henry@zoo.toronto.edu (Henry Spencer) (12/14/90)
Third of the catching-up patches, including a couple of files referenced by stuff in the second patch. This one is mostly work in the various auxiliary systems. The "batcher" program, or whatever is specified as batcher in batchparms, is now run in NEWSARTS rather than in out.going/xxx, while simplifies various things but just might break some custom batchers. (The -d option is no more, as a side effect.) There is a NEWSSITEDIR variable provided to batchers that really want to look in out.going/xxx. Expire understands "never" as a date in explist. Use of the new split() function speeds up expire. Better handling of some classes of errors in expire, including an attempt to deal with null message-IDs. Mkhistory sorts its output by arrival date for NNTP's benefit. Rnews.immed and rnews.batch have been unified into a single rnews, which looks for a flag file NEWSCTL/rnews.newsrun to tell it whether to fire up a newsrun or not, and when space is too tight to receive news, looks for another flag NEWSCTL/rnews.stall to decide whether to throw it away or stall. Newsrun now has a -h option to put processed batches into a hold directory, for debugging, and also nulls out the compress output if compress dies (in case compress has left trash). New program, addmissing, which sweeps the news tree for articles not in history and adds them -- this is both quicker and better than rebuilding the history database from scratch. New programs, histfrom and newsfrom, useful when one of your downstream sites says "uh... I lost everything from Tuesday to Thursday, could you re-send?". Some status reports from newsdaily now do ls -l rather than just ls. Various improvements in documentation, notably a checknews(1) manual page. start of patch 13-Dec-1990 (suggested archive name: `pch13Dec90.Z') this should be run with patch -p0 <thisfile Please do the following (there is no easy way we can convince patch to do this automatically): rm input/rnews.batch The following is a complete list of patches to date. Prereq: 23-Jun-1989 Prereq: 7-Jul-1989 Prereq: 23-Jul-1989 Prereq: 22-Aug-1989 Prereq: 24-Aug-1989 Prereq: 14-Sep-1989 Prereq: 13-Nov-1989 Prereq: 10-Jan-1990 Prereq: 16-Jan-1990 Prereq: 17-Jan-1990 Prereq: 18-Jan-1990 Prereq: 12-Mar-1990 Prereq: 14-Apr-1990 Prereq: 15-Apr-1990 Prereq: 16-Apr-1990 Prereq: 25-May-1990 Prereq: 1-Sep-1990 Prereq: 7-Sep-1990 Prereq: 1-Dec-1990 Prereq: 12-Dec-1990 *** PATCHDATES.old Thu Dec 13 13:23:06 1990 --- PATCHDATES Thu Dec 13 13:23:06 1990 *************** *** 1,20 **** --- 1,21 ---- 23-Jun-1989 7-Jul-1989 23-Jul-1989 22-Aug-1989 24-Aug-1989 14-Sep-1989 13-Nov-1989 10-Jan-1990 16-Jan-1990 17-Jan-1990 18-Jan-1990 12-Mar-1990 14-Apr-1990 15-Apr-1990 16-Apr-1990 25-May-1990 1-Sep-1990 7-Sep-1990 1-Dec-1990 12-Dec-1990 + 13-Dec-1990 Changed files, if any: *** cnpatch/old/batch/Makefile Thu Dec 13 13:23:16 1990 --- batch/Makefile Tue Dec 11 02:05:18 1990 *************** *** 97,103 **** echo 'ln $$* >/dev/null 2>/dev/null' >bin/newslock echo 'cat' >bin/munch ! echo '( echo --- ; cat ) >>output' >bin/xmit echo 'echo 2 ; cat' >bin/munch2 ! echo '( echo +++ ; cat ) >>output' >bin/xmit2 echo 'echo 4' >bin/queuelen echo 'echo 99' >bin/spacefor --- 97,103 ---- echo 'ln $$* >/dev/null 2>/dev/null' >bin/newslock echo 'cat' >bin/munch ! echo "( echo --- ; cat ) >>`pwd`/output" >bin/xmit echo 'echo 2 ; cat' >bin/munch2 ! echo "( echo +++ ; cat ) >>`pwd`/output2" >bin/xmit2 echo 'echo 4' >bin/queuelen echo 'echo 99' >bin/spacefor *************** *** 132,139 **** ( echo --- ; cat test.temp ) >>test.out1 cmp test.good test.out - ./batcher -d `pwd` togo >test.out - cmp test.good test.out rm test.out togo ! rm -rf out.going batchparms test.out test.good mkdir out.going out.going/foo out.going/bar echo 'foo 500 10 batcher munch xmit' >batchparms --- 132,137 ---- ( echo --- ; cat test.temp ) >>test.out1 cmp test.good test.out rm test.out togo ! rm -rf out.going batchparms test.out test.good output output2 mkdir out.going out.going/foo out.going/bar echo 'foo 500 10 batcher munch xmit' >batchparms *************** *** 150,155 **** >batchlog NEWSARTS=`pwd` NEWSCTL=`pwd` ./sendbatches ! test ! -s out.going/foo/output ; ! test ! -s out.going/bar/output ; echo `pwd`/test.1 200 >out.going/foo/togo echo `pwd`/test.2 200 >>out.going/foo/togo --- 148,153 ---- >batchlog NEWSARTS=`pwd` NEWSCTL=`pwd` ./sendbatches ! test ! -s output ; ! test ! -s output2 ; echo `pwd`/test.1 200 >out.going/foo/togo echo `pwd`/test.2 200 >>out.going/foo/togo *************** *** 160,165 **** echo `pwd`/test.3 4000 >test.left NEWSARTS=`pwd` NEWSCTL=`pwd` ./sendbatches ! cmp out.going/foo/output test.out1 ! cmp out.going/bar/output test.out2 test ! -s out.going/foo/togo ; test ! -s out.going/bar/togo ; --- 158,163 ---- echo `pwd`/test.3 4000 >test.left NEWSARTS=`pwd` NEWSCTL=`pwd` ./sendbatches ! cmp output test.out1 ! cmp output2 test.out2 test ! -s out.going/foo/togo ; test ! -s out.going/bar/togo ; *************** *** 171,174 **** --- 169,173 ---- cmp test.good test.log rm -rf test.* togo out.going bin batchparms batchlog batchlog.* + rm -f output output2 clean: *** cnpatch/old/batch/batcher.c Thu Dec 13 13:23:21 1990 --- batch/batcher.c Sun Oct 14 01:01:36 1990 *************** *** 2,14 **** * batcher - send a bunch of news articles as an unbatch script * ! * Usage: batcher [-d dir] listfile * * where listfile is a file containing a list, one per line, of ! * full pathnames of files containing articles. Only the first * field of each line is looked at, so there can be more if needed ! * for other things. ! * ! * The -d option specifies a directory where most articles are ! * likely to be; the program chdirs there to speed things up. */ --- 2,13 ---- * batcher - send a bunch of news articles as an unbatch script * ! * Usage: batcher listfile * * where listfile is a file containing a list, one per line, of ! * names of files containing articles. Only the first * field of each line is looked at, so there can be more if needed ! * for other things. Non-absolute pathnames are understood to lie ! * under the current directory; chdiring to the right place is the ! * parent's problem. */ *************** *** 27,32 **** char *progname; - char *dir = NULL; /* NULL means don't bother chdiring. */ - int dirlen; /* strlen(dir) */ int debug = 0; /* Debugging? */ --- 26,29 ---- *************** *** 44,53 **** progname = argv[0]; ! while ((c = getopt(argc, argv, "d:x")) != EOF) switch (c) { - case 'd': /* Directory containing many articles. */ - dir = optarg; - dirlen = strlen(dir); - break; case 'x': /* Debugging. */ debug++; --- 41,46 ---- progname = argv[0]; ! while ((c = getopt(argc, argv, "x")) != EOF) switch (c) { case 'x': /* Debugging. */ debug++; *************** *** 60,64 **** if (errflg || optind != argc-1) { (void) fprintf(stderr, ! "Usage: batcher [-d dir] listfile\n"); exit(2); } --- 53,57 ---- if (errflg || optind != argc-1) { (void) fprintf(stderr, ! "Usage: batcher listfile\n"); exit(2); } *************** *** 68,75 **** error("unable to open `%s'", argv[optind]); - if (dir != NULL) - if (chdir(dir) < 0) - error("can't chdir to `%s'", dir); - while ((article = fgetms(list)) != NULL) { process(article); --- 61,64 ---- *************** *** 86,104 **** */ process(article) ! char *article; { - char *p; register int artfile; register int count; struct stat sbuf; ! *(article + strcspn(article, "\n\t ")) = '\0'; ! if (dir != NULL && strncmp(article, dir, dirlen) == 0 && ! article[dirlen] == '/') ! p = article+dirlen+1; ! else ! p = article; ! artfile = open(p, 0); if (artfile < 0) { /* --- 75,94 ---- */ process(article) ! register char *article; { register int artfile; register int count; struct stat sbuf; + register char *endp; ! endp = strchr(article, '\t'); ! if (endp == NULL) ! endp = strchr(article, ' '); ! if (endp == NULL) ! endp = strchr(article, '\n'); ! if (endp != NULL) ! *endp = '\0'; ! artfile = open(article, 0); if (artfile < 0) { /* *************** *** 108,112 **** */ if (debug) ! warning("can't find `%s'", p); return; } --- 98,102 ---- */ if (debug) ! warning("can't find `%s'", article); return; } *************** *** 119,127 **** } ! (void) printf("#! rnews %ld\n", sbuf.st_size); ! fflush(stdout); ! while ((count = read(artfile, buffer, sizeof buffer)) > 0) ! if (write(1, buffer, count) != count) error("write failure in `%s'", article); if (count < 0) --- 109,115 ---- } ! printf("#! rnews %ld\n", sbuf.st_size); while ((count = read(artfile, buffer, sizeof buffer)) > 0) ! if (fwrite(buffer, sizeof(char), count, stdout) != count) error("write failure in `%s'", article); if (count < 0) *** cnpatch/old/batch/batchih Thu Dec 13 13:23:28 1990 --- batch/batchih Sun Oct 14 00:47:08 1990 *************** *** 8,18 **** umask $NEWSUMASK - case "$1" - in - -d) - shift ; shift # ignore -d option - ;; - esac - dest=`expr "$NEWSSITE" : "^\(.*\)\.[^.]*$"` case "$dest" --- 8,11 ---- *** cnpatch/old/batch/sendbatches Thu Dec 13 13:23:40 1990 --- batch/sendbatches Wed Nov 7 16:45:58 1990 *************** *** 78,81 **** --- 78,82 ---- PATH=$here:$origpath ; export PATH NEWSSITE=$sys ; export NEWSSITE # For site-specific programs. + NEWSSITEDIR=$here ; export NEWSSITEDIR # ditto # Is there anything to do? *************** *** 145,149 **** for f in $them do ! ( $batcher -d $NEWSARTS $f | $muncher | $sender $sys ) >/tmp/nb$$ 2>&1 if test $? -eq 0 -a ! -s /tmp/nb$$ --- 146,150 ---- for f in $them do ! ( ( cd $NEWSARTS ; $batcher $here/$f ) | $muncher | $sender $sys ) >/tmp/nb$$ 2>&1 if test $? -eq 0 -a ! -s /tmp/nb$$ *** cnpatch/old/expire/expire.c Thu Dec 13 13:23:54 1990 --- expire/expire.c Tue Dec 11 13:38:53 1990 *************** *** 11,14 **** --- 11,15 ---- #include <string.h> #include <errno.h> + #include "fixerrno.h" #include <time.h> #include <signal.h> *************** *** 23,30 **** #include "dbz.h" #ifndef EPOCH ! #define EPOCH ((time_t)0) #endif ! #define DAY ((double)24*60*60) --- 24,34 ---- #include "dbz.h" + /* basic parameters of time, used in back() (and debugging code) */ #ifndef EPOCH ! #define EPOCH ((time_t)0) /* the origin of time_t */ #endif ! #ifndef FOREVER ! #define FOREVER 1e37 /* FOREVER > max value of time_t */ ! #endif #define DAY ((double)24*60*60) *************** *** 109,113 **** char *progname; - extern int errno; extern long atol(); extern double atof(); --- 113,116 ---- *************** *** 134,137 **** --- 137,141 ---- char *doarticle(); void warning(); + void complain(); void printstuff(); void expire(); *************** *** 282,286 **** errno = 0; ! nf = split(ctl, field, 4, "\t "); if (nf != 4) fail("control line `%.20s...' hasn't got 4 fields", ctl); --- 286,290 ---- errno = 0; ! nf = split(ctl, field, 4, ""); if (nf != 4) fail("control line `%.20s...' hasn't got 4 fields", ctl); *************** *** 308,327 **** switch (ndates) { case 3: ! ct->retain = back(atof(dates[0])); ! ct->normal = back(atof(dates[1])); ! ct->purge = back(atof(dates[2])); break; case 2: ! ct->retain = (bounds != NULL) ? bounds->retain : back(0.0); ! ct->normal = back(atof(dates[0])); ! ct->purge = back(atof(dates[1])); break; case 1: ! ct->retain = (bounds != NULL) ? bounds->retain : back(0.0); ! ct->normal = back(atof(dates[0])); ! ct->purge = (bounds != NULL) ? bounds->purge : EPOCH; break; default: ! fail("date processing foulup in `%s'", field[1]); /* NOTREACHED */ break; --- 312,331 ---- switch (ndates) { case 3: ! ct->retain = back(dates[0]); ! ct->normal = back(dates[1]); ! ct->purge = back(dates[2]); break; case 2: ! ct->retain = (bounds != NULL) ? bounds->retain : back("0"); ! ct->normal = back(dates[0]); ! ct->purge = back(dates[1]); break; case 1: ! ct->retain = (bounds != NULL) ? bounds->retain : back("0"); ! ct->normal = back(dates[0]); ! ct->purge = (bounds != NULL) ? bounds->purge : back("never"); break; default: ! fail("invalid date specification `%s'", field[2]); /* NOTREACHED */ break; *************** *** 373,377 **** af = eufopen(afile, "r"); while ((line = fgetms(af)) != NULL) { ! nf = split(line, field, NFACT, " \t"); if (nf != NFACT) fail("wrong number of fields in active for `%s'", field[0]); --- 377,381 ---- af = eufopen(afile, "r"); while ((line = fgetms(af)) != NULL) { ! nf = split(line, field, NFACT, ""); if (nf != NFACT) fail("wrong number of fields in active for `%s'", field[0]); *************** *** 550,557 **** if (nf != 3 && nf != 2) { free(work); ! errno = 0; ! warning("wrong number of fields in `%.40s...'", line); return(line); /* leaving the line in the new history file */ } if (nf == 2) field[2] = NULL; --- 554,569 ---- if (nf != 3 && nf != 2) { free(work); ! complain("wrong number of fields in history `%.40s...'", line); return(line); /* leaving the line in the new history file */ } + if (*field[0] == '\0') { + complain("no message-ID in history `%.40s...' -- expiring", line); + if (strlen(field[1]) < 3) { + complain("\toops -- can't -- fix manually", ""); + free(work); + return(line); /* leaving the line in the new history file */ + } else + sprintf(field[1], "0%c0", *subsep); + } if (nf == 2) field[2] = NULL; *************** *** 564,570 **** expdate = readdate(subfield[1]); if (expdate == NODATE && getdgrump) { ! errno = 0; ! warning("bad expiry date in `%.40s...',", line); ! warning(" specifically, `%s' -- ignored", subfield[1]); } } --- 576,581 ---- expdate = readdate(subfield[1]); if (expdate == NODATE && getdgrump) { ! complain("bad expiry date in `%.40s...',", line); ! complain(" specifically, `%s' -- ignored", subfield[1]); } } *************** *** 572,577 **** if (recdate == NODATE) { free(work); ! errno = 0; ! warning("bad arrival date in `%.40s...' -- fix by hand", line); return(line); } --- 583,587 ---- if (recdate == NODATE) { free(work); ! complain("bad arrival date in `%.40s...' -- fix by hand", line); return(line); } *************** *** 1005,1031 **** /* - - split - divide a line into fields, like awk split() - */ - int /* number of fields */ - split(line, fields, nfmax, sep) - char *line; - char *fields[]; - int nfmax; - char *sep; - { - register int i; - register char *p; - - i = 0; - for (p = strtok(line, sep); p != NULL; p = strtok((char *)NULL, sep)) { - if (i < nfmax) - fields[i] = p; - i++; - } - - return(i); - } - - /* - eufopen - fopen, with fail if doesn't succeed */ --- 1015,1018 ---- *************** *** 1098,1107 **** */ time_t ! back(ndays) ! double ndays; { ! if (ndays > now / DAY) /* past beginning of time */ ! return((time_t)0); ! return(now - ndays*DAY); } --- 1085,1101 ---- */ time_t ! back(ndaystr) ! char *ndaystr; { ! register double goback; /* how far before now it is */ ! ! if (STREQ(ndaystr, "never")) ! goback = FOREVER; /* > now-EPOCH */ ! else ! goback = atof(ndaystr) * DAY; ! ! if (goback > now-EPOCH) /* before EPOCH */ ! return(EPOCH); ! return((time_t)(now - goback)); } *** cnpatch/old/expire/mkhistory Thu Dec 13 13:24:12 1990 --- expire/mkhistory Mon Dec 10 18:16:40 1990 *************** *** 30,34 **** cd $NEWSARTS ! find `ls | egrep -v '\.'` -type f -name '[0-9]*' -print | histinfo | sort | awk -f $NEWSBIN/expire/histdups | histslash >$NEWSCTL/history.n --- 30,35 ---- cd $NEWSARTS ! find `ls | egrep -v '\.'` -type f -name '[0-9]*' -print | histinfo | ! sort -t' ' +1n | awk -f $NEWSBIN/expire/histdups | histslash >$NEWSCTL/history.n *** cnpatch/old/expire/updatemin.c Thu Dec 13 13:24:18 1990 --- expire/updatemin.c Sat Dec 8 23:19:41 1990 *************** *** 7,10 **** --- 7,12 ---- #include <string.h> #include <signal.h> + #include <errno.h> + #include "fixerrno.h" #ifndef BERKDIR *************** *** 71,75 **** cd(fullartfile((char *)NULL)); while ((line = fgetms(a)) != NULL) { ! nf = split(line, field, NF, " \t\n"); if (nf != NF) nfail("bad number of fields in `%s ...'", field[0]); --- 73,77 ---- cd(fullartfile((char *)NULL)); while ((line = fgetms(a)) != NULL) { ! nf = split(line, field, NF, ""); if (nf != NF) nfail("bad number of fields in `%s ...'", field[0]); *************** *** 138,164 **** return(min); - } - - /* - - split - divide a line into fields, like awk split() - */ - int /* number of fields */ - split(line, fields, nfmax, sep) - char *line; - char *fields[]; - int nfmax; - char *sep; - { - register int i; - register char *p; - - i = 0; - for (p = strtok(line, sep); p != NULL; p = strtok((char *)NULL, sep)) { - if (i < nfmax) - fields[i] = p; - i++; - } - - return(i); } --- 140,143 ---- *** cnpatch/old/input/Makefile Thu Dec 13 13:24:21 1990 --- input/Makefile Mon Sep 10 18:26:05 1990 *************** *** 1,9 **** - # You get your choice of rnews.immed or rnews.batch; rnews.immed tries - # to start processing immediately, while rnews.batch waits for somebody - # else (cron) to do it. Running rnews.immed might, perhaps, be reasonable - # if your news load is light and you are wildly impatient about processing - # incoming news. Otherwise, leave this alone. - RNEWS = rnews.batch - DEFINES = COPTS = -O --- 1,2 ---- *************** *** 21,26 **** # workaround for System V make bug SHELL = /bin/sh - DTR = README Makefile newsrun newsrunning newsspool.c c7decode.c rnews.batch \ - rnews.8 bdecode.c recenews recpnews all: $(THEM) --- 14,17 ---- *************** *** 96,108 **** echo 'here is a phony third batch' >>$@ - rnews.immed: rnews.batch - sed '/qqq/s/newsspool/& -i/' rnews.batch >$@ - - rnews: rnews.batch rnews.immed - cp $(RNEWS) rnews - chmod +x rnews - setup: all ! chmod +x rnews.batch rnews.immed rm -rf bin mkdir bin --- 87,92 ---- echo 'here is a phony third batch' >>$@ setup: all ! chmod +x rnews rm -rf bin mkdir bin *************** *** 121,131 **** r: all test.1 test.2 test.3 test.3c test.out setup chmod +x $(THEM) ! NEWSARTS=`pwd` NEWSCTL=`pwd` ./rnews.batch <test.1 cmp in.coming/* test.1 ! NEWSARTS=`pwd` NEWSCTL=`pwd` ./rnews.batch <test.2 sed 1d test.2 >tmp.2 cmp `ls -t in.coming | sed -n '1s;^;in.coming/;p'` tmp.2 rm tmp.2 ! NEWSARTS=`pwd` NEWSCTL=`pwd` ./rnews.batch <test.3 cmp `ls -t in.coming | sed -n '1s;^;in.coming/;p'` test.3c NEWSARTS=`pwd` NEWSCTL=`pwd` ./newsrunning off --- 105,115 ---- r: all test.1 test.2 test.3 test.3c test.out setup chmod +x $(THEM) ! NEWSARTS=`pwd` NEWSCTL=`pwd` ./rnews <test.1 cmp in.coming/* test.1 ! NEWSARTS=`pwd` NEWSCTL=`pwd` ./rnews <test.2 sed 1d test.2 >tmp.2 cmp `ls -t in.coming | sed -n '1s;^;in.coming/;p'` tmp.2 rm tmp.2 ! NEWSARTS=`pwd` NEWSCTL=`pwd` ./rnews <test.3 cmp `ls -t in.coming | sed -n '1s;^;in.coming/;p'` test.3c NEWSARTS=`pwd` NEWSCTL=`pwd` ./newsrunning off *************** *** 138,144 **** test " `echo in.coming/*`" = ' in.coming/bad' ; rm tmp.1 ! NEWSARTS=`pwd` NEWSCTL=`pwd` ./rnews.batch <test.1 ! NEWSARTS=`pwd` NEWSCTL=`pwd` ./rnews.batch <test.2 ! NEWSARTS=`pwd` NEWSCTL=`pwd` NEWSBIN=`pwd`/bin ./rnews.immed <test.3 cmp tmp.1 test.out test " `echo in.coming/*`" = ' in.coming/bad' ; --- 122,129 ---- test " `echo in.coming/*`" = ' in.coming/bad' ; rm tmp.1 ! NEWSARTS=`pwd` NEWSCTL=`pwd` ./rnews <test.1 ! NEWSARTS=`pwd` NEWSCTL=`pwd` ./rnews <test.2 ! echo >rnews.newsrun ! NEWSARTS=`pwd` NEWSCTL=`pwd` NEWSBIN=`pwd`/bin ./rnews <test.3 cmp tmp.1 test.out test " `echo in.coming/*`" = ' in.coming/bad' ; *************** *** 147,154 **** clean: ! rm -f *.o newsspool c7decode tmp.? test.* dtr lint rnews rnews.immed rm -f bdecode rm -rf in.coming bin - - dtr: $(DTR) - makedtr $(DTR) >dtr --- 132,136 ---- clean: ! rm -f *.o newsspool c7decode tmp.? test.* dtr lint rnews.newsrun rm -f bdecode rm -rf in.coming bin *** cnpatch/old/input/README Thu Dec 13 13:24:24 1990 --- input/README Mon Sep 10 19:00:14 1990 *************** *** 6,14 **** the relay subsystem for processing. Newsrunning can be used to turn this function on and off if you want to avoid news processing during busy hours. ! ! Using rnews.immed instead of rnews.batch as rnews will arrange for newsrun ! to be run after each newsspool -- this is expensive but cuts down latency. ! If you've got a fast machine and a light newsfeed and totally lack patience, ! like Geoff :-), this might be a good idea; otherwise, no. "make r" builds everything and runs a full regression test on most everything. --- 6,11 ---- the relay subsystem for processing. Newsrunning can be used to turn this function on and off if you want to avoid news processing during busy hours. ! There is a configuration option to arrange for newsrun to be run after ! each newsspool -- this is expensive but cuts down latency. "make r" builds everything and runs a full regression test on most everything. *** cnpatch/old/input/newsrun Thu Dec 13 13:24:28 1990 --- input/newsrun Tue Sep 18 16:20:54 1990 *************** *** 8,11 **** --- 8,16 ---- umask $NEWSUMASK + hold= + case "$1" in + -h) hold=y ;; + esac + here="$NEWSARTS/in.coming" cd $here *************** *** 86,89 **** --- 91,100 ---- fi + # Save a copy in hold if requested. + if test " $hold" = " y" -a -d hold + then + ln $f hold/$f + fi + # Decompress if necessary. People who get lots of # uncompressed batches and never use c7 encoding might *************** *** 97,100 **** --- 108,112 ---- : okay else + >$text # compress might have left garbage text=$f fi *** cnpatch/old/input/newsspool.c Thu Dec 13 13:24:36 1990 --- input/newsspool.c Sun Oct 28 02:18:58 1990 *************** *** 4,8 **** * The -i option relies on the parent setting (and exporting) $PATH. * ! * $Log$ */ --- 4,8 ---- * The -i option relies on the parent setting (and exporting) $PATH. * ! * -Log- */ *************** *** 15,18 **** --- 12,19 ---- #include <string.h> #include <errno.h> + #include "fixerrno.h" + #ifndef EEXIST + #define EEXIST 0 + #endif #include "libc.h" #include "news.h" *************** *** 198,202 **** register int ntries; time_t now; - extern int errno; if (nfclose(f) == EOF) --- 199,202 ---- *** cnpatch/old/man/expire.8 Thu Dec 13 13:24:50 1990 --- man/expire.8 Sun Oct 28 02:33:45 1990 *************** *** 7,11 **** .\" =()<.ds m @<NEWSMASTER>@>()= .ds m usenet ! .TH EXPIRE 8 "12 May 1990" .BY "C News" .SH NAME --- 7,11 ---- .\" =()<.ds m @<NEWSMASTER>@>()= .ds m usenet ! .TH EXPIRE 8 "28 Oct 1990" .BY "C News" .SH NAME *************** *** 305,306 **** --- 305,312 ---- Protection against old articles reappearing is thus somewhat limited for a while after the history file is rebuilt. + .PP + .I Expire + uses + .IR access (2) + to test for the presence of archiving directories, + which can cause anomalies if it is run setuid (normally it's not). *** cnpatch/old/man/news.5 Thu Dec 13 13:24:56 1990 --- man/news.5 Fri Sep 14 11:29:05 1990 *************** *** 29,33 **** .LP .. ! .TH NEWS 5 "31 Aug 1990" .BY "C News" .SH NAME --- 29,33 ---- .LP .. ! .TH NEWS 5 "12 Sept 1990" .BY "C News" .SH NAME *************** *** 116,120 **** Organization: Bell Labs, Murray Hill .sp 0.3 ! The body of the article comes here, after a blank line. .Ee A --- 116,120 ---- Organization: Bell Labs, Murray Hill .sp 0.3 ! The body of the article comes here, after an empty line. .Ee A *************** *** 169,173 **** .B # as the first character in a line denotes a comment. ! Blank lines are ignored. A logical line may be continued to the next physical line by putting a --- 169,173 ---- .B # as the first character in a line denotes a comment. ! Empty lines are ignored. A logical line may be continued to the next physical line by putting a *** cnpatch/old/man/newsbatch.8 Thu Dec 13 13:25:08 1990 --- man/newsbatch.8 Wed Nov 7 16:48:06 1990 *************** *** 7,11 **** .\" =()<.ds m @<NEWSMASTER>@>()= .ds m usenet ! .TH NEWSBATCH 8 "12 May 1990" .BY "C News" .SH NAME --- 7,11 ---- .\" =()<.ds m @<NEWSMASTER>@>()= .ds m usenet ! .TH NEWSBATCH 8 "7 Nov 1990" .BY "C News" .SH NAME *************** *** 31,52 **** .br .B \&.../batcher - [ - .B \-d - directory - ] listfile .br .B \&.../batchih - [ - .B \-d - directory - ] listfile .br .B \&.../batchsm - [ - .B \-d - directory - ] listfile .br --- 31,40 ---- *************** *** 117,121 **** be appended to a \fItogo\fR file in its \fIsite\fR directory. The batcher expects the lines in \fItogo\fR to have two fields, a filename ! (as a full pathname) of an article and its size in bytes. A missing size field is arbitrarily assumed to be a default average. .PP --- 105,110 ---- be appended to a \fItogo\fR file in its \fIsite\fR directory. The batcher expects the lines in \fItogo\fR to have two fields, a filename ! (as a full pathname, or relative to \fI\*a\fR) ! of an article and its size in bytes. A missing size field is arbitrarily assumed to be a default average. .PP *************** *** 126,132 **** This permits per-site and per-news-database overrides of the default behaviors. ! \fISendbatches\fR provides all these programs with an environment ! variable \fBNEWSSITE\fR, containing the name of the site that batches ! are being prepared for, in case this is useful in customization. .PP The names of most of the auxiliary programs, and some other parameters, --- 115,124 ---- This permits per-site and per-news-database overrides of the default behaviors. ! \fISendbatches\fR provides all these programs with environment ! variables \fBNEWSSITE\fR, containing the name of the site that batches ! are being prepared for, ! and \fBNEWSSITEDIR\fR, containing the full pathname of the \fIsite\fR ! directory, ! in case these are useful in customization. .PP The names of most of the auxiliary programs, and some other parameters, *************** *** 172,175 **** --- 164,171 ---- needed, and a batch transmitter (typically \fIviauux\fR), which sends the batch on its way (e.g. enqueues it for transmission). + The preparer is run with \fI\*a\fR as the current directory, + so non-absolute pathnames in the chunk are valid filenames; + the others are run in the \fIsite\fR directory for easier access to + site-specific customization files. .PP Batch preparers in the standard distribution are: *************** *** 178,193 **** normal batching .IP batchih ! \fIihave\fR part of ihave/sendme .IP batchsm \fIsendme\fR part of ihave/sendme .RE - .PP - If \fIbatcher\fR is invoked with a \fB\-d\fR option, it \fIchdir\fRs to - the \fIdirectory\fR first, and strips it off the beginning of any article - filename that starts with it. - Since most articles will come from \fI\*a\fR, specifying - \fB\-d\ \*a\fR to \fIbatcher\fR will speed it up noticeably. - Since \fIsendbatches\fR does this, other preparers had better be ready - to ignore an initial \fB\-d\fR option. .PP .I Batchih --- 174,181 ---- normal batching .IP batchih ! \fIihave\fR part of old ihave/sendme (not to be confused with NNTP) .IP batchsm \fIsendme\fR part of ihave/sendme .RE .PP .I Batchih *** cnpatch/old/man/newsmaint.8 Thu Dec 13 13:25:20 1990 --- man/newsmaint.8 Sun Oct 14 04:17:08 1990 *************** *** 7,11 **** .\" =()<.ds m @<NEWSMASTER>@>()= .ds m usenet ! .TH NEWSMAINT 8 "8 May 1990" .BY "C News" .SH NAME --- 7,11 ---- .\" =()<.ds m @<NEWSMASTER>@>()= .ds m usenet ! .TH NEWSMAINT 8 "13 Oct 1990" .BY "C News" .SH NAME *************** *** 27,30 **** --- 27,34 ---- .br act.to.times \- create active.times file for news readers + .br + histfrom, newsfrom \- list news arriving over specific time range + .br + addmissing \- add missing news articles to history file .SH SYNOPSIS .B \*b/maint/newshist *************** *** 57,60 **** --- 61,72 ---- .B \*b/maint/act.to.times activefile + .br + .B \*b/maint/histfrom + startdate [ enddate ] + .br + .B \*b/maint/newsfrom + startdate [ enddate ] + .br + .B \*b/maint/addmissing .SH DESCRIPTION These programs are utilities useful in maintaining a C News *************** *** 134,137 **** --- 146,182 ---- which should normally be .IR \*c/active . + .PP + .I Histfrom + emits (on standard output) all history lines for unexpired articles + arriving between + .I startdate + and + .IR enddate . + The dates must be acceptable to + .I getdate + (see + .IR newsaux (8)). + The default + .I enddate + is ``now''. + .PP + .I Newsfrom + runs + .I histfrom + and processes its output to produce a list of absolute pathnames of + articles arriving between + .I startdate + and + .I enddate + (default ``now''). + .PP + .I Addmissing + scans the + .I history + file and the news-article tree to find articles that are not in the history + file, and adds them to the file, + rebuilding the + .I dbz + database to match. .SH FILES .ta 6c *************** *** 138,143 **** .nf \*c/history history file ! \*c/history.pag \fIdbm\fR database for history file ! \*c/history.dir \fIdbm\fR database for history file \*c/log current news log \*c/log.o previous news log --- 183,188 ---- .nf \*c/history history file ! \*c/history.pag \fIdbz\fR database for history file ! \*c/history.dir \fIdbz\fR database for history file \*c/log current news log \*c/log.o previous news log *************** *** 165,168 **** --- 210,224 ---- the \fIbatchparms\fR file, and if none is found, warns the user to create one. + .PP + .I Addmissing + complains and exits if it is unable to lock the news system, if + .I expire + appears to be running, + or if there is not enough space for a temporary copy of the + .I history + file and database. + It also notes the presence of articles which did not have intelligible + headers (which are placed into history with an entry specifying immediate + expiry). .SH HISTORY Written at U of Toronto by Henry Spencer and Geoff Collyer. *************** *** 179,182 **** --- 235,245 ---- to get batching going for a new feed, which is naive if you are running \fIsendbatches\fR selectively, i.e. with arguments. + .PP + .I Addmissing + does not cope properly with articles which are + already in history but with an incomplete + list of pathnames; + the result is duplicate message-IDs in history until the article(s) expire, + which is relatively harmless but may interfere with access to those articles. .PP Various nuisances can result if the maintenance utilities are run as *** cnpatch/old/man/postnews.1 Thu Dec 13 13:25:29 1990 --- man/postnews.1 Wed Nov 21 15:09:32 1990 *************** *** 1,5 **** .\" =()<.ds c @<NEWSCTL>@>()= .ds c /usr/lib/news ! .TH POSTNEWS 1 "9 June 1989" .BY "C News" .SH NAME --- 1,5 ---- .\" =()<.ds c @<NEWSCTL>@>()= .ds c /usr/lib/news ! .TH POSTNEWS 1 "21 Nov 1990" .BY "C News" .SH NAME *************** *** 7,11 **** .SH SYNOPSIS .B postnews ! [newsgroup] .SH DESCRIPTION .I Postnews --- 7,14 ---- .SH SYNOPSIS .B postnews ! [ ! .B \-h ! headerfile ! ] [newsgroup] .SH DESCRIPTION .I Postnews *************** *** 24,27 **** --- 27,36 ---- followed, \fIpostnews\fR gives the user a choice between abandoning the posting or re-editing it. + .PP + The + .B \-h + option specifies that a suitable header has already been constructed and + can be found in \fIheaderfile\fR, in which case \fIpostnews\fR bypasses + the initial prompting for header contents. .SH FILES \*c/postdefltgroup default newsgroup (nonexistent means no default) *** cnpatch/old/man/rnews.8 Thu Dec 13 13:25:34 1990 --- man/rnews.8 Tue Sep 18 17:44:22 1990 *************** *** 7,11 **** .\" =()<.ds m @<NEWSMASTER>@>()= .ds m usenet ! .TH RNEWS 8 "3 Nov 1989" .BY "C News" .SH NAME --- 7,11 ---- .\" =()<.ds m @<NEWSMASTER>@>()= .ds m usenet ! .TH RNEWS 8 "18 Sept 1990" .BY "C News" .SH NAME *************** *** 111,112 **** --- 111,117 ---- It hasn't been tested very well in the C News environment. \fIBdecode\fR (see \fInewsmail\fR(8)) is probably better for most uses. + .PP + .I Relaynews + is a bit too enthusiastic about declaring batches bad, so things end up + in \fIbad\fR for no terribly good reason. + Actually, it's rare for the contents of \fIbad\fR to be very interesting. *** cnpatch/old/misc/Makefile Thu Dec 13 13:25:39 1990 --- misc/Makefile Sun Oct 14 04:09:01 1990 *************** *** 12,16 **** MAINTBIN=newshist ! MAINT = $(MAINTBIN) newsdaily newswatch newsboot locknews addgroup delgroup adddirs addfeed act.to.times UTILBIN = gngp newslock ctime getdate canonhdr UTILS = $(UTILBIN) sizeof newshostname --- 12,17 ---- MAINTBIN=newshist ! MAINT = $(MAINTBIN) newsdaily newswatch newsboot locknews addgroup delgroup \ ! adddirs addfeed act.to.times histfrom newsfrom addmissing UTILBIN = gngp newslock ctime getdate canonhdr UTILS = $(UTILBIN) sizeof newshostname *** cnpatch/old/misc/ctime.c Thu Dec 13 13:25:48 1990 --- misc/ctime.c Sun Oct 14 03:56:24 1990 *************** *** 15,24 **** char *progname; - extern int errno; - extern char *strcpy(); - extern char *strncpy(); - extern char *strcat(); - extern char *strchr(); - extern char *strtok(); extern long atol(); extern char *malloc(); --- 15,18 ---- *** cnpatch/old/misc/getdate.c Thu Dec 13 13:25:53 1990 --- misc/getdate.c Sat Dec 1 19:12:39 1990 *************** *** 12,24 **** struct timeb ftnow; char *progname; - extern int errno; - extern char *strcpy(); - extern char *strncpy(); - extern char *strcat(); - extern char *strchr(); - extern char *strtok(); extern long atol(); extern char *malloc(); --- 12,19 ---- struct timeb ftnow; + int exitstatus = 0; char *progname; extern long atol(); extern char *malloc(); *************** *** 60,64 **** for (; optind < argc; optind++) process(argv[optind]); ! exit(0); } --- 55,59 ---- for (; optind < argc; optind++) process(argv[optind]); ! exit(exitstatus); } *************** *** 72,79 **** time_t it; ! it = getdate(tm, &ftnow); ! if (it < 0) ! (void) fprintf(stderr, "%s: `%s' not a valid date\n", progname, tm); else (void) printf("%ld\n", it); } --- 67,78 ---- time_t it; ! if (strcmp(tm, "now") == 0) ! it = time((time_t *)NULL); else + it = getdate(tm, &ftnow); + if (it < 0) { + (void) fprintf(stderr, "%s: `%s' not a valid date\n", progname, tm); + exitstatus = 1; + } else (void) printf("%ld\n", it); } *** cnpatch/old/misc/newsdaily Thu Dec 13 13:26:00 1990 --- misc/newsdaily Sun Oct 7 19:01:52 1990 *************** *** 58,62 **** ( echo 'old input files:' ! cat $tmp echo ) >>$gripes --- 58,62 ---- ( echo 'old input files:' ! ls -ldtr `cat $tmp` echo ) >>$gripes *************** *** 66,71 **** then ( ! echo 'recent bad input batches:' ! cat $tmp echo ) >>$gripes --- 66,71 ---- then ( ! echo 'recent bad input batches (perhaps worth investigation):' ! ls -ldtr `cat $tmp` echo ) >>$gripes *** cnpatch/old/notebook/ihave Thu Dec 13 13:26:23 1990 --- notebook/ihave Wed Sep 19 17:07:54 1990 *************** *** 17,20 **** --- 17,27 ---- unlike NNTP. .PP + The above perhaps needs emphasizing: + \fIihave/sendme has nothing whatsoever to do with NNTP\fR. + NNTP has ``ihave'' and ``sendme'' messages, which have somewhat similar + functions to the ihave/sendme protocol (hence the similarity of names), + but the implementations are completely unrelated. + If you want to set up an NNTP feed, you are reading the wrong document. + .PP Ihave/sendme is sketched in extremely vague terms in *** cnpatch/old/notebook/problems Thu Dec 13 13:26:37 1990 --- notebook/problems Mon Dec 10 12:50:24 1990 *************** *** 1,3 **** ! .DA "21 July 1990" .TL Known Porting Problems With C News --- 1,3 ---- ! .DA "10 Dec 1990" .TL Known Porting Problems With C News *************** *** 83,86 **** --- 83,114 ---- environment variables) is exactly the wrong size. Perhaps this has been fixed in 4.1. + .PP + It is reliably reported that the VAX 3.1 Ultrix shell is somewhat broken + and gives various kinds of trouble. + Switching to \fI/bin/sh5\fR (note that this requires fixing the first line + of a zillion shell files) is reported to banish the problems. + .PP + In some environments, when processing batches which are not compressed, + extremely mysterious malfunctions occur and messages reporting a + \fIrelaynews\fR exit status of 1005 (and the like) are seen. + This is believed to be a subtle shell bug that shows up + when a pipeline is used as the condition in a shell \fIif\fR. + Unless you use \fIc7encode\fRd transmission\(emwhich is very rare\(ema + workaround is to comment out + the following lines in \fIinput/newsrun\fR: + .DS + elif c7decode <$f 2>/dev/null | compress \-d >$text 2>/dev/null + then + : okay + .DE + A minor bonus is that this will improve processing speed (!). + We plan improvements to input processing, and this problem should be + eliminated as a side effect of that. + .PP + (Note, this is \fInot\fR the well-known bug of many old shells, + in which the choice of process whose exit status is picked up is + semi-random in such situations. + Exercise for the reader: show that + \fIthat\fR is not a problem in \fInewsrun\fR.) .SH Make vs. Test *************** *** 148,176 **** always safe. .SH - Readnews vs. System V - .PP - The terminal handling in the Australian \fIreadnews\fR that we include - is known not to work under a ``pure'' System V. - There are stubs for System V code but it's never been implemented. - Most current System V implementations from serious suppliers have enough - BSD compatibility that this is not an issue. - It is reported that AT&T 3B2 systems don't, to the point where - \fIreadnews\fR simply doesn't work. - .SH - Readnews Simplicity - .PP - Generally, the Australian \fIreadnews\fR - is a bit on the simple and buggy side. - There are various things wrong with it that we're not very interested - in fixing\(emwe consider it useful primarily for naive users who just - want to read system announcements, and recommend that sophisticated - users use something else. - One known interoperability issue is that it is fussier about the format - of \fI.newsrc\fR files than fancier readers. - .PP - Problems will arise if the news directories are in non-standard places, - because \fIreadnews\fR is not well integrated into our configuration - machinery; we want to fix this someday. - .SH void .PP --- 176,179 ---- *************** *** 287,294 **** Some issues inevitably got resolved in favor of current compilers, and may cause complaints from ANSI C compilers. .SH GNU C .PP ! If compiling with the GNU compiler, you will need the \fB\-traditional\fR flag. Beware, also, that if you are using your system's \fIdbm\fR library, --- 290,304 ---- Some issues inevitably got resolved in favor of current compilers, and may cause complaints from ANSI C compilers. + .PP + Work is in progress on moving us closer to ANSI compatibility. + Beware that if \fB__STDC__\fR is defined + by your compiler but it is \fInot\fR ANSI compatible, you are on your + own as far as we're concerned, even if the value is specified as 0. + (We can't just use ``#if\ __STDC__'' because some preprocessors choke + on the appearance of an unknown identifier in #if.) .SH GNU C .PP ! If compiling with the GNU compiler, you may need the \fB\-traditional\fR flag. Beware, also, that if you are using your system's \fIdbm\fR library, *************** *** 299,303 **** \fB\-fpcc-struct-return\fR option cures this. .SH ! Eccentric Awks .PP A number of problems can arise if your \fIawk\fR has bugs, since the shell --- 309,313 ---- \fB\-fpcc-struct-return\fR option cures this. .SH ! Awk Problems .PP A number of problems can arise if your \fIawk\fR has bugs, since the shell *************** *** 306,309 **** --- 316,337 ---- We've fixed the worst trouble spots, but would appreciate detailed information on any more. + .PP + One known problem that is hard to avoid is line-length limits in \fIawk\fR. + In particular, + for several purposes in control-message + handling C News wants to build a ``canonical'' representation + of the \fIsys\fR file, with backslashed newlines removed. + This is done by + NEWSBIN\fI/relay/canonsys.awk\fR. + Most \fIawk\fRs have limits + on line length, and some of the limits are too low to cope with long + multiply-continued \fIsys\fR lines. + 512-byte limits, found in a number of old \fIawk\fRs, are particularly + troublesome. + A quick look indicates that this will interfere, to some uncertain extent, + with \fIcheckgroups\fR and \fIsendsys\fR. + Big deal. :-) + The complaint may also appear from + \fInewgroup\fR, but there it should be harmless. .SH Systems Without Hard Links *************** *** 520,523 **** --- 548,560 ---- .DE This command should output nothing. + .PP + A slightly related problem is that not everyone calls the run-remote-shell + command \fIrsh\fR; on System V in particular, \fIrsh\fR means something + different. + For the moment we have opted to ignore this issue, as the possibilities + for gratuitous differences boggle the mind. + People facing this problem may wish to place an + \fIrsh\fR shell file in the search path to invoke the right command in the + right way, whatever that is. .SH Values of Logical Operators *************** *** 529,533 **** *p++ = isascii(c) && isalnum(c); .DE ! because they don't know how to generate a numeric value for `&&'. One or two places in C News use constructs like this. If you run into this, you might --- 566,571 ---- *p++ = isascii(c) && isalnum(c); .DE ! because they don't know how to generate a numeric value for `&&', ! or because they don't know how to turn that value into a `char'. One or two places in C News use constructs like this. If you run into this, you might *************** *** 536,537 **** --- 574,619 ---- with something like ``(...)\ ?\ 1\ :\ 0'' to get the troublesome operator back into a conditional context. + .SH + Empty Lines + .PP + Some backward operating systems (through which your C News distribution + may have passed on its way to you), and perhaps some stupid text-handling + software even on sane operating systems, do not recognize the notion of + an empty line. + They think all lines must have at least one character in them; the closest + they can come to an empty line is a line consisting of a single blank. + This matters because \fIrelaynews\fR will tolerate white space only in + certain places in the \fIsys\fR file, and in particular, it tolerates + empty lines but not lines consisting solely of white space. + The result will be complaints (in \fIerrlog\fR) about white space in + the \fIsys\fR line for a system named ``\ ''. + .SH + active-File Date + .\" thanks to Simon J. Gerraty for this weirdie + .PP + On some Bull systems, at least ones running DPX/2 B.O.S. 1.0, + apparently \fIrelaynews\fR updates the + contents of the \fIactive\fR file correctly, but the file's date remains + unchanged! + This appears to be a kernel bug. + It reportedly upsets some news readers. + A workaround, said to be effective, is to add the line + .DS + utime(ctlfile(actrelnm), (time_t *)NULL); + .DE + after the call to \fInnafree\fR in \fIactfsync\fR + in \fIlibbig/active.big.c\fR. + .SH + enum Operators + .\" thanks to Scott Reynolds for this one + .PP + Some compilers have difficulty compiling the + \fIreadnews\fR we supply, + because they object to applying the `!' operator to an \fIenum\fR type. + Changing the definition of \fIbooltype\fR in \fIrna/defs.h\fR to + .DS + typedef int bool; + #define false 0 + #define true 1 + .DE + is reported to solve this. Files that are new: new input/rnews (patch can't create, so diff against null): Index: input/rnews *** cnpatch/old/input/rnews Thu Dec 13 13:26:53 1990 --- input/rnews Mon Sep 10 18:58:14 1990 *************** *** 0 **** --- 1,51 ---- + #! /bin/sh + # Incoming-news spooling. + # We ignore arguments -- it looks tempting to put "$*" after cat and + # newsspool, but there are security problems. + + # =()<. ${NEWSCONFIG-@<NEWSCONFIG>@}>()= + . ${NEWSCONFIG-/usr/lib/news/bin/config} + + PATH=$NEWSCTL/bin:$NEWSBIN/input:$NEWSBIN:$NEWSPATH ; export PATH + umask $NEWSUMASK + + # check space, assuming a pretty large batch (no cheap way to find real size) + counter=1 + while test " `spacefor 250000 incoming`" -le 0 + do + sleep 300 + if test " $counter" -gt 1111 # four tries is plenty + then + # oh no! -- nothing we can do, really... + cat >/dev/null + echo incoming news discarded due to space shortage | + mail "$NEWSMASTER" + exit 1 + fi + if test ! -f $NEWSCTL/rnews.stall + then + counter="1$counter" + fi + done + + # spool it + if test -f $NEWSCTL/rnews.newsrun + then + opt=-i + else + opt= + fi + if newsspool $opt >/tmp/ngripe.$$ 2>&1 + then + rm -f /tmp/ngripe.$$ + exit 0 + else + # there really isn't any way to save the data if newsspool fails, + # not without causing other problems + ( + echo newsspool failed!!! + cat /tmp/ngripe.$$ + ) | mail "$NEWSMASTER" + rm -f /tmp/ngripe.$$ + exit 1 + fi new man/checknews.1 (patch can't create, so diff against null): Index: man/checknews.1 *** cnpatch/old/man/checknews.1 Thu Dec 13 13:26:54 1990 --- man/checknews.1 Fri Nov 23 15:50:46 1990 *************** *** 0 **** --- 1,42 ---- + .\" =()<.ds c @<NEWSCTL>@>()= + .ds c /usr/lib/news + .TH CHECKNEWS 1 "23 Nov 1990" + .BY "C News" + .SH NAME + checknews \- check for unread news + .SH SYNOPSIS + .B checknews + .SH DESCRIPTION + .I Checknews + informs the user whether he has unread news. + It is suitable for running from a + .B .profile + (or + .B .login + if you use a certain non-standard shell) + file. + .PP + .I Checknews + normally runs + .IR readnews (1) + with the + .B \-c + option. + As a heuristic concession to practicality, + if the user has a + .B .rnlast + file in his $HOME directory, + .I checknews + runs + .B rn\ \-c\ \-s3 + instead. + .SH FILES + $HOME/.newsrc + .br + \*c/active + .SH SEE ALSO + readnews(1), rn(1) + .SH HISTORY + Written at U of Toronto by Henry Spencer. + .SH BUGS + There are zillions of other news readers in the world. new misc/addmissing (patch can't create, so diff against null): Index: misc/addmissing *** cnpatch/old/misc/addmissing Thu Dec 13 13:26:55 1990 --- misc/addmissing Mon Oct 15 17:50:23 1990 *************** *** 0 **** --- 1,56 ---- + #! /bin/sh + # addmissing - add missing articles to history + # =()<. ${NEWSCONFIG-@<NEWSCONFIG>@}>()= + . ${NEWSCONFIG-/usr/lib/news/bin/config} + + PATH=$NEWSCTL/bin:$NEWSBIN/expire:$NEWSBIN:$NEWSPATH ; export PATH + umask $NEWSUMASK + + lock="$NEWSCTL/LOCK" + ltemp="$NEWSCTL/L.$$" + echo $$ >$ltemp + trap "rm -f $ltemp ; exit 0" 0 1 2 15 + while true + do + if newslock $ltemp $lock + then + trap "rm -f $ltemp $lock ; exit 0" 0 1 2 15 + break + fi + sleep 30 + done + if newslock $ltemp $NEWSCTL/LOCKexpire + then + trap "rm -f $ltemp $lock $NEWSCTL/LOCKexpire ; exit 0" 0 1 2 15 + else + echo "$0: expire is running, $0 aborted" >&2 + exit 1 + fi + + cd $NEWSCTL + size="`sizeof history history.pag history.dir`" + if test " `spacefor $size control`" -le 0 + then + echo "$0: not enough space for temporaries, $0 aborted" >&2 + exit 1 + fi + + egrep ' .* ' history | sed 's/.* //' | tr '. ' '/ + ' | egrep '/' | sort -u >/tmp/hist$$ + + cd $NEWSARTS + find `ls | egrep -v '\.'` -type f -name '[0-9]*' -print | sort >/tmp/tree$$ + + comm -13 /tmp/hist$$ /tmp/tree$$ | histinfo | sort | + awk -f $NEWSBIN/expire/histdups | histslash >/tmp/new$$ + + cd $NEWSCTL + if egrep '^<[^>]*@trash> ' /tmp/new$$ >/dev/null + then + echo "$0: (warning) empty/trash articles found, will expire at once" >&2 + echo "$0: (grep history file for '@trash' to see them)" >&2 + fi + + dbz -a history /tmp/new$$ + + rm -f /tmp/hist$$ /tmp/tree$$ /tmp/new$$ new misc/histfrom (patch can't create, so diff against null): Index: misc/histfrom *** cnpatch/old/misc/histfrom Thu Dec 13 13:26:56 1990 --- misc/histfrom Mon Sep 24 17:07:48 1990 *************** *** 0 **** --- 1,33 ---- + #! /bin/sh + # histfrom - get history-file entries between two dates + + # =()<. ${NEWSCONFIG-@<NEWSCONFIG>@}>()= + . ${NEWSCONFIG-/usr/lib/news/bin/config} + + PATH=$NEWSCTL/bin:$NEWSBIN:$NEWSPATH ; export PATH + umask $NEWSUMASK + + case $# in + 1) start="$1" + end="now" + ;; + 2) start="$1" + end="$2" + ;; + *) echo "Usage: $0 startdate [enddate]" >&2 + exit 2 + ;; + esac + if start=`/bin/getdate "$start"` && end=`/bin/getdate "$end"` + then + : okay + else + exit 1 + fi + + egrep ' .* ' $NEWSCTL/history | awk -F' ' '{ + n = split($2, subf, "~") + arrived = subf[1] + if (arrived >= '$start' && arrived <= '$end') + print + }' new misc/newsfrom (patch can't create, so diff against null): Index: misc/newsfrom *** cnpatch/old/misc/newsfrom Thu Dec 13 13:26:56 1990 --- misc/newsfrom Mon Sep 24 17:07:48 1990 *************** *** 0 **** --- 1,10 ---- + #! /bin/sh + # newsfrom - get pathnames of articles arriving between two dates + + # =()<. ${NEWSCONFIG-@<NEWSCONFIG>@}>()= + . ${NEWSCONFIG-/usr/lib/news/bin/config} + + PATH=$NEWSCTL/bin:$NEWSBIN/maint:$NEWSBIN:$NEWSPATH ; export PATH + umask $NEWSUMASK + + histfrom "$@" | tr . / | sed 's/.* //;s/ .*//;s:^:'"$NEWSARTS"'/:' end of patch 13-Dec-1990 -- "The average pointer, statistically, |Henry Spencer at U of Toronto Zoology points somewhere in X." -Hugh Redelmeier| henry@zoo.toronto.edu utzoo!henry
ronald@robobar.co.uk (Ronald S H Khoo) (12/14/90)
henry@zoo.toronto.edu (Henry Spencer) writes: > *** cnpatch/old/input/newsspool.c Thu Dec 13 13:24:36 1990 > --- input/newsspool.c Sun Oct 28 02:18:58 1990 > *************** > *** 4,11 **** > * The -i option relies on the parent setting (and exporting) $PATH. > * Patch seems to prefer that the 4,11 should be changed to read 4,8. I'd be *real* interested to know how it got to be 4,11 in the first place! -- ronald@robobar.co.uk +44 81 991 1142 (O) +44 71 229 7741 (H)
henry@zoo.toronto.edu (Henry Spencer) (12/15/90)
In article <1990Dec14.140903.28607@robobar.co.uk> ronald@robobar.co.uk (Ronald S H Khoo) writes: >Patch seems to prefer that the 4,11 should be changed to read 4,8. See the corrected version of the patch. >I'd be *real* interested to know how it got to be 4,11 in the first place! Basically, hand adjustment to cope with some difficulties with RCS. I got the other places where this occurred but missed this one. -- "The average pointer, statistically, |Henry Spencer at U of Toronto Zoology points somewhere in X." -Hugh Redelmeier| henry@zoo.toronto.edu utzoo!henry