From owner-freebsd-ports Sun Jul 12 06:42:36 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id GAA26250 for freebsd-ports-outgoing; Sun, 12 Jul 1998 06:42:36 -0700 (PDT) (envelope-from owner-freebsd-ports@FreeBSD.ORG) Received: from ady.warpnet.ro (ady.warpnet.ro [193.230.201.1]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id GAA26230 for ; Sun, 12 Jul 1998 06:42:11 -0700 (PDT) (envelope-from ady@warpnet.ro) Received: from localhost (ady@localhost) by ady.warpnet.ro (8.8.8/8.8.8) with SMTP id QAA00619; Sun, 12 Jul 1998 16:41:12 +0300 (EEST) (envelope-from ady@warpnet.ro) Date: Sun, 12 Jul 1998 16:41:12 +0300 (EEST) From: Adrian Penisoara Reply-To: Adrian Penisoara To: Андрей Чернов cc: FreeBSD ports Subject: Pine 3.96/4.00 ports -- patches Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-ports@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi, [ CC:-ed to freebsd-ports in case someone else has suggestions too ] I will try to discuss patches migration from Pine 3.96 to Pine 4.00 in the port I submitted in ports/7248: There are 17+1 patches (patch-aa ... patch-aq + patch-ar). patch-aa (in pine3, NOT in pine4) --------------------------------- * Pine 3.96: creates pico/makefile.bsf * Pine 4.00: sources already come with an pico/makefile.bsf that seems to me that it does (almost ?) the same job - there is a issue wih -DTERMCAP_WINS: it breaks the building if I try to use it: cc -fpic -DPIC -c -O -DDEBUG -DBSDI -DBSDI2 -DPOSIX -DJOB_CONTROL -DMOUSE -DTERMCAP_WINS attach.c -o attach.so In file included from attach.c:30: headers.h:24: parse error before `not' In file included from /usr/include/sys/types.h:49, from /usr/include/stdio.h:44, from headers.h:29, from attach.c:30: /usr/include/machine/types.h:43: warning: data definition has no type or storage class *** Error code 1 Stop. What does -DTERMCAP_WINS do ? You say "Trust termcap more then built-in defaults" in the CVS log -- please explain me what it does and how can we convince Pine to build it correctly. patch-ab (in pine3, in pine4) ----------------------------- * Pine 3.96: file pico/os_unix.c, patches return expression in errstr() - casting (char *) + reinforces operator precedence (!) * Pine 4.00: file pico/osdep/unix - errstr() already has (char *) casting and I believe that operator precedence reinforcement is not needed; - patches fexist() call with (long *) cast in ffelbowroom(), conforming fexist()'s prototype; patch-ac (in pine3, in pine4) ----------------------------- * Pine 3.96: creates pine/osdep/os-bsf.h * Pine 4.00: file pine/osdep/os-bsf.h, which already comes with source - changes DEFAULT_DEBUG from 2 to 0 (in order to not generate ~/.pine-debug# files) XXX - [ #define BACKGROUND_POST ] XXX - SENDNEWS -- what is the correct path for 'inews' ? - part of the (3.96) file was moved to pico/osdep/os-bsf.h patch-ad (in pine3, NOT in pine4) --------------------------------- * Pine 3.96: creates pine/osdep/os-bsf.ic * Pine 4.00: sources already come with pine/osdep/os-bsf.ic - I don't see differences that should cause problems (be carefull some names of the includers have changed) patch-ae (in pine3, NOT in pine4) --------------------------------- * Pine 3.96: creates pine/makefile.bsf * Pine 4.00: sources already come with a pine/makefile.bsf file - "STDLIBS= -ltermlib" doesn't need '-lcrypt' because LIBS includes `cat $(CCLIENTDIR)/LDFLAGS` (which in turn has -lcrypt); in fact Pine itself shouldn't have anything to do with -lcrypt, only the c-client library should need -lcrypt. - again, -DTERMCAP_WINS issues ! patch-af (in pine3, NOT in pine4) --------------------------------- * Pine 3.96: file imap/ANSI/c-client/Makefile * Pine 4.00: file should be imap/src/osdep/unix/Makefile - Oops, this should have been the content of patch-ag !! patch-ag (in pine3, in pine4) ----------------------------- * Pine 3.96: file imap/ANSI/c-client/mh.c - fixes some incorrect scandir() calls by inserting (void *) cast ? man 3 scandir says it's int (*compar)(const void *, const void *) both in 2.2.6 and 3.0, and mh_numsort()'s prototype really conforms ! So no patch is needed, really... * Pine 4.00: file should have been imap/src/osdep/unix/mh.c but in fact file is imap/src/osdep/unix/Makefile (my fault, see patch-af...) XXX - comments out 'CC=' & 'CFLAGS='; but really shouldn't this be 'CFLAGS+=...' ??; I've mangled 'BASECFLAGS=' at the bsf target for that reason; I will correct this in my next patch... patch-ah (in pine3, NOT in pine4) --------------------------------- * Pine 3.96: file imap/ANSI/c-client/news.c - same bogus fix like path-ag ?!? Again, news_numsort() prototype conforms to what man 3 scandir says, so no patch needed. * Pine 4.00: file should have been imap/src/osdep/unix/news.c patch-ai (in pine3, in pine4) ----------------------------- * Pine 3.96: file pine/send.c - forces ENC8BIT when TYPETEXT body * Pine 4.00: file pine/send.c - maintains the patch, just some reformatting patch-aj (in pine3, in pine4) ----------------------------- * Pine 3.96: file imap/ANSI/c-client/os_bsi.h - includes instead of when in FreeBSD (but doesn't OpenBSD/NetBSD confom too ?) * Pine 4.00: file imap/src/osdep/unix/os_bsi.h - maintains the patch; same observation patch-ak (in pine3, in pine4) ----------------------------- * Pine 3.96: file imap/Makefile - introduce separate bsf target, excludes ipopd/imapd from compiling * Pine 4.00: file imap/Makefile - same basic ideea, only we already are included the same general building target patch-al (in pine3, in pine4) ----------------------------- * Pine 3.96: file build - fix for not building ipopd/imapd daemons, corrects clean target * Pine 4.00: file build - same ideea, but comment instead deleting; maintained "size ..." but adjusted patch-am (in pine3, NOT in pine4) --------------------------------- * Pine 3.96: file pine/pine.h - from CVS log: "Move TAG_EMBED from valid Russian letter '\377' to more neutral '\001'" and "Tag quoting not implemented properly cause '\377' chars skipped" * Pine 4.00: file should be pine/pine.h XXX - my opinion: good reason, but I don't like the patch, it's rather a workaround than a good fix, it conflicts with TAG_INVON; we should ask the Pine people about this; I will insert the patch anyway (but probably with a different constant) ! patch-an (in pine3, in pine4) ----------------------------- * Pine 3.96: file pico/os_unix.h - fixes MAILDIR * Pine 4.00: file pico/osdep/os-bsf.h special for FreeBSD - includes instead of - strange, it still needs MAILDIR fix ?!? :) patch-ao & patch-ap (in pine3, NOT in pine4) --------------------------------- * Pine 3.96: file pico/tcap.c & pine/ttyout.c - CVS log: "Teach pine/pico about End key" -- sorry, but it doesn't work for me, it just says "[Invalid key pressed. Use ? for help]" Does it work for you ? * Pine 4.00: file should be /pico/osdep/term.cap - the new file has a new structure; it seems that it will accept KEY_END, but it's related to termcap_wins/-DTERMCAP_WINS ?!? It doesn't work now... patch-aq (in pine3, NOT in pine4) --------------------------------- * Pine 3.96: files pico/ebind.h & pico/main.h - disables spelling * Pine 4.00: files pico/ebind.h & pico/main.h XXX - I thought we had a spellchecker in FreeBSD, but it seems not; and you say ispell doesn't work... Anyway, I believe The Right Way(tm) to do this is to patch pico/osdep/os-bsf.h and undefine SPELLER ! I'll make the change in the next patch. patch-ar (NOT in pine3, in pine4) --------------------------------- * Pine 3.96: no file * Pine 4.00: file imap/src/osdep/unix/unix.c - Restrictive permissions don't allow locking in /var/mail/, so pine was complainig with "Mailbox vulnerable - error creating ..."; however this is not The Right Way(tm) to do it, but until I find a better solution... So, in the end, the port isn't 100% finished but it compiles and runs flawlessly. I need help with -DTERMCAP_WINS and the other termcap issues ! Thanks ! Ady (@freebsd.ady.ro) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-ports" in the body of the message