-/* $Id: mandocdb.c,v 1.1 2011/07/14 10:57:02 kristaps Exp $ */
+/* $Id: mandocdb.c,v 1.23 2011/12/03 18:47:09 kristaps Exp $ */
/*
* Copyright (c) 2011 Kristaps Dzonsons <kristaps@bsd.lv>
+ * Copyright (c) 2011 Ingo Schwarze <schwarze@openbsd.org>
*
* Permission to use, copy, modify, and distribute this software for any
* purpose with or without fee is hereby granted, provided that the above
#endif
#include <sys/param.h>
+#include <sys/types.h>
+#include <sys/stat.h>
#include <assert.h>
+#include <dirent.h>
#include <fcntl.h>
#include <getopt.h>
#include <stdio.h>
#include <stdint.h>
#include <stdlib.h>
#include <string.h>
+#include <unistd.h>
-#ifdef __linux__
+#if defined(__linux__)
+# include <endian.h>
# include <db_185.h>
+#elif defined(__APPLE__)
+# include <libkern/OSByteOrder.h>
+# include <db.h>
#else
# include <db.h>
#endif
#include "man.h"
#include "mdoc.h"
#include "mandoc.h"
+#include "mandocdb.h"
+#include "manpath.h"
-#define MANDOC_DB "mandoc.db"
-#define MANDOC_IDX "mandoc.index"
#define MANDOC_BUFSZ BUFSIZ
-#define MANDOC_FLAGS O_CREAT|O_TRUNC|O_RDWR
#define MANDOC_SLOP 1024
-/* Bit-fields. See makewhatis.1. */
-
-#define TYPE_NAME 0x01
-#define TYPE_FUNCTION 0x02
-#define TYPE_UTILITY 0x04
-#define TYPE_INCLUDES 0x08
-#define TYPE_VARIABLE 0x10
-#define TYPE_STANDARD 0x20
-#define TYPE_AUTHOR 0x40
-#define TYPE_CONFIG 0x80
-#define TYPE_DESC 0x100
-#define TYPE_XREF 0x200
-#define TYPE_PATH 0x400
-#define TYPE_ENV 0x800
-#define TYPE_ERR 0x1000
+#define MANDOC_SRC 0x1
+#define MANDOC_FORM 0x2
+
+/* Tiny list for files. No need to bring in QUEUE. */
+
+struct of {
+ char *fname; /* heap-allocated */
+ char *sec;
+ char *arch;
+ char *title;
+ int src_form;
+ struct of *next; /* NULL for last one */
+ struct of *first; /* first in list */
+};
/* Buffer for storing growable data. */
struct buf {
char *cp;
- size_t len;
- size_t size;
+ size_t len; /* current length */
+ size_t size; /* total buffer size */
};
/* Operation we're going to perform. */
enum op {
OP_NEW = 0, /* new database */
- OP_UPDATE, /* update entries in existing database */
+ OP_UPDATE, /* delete/add entries in existing database */
OP_DELETE /* delete entries from existing database */
};
static void buf_appendb(struct buf *,
const void *, size_t);
static void dbt_put(DB *, const char *, DBT *, DBT *);
-static void hash_put(DB *, const struct buf *, int);
+static void hash_put(DB *, const struct buf *, uint64_t);
static void hash_reset(DB **);
+static void index_merge(const struct of *, struct mparse *,
+ struct buf *, struct buf *, DB *,
+ DB *, const char *, DB *, const char *,
+ recno_t, const recno_t *, size_t);
+static void index_prune(const struct of *, DB *,
+ const char *, DB *, const char *,
+ recno_t *, recno_t **, size_t *);
+static void ofile_argbuild(int, char *[], struct of **);
+static int ofile_dirbuild(const char *, const char *,
+ const char *, int, struct of **);
+static void ofile_free(struct of *);
+static void pformatted(DB *, struct buf *, struct buf *,
+ const struct of *);
static int pman_node(MAN_ARGS);
static void pmdoc_node(MDOC_ARGS);
static void pmdoc_An(MDOC_ARGS);
};
static const char *progname;
+static int use_all; /* Use all directories and files. */
+static int verb; /* Output verbosity level. */
int
main(int argc, char *argv[])
{
struct mparse *mp; /* parse sequence */
- struct mdoc *mdoc; /* resulting mdoc */
- struct man *man; /* resulting man */
+ struct manpaths dirs;
enum op op; /* current operation */
- char *fn; /* current file being parsed */
- const char *msec, /* manual section */
- *mtitle, /* manual title */
- *arch, /* manual architecture */
- *dir; /* result dir (default: cwd) */
+ const char *dir;
char ibuf[MAXPATHLEN], /* index fname */
- fbuf[MAXPATHLEN], /* btree fname */
- vbuf[8]; /* stringified record number */
- int ch, seq, sseq, verb, i;
+ fbuf[MAXPATHLEN]; /* btree fname */
+ int ch, i, flags;
DB *idx, /* index database */
*db, /* keyword database */
*hash; /* temporary keyword hashtable */
- DBT key, val;
- enum mandoclevel ec; /* exit status */
- size_t sv;
BTREEINFO info; /* btree configuration */
- recno_t rec,
- maxrec; /* supremum of all records */
- recno_t *recs; /* buffer of empty records */
- size_t recsz, /* buffer size of recs */
- reccur; /* valid number of recs */
+ recno_t maxrec; /* last record number in the index */
+ recno_t *recs; /* the numbers of all empty records */
+ size_t sz1, sz2,
+ recsz, /* number of allocated slots in recs */
+ reccur; /* current number of empty records */
struct buf buf, /* keyword buffer */
dbuf; /* description buffer */
+ struct of *of; /* list of files for processing */
extern int optind;
extern char *optarg;
else
++progname;
- dir = "";
+ memset(&dirs, 0, sizeof(struct manpaths));
+
verb = 0;
+ use_all = 0;
+ of = NULL;
db = idx = NULL;
mp = NULL;
hash = NULL;
recsz = reccur = 0;
maxrec = 0;
op = OP_NEW;
- ec = MANDOCLEVEL_SYSERR;
-
- memset(&buf, 0, sizeof(struct buf));
- memset(&dbuf, 0, sizeof(struct buf));
+ dir = NULL;
- while (-1 != (ch = getopt(argc, argv, "d:ruv")))
+ while (-1 != (ch = getopt(argc, argv, "ad:u:v")))
switch (ch) {
+ case ('a'):
+ use_all = 1;
+ break;
case ('d'):
dir = optarg;
- break;
- case ('r'):
- op = OP_DELETE;
+ op = OP_UPDATE;
break;
case ('u'):
- op = OP_UPDATE;
+ dir = optarg;
+ op = OP_DELETE;
break;
case ('v'):
verb++;
argc -= optind;
argv += optind;
- ibuf[0] = ibuf[MAXPATHLEN - 2] =
- fbuf[0] = fbuf[MAXPATHLEN - 2] = '\0';
+ memset(&info, 0, sizeof(BTREEINFO));
+ info.flags = R_DUP;
+
+ mp = mparse_alloc(MPARSE_AUTO, MANDOCLEVEL_FATAL, NULL, NULL);
- strlcat(fbuf, dir, MAXPATHLEN);
- strlcat(fbuf, MANDOC_DB, MAXPATHLEN);
+ memset(&buf, 0, sizeof(struct buf));
+ memset(&dbuf, 0, sizeof(struct buf));
- strlcat(ibuf, dir, MAXPATHLEN);
- strlcat(ibuf, MANDOC_IDX, MAXPATHLEN);
+ buf.size = dbuf.size = MANDOC_BUFSZ;
- if ('\0' != fbuf[MAXPATHLEN - 2] ||
- '\0' != ibuf[MAXPATHLEN - 2]) {
- fprintf(stderr, "%s: Path too long\n", dir);
- goto out;
- }
+ buf.cp = mandoc_malloc(buf.size);
+ dbuf.cp = mandoc_malloc(dbuf.size);
- /*
- * For the keyword database, open a BTREE database that allows
- * duplicates.
- * For the index database, use a standard RECNO database type.
- * Truncate the database if we're creating a new one.
- */
+ flags = OP_NEW == op ? O_CREAT|O_TRUNC|O_RDWR : O_CREAT|O_RDWR;
- memset(&info, 0, sizeof(BTREEINFO));
- info.flags = R_DUP;
+ if (OP_UPDATE == op || OP_DELETE == op) {
+ ibuf[0] = fbuf[0] = '\0';
- if (OP_NEW == op) {
- db = dbopen(fbuf, MANDOC_FLAGS, 0644, DB_BTREE, &info);
- idx = dbopen(ibuf, MANDOC_FLAGS, 0644, DB_RECNO, NULL);
- } else {
- db = dbopen(fbuf, O_CREAT|O_RDWR, 0644, DB_BTREE, &info);
- idx = dbopen(ibuf, O_CREAT|O_RDWR, 0644, DB_RECNO, NULL);
- }
+ strlcat(fbuf, dir, MAXPATHLEN);
+ strlcat(fbuf, "/", MAXPATHLEN);
+ sz1 = strlcat(fbuf, MANDOC_DB, MAXPATHLEN);
+
+ strlcat(ibuf, dir, MAXPATHLEN);
+ strlcat(ibuf, "/", MAXPATHLEN);
+ sz2 = strlcat(ibuf, MANDOC_IDX, MAXPATHLEN);
+
+ if (sz1 >= MAXPATHLEN || sz2 >= MAXPATHLEN) {
+ fprintf(stderr, "%s: Path too long\n", dir);
+ exit((int)MANDOCLEVEL_BADARG);
+ }
+
+ db = dbopen(fbuf, flags, 0644, DB_BTREE, &info);
+ idx = dbopen(ibuf, flags, 0644, DB_RECNO, NULL);
+
+ if (NULL == db) {
+ perror(fbuf);
+ exit((int)MANDOCLEVEL_SYSERR);
+ } else if (NULL == idx) {
+ perror(ibuf);
+ exit((int)MANDOCLEVEL_SYSERR);
+ }
+
+ if (verb > 2) {
+ printf("%s: Opened\n", fbuf);
+ printf("%s: Opened\n", ibuf);
+ }
+
+ ofile_argbuild(argc, argv, &of);
+ if (NULL == of)
+ goto out;
+
+ of = of->first;
+
+ index_prune(of, db, fbuf, idx, ibuf,
+ &maxrec, &recs, &recsz);
+
+ /*
+ * Go to the root of the respective manual tree
+ * such that .so links work. In case of failure,
+ * just prod on, even though .so links won't work.
+ */
+
+ if (OP_UPDATE == op) {
+ chdir(dir);
+ index_merge(of, mp, &dbuf, &buf, hash,
+ db, fbuf, idx, ibuf,
+ maxrec, recs, reccur);
+ }
- if (NULL == db) {
- perror(fbuf);
- goto out;
- } else if (NULL == db) {
- perror(ibuf);
goto out;
}
/*
- * If we're going to delete or update a database, remove the
- * entries now (both the index and all keywords pointing to it).
- * This doesn't actually remove them: it only sets their record
- * value lengths to zero.
- * While doing so, add the empty records to a list we'll access
- * later in re-adding entries to the database.
+ * Configure the directories we're going to scan.
+ * If we have command-line arguments, use them.
+ * If not, we use man(1)'s method (see mandocdb.8).
*/
- if (OP_DELETE == op || OP_UPDATE == op) {
- seq = R_FIRST;
- while (0 == (ch = (*idx->seq)(idx, &key, &val, seq))) {
- seq = R_NEXT;
- maxrec = *(recno_t *)key.data;
- if (0 == val.size && OP_UPDATE == op) {
- if (reccur >= recsz) {
- recsz += MANDOC_SLOP;
- recs = mandoc_realloc
- (recs, recsz * sizeof(recno_t));
- }
- recs[(int)reccur] = maxrec;
- reccur++;
- continue;
- }
+ if (argc > 0) {
+ dirs.paths = mandoc_malloc(argc * sizeof(char *));
+ dirs.sz = argc;
+ for (i = 0; i < argc; i++)
+ dirs.paths[i] = mandoc_strdup(argv[i]);
+ } else
+ manpath_parse(&dirs, NULL, NULL);
- fn = (char *)val.data;
- for (i = 0; i < argc; i++)
- if (0 == strcmp(fn, argv[i]))
- break;
+ for (i = 0; i < dirs.sz; i++) {
+ ibuf[0] = fbuf[0] = '\0';
- if (i == argc)
- continue;
+ strlcat(fbuf, dirs.paths[i], MAXPATHLEN);
+ strlcat(fbuf, "/", MAXPATHLEN);
+ sz1 = strlcat(fbuf, MANDOC_DB, MAXPATHLEN);
- sseq = R_FIRST;
- while (0 == (ch = (*db->seq)(db, &key, &val, sseq))) {
- sseq = R_NEXT;
- assert(8 == val.size);
- if (maxrec != *(recno_t *)(val.data + 4))
- continue;
- if (verb > 1)
- printf("%s: Deleted keyword: %s\n",
- fn, (char *)key.data);
- ch = (*db->del)(db, &key, R_CURSOR);
- if (ch < 0)
- break;
- }
- if (ch < 0) {
- perror(fbuf);
- exit((int)MANDOCLEVEL_SYSERR);
- }
+ strlcat(ibuf, dirs.paths[i], MAXPATHLEN);
+ strlcat(ibuf, "/", MAXPATHLEN);
+ sz2 = strlcat(ibuf, MANDOC_IDX, MAXPATHLEN);
- if (verb)
- printf("%s: Deleted index\n", fn);
+ if (sz1 >= MAXPATHLEN || sz2 >= MAXPATHLEN) {
+ fprintf(stderr, "%s: Path too long\n",
+ dirs.paths[i]);
+ exit((int)MANDOCLEVEL_BADARG);
+ }
- val.size = 0;
- ch = (*idx->put)(idx, &key, &val, R_CURSOR);
- if (ch < 0) {
- perror(ibuf);
- exit((int)MANDOCLEVEL_SYSERR);
- }
+ if (db)
+ (*db->close)(db);
+ if (idx)
+ (*idx->close)(idx);
- if (OP_UPDATE == op) {
- if (reccur >= recsz) {
- recsz += MANDOC_SLOP;
- recs = mandoc_realloc
- (recs, recsz * sizeof(recno_t));
- }
- recs[(int)reccur] = maxrec;
- reccur++;
- }
+ db = dbopen(fbuf, flags, 0644, DB_BTREE, &info);
+ idx = dbopen(ibuf, flags, 0644, DB_RECNO, NULL);
+
+ if (NULL == db) {
+ perror(fbuf);
+ exit((int)MANDOCLEVEL_SYSERR);
+ } else if (NULL == idx) {
+ perror(ibuf);
+ exit((int)MANDOCLEVEL_SYSERR);
}
- maxrec++;
- }
- if (OP_DELETE == op) {
- ec = MANDOCLEVEL_OK;
- goto out;
+ if (verb > 2) {
+ printf("%s: Truncated\n", fbuf);
+ printf("%s: Truncated\n", ibuf);
+ }
+
+ ofile_free(of);
+ of = NULL;
+
+ if ( ! ofile_dirbuild(dirs.paths[i], NULL, NULL,
+ 0, &of))
+ exit((int)MANDOCLEVEL_SYSERR);
+
+ if (NULL == of)
+ continue;
+
+ of = of->first;
+
+ /*
+ * Go to the root of the respective manual tree
+ * such that .so links work. In case of failure,
+ * just prod on, even though .so links won't work.
+ */
+
+ chdir(dirs.paths[i]);
+ index_merge(of, mp, &dbuf, &buf, hash, db, fbuf,
+ idx, ibuf, maxrec, recs, reccur);
}
- /*
- * Add records to the database.
- * Try parsing each manual given on the command line.
- * If we fail, then emit an error and keep on going.
- * Take resulting trees and push them down into the database code.
- * Use the auto-parser and don't report any errors.
- */
+out:
+ if (db)
+ (*db->close)(db);
+ if (idx)
+ (*idx->close)(idx);
+ if (hash)
+ (*hash->close)(hash);
+ if (mp)
+ mparse_free(mp);
- mp = mparse_alloc(MPARSE_AUTO, MANDOCLEVEL_FATAL, NULL, NULL);
+ manpath_free(&dirs);
+ ofile_free(of);
+ free(buf.cp);
+ free(dbuf.cp);
+ free(recs);
- buf.size = dbuf.size = MANDOC_BUFSZ;
- buf.cp = mandoc_malloc(buf.size);
- dbuf.cp = mandoc_malloc(dbuf.size);
+ return(MANDOCLEVEL_OK);
+}
- for (rec = 0, i = 0; i < argc; i++) {
- fn = argv[i];
- if (OP_UPDATE == op) {
- if (reccur > 0) {
- --reccur;
- rec = recs[(int)reccur];
- } else if (maxrec > 0) {
- rec = maxrec;
- maxrec = 0;
- } else
- rec++;
+void
+index_merge(const struct of *of, struct mparse *mp,
+ struct buf *dbuf, struct buf *buf, DB *hash,
+ DB *db, const char *dbf, DB *idx, const char *idxf,
+ recno_t maxrec, const recno_t *recs, size_t reccur)
+{
+ recno_t rec;
+ int ch;
+ DBT key, val;
+ struct mdoc *mdoc;
+ struct man *man;
+ const char *fn, *msec, *mtitle, *arch;
+ size_t sv;
+ unsigned seq;
+ struct db_val vbuf;
+
+ for (rec = 0; of; of = of->next) {
+ fn = of->fname;
+
+ /*
+ * Reclaim an empty index record, if available.
+ */
+
+ if (reccur > 0) {
+ --reccur;
+ rec = recs[(int)reccur];
+ } else if (maxrec > 0) {
+ rec = maxrec;
+ maxrec = 0;
} else
rec++;
mparse_reset(mp);
hash_reset(&hash);
+ mdoc = NULL;
+ man = NULL;
- if (mparse_readfd(mp, -1, fn) >= MANDOCLEVEL_FATAL) {
- fprintf(stderr, "%s: Parse failure\n", fn);
- continue;
+ /*
+ * Try interpreting the file as mdoc(7) or man(7)
+ * source code, unless it is already known to be
+ * formatted. Fall back to formatted mode.
+ */
+
+ if ((MANDOC_SRC & of->src_form ||
+ ! (MANDOC_FORM & of->src_form)) &&
+ MANDOCLEVEL_FATAL > mparse_readfd(mp, -1, fn))
+ mparse_result(mp, &mdoc, &man);
+
+ if (NULL != mdoc) {
+ msec = mdoc_meta(mdoc)->msec;
+ arch = mdoc_meta(mdoc)->arch;
+ mtitle = mdoc_meta(mdoc)->title;
+ } else if (NULL != man) {
+ msec = man_meta(man)->msec;
+ arch = NULL;
+ mtitle = man_meta(man)->title;
+ } else {
+ msec = of->sec;
+ arch = of->arch;
+ mtitle = of->title;
}
- mparse_result(mp, &mdoc, &man);
- if (NULL == mdoc && NULL == man)
- continue;
+ /*
+ * By default, skip a file if the manual section
+ * and architecture given in the file disagree
+ * with the directory where the file is located.
+ */
+
+ if (0 == use_all) {
+ assert(of->sec);
+ assert(msec);
+ if (strcmp(msec, of->sec))
+ continue;
- msec = NULL != mdoc ?
- mdoc_meta(mdoc)->msec : man_meta(man)->msec;
- mtitle = NULL != mdoc ?
- mdoc_meta(mdoc)->title : man_meta(man)->title;
- arch = NULL != mdoc ? mdoc_meta(mdoc)->arch : NULL;
+ if (NULL == arch) {
+ if (NULL != of->arch)
+ continue;
+ } else if (NULL == of->arch ||
+ strcmp(arch, of->arch))
+ continue;
+ }
if (NULL == arch)
arch = "";
+ /*
+ * By default, skip a file if the title given
+ * in the file disagrees with the file name.
+ * If both agree, use the file name as the title,
+ * because the one in the file usually is all caps.
+ */
+
+ assert(of->title);
+ assert(mtitle);
+
+ if (0 == strcasecmp(mtitle, of->title))
+ mtitle = of->title;
+ else if (0 == use_all)
+ continue;
+
/*
* The index record value consists of a nil-terminated
* filename, a nil-terminated manual section, and a
* going to write a nil byte in its place.
*/
- dbuf.len = 0;
- buf_appendb(&dbuf, fn, strlen(fn) + 1);
- buf_appendb(&dbuf, msec, strlen(msec) + 1);
- buf_appendb(&dbuf, mtitle, strlen(mtitle) + 1);
- buf_appendb(&dbuf, arch, strlen(arch) + 1);
+ dbuf->len = 0;
+ buf_append(dbuf, mdoc ? "mdoc" : (man ? "man" : "cat"));
+ buf_appendb(dbuf, fn, strlen(fn) + 1);
+ buf_appendb(dbuf, msec, strlen(msec) + 1);
+ buf_appendb(dbuf, mtitle, strlen(mtitle) + 1);
+ buf_appendb(dbuf, arch, strlen(arch) + 1);
- sv = dbuf.len;
+ sv = dbuf->len;
/* Fix the record number in the btree value. */
if (mdoc)
- pmdoc_node(hash, &buf, &dbuf,
+ pmdoc_node(hash, buf, dbuf,
mdoc_node(mdoc), mdoc_meta(mdoc));
- else
- pman_node(hash, &buf, &dbuf, man_node(man));
+ else if (man)
+ pman_node(hash, buf, dbuf, man_node(man));
+ else
+ pformatted(hash, buf, dbuf, of);
/*
* Copy from the in-memory hashtable of pending keywords
* into the database.
*/
-
- memset(vbuf, 0, sizeof(uint32_t));
- memcpy(vbuf + 4, &rec, sizeof(uint32_t));
+ vbuf.rec = htobe32(rec);
seq = R_FIRST;
while (0 == (ch = (*hash->seq)(hash, &key, &val, seq))) {
seq = R_NEXT;
-
- memcpy(vbuf, val.data, sizeof(uint32_t));
- val.size = sizeof(vbuf);
- val.data = vbuf;
-
- if (verb > 1)
- printf("%s: Added keyword: %s, 0x%x\n",
- fn, (char *)key.data,
- *(int *)val.data);
- dbt_put(db, fbuf, &key, &val);
+ vbuf.mask = htobe64(*(uint64_t *)val.data);
+ val.size = sizeof(struct db_val);
+ val.data = &vbuf;
+ dbt_put(db, dbf, &key, &val);
}
if (ch < 0) {
perror("hash");
* set, put an empty one in now.
*/
- if (dbuf.len == sv)
- buf_appendb(&dbuf, "", 1);
+ if (dbuf->len == sv)
+ buf_appendb(dbuf, "", 1);
key.data = &rec;
key.size = sizeof(recno_t);
- val.data = dbuf.cp;
- val.size = dbuf.len;
+ val.data = dbuf->cp;
+ val.size = dbuf->len;
- if (verb > 0)
+ if (verb)
printf("%s: Added index\n", fn);
- dbt_put(idx, ibuf, &key, &val);
+ dbt_put(idx, idxf, &key, &val);
}
+}
- ec = MANDOCLEVEL_OK;
-out:
- if (db)
- (*db->close)(db);
- if (idx)
- (*idx->close)(idx);
- if (hash)
- (*hash->close)(hash);
- if (mp)
- mparse_free(mp);
+/*
+ * Scan through all entries in the index file `idx' and prune those
+ * entries in `ofile'.
+ * Pruning consists of removing from `db', then invalidating the entry
+ * in `idx' (zeroing its value size).
+ */
+static void
+index_prune(const struct of *ofile, DB *db, const char *dbf,
+ DB *idx, const char *idxf,
+ recno_t *maxrec, recno_t **recs, size_t *recsz)
+{
+ const struct of *of;
+ const char *fn, *cp;
+ struct db_val *vbuf;
+ unsigned seq, sseq;
+ DBT key, val;
+ size_t reccur;
+ int ch;
- free(buf.cp);
- free(dbuf.cp);
- free(recs);
+ reccur = 0;
+ seq = R_FIRST;
+ while (0 == (ch = (*idx->seq)(idx, &key, &val, seq))) {
+ seq = R_NEXT;
+ *maxrec = *(recno_t *)key.data;
+ cp = val.data;
+
+ /* Deleted records are zero-sized. Skip them. */
+
+ if (0 == val.size)
+ goto cont;
+
+ /*
+ * Make sure we're sane.
+ * Read past our mdoc/man/cat type to the next string,
+ * then make sure it's bounded by a NUL.
+ * Failing any of these, we go into our error handler.
+ */
+
+ if (NULL == (fn = memchr(cp, '\0', val.size)))
+ break;
+ if (++fn - cp >= (int)val.size)
+ break;
+ if (NULL == memchr(fn, '\0', val.size - (fn - cp)))
+ break;
+
+ /*
+ * Search for the file in those we care about.
+ * XXX: build this into a tree. Too slow.
+ */
+
+ for (of = ofile; of; of = of->next)
+ if (0 == strcmp(fn, of->fname))
+ break;
+
+ if (NULL == of)
+ continue;
+
+ /*
+ * Search through the keyword database, throwing out all
+ * references to our file.
+ */
+
+ sseq = R_FIRST;
+ while (0 == (ch = (*db->seq)(db, &key, &val, sseq))) {
+ sseq = R_NEXT;
+ if (sizeof(struct db_val) != val.size)
+ break;
+
+ vbuf = val.data;
+ if (*maxrec != betoh32(vbuf->rec))
+ continue;
+
+ if ((ch = (*db->del)(db, &key, R_CURSOR)) < 0)
+ break;
+ }
+
+ if (ch < 0) {
+ perror(dbf);
+ exit((int)MANDOCLEVEL_SYSERR);
+ } else if (1 != ch) {
+ fprintf(stderr, "%s: Corrupt database\n", dbf);
+ exit((int)MANDOCLEVEL_SYSERR);
+ }
+
+ if (verb)
+ printf("%s: Deleted index\n", fn);
+
+ val.size = 0;
+ ch = (*idx->put)(idx, &key, &val, R_CURSOR);
+
+ if (ch < 0)
+ break;
+cont:
+ if (reccur >= *recsz) {
+ *recsz += MANDOC_SLOP;
+ *recs = mandoc_realloc
+ (*recs, *recsz * sizeof(recno_t));
+ }
+
+ (*recs)[(int)reccur] = *maxrec;
+ reccur++;
+ }
+
+ if (ch < 0) {
+ perror(idxf);
+ exit((int)MANDOCLEVEL_SYSERR);
+ } else if (1 != ch) {
+ fprintf(stderr, "%s: Corrupt index\n", idxf);
+ exit((int)MANDOCLEVEL_SYSERR);
+ }
- return((int)ec);
+ (*maxrec)++;
}
/*
return;
buf_appendmdoc(buf, n->child, 0);
- hash_put(hash, buf, TYPE_AUTHOR);
+ hash_put(hash, buf, TYPE_An);
}
static void
if (NULL != (hash = *db))
(*hash->close)(hash);
- *db = dbopen(NULL, MANDOC_FLAGS, 0644, DB_HASH, NULL);
+ *db = dbopen(NULL, O_CREAT|O_RDWR, 0644, DB_HASH, NULL);
if (NULL == *db) {
perror("hash");
exit((int)MANDOCLEVEL_SYSERR);
buf_appendb(buf, start, (size_t)(end - start + 1));
buf_appendb(buf, "", 1);
- hash_put(hash, buf, TYPE_INCLUDES);
+ hash_put(hash, buf, TYPE_In);
}
/* ARGSUSED */
return;
buf_appendmdoc(buf, n->child, 0);
- hash_put(hash, buf, TYPE_CONFIG);
+ hash_put(hash, buf, TYPE_Cd);
}
/* ARGSUSED */
return;
buf_append(buf, n->child->string);
- hash_put(hash, buf, TYPE_INCLUDES);
+ hash_put(hash, buf, TYPE_In);
}
/* ARGSUSED */
cp++;
buf_append(buf, cp);
- hash_put(hash, buf, TYPE_FUNCTION);
+ hash_put(hash, buf, TYPE_Fn);
}
/* ARGSUSED */
return;
buf_append(buf, n->child->string);
- hash_put(hash, buf, TYPE_STANDARD);
+ hash_put(hash, buf, TYPE_St);
}
/* ARGSUSED */
} else
buf_appendb(buf, ".", 2);
- hash_put(hash, buf, TYPE_XREF);
+ hash_put(hash, buf, TYPE_Xr);
}
/* ARGSUSED */
buf_appendb(buf, start, sz);
buf_appendb(buf, "", 1);
- hash_put(hash, buf, TYPE_VARIABLE);
+ hash_put(hash, buf, TYPE_Va);
}
/* ARGSUSED */
return;
buf_append(buf, n->child->string);
- hash_put(hash, buf, TYPE_FUNCTION);
+ hash_put(hash, buf, TYPE_Fn);
}
buf_appendmdoc(dbuf, n->child, 1);
buf_appendmdoc(buf, n->child, 0);
- hash_put(hash, buf, TYPE_DESC);
+ hash_put(hash, buf, TYPE_Nd);
}
/* ARGSUSED */
return;
buf_appendmdoc(buf, n->child, 0);
- hash_put(hash, buf, TYPE_ERR);
+ hash_put(hash, buf, TYPE_Er);
}
/* ARGSUSED */
return;
buf_appendmdoc(buf, n->child, 0);
- hash_put(hash, buf, TYPE_ENV);
+ hash_put(hash, buf, TYPE_Ev);
}
/* ARGSUSED */
return;
buf_appendmdoc(buf, n->child, 0);
- hash_put(hash, buf, TYPE_PATH);
+ hash_put(hash, buf, TYPE_Pa);
}
/* ARGSUSED */
if (SEC_NAME == n->sec) {
buf_appendmdoc(buf, n->child, 0);
- hash_put(hash, buf, TYPE_NAME);
+ hash_put(hash, buf, TYPE_Nm);
return;
} else if (SEC_SYNOPSIS != n->sec || MDOC_HEAD != n->type)
return;
buf_append(buf, m->name);
buf_appendmdoc(buf, n->child, 0);
- hash_put(hash, buf, TYPE_UTILITY);
+ hash_put(hash, buf, TYPE_Nm);
}
static void
-hash_put(DB *db, const struct buf *buf, int mask)
+hash_put(DB *db, const struct buf *buf, uint64_t mask)
{
DBT key, val;
int rc;
perror("hash");
exit((int)MANDOCLEVEL_SYSERR);
} else if (0 == rc)
- mask |= *(int *)val.data;
+ mask |= *(uint64_t *)val.data;
val.data = &mask;
- val.size = sizeof(int);
+ val.size = sizeof(uint64_t);
if ((rc = (*db->put)(db, &key, &val, 0)) < 0) {
perror("hash");
buf_appendb(buf, start, sz);
buf_appendb(buf, "", 1);
- hash_put(hash, buf, TYPE_NAME);
+ hash_put(hash, buf, TYPE_Nm);
if (' ' == start[(int)sz]) {
start += (int)sz + 1;
buf_appendb(dbuf, start, sz);
buf_appendb(buf, start, sz);
- hash_put(hash, buf, TYPE_DESC);
+ hash_put(hash, buf, TYPE_Nd);
}
}
- if (pman_node(hash, buf, dbuf, n->child))
- return(1);
- if (pman_node(hash, buf, dbuf, n->next))
- return(1);
+ for (n = n->child; n; n = n->next)
+ if (pman_node(hash, buf, dbuf, n))
+ return(1);
return(0);
}
+/*
+ * Parse a formatted manual page.
+ * By necessity, this involves rather crude guesswork.
+ */
+static void
+pformatted(DB *hash, struct buf *buf, struct buf *dbuf,
+ const struct of *of)
+{
+ FILE *stream;
+ char *line, *p;
+ size_t len, plen;
+
+ if (NULL == (stream = fopen(of->fname, "r"))) {
+ perror(of->fname);
+ return;
+ }
+
+ /*
+ * Always use the title derived from the filename up front,
+ * do not even try to find it in the file. This also makes
+ * sure we don't end up with an orphan index record, even if
+ * the file content turns out to be completely unintelligible.
+ */
+
+ buf->len = 0;
+ buf_append(buf, of->title);
+ hash_put(hash, buf, TYPE_Nm);
+
+ while (NULL != (line = fgetln(stream, &len)) && '\n' != *line)
+ /* Skip to first blank line. */ ;
+
+ while (NULL != (line = fgetln(stream, &len)) &&
+ ('\n' == *line || ' ' == *line))
+ /* Skip to first section header. */ ;
+
+ /*
+ * If no page content can be found,
+ * reuse the page title as the page description.
+ */
+
+ if (NULL == (line = fgetln(stream, &len))) {
+ buf_appendb(dbuf, buf->cp, buf->size);
+ hash_put(hash, buf, TYPE_Nd);
+ fclose(stream);
+ return;
+ }
+ fclose(stream);
+
+ /*
+ * If there is a dash, skip to the text following it.
+ */
+
+ for (p = line, plen = len; plen; p++, plen--)
+ if ('-' == *p)
+ break;
+ for ( ; plen; p++, plen--)
+ if ('-' != *p && ' ' != *p && 8 != *p)
+ break;
+ if (0 == plen) {
+ p = line;
+ plen = len;
+ }
+
+ /*
+ * Copy the rest of the line, but no more than 70 bytes.
+ */
+
+ if (70 < plen)
+ plen = 70;
+ p[plen-1] = '\0';
+ buf_appendb(dbuf, p, plen);
+ buf->len = 0;
+ buf_appendb(buf, p, plen);
+ hash_put(hash, buf, TYPE_Nd);
+}
+
+static void
+ofile_argbuild(int argc, char *argv[], struct of **of)
+{
+ char buf[MAXPATHLEN];
+ char *sec, *arch, *title, *p;
+ int i, src_form;
+ struct of *nof;
+
+ for (i = 0; i < argc; i++) {
+
+ /*
+ * Try to infer the manual section, architecture and
+ * page title from the path, assuming it looks like
+ * man*[/<arch>]/<title>.<section> or
+ * cat<section>[/<arch>]/<title>.0
+ */
+
+ if (strlcpy(buf, argv[i], sizeof(buf)) >= sizeof(buf)) {
+ fprintf(stderr, "%s: Path too long\n", argv[i]);
+ continue;
+ }
+ sec = arch = title = NULL;
+ src_form = 0;
+ p = strrchr(buf, '\0');
+ while (p-- > buf) {
+ if (NULL == sec && '.' == *p) {
+ sec = p + 1;
+ *p = '\0';
+ if ('0' == *sec)
+ src_form |= MANDOC_FORM;
+ else if ('1' <= *sec && '9' >= *sec)
+ src_form |= MANDOC_SRC;
+ continue;
+ }
+ if ('/' != *p)
+ continue;
+ if (NULL == title) {
+ title = p + 1;
+ *p = '\0';
+ continue;
+ }
+ if (strncmp("man", p + 1, 3)) {
+ src_form |= MANDOC_SRC;
+ arch = p + 1;
+ } else if (strncmp("cat", p + 1, 3)) {
+ src_form |= MANDOC_FORM;
+ arch = p + 1;
+ }
+ break;
+ }
+ if (NULL == title)
+ title = buf;
+
+ /*
+ * Build the file structure.
+ */
+
+ nof = mandoc_calloc(1, sizeof(struct of));
+ nof->fname = mandoc_strdup(argv[i]);
+ if (NULL != sec)
+ nof->sec = mandoc_strdup(sec);
+ if (NULL != arch)
+ nof->arch = mandoc_strdup(arch);
+ nof->title = mandoc_strdup(title);
+ nof->src_form = src_form;
+
+ /*
+ * Add the structure to the list.
+ */
+
+ if (verb > 2)
+ printf("%s: Scheduling\n", argv[i]);
+ if (NULL == *of) {
+ *of = nof;
+ (*of)->first = nof;
+ } else {
+ nof->first = (*of)->first;
+ (*of)->next = nof;
+ *of = nof;
+ }
+ }
+}
+
+/*
+ * Recursively build up a list of files to parse.
+ * We use this instead of ftw() and so on because I don't want global
+ * variables hanging around.
+ * This ignores the mandoc.db and mandoc.index files, but assumes that
+ * everything else is a manual.
+ * Pass in a pointer to a NULL structure for the first invocation.
+ */
+static int
+ofile_dirbuild(const char *dir, const char* psec, const char *parch,
+ int p_src_form, struct of **of)
+{
+ char buf[MAXPATHLEN];
+ struct stat sb;
+ size_t sz;
+ DIR *d;
+ const char *fn, *sec, *arch;
+ char *p, *q, *suffix;
+ struct of *nof;
+ struct dirent *dp;
+ int src_form;
+
+ if (NULL == (d = opendir(dir))) {
+ perror(dir);
+ return(0);
+ }
+
+ while (NULL != (dp = readdir(d))) {
+ fn = dp->d_name;
+
+ if ('.' == *fn)
+ continue;
+
+ src_form = p_src_form;
+
+ if (DT_DIR == dp->d_type) {
+ sec = psec;
+ arch = parch;
+
+ /*
+ * By default, only use directories called:
+ * man<section>/[<arch>/] or
+ * cat<section>/[<arch>/]
+ */
+
+ if (NULL == sec) {
+ if(0 == strncmp("man", fn, 3)) {
+ src_form |= MANDOC_SRC;
+ sec = fn + 3;
+ } else if (0 == strncmp("cat", fn, 3)) {
+ src_form |= MANDOC_FORM;
+ sec = fn + 3;
+ } else if (use_all)
+ sec = fn;
+ else
+ continue;
+ } else if (NULL == arch && (use_all ||
+ NULL == strchr(fn, '.')))
+ arch = fn;
+ else if (0 == use_all)
+ continue;
+
+ buf[0] = '\0';
+ strlcat(buf, dir, MAXPATHLEN);
+ strlcat(buf, "/", MAXPATHLEN);
+ sz = strlcat(buf, fn, MAXPATHLEN);
+
+ if (MAXPATHLEN <= sz) {
+ fprintf(stderr, "%s: Path too long\n", dir);
+ return(0);
+ }
+
+ if (verb > 2)
+ printf("%s: Scanning\n", buf);
+
+ if ( ! ofile_dirbuild(buf, sec, arch,
+ src_form, of))
+ return(0);
+ }
+ if (DT_REG != dp->d_type ||
+ (NULL == psec && !use_all) ||
+ !strcmp(MANDOC_DB, fn) ||
+ !strcmp(MANDOC_IDX, fn))
+ continue;
+
+ /*
+ * By default, skip files where the file name suffix
+ * does not agree with the section directory
+ * they are located in.
+ */
+
+ suffix = strrchr(fn, '.');
+ if (0 == use_all) {
+ if (NULL == suffix)
+ continue;
+ if ((MANDOC_SRC & src_form &&
+ strcmp(suffix + 1, psec)) ||
+ (MANDOC_FORM & src_form &&
+ strcmp(suffix + 1, "0")))
+ continue;
+ }
+ if (NULL != suffix) {
+ if ('0' == suffix[1])
+ src_form |= MANDOC_FORM;
+ else if ('1' <= suffix[1] && '9' >= suffix[1])
+ src_form |= MANDOC_SRC;
+ }
+
+
+ /*
+ * Skip formatted manuals if a source version is
+ * available. Ignore the age: it is very unlikely
+ * that people install newer formatted base manuals
+ * when they used to have source manuals before,
+ * and in ports, old manuals get removed on update.
+ */
+ if (0 == use_all && MANDOC_FORM & src_form &&
+ NULL != psec) {
+ buf[0] = '\0';
+ strlcat(buf, dir, MAXPATHLEN);
+ p = strrchr(buf, '/');
+ if (NULL == p)
+ p = buf;
+ else
+ p++;
+ if (0 == strncmp("cat", p, 3))
+ memcpy(p, "man", 3);
+ strlcat(buf, "/", MAXPATHLEN);
+ sz = strlcat(buf, fn, MAXPATHLEN);
+ if (sz >= MAXPATHLEN) {
+ fprintf(stderr, "%s: Path too long\n", buf);
+ continue;
+ }
+ q = strrchr(buf, '.');
+ if (NULL != q && p < q++) {
+ *q = '\0';
+ sz = strlcat(buf, psec, MAXPATHLEN);
+ if (sz >= MAXPATHLEN) {
+ fprintf(stderr,
+ "%s: Path too long\n", buf);
+ continue;
+ }
+ if (0 == stat(buf, &sb))
+ continue;
+ }
+ }
+
+ buf[0] = '\0';
+ strlcat(buf, dir, MAXPATHLEN);
+ strlcat(buf, "/", MAXPATHLEN);
+ sz = strlcat(buf, fn, MAXPATHLEN);
+ if (sz >= MAXPATHLEN) {
+ fprintf(stderr, "%s: Path too long\n", dir);
+ continue;
+ }
+
+ nof = mandoc_calloc(1, sizeof(struct of));
+ nof->fname = mandoc_strdup(buf);
+ if (NULL != psec)
+ nof->sec = mandoc_strdup(psec);
+ if (NULL != parch)
+ nof->arch = mandoc_strdup(parch);
+ nof->src_form = src_form;
+
+ /*
+ * Remember the file name without the extension,
+ * to be used as the page title in the database.
+ */
+
+ if (NULL != suffix)
+ *suffix = '\0';
+ nof->title = mandoc_strdup(fn);
+
+ /*
+ * Add the structure to the list.
+ */
+
+ if (verb > 2)
+ printf("%s: Scheduling\n", buf);
+ if (NULL == *of) {
+ *of = nof;
+ (*of)->first = nof;
+ } else {
+ nof->first = (*of)->first;
+ (*of)->next = nof;
+ *of = nof;
+ }
+ }
+
+ closedir(d);
+ return(1);
+}
+
+static void
+ofile_free(struct of *of)
+{
+ struct of *nof;
+
+ while (of) {
+ nof = of->next;
+ free(of->fname);
+ free(of->sec);
+ free(of->arch);
+ free(of->title);
+ free(of);
+ of = nof;
+ }
+}
+
static void
usage(void)
{
- fprintf(stderr, "usage: %s [-ruv] [-d path] [file...]\n",
- progname);
+ fprintf(stderr, "usage: %s [-v] "
+ "[-d dir [files...] |"
+ " -u dir [files...] |"
+ " dir...]\n", progname);
}
git.ckatri.com / mandoc.git / blobdiff