo(s) i'm pointing to in a clean install environment if for nothing other than to at least confirm it on other machines/installs than mine... i can't say that my setup is perfect... long way from it, in fact but it just seems that since i've reproduced it in at least two separate installations, that... well... you know? ;) -- NOTE: NEW EMAIL ADDRESS!! _\/ (@@) Waldo Kitty, Waldo's Place USA __ooO_( )_Ooo_ telnet://bbs.wpusa.dynip.com _|_|_|_|_|_|_ http://www.wpusa.dynip.com |_|_|_|_|_| ftp://ftp.wpusa.dynip.com _|_Eat_SPAM_to_email_me!_YUM!__|_ wkitty42 -at- windstream.net --- avast! Antivirus: Outbound message clean. Virus Database (VPS): 090118-0, 01/18/2009 Tested on: 1/19/09 02:00:02 avast! - copyright (c) 1988-2009 ALWIL Software. http://www.avast.com ___ Lazarus mailing list Lazarus@lazarus.freepascal.org http://www.lazarus.freepascal.org/mailman/listinfo/lazarus
Re: [Lazarus] is it me or a bug that i ve been fighting?
Martin Friebe wrote: 2) You say you do not know the exact error: because the popup box doesn't state it = Try running the exe in the debugger, make sure you compile with debug symbols for gdb (-g, maybe also -gl). just following up on this (as my other reply hasn't srrived back from the list yet) but i have just attempted to, with laz 0.9.27 fpc 2.2.3, run my app from the laz gui via F9... i got a box stating Project raised exception class 'External SIGSEGV' at which point i could only click on the [OK] button in that box... then it appeared that my box was locked up but it took a minute or so and another box appeared titled Assembly with what appeared to be a list of addresses but i farkled things up by attempting to scroll the box to see what may have been listed and i lost the starting point that it gave me... so, at that point, i used the Laz GUI's stop (CTRL-F2 IIRC) option to terminate the app but now i'm still waiting on Laz to come back to life... since it has been over 5 minutes, it appears that i'm going to have to reboot the system to get it back and try to go on further :? Or try to compile not as a windows app (switch off -WG) then you get a console window, and a lot of output. (Both options are in Project-compiler options-tab: linking) i may try this option next... dunno know what i'll see as far as output... anything special i should watch for or possibly TEE to a log file?? maybe i should also mention that i'm using 4DOS for a command interpreter replacement, too?? possibly that could have something to do with this problem?? -- NOTE: NEW EMAIL ADDRESS!! _\/ (@@) Waldo Kitty, Waldo's Place USA __ooO_( )_Ooo_ telnet://bbs.wpusa.dynip.com _|_|_|_|_|_|_ http://www.wpusa.dynip.com |_|_|_|_|_| ftp://ftp.wpusa.dynip.com _|_Eat_SPAM_to_email_me!_YUM!__|_ wkitty42 -at- windstream.net --- avast! Antivirus: Outbound message clean. Virus Database (VPS): 090118-0, 01/18/2009 Tested on: 1/19/09 02:14:14 avast! - copyright (c) 1988-2009 ALWIL Software. http://www.avast.com ___ Lazarus mailing list Lazarus@lazarus.freepascal.org http://www.lazarus.freepascal.org/mailman/listinfo/lazarus
Re: [Lazarus] is it me or a bug that i ve been fighting?
waldo kitty wrote: Project raised exception class 'External SIGSEGV' at which point i could only click on the [OK] button in that box... then it appeared that my box was locked up but it took a minute or so and another box appeared titled Assembly with what appeared to be a list of addresses but i farkled things up by attempting to scroll the box to see what may have been listed and i lost the starting point that it gave me... so, at that point, i used the Laz GUI's stop (CTRL-F2 IIRC) option to terminate the app but now i'm still waiting on Laz to come back to life... since it has been over 5 minutes, it appears that i'm going to have to reboot the system to get it back and try to go on further :? and another followup... i finally ended up using enditall to kill the gdb and laz tasks... laz first since i didn't notice gdb until afterward... anyway, i then reloaded laz and ran my app via F9 again... this time, i left the Assembler (not Assembly as previously stated) window alone when it came up and after a minute or so, its output finally settled down such that the output looks like so... BFF8AC13 03508E48 .. 03508E49 .. 03508E4A .. 03508E4B .. 03508E4C .. 03508E4D .. 03508E4E .. 03508E4F .. 03508E50 .. 03508E51 .. 03508E52 .. 03508E53 .. 03508E54 .. 03508E55 .. 03508E56 .. 03508E57 .. 03508E58 .. 03508E59 .. 03508E5A .. 03508E5B .. 03508E5C .. and that's all i can see without mucking about with that box... since i can't resist, i did try to scroll up to see what i could see and i have this for the X entries above the starting and including the first one in the above... BFF8AC0F .. BFF8AC10 BFF8AC10 BFF8AC10 BFF8AC10 BFF8AC10 .. BFF8AC11 .. BFF8AC12 .. BFF8AC13 starting point of above... dunno know if that helps anyone or not.. it doesn't help me as it appears to be addresses but there's no associated content :( -- NOTE: NEW EMAIL ADDRESS!! _\/ (@@) Waldo Kitty, Waldo's Place USA __ooO_( )_Ooo_ telnet://bbs.wpusa.dynip.com _|_|_|_|_|_|_ http://www.wpusa.dynip.com |_|_|_|_|_| ftp://ftp.wpusa.dynip.com _|_Eat_SPAM_to_email_me!_YUM!__|_ wkitty42 -at- windstream.net --- avast! Antivirus: Outbound message clean. Virus Database (VPS): 090118-0, 01/18/2009 Tested on: 1/19/09 02:33:01 avast! - copyright (c) 1988-2009 ALWIL Software. http://www.avast.com ___ Lazarus mailing list Lazarus@lazarus.freepascal.org http://www.lazarus.freepascal.org/mailman/listinfo/lazarus
Re: [Lazarus] is it me or a bug that i ve been fighting?
Martin Friebe wrote: Or try to compile not as a windows app (switch off -WG) then you get a console window, and a lot of output. (Both options are in Project-compiler options-tab: linking) now i've done this and i still got the GUI window of my app which allowed me to click the [connect] so that i could run this in the same way as i have been doing... i did get a msdos window but there was no output in it at all... i still also got that assembler window and every time it has come up, the starting address has been the same but the other numbers under it have been different... now what? :? -- NOTE: NEW EMAIL ADDRESS!! _\/ (@@) Waldo Kitty, Waldo's Place USA __ooO_( )_Ooo_ telnet://bbs.wpusa.dynip.com _|_|_|_|_|_|_ http://www.wpusa.dynip.com |_|_|_|_|_| ftp://ftp.wpusa.dynip.com _|_Eat_SPAM_to_email_me!_YUM!__|_ wkitty42 -at- windstream.net --- avast! Antivirus: Outbound message clean. Virus Database (VPS): 090118-0, 01/18/2009 Tested on: 1/19/09 02:51:09 avast! - copyright (c) 1988-2009 ALWIL Software. http://www.avast.com ___ Lazarus mailing list Lazarus@lazarus.freepascal.org http://www.lazarus.freepascal.org/mailman/listinfo/lazarus
