Re: Cygwin Gcc Compatibility with gcc on other platforms ??

Re: Cygwin Gcc Compatibility with gcc on other platforms ??

[N8TM]

In a message dated 99-07-29 16:53:22 EDT, raman_bakshi@yahoo.com writes:

<< if i write and compile code
 using cygwin gcc (ANSI C code), will that code
 compile (not execute, thats another thing dependent
 on cpu etc) on platforms like linux etc where the GNU
 gcc compiler is available ? >>
Yes, if you truly restrict yourself to ANSI C and don't depend on libraries 
and non-ANSI include files which invariably differ from cygwin.  The cygwin 
developers have done an amazing job of making cygwin into a standard gnu 
compiler host, so you get excellent cross-platform compatibility.  This goal 
fortunately appears to be kept in mind by most of those who are working at 
the same time to get more Windoze compatibility.

--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe@sourceware.cygnus.com

gcc build error--configuration problem?

[Josh Baudhuin]

I just installed the 'full' binary release of 20.1. Then I tried to build
cvs-1.10 using the tar from the net.

The 'configure' script seemed to run w/o a hitch, but when I build, I get a
compile error when it comes to src/rcs.c:

/usr/local/src/cvs-1.10 <21> make
making all in lib
make[1]: Nothing to be done for `all'.
making all in zlib
make[1]: Nothing to be done for `all'.
making all in diff
make[1]: Nothing to be done for `all'.
making all in src
gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O -c rcs.c
rcs.c: In function `apply_rcs_changes':
rcs.c:6994: parse error before `('
make[1]: *** [rcs.o] Error 1
make: *** [all] Error 1

Cf. the source itself:
	. . .
6977
6978 static int
6979 apply_rcs_changes (lines, diffbuf, difflen, name, addvers, delvers)
6980      struct linevector *lines;
6981      const char *diffbuf;
6982      size_t difflen;
6983      const char *name;
6984      RCSVers *addvers;
6985      RCSVers *delvers;
6986 {
6987     const char *p;
6988     const char *q;
6989     int op;
6990     /* The RCS format throws us for a loop in that the deltafrags (if
6991        we define a deltafrag as an add or a delete) need to be applied
6992        in reverse order.  So we stick them into a linked list.  */
6993     struct deltafrag {
6994         enum {ADD, DELETE} type;
6995         unsigned long pos;
6996         unsigned long nlines;
6997         const char *new_lines;
6998         size_t len;
6999         struct deltafrag *next;
7000     };
7001     struct deltafrag *dfhead;
7002     struct deltafrag *df;
	. . .

Seems like there must be a macro in there for ADD or DELETE. Is the
configuration wrong? Do I need to do more to set up the build environment
than the simple install of full.exe? [Doesn't seem to be a CDK install
anymore.]


--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe@sourceware.cygnus.com

Re: gcc build error--configuration problem?

[Mumit Khan]

"Josh Baudhuin" <joshb@Cadence.COM> writes:
> 
> I just installed the 'full' binary release of 20.1. Then I tried to build
> cvs-1.10 using the tar from the net.
> 
> The 'configure' script seemed to run w/o a hitch, but when I build, I get a
> compile error when it comes to src/rcs.c:


[ ... ]

> gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O -c rcs.c
> rcs.c: In function `apply_rcs_changes':
> rcs.c:6994: parse error before `('
> make[1]: *** [rcs.o] Error 1
> make: *** [all] Error 1
> 
[ ... ]

> 
> Seems like there must be a macro in there for ADD or DELETE. Is the
> configuration wrong? Do I need to do more to set up the build environment
> than the simple install of full.exe? [Doesn't seem to be a CDK install
> anymore.]
> 

The windows32api headers define a DELETE macro, which is obviously not
what RCS expects.

You may need to insert a #undef DELETE after the appropriate to nullify
the windows one and use RCS one.

Damn these namespace polluting macros in system headers.

My preference of course is to change the RCS macros/enums to use RCS_
prefix and be done with it.

Regards,
Mumit


--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe@sourceware.cygnus.com

RE: gcc build error--configuration problem?

[Josh Baudhuin]

Thanks. That sounds right.

-----Original Message-----
From: Mumit Khan [ mailto:khan@xraylith.wisc.EDU ]
Sent: Thursday, 29 July 1999 2:32 pm
To: Josh Baudhuin
Cc: cygwin@sourceware.cygnus.com
Subject: Re: gcc build error--configuration problem?


"Josh Baudhuin" <joshb@Cadence.COM> writes:
>
> I just installed the 'full' binary release of 20.1. Then I tried to build
> cvs-1.10 using the tar from the net.
>
> The 'configure' script seemed to run w/o a hitch, but when I build, I get
a
> compile error when it comes to src/rcs.c:


[ ... ]

> gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O -c rcs.c
> rcs.c: In function `apply_rcs_changes':
> rcs.c:6994: parse error before `('
> make[1]: *** [rcs.o] Error 1
> make: *** [all] Error 1
>
[ ... ]

>
> Seems like there must be a macro in there for ADD or DELETE. Is the
> configuration wrong? Do I need to do more to set up the build environment
> than the simple install of full.exe? [Doesn't seem to be a CDK install
> anymore.]
>

The windows32api headers define a DELETE macro, which is obviously not
what RCS expects.

You may need to insert a #undef DELETE after the appropriate to nullify
the windows one and use RCS one.

Damn these namespace polluting macros in system headers.

My preference of course is to change the RCS macros/enums to use RCS_
prefix and be done with it.

Regards,
Mumit



--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe@sourceware.cygnus.com

RE: gcc build error--configuration problem?

[Josh Baudhuin]

Thanks. That sounds right.

-----Original Message-----
From: Mumit Khan [ mailto:khan@xraylith.wisc.EDU ]
Sent: Thursday, 29 July 1999 2:32 pm
To: Josh Baudhuin
Cc: cygwin@sourceware.cygnus.com
Subject: Re: gcc build error--configuration problem?


"Josh Baudhuin" <joshb@Cadence.COM> writes:
>
> I just installed the 'full' binary release of 20.1. Then I tried to build
> cvs-1.10 using the tar from the net.
>
> The 'configure' script seemed to run w/o a hitch, but when I build, I get
a
> compile error when it comes to src/rcs.c:


[ ... ]

> gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O -c rcs.c
> rcs.c: In function `apply_rcs_changes':
> rcs.c:6994: parse error before `('
> make[1]: *** [rcs.o] Error 1
> make: *** [all] Error 1
>
[ ... ]

>
> Seems like there must be a macro in there for ADD or DELETE. Is the
> configuration wrong? Do I need to do more to set up the build environment
> than the simple install of full.exe? [Doesn't seem to be a CDK install
> anymore.]
>

The windows32api headers define a DELETE macro, which is obviously not
what RCS expects.

You may need to insert a #undef DELETE after the appropriate to nullify
the windows one and use RCS one.

Damn these namespace polluting macros in system headers.

My preference of course is to change the RCS macros/enums to use RCS_
prefix and be done with it.

Regards,
Mumit



--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe@sourceware.cygnus.com

Re: gcc build error--configuration problem?

[Mumit Khan]

"Josh Baudhuin" <joshb@Cadence.COM> writes:
> 
> I just installed the 'full' binary release of 20.1. Then I tried to build
> cvs-1.10 using the tar from the net.
> 
> The 'configure' script seemed to run w/o a hitch, but when I build, I get a
> compile error when it comes to src/rcs.c:


[ ... ]

> gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O -c rcs.c
> rcs.c: In function `apply_rcs_changes':
> rcs.c:6994: parse error before `('
> make[1]: *** [rcs.o] Error 1
> make: *** [all] Error 1
> 
[ ... ]

> 
> Seems like there must be a macro in there for ADD or DELETE. Is the
> configuration wrong? Do I need to do more to set up the build environment
> than the simple install of full.exe? [Doesn't seem to be a CDK install
> anymore.]
> 

The windows32api headers define a DELETE macro, which is obviously not
what RCS expects.

You may need to insert a #undef DELETE after the appropriate to nullify
the windows one and use RCS one.

Damn these namespace polluting macros in system headers.

My preference of course is to change the RCS macros/enums to use RCS_
prefix and be done with it.

Regards,
Mumit


--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe@sourceware.cygnus.com

gcc build error--configuration problem?

[Josh Baudhuin]

I just installed the 'full' binary release of 20.1. Then I tried to build
cvs-1.10 using the tar from the net.

The 'configure' script seemed to run w/o a hitch, but when I build, I get a
compile error when it comes to src/rcs.c:

/usr/local/src/cvs-1.10 <21> make
making all in lib
make[1]: Nothing to be done for `all'.
making all in zlib
make[1]: Nothing to be done for `all'.
making all in diff
make[1]: Nothing to be done for `all'.
making all in src
gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O -c rcs.c
rcs.c: In function `apply_rcs_changes':
rcs.c:6994: parse error before `('
make[1]: *** [rcs.o] Error 1
make: *** [all] Error 1

Cf. the source itself:
	. . .
6977
6978 static int
6979 apply_rcs_changes (lines, diffbuf, difflen, name, addvers, delvers)
6980      struct linevector *lines;
6981      const char *diffbuf;
6982      size_t difflen;
6983      const char *name;
6984      RCSVers *addvers;
6985      RCSVers *delvers;
6986 {
6987     const char *p;
6988     const char *q;
6989     int op;
6990     /* The RCS format throws us for a loop in that the deltafrags (if
6991        we define a deltafrag as an add or a delete) need to be applied
6992        in reverse order.  So we stick them into a linked list.  */
6993     struct deltafrag {
6994         enum {ADD, DELETE} type;
6995         unsigned long pos;
6996         unsigned long nlines;
6997         const char *new_lines;
6998         size_t len;
6999         struct deltafrag *next;
7000     };
7001     struct deltafrag *dfhead;
7002     struct deltafrag *df;
	. . .

Seems like there must be a macro in there for ADD or DELETE. Is the
configuration wrong? Do I need to do more to set up the build environment
than the simple install of full.exe? [Doesn't seem to be a CDK install
anymore.]


--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe@sourceware.cygnus.com

Re: Cygwin Gcc Compatibility with gcc on other platforms ??

[N8TM]

In a message dated 99-07-29 16:53:22 EDT, raman_bakshi@yahoo.com writes:

<< if i write and compile code
 using cygwin gcc (ANSI C code), will that code
 compile (not execute, thats another thing dependent
 on cpu etc) on platforms like linux etc where the GNU
 gcc compiler is available ? >>
Yes, if you truly restrict yourself to ANSI C and don't depend on libraries 
and non-ANSI include files which invariably differ from cygwin.  The cygwin 
developers have done an amazing job of making cygwin into a standard gnu 
compiler host, so you get excellent cross-platform compatibility.  This goal 
fortunately appears to be kept in mind by most of those who are working at 
the same time to get more Windoze compatibility.

--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe@sourceware.cygnus.com

Re: gcc build error--configuration problem?

[Earnie Boyd]

--- Mumit Khan <khan@xraylith.wisc.EDU> wrote:
> "Josh Baudhuin" <joshb@Cadence.COM> writes:
-8<-
> > gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O -c rcs.c
> > rcs.c: In function `apply_rcs_changes':
> > rcs.c:6994: parse error before `('
> > make[1]: *** [rcs.o] Error 1
> > make: *** [all] Error 1
> > 
> [ ... ]
> 
> > 
> > Seems like there must be a macro in there for ADD or DELETE. Is the
> > configuration wrong? Do I need to do more to set up the build environment
> > than the simple install of full.exe? [Doesn't seem to be a CDK install
> > anymore.]
> > 
> 
> The windows32api headers define a DELETE macro, which is obviously not
> what RCS expects.
> 
> You may need to insert a #undef DELETE after the appropriate to nullify
> the windows one and use RCS one.
> 
> Damn these namespace polluting macros in system headers.
> 
> My preference of course is to change the RCS macros/enums to use RCS_
> prefix and be done with it.

Yep.  Exactly what I did when I built it.  I might add that I've not had any
problems using it with text files.  I haven't had the need for binary files so
I can't say what will happen if you use the -b switch.  It appears to be coded
for it though.
===
Earnie Boyd < mailto:earnie_boyd@yahoo.com >

Newbies, please visit
< http://www.freeyellow.com/members5/gw32/index.html >

(If you respond to the list, then please don't cc me)
_____________________________________________________________
Do You Yahoo!?
Free instant messaging and more at http://messenger.yahoo.com


--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe@sourceware.cygnus.com

Re: gcc build error--configuration problem?

[Earnie Boyd]

--- Mumit Khan <khan@xraylith.wisc.EDU> wrote:
> "Josh Baudhuin" <joshb@Cadence.COM> writes:
-8<-
> > gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O -c rcs.c
> > rcs.c: In function `apply_rcs_changes':
> > rcs.c:6994: parse error before `('
> > make[1]: *** [rcs.o] Error 1
> > make: *** [all] Error 1
> > 
> [ ... ]
> 
> > 
> > Seems like there must be a macro in there for ADD or DELETE. Is the
> > configuration wrong? Do I need to do more to set up the build environment
> > than the simple install of full.exe? [Doesn't seem to be a CDK install
> > anymore.]
> > 
> 
> The windows32api headers define a DELETE macro, which is obviously not
> what RCS expects.
> 
> You may need to insert a #undef DELETE after the appropriate to nullify
> the windows one and use RCS one.
> 
> Damn these namespace polluting macros in system headers.
> 
> My preference of course is to change the RCS macros/enums to use RCS_
> prefix and be done with it.

Yep.  Exactly what I did when I built it.  I might add that I've not had any
problems using it with text files.  I haven't had the need for binary files so
I can't say what will happen if you use the -b switch.  It appears to be coded
for it though.
===
Earnie Boyd < mailto:earnie_boyd@yahoo.com >

Newbies, please visit
< http://www.freeyellow.com/members5/gw32/index.html >

(If you respond to the list, then please don't cc me)
_____________________________________________________________
Do You Yahoo!?
Free instant messaging and more at http://messenger.yahoo.com


--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe@sourceware.cygnus.com