Thread: ODBC Patch
This patch fixes resource.h which is generated automatically by MS VC++ and broken by pgindent :-( Regards, Dave.
Helps if I actually attach it! > -----Original Message----- > From: Dave Page [mailto:dpage@vale-housing.co.uk] > Sent: 12 February 2001 08:00 > To: pgsql-patches@postgresql.org > Subject: [PATCHES] ODBC Patch > > > This patch fixes resource.h which is generated automatically > by MS VC++ and > broken by pgindent :-( > > Regards, > > Dave. >
Attachment
Dave, can you test the attached patch? I assume the problem was that pgindent does not recognize // as comments, and in fact, we don't want to use them. This should fix the problem. [ Charset ISO-8859-1 unsupported, converting... ] > Helps if I actually attach it! > > > -----Original Message----- > > From: Dave Page [mailto:dpage@vale-housing.co.uk] > > Sent: 12 February 2001 08:00 > > To: pgsql-patches@postgresql.org > > Subject: [PATCHES] ODBC Patch > > > > > > This patch fixes resource.h which is generated automatically > > by MS VC++ and > > broken by pgindent :-( > > > > Regards, > > > > Dave. > > > [ Attachment, skipping... ] -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 853-3000 + If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup. | Drexel Hill, Pennsylvania 19026 ? libpsqlodbc.so.0.26 Index: resource.h =================================================================== RCS file: /home/projects/pgsql/cvsroot/pgsql/src/interfaces/odbc/resource.h,v retrieving revision 1.15 diff -c -r1.15 resource.h *** resource.h 2001/02/11 04:56:58 1.15 --- resource.h 2001/02/12 12:57:08 *************** *** 1,13 **** ! // ! { ! { ! NO_DEPENDENCIES ! } ! } - //Microsoft Developer Studio generated include file. - // Used by psqlodbc.rc - // #define IDS_BADDSN 1 #define IDS_MSGTITLE 2 #define DLG_OPTIONS_DRV 102 --- 1,9 ---- ! /* ! * {{NO_DEPENDENCIES}} ! * Microsoft Developer Studio generated include file. ! * Used by psqlodbc.rc ! */ #define IDS_BADDSN 1 #define IDS_MSGTITLE 2 #define DLG_OPTIONS_DRV 102
> -----Original Message----- > From: Bruce Momjian [mailto:pgman@candle.pha.pa.us] > Sent: 12 February 2001 12:59 > To: Dave Page > Cc: 'pgsql-patches@postgresql.org' > Subject: Re: [PATCHES] ODBC Patch > > > Dave, can you test the attached patch? I assume the problem was that > pgindent does not recognize // as comments, and in fact, we don't want > to use them. > > This should fix the problem. I've not tested it (I'm just configuring my new laptop :-) ) but that should work (and has done - if you look back in CVS I made that mod before). The problem is that MS VC++ autogenerates resource.h whenever psqlodbc.rc is updated in the IDE. Is it possible to get pgindent to ignore that file or would that cause more problems? Regards, Dave.
> > Dave, can you test the attached patch? I assume the problem was that > > pgindent does not recognize // as comments, and in fact, we don't want > > to use them. > > > > This should fix the problem. > > I've not tested it (I'm just configuring my new laptop :-) ) but that should > work (and has done - if you look back in CVS I made that mod before). The > problem is that MS VC++ autogenerates resource.h whenever psqlodbc.rc is > updated in the IDE. Is it possible to get pgindent to ignore that file or > would that cause more problems? Sure, we can ignore it, but do we want an MS-generated file to be tracked in our CVS, and especially one with // comments? If resource.h was included only by MS compilers, that would be better, but I see every platform including the file. Can you send me that resource.h file again, and I will merge in any change made by *.rc, but keep the /* */ comments, OK? -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 853-3000 + If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup. | Drexel Hill, Pennsylvania 19026
Bruce Momjian <pgman@candle.pha.pa.us> writes:
>> problem is that MS VC++ autogenerates resource.h whenever psqlodbc.rc is
>> updated in the IDE. Is it possible to get pgindent to ignore that file or
>> would that cause more problems?
> Sure, we can ignore it, but do we want an MS-generated file to be
> tracked in our CVS, and especially one with // comments?
> If resource.h was included only by MS compilers, that would be better,
> but I see every platform including the file.
Good point. This will not be buildable on machines whose compilers do
not like //.
> Can you send me that resource.h file again, and I will merge in any
> change made by *.rc, but keep the /* */ comments, OK?
I don't think that's a feasible long-term maintenance approach either.
Will you or Dave remember to do this any time psqlodbc.rc changes?
Seems like the correct answer is for pgindent to automatically transform
"// comment" to "/* comment */".
regards, tom lane
> -----Original Message----- > From: Bruce Momjian [mailto:pgman@candle.pha.pa.us] > Sent: 12 February 2001 14:12 > To: Dave Page > Cc: 'pgsql-patches@postgresql.org' > Subject: Re: [PATCHES] ODBC Patch > > > > > Dave, can you test the attached patch? I assume the > problem was that > > > pgindent does not recognize // as comments, and in fact, > we don't want > > > to use them. > > > > > > This should fix the problem. > > > > I've not tested it (I'm just configuring my new laptop :-) > ) but that should > > work (and has done - if you look back in CVS I made that > mod before). The > > problem is that MS VC++ autogenerates resource.h whenever > psqlodbc.rc is > > updated in the IDE. Is it possible to get pgindent to > ignore that file or > > would that cause more problems? > > Sure, we can ignore it, but do we want an MS-generated file to be > tracked in our CVS, and especially one with // comments? > > If resource.h was included only by MS compilers, that would be better, > but I see every platform including the file. > > Can you send me that resource.h file again, and I will merge in any > change made by *.rc, but keep the /* */ comments, OK? Here's the (working) file with C style comments. I've made comments on the subject in a reply to Tom's message... Regards, Dave.
Attachment
> -----Original Message----- > From: Tom Lane [mailto:tgl@sss.pgh.pa.us] > Sent: 12 February 2001 15:19 > To: Bruce Momjian > Cc: Dave Page; 'pgsql-patches@postgresql.org' > Subject: Re: [PATCHES] ODBC Patch > > > Bruce Momjian <pgman@candle.pha.pa.us> writes: > >> problem is that MS VC++ autogenerates resource.h whenever > psqlodbc.rc is > >> updated in the IDE. Is it possible to get pgindent to > ignore that file or > >> would that cause more problems? > > > Sure, we can ignore it, but do we want an MS-generated file to be > > tracked in our CVS, and especially one with // comments? > > If resource.h was included only by MS compilers, that would > be better, > > but I see every platform including the file. > > Good point. This will not be buildable on machines whose compilers do > not like //. > > > Can you send me that resource.h file again, and I will merge in any > > change made by *.rc, but keep the /* */ comments, OK? > > I don't think that's a feasible long-term maintenance approach either. > Will you or Dave remember to do this any time psqlodbc.rc changes? I can certainly do this anytime I update the driver, but I obviously can't keep track of other people who change it. IMHO it (psqlodbc.rc) should be updated for every patch submitted as it (along with psqlodbc.h) contain the revision number which is essential for the ODBC driver which is often updated outside of the main PostgreSQL releases. > Seems like the correct answer is for pgindent to > automatically transform > "// comment" to "/* comment */". This sounds much more sensible. Regards, Dave.
Applied. Thanks. [ Charset ISO-8859-1 unsupported, converting... ] > > > > -----Original Message----- > > From: Bruce Momjian [mailto:pgman@candle.pha.pa.us] > > Sent: 12 February 2001 14:12 > > To: Dave Page > > Cc: 'pgsql-patches@postgresql.org' > > Subject: Re: [PATCHES] ODBC Patch > > > > > > > > Dave, can you test the attached patch? I assume the > > problem was that > > > > pgindent does not recognize // as comments, and in fact, > > we don't want > > > > to use them. > > > > > > > > This should fix the problem. > > > > > > I've not tested it (I'm just configuring my new laptop :-) > > ) but that should > > > work (and has done - if you look back in CVS I made that > > mod before). The > > > problem is that MS VC++ autogenerates resource.h whenever > > psqlodbc.rc is > > > updated in the IDE. Is it possible to get pgindent to > > ignore that file or > > > would that cause more problems? > > > > Sure, we can ignore it, but do we want an MS-generated file to be > > tracked in our CVS, and especially one with // comments? > > > > If resource.h was included only by MS compilers, that would be better, > > but I see every platform including the file. > > > > Can you send me that resource.h file again, and I will merge in any > > change made by *.rc, but keep the /* */ comments, OK? > > Here's the (working) file with C style comments. > > I've made comments on the subject in a reply to Tom's message... > > Regards, Dave. > [ Attachment, skipping... ] -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 853-3000 + If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup. | Drexel Hill, Pennsylvania 19026
> > > Can you send me that resource.h file again, and I will merge in any
> > > change made by *.rc, but keep the /* */ comments, OK?
> >
> > I don't think that's a feasible long-term maintenance approach either.
> > Will you or Dave remember to do this any time psqlodbc.rc changes?
>
> I can certainly do this anytime I update the driver, but I obviously can't
> keep track of other people who change it. IMHO it (psqlodbc.rc) should be
> updated for every patch submitted as it (along with psqlodbc.h) contain the
> revision number which is essential for the ODBC driver which is often
> updated outside of the main PostgreSQL releases.
It is my understanding that resource.h is changed only when a new
option is added t psqlodbc.rc. It does not change with a version number
change because resource.h doesn't contain a version number.
>
> > Seems like the correct answer is for pgindent to
> > automatically transform
> > "// comment" to "/* comment */".
>
> This sounds much more sensible.
Sure, I can do that. I can have pgindent convert:
// comment
to
/* comment */
Dave, can you send me the MS-generated resource.h for me to test with
and cvs commit? Thanks.
--
Bruce Momjian | http://candle.pha.pa.us
pgman@candle.pha.pa.us | (610) 853-3000
+ If your life is a hard drive, | 830 Blythe Avenue
+ Christ can be your backup. | Drexel Hill, Pennsylvania 19026
> -----Original Message----- > From: Bruce Momjian [mailto:pgman@candle.pha.pa.us] > Sent: 12 February 2001 18:32 > To: Dave Page > Cc: 'Tom Lane'; 'pgsql-patches@postgresql.org' > Subject: Re: [PATCHES] ODBC Patch > > > > It is my understanding that resource.h is changed only when a new > option is added t psqlodbc.rc. It does not change with a > version number > change because resource.h doesn't contain a version number. > I've just tested with MS VC++ SP4 and it will not update psqlodbc.rc without rewriting resource.h even if you only change the version number. If you say No when prompted to update resource.h, it then doesn't update psqlodbc.rc either. > Dave, can you send me the MS-generated resource.h for me to test with > and cvs commit? Thanks. Attached. Regards, Dave.
Attachment
OK, attached is the resource.h file, run through pgindent. Looks good. Committed. [ Charset ISO-8859-1 unsupported, converting... ] > > > > -----Original Message----- > > From: Bruce Momjian [mailto:pgman@candle.pha.pa.us] > > Sent: 12 February 2001 18:32 > > To: Dave Page > > Cc: 'Tom Lane'; 'pgsql-patches@postgresql.org' > > Subject: Re: [PATCHES] ODBC Patch > > > > > > > > It is my understanding that resource.h is changed only when a new > > option is added t psqlodbc.rc. It does not change with a > > version number > > change because resource.h doesn't contain a version number. > > > > I've just tested with MS VC++ SP4 and it will not update psqlodbc.rc without > rewriting resource.h even if you only change the version number. If you say > No when prompted to update resource.h, it then doesn't update psqlodbc.rc > either. > > > Dave, can you send me the MS-generated resource.h for me to test with > > and cvs commit? Thanks. > > Attached. > > Regards, Dave. > -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 853-3000 + If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup. | Drexel Hill, Pennsylvania 19026 /* {{NO_DEPENDENCIES}} */ /* Microsoft Developer Studio generated include file. */ /* Used by psqlodbc.rc */ /* */ #define IDS_BADDSN 1 #define IDS_MSGTITLE 2 #define DLG_OPTIONS_DRV 102 #define DLG_OPTIONS_DS 103 #define IDC_DSNAME 400 #define IDC_DSNAMETEXT 401 #define IDC_DESC 404 #define IDC_SERVER 407 #define IDC_DATABASE 408 #define DLG_CONFIG 1001 #define IDC_PORT 1002 #define IDC_USER 1006 #define IDC_PASSWORD 1009 #define DS_READONLY 1011 #define DS_SHOWOIDCOLUMN 1012 #define DS_FAKEOIDINDEX 1013 #define DRV_COMMLOG 1014 #define IDC_DATASOURCE 1018 #define DRV_OPTIMIZER 1019 #define DS_CONNSETTINGS 1020 #define IDC_DRIVER 1021 #define DRV_CONNSETTINGS 1031 #define DRV_UNIQUEINDEX 1032 #define DRV_UNKNOWN_MAX 1035 #define DRV_UNKNOWN_DONTKNOW 1036 #define DRV_READONLY 1037 #define IDC_DESCTEXT 1039 #define DRV_MSG_LABEL 1040 #define DRV_UNKNOWN_LONGEST 1041 #define DRV_TEXT_LONGVARCHAR 1043 #define DRV_UNKNOWNS_LONGVARCHAR 1044 #define DRV_CACHE_SIZE 1045 #define DRV_VARCHAR_SIZE 1046 #define DRV_LONGVARCHAR_SIZE 1047 #define IDDEFAULTS 1048 #define DRV_USEDECLAREFETCH 1049 #define DRV_BOOLS_CHAR 1050 #define DS_SHOWSYSTEMTABLES 1051 #define DRV_EXTRASYSTABLEPREFIXES 1051 #define DS_ROWVERSIONING 1052 #define DRV_PARSE 1052 #define DRV_CANCELASFREESTMT 1053 #define IDC_OPTIONS 1054 #define DRV_KSQO 1055 #define DS_PG64 1057 /* Next default values for new objects */ /* */ #ifdef APSTUDIO_INVOKED #ifndef APSTUDIO_READONLY_SYMBOLS #define _APS_NEXT_RESOURCE_VALUE 104 #define _APS_NEXT_COMMAND_VALUE 40001 #define _APS_NEXT_CONTROL_VALUE 1060 #define _APS_NEXT_SYMED_VALUE 101 #endif #endif