c1a85efa3a
The makewhatis utility is now ok with WARNS=3. Reviewed by: bde
1030 lines
23 KiB
C
1030 lines
23 KiB
C
/*-
|
|
* Copyright (c) 2002 John Rochester
|
|
* All rights reserved.
|
|
*
|
|
* Redistribution and use in source and binary forms, with or without
|
|
* modification, are permitted provided that the following conditions
|
|
* are met:
|
|
* 1. Redistributions of source code must retain the above copyright
|
|
* notice, this list of conditions and the following disclaimer,
|
|
* in this position and unchanged.
|
|
* 2. Redistributions in binary form must reproduce the above copyright
|
|
* notice, this list of conditions and the following disclaimer in the
|
|
* documentation and/or other materials provided with the distribution.
|
|
* 3. The name of the author may not be used to endorse or promote products
|
|
* derived from this software without specific prior written permission
|
|
*
|
|
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
|
|
* IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
|
|
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
|
|
* IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
|
|
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
|
|
* NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
|
|
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
|
|
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
|
|
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
|
|
* THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
|
*/
|
|
|
|
#include <sys/cdefs.h>
|
|
__FBSDID("$FreeBSD$");
|
|
|
|
#include <sys/types.h>
|
|
#include <sys/stat.h>
|
|
#include <sys/param.h>
|
|
#include <sys/queue.h>
|
|
|
|
#include <ctype.h>
|
|
#include <dirent.h>
|
|
#include <err.h>
|
|
#include <stdio.h>
|
|
#include <stdlib.h>
|
|
#include <string.h>
|
|
#include <stringlist.h>
|
|
#include <unistd.h>
|
|
#include <zlib.h>
|
|
|
|
#define DEFAULT_MANPATH "/usr/share/man"
|
|
#define LINE_ALLOC 4096
|
|
|
|
static char blank[] = "";
|
|
|
|
/*
|
|
* Information collected about each man page in a section.
|
|
*/
|
|
struct page_info {
|
|
char * filename;
|
|
char * name;
|
|
char * suffix;
|
|
int gzipped;
|
|
ino_t inode;
|
|
};
|
|
|
|
/*
|
|
* An entry kept for each visited directory.
|
|
*/
|
|
struct visited_dir {
|
|
dev_t device;
|
|
ino_t inode;
|
|
SLIST_ENTRY(visited_dir) next;
|
|
};
|
|
|
|
/*
|
|
* an expanding string
|
|
*/
|
|
struct sbuf {
|
|
char * content; /* the start of the buffer */
|
|
char * end; /* just past the end of the content */
|
|
char * last; /* the last allocated character */
|
|
};
|
|
|
|
/*
|
|
* Removes the last amount characters from the sbuf.
|
|
*/
|
|
#define sbuf_retract(sbuf, amount) \
|
|
((sbuf)->end -= (amount))
|
|
/*
|
|
* Returns the length of the sbuf content.
|
|
*/
|
|
#define sbuf_length(sbuf) \
|
|
((sbuf)->end - (sbuf)->content)
|
|
|
|
typedef char *edited_copy(char *from, char *to, int length);
|
|
|
|
static int append; /* -a flag: append to existing whatis */
|
|
static int verbose; /* -v flag: be verbose with warnings */
|
|
static int indent = 24; /* -i option: description indentation */
|
|
static const char *whatis_name="whatis";/* -n option: the name */
|
|
static char *common_output; /* -o option: the single output file */
|
|
static char *locale; /* user's locale if -L is used */
|
|
static char *lang_locale; /* short form of locale */
|
|
static char *machine;
|
|
|
|
static int exit_code; /* exit code to use when finished */
|
|
static SLIST_HEAD(, visited_dir) visited_dirs =
|
|
SLIST_HEAD_INITIALIZER(visited_dirs);
|
|
|
|
/*
|
|
* While the whatis line is being formed, it is stored in whatis_proto.
|
|
* When finished, it is reformatted into whatis_final and then appended
|
|
* to whatis_lines.
|
|
*/
|
|
static struct sbuf *whatis_proto;
|
|
static struct sbuf *whatis_final;
|
|
static StringList *whatis_lines; /* collected output lines */
|
|
|
|
static char tmp_file[MAXPATHLEN]; /* path of temporary file, if any */
|
|
|
|
/* A set of possible names for the NAME man page section */
|
|
static const char *name_section_titles[] = {
|
|
"NAME", "Name", "NAMN", "BEZEICHNUNG", "\xcc\xbe\xbe\xce",
|
|
"\xee\xe1\xfa\xf7\xe1\xee\xe9\xe5", NULL
|
|
};
|
|
|
|
/* A subset of the mdoc(7) commands to ignore */
|
|
static char mdoc_commands[] = "ArDvErEvFlLiNmPa";
|
|
|
|
/*
|
|
* Frees a struct page_info and its content.
|
|
*/
|
|
static void
|
|
free_page_info(struct page_info *info)
|
|
{
|
|
free(info->filename);
|
|
free(info->name);
|
|
free(info->suffix);
|
|
free(info);
|
|
}
|
|
|
|
/*
|
|
* Allocates and fills in a new struct page_info given the
|
|
* name of the man section directory and the dirent of the file.
|
|
* If the file is not a man page, returns NULL.
|
|
*/
|
|
static struct page_info *
|
|
new_page_info(char *dir, struct dirent *dirent)
|
|
{
|
|
struct page_info *info;
|
|
int basename_length;
|
|
char *suffix;
|
|
struct stat st;
|
|
|
|
info = (struct page_info *) malloc(sizeof(struct page_info));
|
|
if (info == NULL)
|
|
err(1, "malloc");
|
|
basename_length = strlen(dirent->d_name);
|
|
suffix = &dirent->d_name[basename_length];
|
|
asprintf(&info->filename, "%s/%s", dir, dirent->d_name);
|
|
if ((info->gzipped = basename_length >= 4 && strcmp(&dirent->d_name[basename_length - 3], ".gz") == 0)) {
|
|
suffix -= 3;
|
|
*suffix = '\0';
|
|
}
|
|
for (;;) {
|
|
if (--suffix == dirent->d_name || !isalnum(*suffix)) {
|
|
if (*suffix == '.')
|
|
break;
|
|
if (verbose)
|
|
warnx("%s: invalid man page name", info->filename);
|
|
free(info->filename);
|
|
free(info);
|
|
return NULL;
|
|
}
|
|
}
|
|
*suffix++ = '\0';
|
|
info->name = strdup(dirent->d_name);
|
|
info->suffix = strdup(suffix);
|
|
if (stat(info->filename, &st) < 0) {
|
|
warn("%s", info->filename);
|
|
free_page_info(info);
|
|
return NULL;
|
|
}
|
|
if (!S_ISREG(st.st_mode)) {
|
|
if (verbose && !S_ISDIR(st.st_mode))
|
|
warnx("%s: not a regular file", info->filename);
|
|
free_page_info(info);
|
|
return NULL;
|
|
}
|
|
info->inode = st.st_ino;
|
|
return info;
|
|
}
|
|
|
|
/*
|
|
* Reset an sbuf's length to 0.
|
|
*/
|
|
static void
|
|
sbuf_clear(struct sbuf *sbuf)
|
|
{
|
|
sbuf->end = sbuf->content;
|
|
}
|
|
|
|
/*
|
|
* Allocate a new sbuf.
|
|
*/
|
|
static struct sbuf *
|
|
new_sbuf(void)
|
|
{
|
|
struct sbuf *sbuf = (struct sbuf *) malloc(sizeof(struct sbuf));
|
|
sbuf->content = (char *) malloc(LINE_ALLOC);
|
|
sbuf->last = sbuf->content + LINE_ALLOC - 1;
|
|
sbuf_clear(sbuf);
|
|
return sbuf;
|
|
}
|
|
|
|
/*
|
|
* Ensure that there is enough room in the sbuf for chars more characters.
|
|
*/
|
|
static void
|
|
sbuf_need(struct sbuf *sbuf, int nchars)
|
|
{
|
|
/* let's assume we only need to double it, but check just in case */
|
|
while (sbuf->end + nchars > sbuf->last) {
|
|
int alloc;
|
|
char *new_content;
|
|
|
|
alloc = (sbuf->last - sbuf->content + 1) * 2;
|
|
new_content = (char *) malloc(alloc);
|
|
memcpy(new_content, sbuf->content, sbuf->end - sbuf->content);
|
|
sbuf->end = new_content + (sbuf->end - sbuf->content);
|
|
free(sbuf->content);
|
|
sbuf->content = new_content;
|
|
}
|
|
}
|
|
|
|
/*
|
|
* Appends a string of a given length to the sbuf.
|
|
*/
|
|
static void
|
|
sbuf_append(struct sbuf *sbuf, const char *text, int length)
|
|
{
|
|
if (length > 0) {
|
|
sbuf_need(sbuf, length);
|
|
memcpy(sbuf->end, text, length);
|
|
sbuf->end += length;
|
|
}
|
|
}
|
|
|
|
/*
|
|
* Appends a null-terminated string to the sbuf.
|
|
*/
|
|
static void
|
|
sbuf_append_str(struct sbuf *sbuf, char *text)
|
|
{
|
|
sbuf_append(sbuf, text, strlen(text));
|
|
}
|
|
|
|
/*
|
|
* Appends an edited null-terminated string to the sbuf.
|
|
*/
|
|
static void
|
|
sbuf_append_edited(struct sbuf *sbuf, char *text, edited_copy copy)
|
|
{
|
|
int length = strlen(text);
|
|
if (length > 0) {
|
|
sbuf_need(sbuf, length);
|
|
sbuf->end = copy(text, sbuf->end, length);
|
|
}
|
|
}
|
|
|
|
/*
|
|
* Strips any of a set of chars from the end of the sbuf.
|
|
*/
|
|
static void
|
|
sbuf_strip(struct sbuf *sbuf, const char *set)
|
|
{
|
|
while (sbuf->end > sbuf->content && strchr(set, sbuf->end[-1]) != NULL)
|
|
sbuf->end--;
|
|
}
|
|
|
|
/*
|
|
* Returns the null-terminated string built by the sbuf.
|
|
*/
|
|
static char *
|
|
sbuf_content(struct sbuf *sbuf)
|
|
{
|
|
*sbuf->end = '\0';
|
|
return sbuf->content;
|
|
}
|
|
|
|
/*
|
|
* Returns true if no man page exists in the directory with
|
|
* any of the names in the StringList.
|
|
*/
|
|
static int
|
|
no_page_exists(char *dir, StringList *names, char *suffix)
|
|
{
|
|
char path[MAXPATHLEN];
|
|
size_t i;
|
|
|
|
for (i = 0; i < names->sl_cur; i++) {
|
|
snprintf(path, sizeof path, "%s/%s.%s.gz", dir, names->sl_str[i], suffix);
|
|
if (access(path, F_OK) < 0) {
|
|
path[strlen(path) - 3] = '\0';
|
|
if (access(path, F_OK) < 0)
|
|
continue;
|
|
}
|
|
return 0;
|
|
}
|
|
return 1;
|
|
}
|
|
|
|
static void
|
|
trap_signal(int sig __unused)
|
|
{
|
|
if (tmp_file[0] != '\0')
|
|
unlink(tmp_file);
|
|
exit(1);
|
|
}
|
|
|
|
/*
|
|
* Attempts to open an output file. Returns NULL if unsuccessful.
|
|
*/
|
|
static FILE *
|
|
open_output(char *name)
|
|
{
|
|
FILE *output;
|
|
|
|
whatis_lines = sl_init();
|
|
if (append) {
|
|
char line[LINE_ALLOC];
|
|
|
|
output = fopen(name, "r");
|
|
if (output == NULL) {
|
|
warn("%s", name);
|
|
exit_code = 1;
|
|
return NULL;
|
|
}
|
|
while (fgets(line, sizeof line, output) != NULL) {
|
|
line[strlen(line) - 1] = '\0';
|
|
sl_add(whatis_lines, strdup(line));
|
|
}
|
|
}
|
|
if (common_output == NULL) {
|
|
snprintf(tmp_file, sizeof tmp_file, "%s.tmp", name);
|
|
name = tmp_file;
|
|
}
|
|
output = fopen(name, "w");
|
|
if (output == NULL) {
|
|
warn("%s", name);
|
|
exit_code = 1;
|
|
return NULL;
|
|
}
|
|
return output;
|
|
}
|
|
|
|
static int
|
|
linesort(const void *a, const void *b)
|
|
{
|
|
return strcmp((const char *)(*(const char **)a), (const char *)(*(const char **)b));
|
|
}
|
|
|
|
/*
|
|
* Writes the unique sorted lines to the output file.
|
|
*/
|
|
static void
|
|
finish_output(FILE *output, char *name)
|
|
{
|
|
size_t i;
|
|
char *prev = NULL;
|
|
|
|
qsort(whatis_lines->sl_str, whatis_lines->sl_cur, sizeof(char *), linesort);
|
|
for (i = 0; i < whatis_lines->sl_cur; i++) {
|
|
char *line = whatis_lines->sl_str[i];
|
|
if (i > 0 && strcmp(line, prev) == 0)
|
|
continue;
|
|
prev = line;
|
|
fputs(line, output);
|
|
putc('\n', output);
|
|
}
|
|
fclose(output);
|
|
sl_free(whatis_lines, 1);
|
|
if (common_output == NULL) {
|
|
rename(tmp_file, name);
|
|
unlink(tmp_file);
|
|
}
|
|
}
|
|
|
|
static FILE *
|
|
open_whatis(char *mandir)
|
|
{
|
|
char filename[MAXPATHLEN];
|
|
|
|
snprintf(filename, sizeof filename, "%s/%s", mandir, whatis_name);
|
|
return open_output(filename);
|
|
}
|
|
|
|
static void
|
|
finish_whatis(FILE *output, char *mandir)
|
|
{
|
|
char filename[MAXPATHLEN];
|
|
|
|
snprintf(filename, sizeof filename, "%s/%s", mandir, whatis_name);
|
|
finish_output(output, filename);
|
|
}
|
|
|
|
/*
|
|
* Tests to see if the given directory has already been visited.
|
|
*/
|
|
static int
|
|
already_visited(char *dir)
|
|
{
|
|
struct stat st;
|
|
struct visited_dir *visit;
|
|
|
|
if (stat(dir, &st) < 0) {
|
|
warn("%s", dir);
|
|
exit_code = 1;
|
|
return 1;
|
|
}
|
|
SLIST_FOREACH(visit, &visited_dirs, next) {
|
|
if (visit->inode == st.st_ino &&
|
|
visit->device == st.st_dev) {
|
|
warnx("already visited %s", dir);
|
|
return 1;
|
|
}
|
|
}
|
|
visit = (struct visited_dir *) malloc(sizeof(struct visited_dir));
|
|
visit->device = st.st_dev;
|
|
visit->inode = st.st_ino;
|
|
SLIST_INSERT_HEAD(&visited_dirs, visit, next);
|
|
return 0;
|
|
}
|
|
|
|
/*
|
|
* Removes trailing spaces from a string, returning a pointer to just
|
|
* beyond the new last character.
|
|
*/
|
|
static char *
|
|
trim_rhs(char *str)
|
|
{
|
|
char *rhs = &str[strlen(str)];
|
|
while (--rhs > str && isspace(*rhs))
|
|
;
|
|
*++rhs = '\0';
|
|
return rhs;
|
|
}
|
|
|
|
/*
|
|
* Returns a pointer to the next non-space character in the string.
|
|
*/
|
|
static char *
|
|
skip_spaces(char *s)
|
|
{
|
|
while (*s != '\0' && isspace(*s))
|
|
s++;
|
|
return s;
|
|
}
|
|
|
|
/*
|
|
* Returns whether the string contains only digits.
|
|
*/
|
|
static int
|
|
only_digits(char *line)
|
|
{
|
|
if (!isdigit(*line++))
|
|
return 0;
|
|
while (isdigit(*line))
|
|
line++;
|
|
return *line == '\0';
|
|
}
|
|
|
|
/*
|
|
* Returns whether the line is of one of the forms:
|
|
* .Sh NAME
|
|
* .Sh "NAME"
|
|
* etc.
|
|
* assuming that section_start is ".Sh".
|
|
*/
|
|
static int
|
|
name_section_line(char *line, const char *section_start)
|
|
{
|
|
char *rhs;
|
|
const char **title;
|
|
|
|
if (strncmp(line, section_start, 3) != 0)
|
|
return 0;
|
|
line = skip_spaces(line + 3);
|
|
rhs = trim_rhs(line);
|
|
if (*line == '"') {
|
|
line++;
|
|
if (*--rhs == '"')
|
|
*rhs = '\0';
|
|
}
|
|
for (title = name_section_titles; *title != NULL; title++)
|
|
if (strcmp(*title, line) == 0)
|
|
return 1;
|
|
return 0;
|
|
}
|
|
|
|
/*
|
|
* Copies characters while removing the most common nroff/troff
|
|
* markup:
|
|
* \(em, \(mi, \s[+-N], \&
|
|
* \fF, \f(fo, \f[font]
|
|
* \*s, \*(st, \*[stringvar]
|
|
*/
|
|
static char *
|
|
de_nroff_copy(char *from, char *to, int fromlen)
|
|
{
|
|
char *from_end = &from[fromlen];
|
|
while (from < from_end) {
|
|
switch (*from) {
|
|
case '\\':
|
|
switch (*++from) {
|
|
case '(':
|
|
if (strncmp(&from[1], "em", 2) == 0 ||
|
|
strncmp(&from[1], "mi", 2) == 0) {
|
|
from += 3;
|
|
continue;
|
|
}
|
|
break;
|
|
case 's':
|
|
if (*++from == '-')
|
|
from++;
|
|
while (isdigit(*from))
|
|
from++;
|
|
continue;
|
|
case 'f':
|
|
case '*':
|
|
if (*++from == '(')
|
|
from += 3;
|
|
else if (*from == '[') {
|
|
while (*++from != ']' && from < from_end);
|
|
from++;
|
|
} else
|
|
from++;
|
|
continue;
|
|
case '&':
|
|
from++;
|
|
continue;
|
|
}
|
|
break;
|
|
}
|
|
*to++ = *from++;
|
|
}
|
|
return to;
|
|
}
|
|
|
|
/*
|
|
* Appends a string with the nroff formatting removed.
|
|
*/
|
|
static void
|
|
add_nroff(char *text)
|
|
{
|
|
sbuf_append_edited(whatis_proto, text, de_nroff_copy);
|
|
}
|
|
|
|
/*
|
|
* Appends "name(suffix), " to whatis_final.
|
|
*/
|
|
static void
|
|
add_whatis_name(char *name, char *suffix)
|
|
{
|
|
if (*name != '\0') {
|
|
sbuf_append_str(whatis_final, name);
|
|
sbuf_append(whatis_final, "(", 1);
|
|
sbuf_append_str(whatis_final, suffix);
|
|
sbuf_append(whatis_final, "), ", 3);
|
|
}
|
|
}
|
|
|
|
/*
|
|
* Processes an old-style man(7) line. This ignores commands with only
|
|
* a single number argument.
|
|
*/
|
|
static void
|
|
process_man_line(char *line)
|
|
{
|
|
if (*line == '.') {
|
|
while (isalpha(*++line))
|
|
;
|
|
line = skip_spaces(line);
|
|
if (only_digits(line))
|
|
return;
|
|
} else
|
|
line = skip_spaces(line);
|
|
if (*line != '\0') {
|
|
add_nroff(line);
|
|
sbuf_append(whatis_proto, " ", 1);
|
|
}
|
|
}
|
|
|
|
/*
|
|
* Processes a new-style mdoc(7) line.
|
|
*/
|
|
static void
|
|
process_mdoc_line(char *line)
|
|
{
|
|
int xref;
|
|
int arg = 0;
|
|
char *line_end = &line[strlen(line)];
|
|
int orig_length = sbuf_length(whatis_proto);
|
|
char *next;
|
|
|
|
if (*line == '\0')
|
|
return;
|
|
if (line[0] != '.' || !isupper(line[1]) || !islower(line[2])) {
|
|
add_nroff(skip_spaces(line));
|
|
sbuf_append(whatis_proto, " ", 1);
|
|
return;
|
|
}
|
|
xref = strncmp(line, ".Xr", 3) == 0;
|
|
line += 3;
|
|
while ((line = skip_spaces(line)) < line_end) {
|
|
if (*line == '"') {
|
|
next = ++line;
|
|
for (;;) {
|
|
next = strchr(next, '"');
|
|
if (next == NULL)
|
|
break;
|
|
strcpy(next, &next[1]);
|
|
line_end--;
|
|
if (*next != '"')
|
|
break;
|
|
next++;
|
|
}
|
|
} else
|
|
next = strpbrk(line, " \t");
|
|
if (next != NULL)
|
|
*next++ = '\0';
|
|
else
|
|
next = line_end;
|
|
if (isupper(*line) && islower(line[1]) && line[2] == '\0') {
|
|
if (strcmp(line, "Ns") == 0) {
|
|
arg = 0;
|
|
line = next;
|
|
continue;
|
|
}
|
|
if (strstr(mdoc_commands, line) != NULL) {
|
|
line = next;
|
|
continue;
|
|
}
|
|
}
|
|
if (arg > 0 && strchr(",.:;?!)]", *line) == 0) {
|
|
if (xref) {
|
|
sbuf_append(whatis_proto, "(", 1);
|
|
add_nroff(line);
|
|
sbuf_append(whatis_proto, ")", 1);
|
|
xref = 0;
|
|
line = blank;
|
|
} else
|
|
sbuf_append(whatis_proto, " ", 1);
|
|
}
|
|
add_nroff(line);
|
|
arg++;
|
|
line = next;
|
|
}
|
|
if (sbuf_length(whatis_proto) > orig_length)
|
|
sbuf_append(whatis_proto, " ", 1);
|
|
}
|
|
|
|
/*
|
|
* Collects a list of comma-separated names from the text.
|
|
*/
|
|
static void
|
|
collect_names(StringList *names, char *text)
|
|
{
|
|
char *arg;
|
|
|
|
for (;;) {
|
|
arg = text;
|
|
text = strchr(text, ',');
|
|
if (text != NULL)
|
|
*text++ = '\0';
|
|
sl_add(names, arg);
|
|
if (text == NULL)
|
|
return;
|
|
if (*text == ' ')
|
|
text++;
|
|
}
|
|
}
|
|
|
|
enum { STATE_UNKNOWN, STATE_MANSTYLE, STATE_MDOCNAME, STATE_MDOCDESC };
|
|
|
|
/*
|
|
* Processes a man page source into a single whatis line and adds it
|
|
* to whatis_lines.
|
|
*/
|
|
static void
|
|
process_page(struct page_info *page, char *section_dir)
|
|
{
|
|
gzFile *in;
|
|
char buffer[4096];
|
|
char *line;
|
|
StringList *names;
|
|
char *descr;
|
|
int state = STATE_UNKNOWN;
|
|
size_t i;
|
|
|
|
sbuf_clear(whatis_proto);
|
|
if ((in = gzopen(page->filename, "r")) == NULL) {
|
|
warn("%s", page->filename);
|
|
exit_code = 1;
|
|
return;
|
|
}
|
|
while (gzgets(in, buffer, sizeof buffer) != NULL) {
|
|
line = buffer;
|
|
if (strncmp(line, ".\\\"", 3) == 0) /* ignore comments */
|
|
continue;
|
|
switch (state) {
|
|
/*
|
|
* haven't reached the NAME section yet.
|
|
*/
|
|
case STATE_UNKNOWN:
|
|
if (name_section_line(line, ".SH"))
|
|
state = STATE_MANSTYLE;
|
|
else if (name_section_line(line, ".Sh"))
|
|
state = STATE_MDOCNAME;
|
|
continue;
|
|
/*
|
|
* Inside an old-style .SH NAME section.
|
|
*/
|
|
case STATE_MANSTYLE:
|
|
if (strncmp(line, ".SH", 3) == 0)
|
|
break;
|
|
trim_rhs(line);
|
|
if (strcmp(line, ".") == 0)
|
|
continue;
|
|
if (strncmp(line, ".IX", 3) == 0) {
|
|
line += 3;
|
|
line = skip_spaces(line);
|
|
}
|
|
process_man_line(line);
|
|
continue;
|
|
/*
|
|
* Inside a new-style .Sh NAME section (the .Nm part).
|
|
*/
|
|
case STATE_MDOCNAME:
|
|
trim_rhs(line);
|
|
if (strncmp(line, ".Nm", 3) == 0) {
|
|
process_mdoc_line(line);
|
|
continue;
|
|
} else {
|
|
if (strcmp(line, ".") == 0)
|
|
continue;
|
|
sbuf_append(whatis_proto, "- ", 2);
|
|
state = STATE_MDOCDESC;
|
|
}
|
|
/* fall through */
|
|
/*
|
|
* Inside a new-style .Sh NAME section (after the .Nm-s).
|
|
*/
|
|
case STATE_MDOCDESC:
|
|
if (strncmp(line, ".Sh", 3) == 0)
|
|
break;
|
|
trim_rhs(line);
|
|
if (strcmp(line, ".") == 0)
|
|
continue;
|
|
process_mdoc_line(line);
|
|
continue;
|
|
}
|
|
break;
|
|
}
|
|
gzclose(in);
|
|
sbuf_strip(whatis_proto, " \t.-");
|
|
line = sbuf_content(whatis_proto);
|
|
/*
|
|
* line now contains the appropriate data, but without
|
|
* the proper indentation or the section appended to each name.
|
|
*/
|
|
descr = strstr(line, " - ");
|
|
if (descr == NULL) {
|
|
descr = strchr(line, ' ');
|
|
if (descr == NULL) {
|
|
if (verbose)
|
|
fprintf(stderr, " ignoring junk description \"%s\"\n", line);
|
|
return;
|
|
}
|
|
*descr++ = '\0';
|
|
} else {
|
|
*descr = '\0';
|
|
descr += 3;
|
|
}
|
|
names = sl_init();
|
|
collect_names(names, line);
|
|
sbuf_clear(whatis_final);
|
|
if (!sl_find(names, page->name) && no_page_exists(section_dir, names, page->suffix)) {
|
|
/*
|
|
* Add the page name since that's the only thing that
|
|
* man(1) will find.
|
|
*/
|
|
add_whatis_name(page->name, page->suffix);
|
|
}
|
|
for (i = 0; i < names->sl_cur; i++)
|
|
add_whatis_name(names->sl_str[i], page->suffix);
|
|
sl_free(names, 0);
|
|
sbuf_retract(whatis_final, 2); /* remove last ", " */
|
|
while (sbuf_length(whatis_final) < indent)
|
|
sbuf_append(whatis_final, " ", 1);
|
|
sbuf_append(whatis_final, " - ", 3);
|
|
sbuf_append_str(whatis_final, skip_spaces(descr));
|
|
sl_add(whatis_lines, strdup(sbuf_content(whatis_final)));
|
|
}
|
|
|
|
/*
|
|
* Sorts pages first by inode number, then by name.
|
|
*/
|
|
static int
|
|
pagesort(const void *a, const void *b)
|
|
{
|
|
struct page_info *p1 = *(struct page_info **) a;
|
|
struct page_info *p2 = *(struct page_info **) b;
|
|
if (p1->inode == p2->inode)
|
|
return strcmp(p1->name, p2->name);
|
|
return p1->inode - p2->inode;
|
|
}
|
|
|
|
/*
|
|
* Processes a single man section.
|
|
*/
|
|
static void
|
|
process_section(char *section_dir)
|
|
{
|
|
struct dirent **entries;
|
|
int nentries;
|
|
struct page_info **pages;
|
|
int npages = 0;
|
|
int i;
|
|
ino_t prev_inode = 0;
|
|
|
|
if (verbose)
|
|
fprintf(stderr, " %s\n", section_dir);
|
|
|
|
/*
|
|
* scan the man section directory for pages
|
|
*/
|
|
nentries = scandir(section_dir, &entries, NULL, alphasort);
|
|
if (nentries < 0) {
|
|
warn("%s", section_dir);
|
|
exit_code = 1;
|
|
return;
|
|
}
|
|
/*
|
|
* collect information about man pages
|
|
*/
|
|
pages = (struct page_info **) calloc(nentries, sizeof(struct page_info *));
|
|
for (i = 0; i < nentries; i++) {
|
|
struct page_info *info = new_page_info(section_dir, entries[i]);
|
|
if (info != NULL)
|
|
pages[npages++] = info;
|
|
free(entries[i]);
|
|
}
|
|
free(entries);
|
|
qsort(pages, npages, sizeof(struct page_info *), pagesort);
|
|
/*
|
|
* process each unique page
|
|
*/
|
|
for (i = 0; i < npages; i++) {
|
|
struct page_info *page = pages[i];
|
|
if (page->inode != prev_inode) {
|
|
prev_inode = page->inode;
|
|
if (verbose)
|
|
fprintf(stderr, " reading %s\n", page->filename);
|
|
process_page(page, section_dir);
|
|
} else if (verbose)
|
|
fprintf(stderr, " skipping %s, duplicate\n", page->filename);
|
|
free_page_info(page);
|
|
}
|
|
free(pages);
|
|
}
|
|
|
|
/*
|
|
* Returns whether the directory entry is a man page section.
|
|
*/
|
|
static int
|
|
select_sections(struct dirent *entry)
|
|
{
|
|
char *p = &entry->d_name[3];
|
|
|
|
if (strncmp(entry->d_name, "man", 3) != 0)
|
|
return 0;
|
|
while (*p != '\0') {
|
|
if (!isalnum(*p++))
|
|
return 0;
|
|
}
|
|
return 1;
|
|
}
|
|
|
|
/*
|
|
* Processes a single top-level man directory by finding all the
|
|
* sub-directories named man* and processing each one in turn.
|
|
*/
|
|
static void
|
|
process_mandir(char *dir_name)
|
|
{
|
|
struct dirent **entries;
|
|
int nsections;
|
|
FILE *fp = NULL;
|
|
int i;
|
|
struct stat st;
|
|
|
|
if (already_visited(dir_name))
|
|
return;
|
|
if (verbose)
|
|
fprintf(stderr, "man directory %s\n", dir_name);
|
|
nsections = scandir(dir_name, &entries, select_sections, alphasort);
|
|
if (nsections < 0) {
|
|
warn("%s", dir_name);
|
|
exit_code = 1;
|
|
return;
|
|
}
|
|
if (common_output == NULL && (fp = open_whatis(dir_name)) == NULL)
|
|
return;
|
|
for (i = 0; i < nsections; i++) {
|
|
char section_dir[MAXPATHLEN];
|
|
snprintf(section_dir, sizeof section_dir, "%s/%s", dir_name, entries[i]->d_name);
|
|
process_section(section_dir);
|
|
snprintf(section_dir, sizeof section_dir, "%s/%s/%s", dir_name,
|
|
entries[i]->d_name, machine);
|
|
if (stat(section_dir, &st) == 0 && S_ISDIR(st.st_mode))
|
|
process_section(section_dir);
|
|
free(entries[i]);
|
|
}
|
|
free(entries);
|
|
if (common_output == NULL)
|
|
finish_whatis(fp, dir_name);
|
|
}
|
|
|
|
/*
|
|
* Processes one argument, which may be a colon-separated list of
|
|
* directories.
|
|
*/
|
|
static void
|
|
process_argument(const char *arg)
|
|
{
|
|
char *dir;
|
|
char *mandir;
|
|
char *parg;
|
|
|
|
parg = strdup(arg);
|
|
if (parg == NULL)
|
|
err(1, "out of memory");
|
|
while ((dir = strsep(&parg, ":")) != NULL) {
|
|
if (locale != NULL) {
|
|
asprintf(&mandir, "%s/%s", dir, locale);
|
|
process_mandir(mandir);
|
|
free(mandir);
|
|
if (lang_locale != NULL) {
|
|
asprintf(&mandir, "%s/%s", dir, lang_locale);
|
|
process_mandir(mandir);
|
|
free(mandir);
|
|
}
|
|
} else {
|
|
process_mandir(dir);
|
|
}
|
|
}
|
|
free(parg);
|
|
}
|
|
|
|
|
|
int
|
|
main(int argc, char **argv)
|
|
{
|
|
int opt;
|
|
extern int optind;
|
|
extern char *optarg;
|
|
FILE *fp = NULL;
|
|
|
|
while ((opt = getopt(argc, argv, "ai:n:o:vL")) != -1) {
|
|
switch (opt) {
|
|
case 'a':
|
|
append++;
|
|
break;
|
|
case 'i':
|
|
indent = atoi(optarg);
|
|
break;
|
|
case 'n':
|
|
whatis_name = optarg;
|
|
break;
|
|
case 'o':
|
|
common_output = optarg;
|
|
break;
|
|
case 'v':
|
|
verbose++;
|
|
break;
|
|
case 'L':
|
|
locale = getenv("LC_ALL");
|
|
if (locale == NULL)
|
|
locale = getenv("LC_CTYPE");
|
|
if (locale == NULL)
|
|
locale = getenv("LANG");
|
|
if (locale != NULL) {
|
|
char *sep = strchr(locale, '_');
|
|
if (sep != NULL && isupper(sep[1]) &&
|
|
isupper(sep[2])) {
|
|
asprintf(&lang_locale, "%.*s%s", sep - locale, locale, &sep[3]);
|
|
}
|
|
}
|
|
break;
|
|
default:
|
|
fprintf(stderr, "usage: %s [-a] [-i indent] [-n name] [-o output_file] [-v] [-L] [directories...]\n", argv[0]);
|
|
exit(1);
|
|
}
|
|
}
|
|
|
|
signal(SIGINT, trap_signal);
|
|
signal(SIGHUP, trap_signal);
|
|
signal(SIGQUIT, trap_signal);
|
|
signal(SIGTERM, trap_signal);
|
|
SLIST_INIT(&visited_dirs);
|
|
whatis_proto = new_sbuf();
|
|
whatis_final = new_sbuf();
|
|
|
|
if ((machine = getenv("MACHINE")) == NULL)
|
|
machine = MACHINE;
|
|
|
|
if (common_output != NULL && (fp = open_output(common_output)) == NULL)
|
|
err(1, "%s", common_output);
|
|
if (optind == argc) {
|
|
const char *manpath = getenv("MANPATH");
|
|
if (manpath == NULL)
|
|
manpath = DEFAULT_MANPATH;
|
|
process_argument(manpath);
|
|
} else {
|
|
while (optind < argc)
|
|
process_argument(argv[optind++]);
|
|
}
|
|
if (common_output != NULL)
|
|
finish_output(fp, common_output);
|
|
exit(exit_code);
|
|
}
|