Thread

  1. Re: Leaking definitions to user programs

    Christof Petig <christof.petig@wtal.de> — 2000-11-03T14:53:33Z

    Hi Bruce, Hi Michael,
    
    here is the really short patch for shutting out all postgres definitions
    from ecpg
    programs. (e.g. Datum, Pointer, DEBUG, ERROR).
    Someone really should take a look into libpq and do the same.
    But I had to copy a small part of c.h (bool,true,false,TRUE,FALSE) into
    ecpg/include/libecpg.h. And ... there is a possible bug in c.h. You
    can't check a
    typedef via #ifndef.
    
    typedef char bool;
    ...
    #ifndef bool
    typedef char bool;
    #endif
    
    will fail. But I don't know any decent solution to that problem!
    Perhaps c.h should be broken into seperate parts.
    
    Christof
    
    PS: to Jacek: you need this patch to compile libcommon++.a!
    
    Bruce Momjian wrote:
    
    > Thanks.
    >
    > > > Yes, leaking into user programs is a bad practice.  Is there a
    > > > solution/patch for that?
    > >
    > > A solution would be a simple patch which is not available yet. But I plan on
    > > doing one (some other things still have higher priority).
    > >
    > > Christof
  2. Re: Leaking definitions to user programs

    Bruce Momjian <pgman@candle.pha.pa.us> — 2001-01-20T04:50:55Z

    You may find 7.1beta has fixed this.  I know some include files were
    rearranged in 7.1.
    
    > Hi Bruce, Hi Michael,
    > 
    > here is the really short patch for shutting out all postgres definitions
    > from ecpg
    > programs. (e.g. Datum, Pointer, DEBUG, ERROR).
    > Someone really should take a look into libpq and do the same.
    > But I had to copy a small part of c.h (bool,true,false,TRUE,FALSE) into
    > ecpg/include/libecpg.h. And ... there is a possible bug in c.h. You
    > can't check a
    > typedef via #ifndef.
    > 
    > typedef char bool;
    > ...
    > #ifndef bool
    > typedef char bool;
    > #endif
    > 
    > will fail. But I don't know any decent solution to that problem!
    > Perhaps c.h should be broken into seperate parts.
    > 
    > Christof
    > 
    > PS: to Jacek: you need this patch to compile libcommon++.a!
    > 
    > Bruce Momjian wrote:
    > 
    > > Thanks.
    > >
    > > > > Yes, leaking into user programs is a bad practice.  Is there a
    > > > > solution/patch for that?
    > > >
    > > > A solution would be a simple patch which is not available yet. But I plan on
    > > > doing one (some other things still have higher priority).
    > > >
    > > > Christof
    
    [ application/x-gzip is not supported, 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