/* ** This software is Copyright (c) 1989, 1990, 1991 by Kent Landfield. ** ** Permission is hereby granted to copy, distribute or otherwise ** use any part of this package as long as you do not try to make ** money from it or pretend that you wrote it. This copyright ** notice must be maintained in any copy made. ** */ #if !defined(lint) && !defined(SABER) static char SID[] = "@(#)header.c 2.3 5/9/91"; #endif #include #include #include #include "article.h" int fprintf(); int its(s1) register char *s1; { int strlen(); int strncmp(); if (strncmp(s,s1,strlen(s1)) == 0) return(TRUE); return(FALSE); } int line_type() { if (its("Path: ")) return PATH; if (its("From: ")) return FROM; if (its("Newsgroups: ")) return NEWSGROUP; if (its("Subject: ")) return SUBJECT; if (its("Keywords: ")) return KEYWORDS; if (its("Date: ")) return DATE; if (its("Message-ID: ")) return MSG_ID; if (its("Lines: ")) return NUMLINES; if (its("Approved: ")) return APPROVED; /* The following are the auxilliary headers used by */ /* the moderators of the sources groups. In some */ /* cases, line checks are done with "historical" */ /* auxilliary headers. Most of the moderators have */ /* now standardized on comp.sources.unix's format. */ /* Archive header lines for comp.sources.unix */ /* for comp.sources.amiga, comp.sources.atari.st, */ /* comp.sources.misc, and comp.sources.x. */ if (its("Submitted-by: ")) return SUBMITTED_BY; if (its("Posting-number: ")) return POSTING_NUMBER; if (its("Archive-name: ")) return ARCH_NAME; /* Auxiliary header used as a backward reference */ /* to the location of the initially posted sources */ /* in the event of a patch. This line only exists */ /* if the current article is a patch. */ if (its("Patch-To: ")) return PATCH_TO; if (its("Patch-to: ")) return PATCH_TO; /* The X-Checksum-Snefru archive was being used by */ /* comp.sources.unix and comp.sources.misc to verify */ /* articles. Once snefru was broken, the moderators */ /* of c.s.u and c.s.m switched to use X-Md4-Signature: */ /* which is based on MD4 (RFC1186). */ if (its("X-Checksum-Snefru: ")) return X_CHECKSUM; if (its("X-Md4-Signature: ")) return X_CHECKSUM; /* The following lines are used by comp.sources.games */ /* currently to indicate the environmental keywords */ /* indicating what is required by the software to run. */ /* Supersedes: is also used by comp.sources.misc to */ /* indicate when a posting totally replaces another. */ if (its("Environment: ")) return ENVIRONMENT; if (its("Supersedes: ")) return SUPERSEDES; /* Archive header lines for historical purposes */ /* once used in comp.sources.misc */ if (its("Submitted-By: ")) return SUBMITTED_BY; if (its("comp.sources.misc: ")) return POSTING_NUMBER; if (its("Archive-Name: ")) return ARCH_NAME; /* Archive header lines for supporting the */ /* uunet comp.sources.misc article archives */ if (its("subject: ")) return SUBJECT; if (its("from: ")) return FROM; /* Archive header lines used in comp.sources.games */ /* Archive-name is the same as comp.sources.unix */ if (its("Submitted by: ")) return SUBMITTED_BY; if (its("Comp.sources.games: ")) return POSTING_NUMBER; /* Archive header lines for historical purposes */ /* once used in mod.sources articles. */ if (its("Mod.sources: ")) return POSTING_NUMBER; /* The following lines are used by comp.archives */ /* to as auxiliary header lines. Archive-name */ /* is also used. */ if (its("Original-posting-by: ")) return ORIGINAL_POSTING_BY; if (its("Original-subject: ")) return ORIGINAL_SUBJECT; if (its("Archive-site: ")) return ARCHIVE_SITE; if (its("Archive-directory: ")) return ARCHIVE_DIRECTORY; if (its("Reposted-by: ")) return REPOSTED_BY; if (its("Archive: ")) return ARCHIVE_; /* The following lines are used by comp.sources.apple2 */ /* to indicate the type of the system the software */ /* is to run on and the version of the software. */ if (its("Architecture: ")) return ARCHITECTURE; if (its("Version-number: ")) return VERSION_NUMBER; /* The remainder are other types of lines included */ /* headers and are includes for formatting output */ /* and for potential future use. */ if (its("References: ")) return REFERENCES; if (its("Organization: ")) return ORGANIZATION; if (its("Distribution: ")) return DISTRIBUTION; if (its("Xref: ")) return XREF; if (its("Expires: ")) return EXPIRE; if (its("Article-I.D.: ")) return ARTICLEID; if (its("Reply-To: ")) return REPLY_TO; if (its("Control: ")) return CONTROL; if (its("Sender: ")) return SENDER; if (its("Followup-To: ")) return FOLLOWUP_TO; if (its("Summary: ")) return SUMMARY; return OTHER; } void data(hpfield, size, fldname, header_id) char *hpfield; int size; char *fldname; int header_id; { register char *ptr; register char *p; char *strncpy(); char *strcpy(); char *strchr(); for (ptr = strchr(s, ':'); isspace(*++ptr); ) ; if (*ptr != '\0') { (void) strncpy(hpfield, ptr, size - 1); /* * Strip trailing newlines, blanks, and tabs from hpfield. */ for (p = hpfield; *p; ++p) ; while (--p >= hpfield && (*p == '\n' || *p == ' ' || *p == '\t')) ; *++p = '\0'; } (void) strcpy(header.header[header_id], hpfield); if (debug) (void) fprintf(logfp,"%s: %s\n",fldname, hpfield); } void dump_article() { char *type_str; switch(article.rectype) { case PATCH : type_str = "PATCH"; break; case ADMINISTRATION: type_str = "ADMINISTRATION"; break; case INFORMATIONAL : type_str = "INFORMATIONAL"; break; default : type_str = "NORMAL"; break; } (void) fprintf(logfp,"Article: [%s]\n",article.newsarticle); (void) fprintf(logfp," newsgroup: [%s]\n",article.newsgroup); (void) fprintf(logfp," filename: [%s]\n",article.filename); (void) fprintf(logfp," volume: [%d]\n",article.volume); (void) fprintf(logfp," issue: [%d]\n",article.issue); (void) fprintf(logfp," record type: [%s]\n", type_str); if (article.rectype == PATCH) { (void) fprintf(logfp," patch volume: [%d]\n",article.patch_volume); (void) fprintf(logfp," patch issue: [%s]\n",article.patch_issue); } (void) fprintf(logfp," reposted: [%s]\n", article.repost ? "YES": "NO"); (void) fprintf(logfp," description: [%s]\n",article.description); (void) fprintf(logfp," author's name: [%s]\n",article.author_name); (void) fprintf(logfp," author's logon: [%s]\n\n",article.author_signon); } void init_article() { article.newsgroup[0] = '\0'; article.newsarticle[0] = '\0'; article.filename[0] = '\0'; article.volume = -1; article.issue = -1; article.rectype = NORMAL; article.repost = FALSE; article.patch_volume = -1; article.package_name[0] = '\0'; article.patch_issue[0] = '\0'; article.description[0] = '\0'; article.author_name[0] = '\0'; article.author_signon[0] = '\0'; header.from[0] = '\0'; /* From: */ header.path[0] = '\0'; /* Path: */ header.nbuf[0] = '\0'; /* Newsgroups: */ header.subject[0] = '\0'; /* Subject: */ header.ident[0] = '\0'; /* Message-ID: */ header.replyto[0] = '\0'; /* Reply-To: */ header.references[0] = '\0'; /* References: */ header.subdate[0] = '\0'; /* Date: (submission) */ header.subtime = 0; /* subdate in secs */ header.expdate[0] = '\0'; /* Expires: */ header.ctlmsg[0] = '\0'; /* Control: */ header.sender[0] = '\0'; /* Sender: */ header.followup_to[0] = '\0'; /* Followup-to: */ header.distribution[0] = '\0'; /* Distribution: */ header.organization[0] = '\0'; /* Organization: */ header.numlines[0] = '\0'; /* Lines: */ header.intnumlines = 0; /* Integer Version */ header.keywords[0] = '\0'; /* Keywords: */ header.summary[0] = '\0'; /* Summary: */ header.approved[0] = '\0'; /* Approved: */ header.xref[0] = '\0'; /* Xref: */ header.supersedes[0] = '\0'; /* Supersedes: */ header.submitted_by[0] = '\0'; /* Submitted_by: */ header.posting_num[0] = '\0'; /* Posting-number: */ header.archive_name[0] = '\0'; /* Archive-name: */ header.patch_to[0] = '\0'; /* Patch-To: */ header.x_checksum[0] = '\0'; /* X-Checksum-Snefru: or */ /* X-Md4-Signature */ header.orig_poster[0] = '\0'; /* Original-posting-by: */ header.orig_subject[0] = '\0'; /* Original-subject: */ header.archive_site[0] = '\0'; /* Archive-site: */ header.archive_dir[0] = '\0'; /* Archive-directory: */ header.reposter[0] = '\0'; /* Reposted-by: */ header.architecture[0] = '\0'; /* Architecture: */ header.version_number[0] = '\0'; /* Version-number: */ header.archive[0] = '\0'; /* Archive: */ header.environment[0] = '\0'; /* Environment: */ } void store_line() { int atoi(); char *strchr(), *strcpy(), *strstrip(), *substr(); char *dp, *sp; char wrk[256]; switch(line_type()) { case PATH: /*PATH REQUIRED************************/ data(header.path, sizeof(header.path), "PATH:", PATH); break; case FROM: /*FROM REQUIRED************************/ data(header.from, sizeof(header.from), "FROM:", FROM); break; case NEWSGROUP: /*NEWSGROUP REQUIRED**************/ data(header.nbuf, sizeof(header.nbuf), "NEWSGROUPS:", NEWSGROUP); if ((sp = strchr(s,':')) != NULL) { do { ++sp; } while(!isalpha(*sp)); /* remove all crossposting labels */ if ((dp = strchr(sp,',')) != NULL) *dp = '\0'; (void) strcpy(article.newsgroup,sp); } break; case SUBJECT: /*SUBJECT REQUIRED******************/ data(header.subject, sizeof(header.subject), "SUBJECT:", SUBJECT); /* ** Save the subject as the description for articles ** that have no volume/issue format. Later in the ** code, this subject line minus the volume/issue ** is stored back in the .description element if ** the volume/issue indicator is found. */ (void) strcpy(article.description, header.subject); /* ** Check to see if this article is a repost of ** a previously posted article. */ if (substr(s, "REPOST") != NULL) article.repost = TRUE; /* ** Time to get the filename. Assure that it is in a ** volume/issue (v01INF1 or v01ADM1 or v01SRC001 or ** v01i001) format. */ if ((sp = strchr(s,'v')) == NULL) return; /* no volume indicator */ /* ** Is there a number that follows ** the volume indicator ? */ if (*(sp+1) < '0' || *(sp+1) > '9') return; /* The volume number is missing */ /* ** Is there a second ':' as well ? */ (void) strcpy(article.filename,sp); if ((dp = strchr(article.filename,':')) == NULL) return; /* not in v01i001: format */ /* ** terminate the article's filename and ** store the article's description */ *dp = '\0'; (void) strcpy(article.description, strstrip(++dp)); /* ** Store the filename in a work ** buffer so I can stomp on it. */ (void) strcpy(wrk, article.filename); ++sp; /* increment past the 'v' */ /* ** This is an informational posting. */ if ((dp = substr(wrk, "INF")) != NULL) { article.rectype = INFORMATIONAL; article.issue = atoi((dp+3)); } /* ** This is an administration posting. */ else if ((dp = substr(wrk, "ADM")) != NULL) { article.rectype = ADMINISTRATION; article.issue = atoi((dp+3)); } /* ** This is a comp.sources.apple2 posting. ** Why this is needed, I don't know... */ else if ((dp = substr(wrk, "SRC")) != NULL) { article.issue = atoi((dp+3)); } /* ** check to see if there is an issue indicator */ else if ((dp = strchr(wrk,'i')) != NULL) { /* parse the issue from the filename */ article.issue = atoi(dp+1); } else return; /* proven guilty. not volume/issue format */ /* parse the volume from the filename */ *dp = '\0'; article.volume = atoi(sp); break; case DATE: data(header.subdate, sizeof(header.subdate), "DATE:", DATE); break; case EXPIRE: data(header.expdate, sizeof(header.expdate), "EXPIRES:", EXPIRE); break; case MSG_ID: data(header.ident, sizeof(header.ident), "MESSAGE-ID:", MSG_ID); break; case REPLY_TO: data(header.replyto, sizeof(header.replyto), "REPLY-TO:", REPLY_TO); break; case REFERENCES: data(header.references, sizeof(header.references), "REFERENCES:", REFERENCES); break; case SENDER: data(header.sender, sizeof(header.sender), "SENDER:", SENDER); break; case FOLLOWUP_TO: data(header.followup_to, sizeof(header.followup_to),"FOLLOWUP-TO:", FOLLOWUP_TO); break; case CONTROL: data(header.ctlmsg, sizeof(header.ctlmsg),"CONTROL:", CONTROL); break; case DISTRIBUTION: data(header.distribution, sizeof(header.distribution),"DISTRIBUTION:", DISTRIBUTION); break; case ORGANIZATION: data(header.organization, sizeof(header.organization),"ORGANIZATION:", ORGANIZATION); break; case NUMLINES: data(header.numlines, sizeof(header.numlines),"LINES:", NUMLINES); header.intnumlines = atoi(header.numlines); break; case KEYWORDS: data(header.keywords, sizeof(header.keywords), "KEYWORDS:",KEYWORDS); break; case APPROVED: data(header.approved, sizeof(header.approved), "APPROVED:",APPROVED); break; case ENVIRONMENT: data(header.environment, sizeof(header.environment),"ENVIRONMENT:",ENVIRONMENT); break; case SUPERSEDES: data(header.supersedes, sizeof(header.supersedes),"SUPERSEDES:",SUPERSEDES); break; case XREF: data(header.xref, sizeof(header.xref),"XREF:",XREF); break; case SUMMARY: data(header.summary, sizeof(header.summary),"SUMMARY:", SUMMARY); break; case POSTING_NUMBER: data(header.posting_num, sizeof(header.posting_num), "POSTING_NUMBER:", POSTING_NUMBER); /* ** Well we almost have a standard here... ** ** comp.sources.amiga, comp.sources.atari.st, comp.sources.games ** comp.sources.sun, comp.sources.x, comp.sources.unix all use ** Posting-Number: lines as follows: ** Volume #, Issue # ** indicates a source posting while ** Volume #, Info # ** is used for all the periodic and administration postings. ** For informational only... c.s.u once used ** Volume #, Number # ** to indicate a source posting. Really. Check volume 10... ** ** comp.sources.misc uses the same Posting-number: lines for ** source postings but used ** Volume #, Administrivia # ** for all the periodic and administration postings. ** In the future, comp.sources.misc will use the Info format ** in order to move towards a standard. ** ** comp.sources.apple2 uses the following Posting-number: ** lines for the INF, source and Administration postings. ** Volume 1, Information: 1 ** Volume 1, Source:1 ** Volume 1, Administrivia: 17 ** ** If the volume number and issue cannot be determined from ** the subject line then lets give it a try here... */ if (article.volume == -1 || article.issue == -1) { if ((sp = strchr(s,'V')) != NULL) { do { ++sp; } while((*sp) && (!isdigit(*sp))); /* Is there something there... */ if (!*sp) return; /* ** Time to get the volume number.. ** Save it off so that the string can ** be stomped on.. What I am going to do here ** is to assure that I have both a volume and ** an issue number prior to setting any variables ** the describe the article type... */ (void) strcpy(wrk,sp); if ((dp = strchr(wrk,',')) != NULL) *dp++ = '\0'; else /* bad format, bail out... */ return; /* ** Now let's position to the type of the posting.. */ while((*dp) && (!isalpha(*dp))) ++dp; if (!*dp) /* Is there something there... */ return; /* ** Now it is time to get the Info,Issue,Administrivia, ** Information or Source number.... */ sp = dp; do { ++sp; } while((*sp) && (!isdigit(*sp))); if (!*sp) /* Is there something there... */ return; /* ** We have a seemingly valid Posting-number: line. ** Now we can start setting values... */ article.volume = atoi(wrk); article.issue = atoi(sp); /* ** Determine the type of the posting and ** set the rectype variable. */ if (strncmp(dp,"Info",4) == 0) /* covers c.s.u and c.s.a2 as well */ article.rectype = INFORMATIONAL; if (strncmp(dp,"Administrivia",4) == 0) { if (strcmp(article.newsgroup,"comp.sources.misc") == 0) article.rectype = INFORMATIONAL; else /* comp.sources.apple2 */ article.rectype = ADMINISTRATION; } /* ** fill in the volume/issue filename */ if (strcmp(article.newsgroup,"comp.sources.apple2") == 0) { if (article.rectype == INFORMATIONAL) (void) sprintf(article.filename,"v%3.3dINF%3.3d", article.volume, article.issue); else if (article.rectype == ADMINISTRATION) (void) sprintf(article.filename,"v%3.3dADM%3.3d", article.volume, article.issue); else (void) sprintf(article.filename,"v%3.3dSRC%3.3d", article.volume, article.issue); } else { if (article.rectype == INFORMATIONAL) (void) sprintf(article.filename,"v%2.2dINF%1.1d", article.volume, article.issue); else (void) sprintf(article.filename,"v%2.2di%3.3d", article.volume, article.issue); } } } break; case SUBMITTED_BY: data(header.submitted_by, sizeof(header.submitted_by), "SUBMITTED_BY:", SUBMITTED_BY); /* ** Save the author's name and sign on if specified ** Can be in any of the following formats: ** kent@sparky.uucp ** kent@sparky.imd.sterling.com (Kent Landfield) ** Kent Landfield */ if ((sp = strchr(s,':')) != NULL) { (void) strcpy(article.author_signon,(sp+2)); /* ** Has a name been attached to the signon ? */ if ((dp = strchr(article.author_signon,'<')) != NULL) { *(dp-1) = '\0'; (void) strcpy(article.author_name, article.author_signon); (void) strcpy(article.author_signon, ++dp); /* ** Save the name, removing the <>. */ if ((dp = strchr(article.author_signon,'>')) != NULL) *dp = '\0'; } else if ((dp = strchr(article.author_signon,'(')) != NULL) { *(dp-1) = '\0'; /* ** Save the name, removing the (). */ (void) strcpy(article.author_name, ++dp); if ((dp = strchr(article.author_name,')')) != NULL) *dp = '\0'; } } break; case ARCH_NAME: data(header.archive_name,sizeof(header.archive_name),"ARCH_NAME:", ARCH_NAME); break; case PATCH_TO: data(header.patch_to,sizeof(header.patch_to),"PATCH_TO:", PATCH_TO); article.rectype = PATCH; /* set the article type */ /* ** Parse the initially posted article's package-name, volume ** and issue. The format of the auxiliary header is: ** ** Patch-To: rkive: Volume 5, Issue 110 ** Or in the case of multipart initial postings, ** Patch-To: rkive: Volume 5, Issue 110-120 ** Or in the case of multipart non-sequential initial postings, ** Patch-To: rkive: Volume 5, Issue 110, 115, 116-120 ** */ /* ** First get the package-name */ dp = strchr(s,':'); /* This exists or we would not be here */ while ((*++dp == ' ') || *dp == '\t'); /* skip spaces, tabs */ sp = dp; if ((dp = strchr(sp,':')) != NULL) { *dp = '\0'; /* ** Now let's check to see if the moderator got it right... ** Check to assure that a valid archive-name has been ** specified in the Patch-To: line... */ if (strchr(sp,' ') != NULL) { dp = s; /* reset to start of string if no package-name */ (void) strcpy(article.package_name,"Unknown"); } else (void) strcpy(article.package_name,sp); } else { dp = s; /* reset to start of string if no package-name */ (void) strcpy(article.package_name,"Unknown"); } /* ** Next, get the volume number. */ ++dp; while (*dp && (!isdigit(*dp))) ++dp; sp = dp; while (*dp && (isdigit(*dp))) ++dp; *dp = '\0'; article.patch_volume = atoi(sp); /* ** Now get the issue number. ** It is stored as a string but error check that ** just numbers, '-' and ',' are in the string. */ ++dp; while (*dp && (!isdigit(*dp))) ++dp; sp = article.patch_issue; while (*dp && (isdigit(*dp) || *dp == '-' || *dp == ',' || *dp == ' ')) { if (*dp != ' ') *sp++ = *dp; ++dp; } *sp = '\0'; break; case X_CHECKSUM: data(header.x_checksum,sizeof(header.x_checksum),"X_CHECKSUM:", X_CHECKSUM); break; case ORIGINAL_POSTING_BY: data(header.orig_poster,sizeof(header.orig_poster),"ORIGINAL_POSTING_BY:",ORIGINAL_POSTING_BY); break; case ORIGINAL_SUBJECT: data(header.orig_subject,sizeof(header.orig_subject),"ORIGINAL-SUBJECT:",ORIGINAL_SUBJECT); break; case ARCHIVE_SITE: data(header.archive_site,sizeof(header.archive_site),"ARCHIVE_SITE:",ARCHIVE_SITE); break; case ARCHIVE_DIRECTORY: data(header.archive_dir,sizeof(header.archive_dir),"ARCHIVE_DIRECTORY:",ARCHIVE_DIRECTORY); break; case REPOSTED_BY: data(header.reposter,sizeof(header.reposter),"REPOSTED_BY:",REPOSTED_BY); break; case ARCHITECTURE: data(header.architecture,sizeof(header.architecture),"ARCHITECTURE:",ARCHITECTURE); break; case VERSION_NUMBER: data(header.version_number,sizeof(header.version_number),"VERSION_NUMBER:",VERSION_NUMBER); break; case ARCHIVE_: data(header.archive,sizeof(header.archive),"ARCHIVE:",ARCHIVE_); break; } return; }