version 1.90, 2013/12/27 23:41:55 |
version 1.103, 2014/01/06 03:02:46 |
|
|
/* $Id$ */ |
/* $Id$ */ |
/* |
/* |
* Copyright (c) 2011, 2012 Kristaps Dzonsons <kristaps@bsd.lv> |
* Copyright (c) 2011, 2012 Kristaps Dzonsons <kristaps@bsd.lv> |
* Copyright (c) 2011, 2012, 2013 Ingo Schwarze <schwarze@openbsd.org> |
* Copyright (c) 2011, 2012, 2013, 2014 Ingo Schwarze <schwarze@openbsd.org> |
* |
* |
* Permission to use, copy, modify, and distribute this software for any |
* Permission to use, copy, modify, and distribute this software for any |
* purpose with or without fee is hereby granted, provided that the above |
* purpose with or without fee is hereby granted, provided that the above |
|
|
}; |
}; |
|
|
struct str { |
struct str { |
char *utf8; /* key in UTF-8 form */ |
char *rendered; /* key in UTF-8 or ASCII form */ |
const struct mpage *mpage; /* if set, the owning parse */ |
const struct mpage *mpage; /* if set, the owning parse */ |
uint64_t mask; /* bitmask in sequence */ |
uint64_t mask; /* bitmask in sequence */ |
char key[]; /* the string itself */ |
char key[]; /* may contain escape sequences */ |
}; |
}; |
|
|
struct inodev { |
struct inodev { |
|
|
struct mlink *next; /* singly linked list */ |
struct mlink *next; /* singly linked list */ |
}; |
}; |
|
|
struct title { |
|
char *title; /* name(sec/arch) given inside the file */ |
|
char *file; /* file name in case of mismatch */ |
|
}; |
|
|
|
enum stmt { |
enum stmt { |
STMT_DELETE_PAGE = 0, /* delete mpage */ |
STMT_DELETE_PAGE = 0, /* delete mpage */ |
STMT_INSERT_PAGE, /* insert mpage */ |
STMT_INSERT_PAGE, /* insert mpage */ |
Line 143 static void *hash_alloc(size_t, void *); |
|
Line 138 static void *hash_alloc(size_t, void *); |
|
static void hash_free(void *, size_t, void *); |
static void hash_free(void *, size_t, void *); |
static void *hash_halloc(size_t, void *); |
static void *hash_halloc(size_t, void *); |
static void mlink_add(struct mlink *, const struct stat *); |
static void mlink_add(struct mlink *, const struct stat *); |
|
static int mlink_check(struct mpage *, struct mlink *); |
static void mlink_free(struct mlink *); |
static void mlink_free(struct mlink *); |
static void mlinks_undupe(struct mpage *); |
static void mlinks_undupe(struct mpage *); |
static void mpages_free(void); |
static void mpages_free(void); |
static void mpages_merge(struct mchars *, struct mparse *, int); |
static void mpages_merge(struct mchars *, struct mparse *); |
static void parse_cat(struct mpage *); |
static void parse_cat(struct mpage *); |
static void parse_man(struct mpage *, const struct man_node *); |
static void parse_man(struct mpage *, const struct man_node *); |
static void parse_mdoc(struct mpage *, const struct mdoc_node *); |
static void parse_mdoc(struct mpage *, const struct mdoc_node *); |
Line 154 static int parse_mdoc_body(struct mpage *, const stru |
|
Line 150 static int parse_mdoc_body(struct mpage *, const stru |
|
static int parse_mdoc_head(struct mpage *, const struct mdoc_node *); |
static int parse_mdoc_head(struct mpage *, const struct mdoc_node *); |
static int parse_mdoc_Fd(struct mpage *, const struct mdoc_node *); |
static int parse_mdoc_Fd(struct mpage *, const struct mdoc_node *); |
static int parse_mdoc_Fn(struct mpage *, const struct mdoc_node *); |
static int parse_mdoc_Fn(struct mpage *, const struct mdoc_node *); |
static int parse_mdoc_In(struct mpage *, const struct mdoc_node *); |
|
static int parse_mdoc_Nd(struct mpage *, const struct mdoc_node *); |
static int parse_mdoc_Nd(struct mpage *, const struct mdoc_node *); |
static int parse_mdoc_Nm(struct mpage *, const struct mdoc_node *); |
static int parse_mdoc_Nm(struct mpage *, const struct mdoc_node *); |
static int parse_mdoc_Sh(struct mpage *, const struct mdoc_node *); |
static int parse_mdoc_Sh(struct mpage *, const struct mdoc_node *); |
static int parse_mdoc_St(struct mpage *, const struct mdoc_node *); |
|
static int parse_mdoc_Xr(struct mpage *, const struct mdoc_node *); |
static int parse_mdoc_Xr(struct mpage *, const struct mdoc_node *); |
static void putkey(const struct mpage *, |
static void putkey(const struct mpage *, |
const char *, uint64_t); |
const char *, uint64_t); |
Line 166 static void putkeys(const struct mpage *, |
|
Line 160 static void putkeys(const struct mpage *, |
|
const char *, size_t, uint64_t); |
const char *, size_t, uint64_t); |
static void putmdockey(const struct mpage *, |
static void putmdockey(const struct mpage *, |
const struct mdoc_node *, uint64_t); |
const struct mdoc_node *, uint64_t); |
|
static void render_key(struct mchars *, struct str *); |
static void say(const char *, const char *, ...); |
static void say(const char *, const char *, ...); |
static int set_basedir(const char *); |
static int set_basedir(const char *); |
static int treescan(void); |
static int treescan(void); |
static size_t utf8(unsigned int, char [7]); |
static size_t utf8(unsigned int, char [7]); |
static void utf8key(struct mchars *, struct str *); |
|
|
|
static char *progname; |
static char *progname; |
static int use_all; /* use all found files */ |
|
static int nodb; /* no database changes */ |
static int nodb; /* no database changes */ |
|
static int quick; /* abort the parse early */ |
|
static int use_all; /* use all found files */ |
static int verb; /* print what we're doing */ |
static int verb; /* print what we're doing */ |
static int warnings; /* warn about crap */ |
static int warnings; /* warn about crap */ |
|
static int write_utf8; /* write UTF-8 output; else ASCII */ |
static int exitcode; /* to be returned by main */ |
static int exitcode; /* to be returned by main */ |
static enum op op; /* operational mode */ |
static enum op op; /* operational mode */ |
static char basedir[PATH_MAX]; /* current base directory */ |
static char basedir[PATH_MAX]; /* current base directory */ |
Line 216 static const struct mdoc_handler mdocs[MDOC_MAX] = { |
|
Line 212 static const struct mdoc_handler mdocs[MDOC_MAX] = { |
|
{ parse_mdoc_Fn, 0 }, /* Fn */ |
{ parse_mdoc_Fn, 0 }, /* Fn */ |
{ NULL, TYPE_Ft }, /* Ft */ |
{ NULL, TYPE_Ft }, /* Ft */ |
{ NULL, TYPE_Ic }, /* Ic */ |
{ NULL, TYPE_Ic }, /* Ic */ |
{ parse_mdoc_In, TYPE_In }, /* In */ |
{ NULL, TYPE_In }, /* In */ |
{ NULL, TYPE_Li }, /* Li */ |
{ NULL, TYPE_Li }, /* Li */ |
{ parse_mdoc_Nd, TYPE_Nd }, /* Nd */ |
{ parse_mdoc_Nd, TYPE_Nd }, /* Nd */ |
{ parse_mdoc_Nm, TYPE_Nm }, /* Nm */ |
{ parse_mdoc_Nm, TYPE_Nm }, /* Nm */ |
Line 224 static const struct mdoc_handler mdocs[MDOC_MAX] = { |
|
Line 220 static const struct mdoc_handler mdocs[MDOC_MAX] = { |
|
{ NULL, 0 }, /* Ot */ |
{ NULL, 0 }, /* Ot */ |
{ NULL, TYPE_Pa }, /* Pa */ |
{ NULL, TYPE_Pa }, /* Pa */ |
{ NULL, 0 }, /* Rv */ |
{ NULL, 0 }, /* Rv */ |
{ parse_mdoc_St, 0 }, /* St */ |
{ NULL, TYPE_St }, /* St */ |
{ NULL, TYPE_Va }, /* Va */ |
{ NULL, TYPE_Va }, /* Va */ |
{ parse_mdoc_body, TYPE_Va }, /* Vt */ |
{ parse_mdoc_body, TYPE_Va }, /* Vt */ |
{ parse_mdoc_Xr, 0 }, /* Xr */ |
{ parse_mdoc_Xr, 0 }, /* Xr */ |
Line 352 main(int argc, char *argv[]) |
|
Line 348 main(int argc, char *argv[]) |
|
path_arg = NULL; |
path_arg = NULL; |
op = OP_DEFAULT; |
op = OP_DEFAULT; |
|
|
while (-1 != (ch = getopt(argc, argv, "aC:d:ntu:vW"))) |
while (-1 != (ch = getopt(argc, argv, "aC:d:nQT:tu:vW"))) |
switch (ch) { |
switch (ch) { |
case ('a'): |
case ('a'): |
use_all = 1; |
use_all = 1; |
Line 370 main(int argc, char *argv[]) |
|
Line 366 main(int argc, char *argv[]) |
|
case ('n'): |
case ('n'): |
nodb = 1; |
nodb = 1; |
break; |
break; |
|
case ('Q'): |
|
quick = 1; |
|
break; |
|
case ('T'): |
|
if (strcmp(optarg, "utf8")) { |
|
fprintf(stderr, "-T%s: Unsupported " |
|
"output format\n", optarg); |
|
goto usage; |
|
} |
|
write_utf8 = 1; |
|
break; |
case ('t'): |
case ('t'): |
CHECKOP(op, ch); |
CHECKOP(op, ch); |
dup2(STDOUT_FILENO, STDERR_FILENO); |
dup2(STDOUT_FILENO, STDERR_FILENO); |
Line 401 main(int argc, char *argv[]) |
|
Line 408 main(int argc, char *argv[]) |
|
|
|
exitcode = (int)MANDOCLEVEL_OK; |
exitcode = (int)MANDOCLEVEL_OK; |
mp = mparse_alloc(MPARSE_AUTO, |
mp = mparse_alloc(MPARSE_AUTO, |
MANDOCLEVEL_FATAL, NULL, NULL, NULL); |
MANDOCLEVEL_FATAL, NULL, NULL, quick); |
mc = mchars_alloc(); |
mc = mchars_alloc(); |
|
|
ohash_init(&mpages, 6, &mpages_info); |
ohash_init(&mpages, 6, &mpages_info); |
Line 427 main(int argc, char *argv[]) |
|
Line 434 main(int argc, char *argv[]) |
|
if (OP_TEST != op) |
if (OP_TEST != op) |
dbprune(); |
dbprune(); |
if (OP_DELETE != op) |
if (OP_DELETE != op) |
mpages_merge(mc, mp, 0); |
mpages_merge(mc, mp); |
dbclose(1); |
dbclose(1); |
} else { |
} else { |
/* |
/* |
Line 471 main(int argc, char *argv[]) |
|
Line 478 main(int argc, char *argv[]) |
|
if (0 == dbopen(0)) |
if (0 == dbopen(0)) |
goto out; |
goto out; |
|
|
mpages_merge(mc, mp, warnings && !use_all); |
mpages_merge(mc, mp); |
dbclose(0); |
dbclose(0); |
|
|
if (j + 1 < dirs.sz) { |
if (j + 1 < dirs.sz) { |
|
|
ohash_delete(&mlinks); |
ohash_delete(&mlinks); |
return(exitcode); |
return(exitcode); |
usage: |
usage: |
fprintf(stderr, "usage: %s [-anvW] [-C file]\n" |
fprintf(stderr, "usage: %s [-anQvW] [-C file] [-Tutf8]\n" |
" %s [-anvW] dir ...\n" |
" %s [-anQvW] [-Tutf8] dir ...\n" |
" %s [-nvW] -d dir [file ...]\n" |
" %s [-nQvW] [-Tutf8] -d dir [file ...]\n" |
" %s [-nvW] -u dir [file ...]\n" |
" %s [-nvW] -u dir [file ...]\n" |
" %s -t file ...\n", |
" %s [-Q] -t file ...\n", |
progname, progname, progname, |
progname, progname, progname, |
progname, progname); |
progname, progname); |
|
|
|
|
FTSENT *ff; |
FTSENT *ff; |
struct mlink *mlink; |
struct mlink *mlink; |
int dform; |
int dform; |
char *fsec; |
char *dsec, *arch, *fsec, *cp; |
const char *dsec, *arch, *cp, *path; |
const char *path; |
const char *argv[2]; |
const char *argv[2]; |
|
|
argv[0] = "."; |
argv[0] = "."; |
|
|
continue; |
continue; |
} else |
} else |
fsec[-1] = '\0'; |
fsec[-1] = '\0'; |
|
|
mlink = mandoc_calloc(1, sizeof(struct mlink)); |
mlink = mandoc_calloc(1, sizeof(struct mlink)); |
strlcpy(mlink->file, path, sizeof(mlink->file)); |
strlcpy(mlink->file, path, sizeof(mlink->file)); |
mlink->dform = dform; |
mlink->dform = dform; |
if (NULL != dsec) |
mlink->dsec = dsec; |
mlink->dsec = mandoc_strdup(dsec); |
mlink->arch = arch; |
if (NULL != arch) |
mlink->name = ff->fts_name; |
mlink->arch = mandoc_strdup(arch); |
mlink->fsec = fsec; |
mlink->name = mandoc_strdup(ff->fts_name); |
|
if (NULL != fsec) |
|
mlink->fsec = mandoc_strdup(fsec); |
|
mlink_add(mlink, ff->fts_statp); |
mlink_add(mlink, ff->fts_statp); |
continue; |
continue; |
} else if (FTS_D != ff->fts_info && |
} else if (FTS_D != ff->fts_info && |
|
|
* Try to infer this from the name. |
* Try to infer this from the name. |
* If we're not in use_all, enforce it. |
* If we're not in use_all, enforce it. |
*/ |
*/ |
dsec = NULL; |
|
dform = FORM_NONE; |
|
cp = ff->fts_name; |
cp = ff->fts_name; |
if (FTS_DP == ff->fts_info) |
if (FTS_DP == ff->fts_info) |
break; |
break; |
|
|
} else if (0 == strncmp(cp, "cat", 3)) { |
} else if (0 == strncmp(cp, "cat", 3)) { |
dform = FORM_CAT; |
dform = FORM_CAT; |
dsec = cp + 3; |
dsec = cp + 3; |
|
} else { |
|
dform = FORM_NONE; |
|
dsec = NULL; |
} |
} |
|
|
if (NULL != dsec || use_all) |
if (NULL != dsec || use_all) |
|
|
* Possibly our architecture. |
* Possibly our architecture. |
* If we're descending, keep tabs on it. |
* If we're descending, keep tabs on it. |
*/ |
*/ |
arch = NULL; |
|
if (FTS_DP != ff->fts_info && NULL != dsec) |
if (FTS_DP != ff->fts_info && NULL != dsec) |
arch = ff->fts_name; |
arch = ff->fts_name; |
|
else |
|
arch = NULL; |
break; |
break; |
default: |
default: |
if (FTS_DP == ff->fts_info || use_all) |
if (FTS_DP == ff->fts_info || use_all) |
Line 720 filescan(const char *file) |
|
Line 727 filescan(const char *file) |
|
*p++ = '\0'; |
*p++ = '\0'; |
if (0 == strncmp(start, "man", 3)) { |
if (0 == strncmp(start, "man", 3)) { |
mlink->dform = FORM_SRC; |
mlink->dform = FORM_SRC; |
mlink->dsec = mandoc_strdup(start + 3); |
mlink->dsec = start + 3; |
} else if (0 == strncmp(start, "cat", 3)) { |
} else if (0 == strncmp(start, "cat", 3)) { |
mlink->dform = FORM_CAT; |
mlink->dform = FORM_CAT; |
mlink->dsec = mandoc_strdup(start + 3); |
mlink->dsec = start + 3; |
} |
} |
|
|
start = p; |
start = p; |
if (NULL != mlink->dsec && NULL != (p = strchr(start, '/'))) { |
if (NULL != mlink->dsec && NULL != (p = strchr(start, '/'))) { |
*p++ = '\0'; |
*p++ = '\0'; |
mlink->arch = mandoc_strdup(start); |
mlink->arch = start; |
start = p; |
start = p; |
} |
} |
} |
} |
Line 744 filescan(const char *file) |
|
Line 751 filescan(const char *file) |
|
|
|
if ('.' == *p) { |
if ('.' == *p) { |
*p++ = '\0'; |
*p++ = '\0'; |
mlink->fsec = mandoc_strdup(p); |
mlink->fsec = p; |
} |
} |
|
|
/* |
/* |
Line 756 filescan(const char *file) |
|
Line 763 filescan(const char *file) |
|
mlink->name = p + 1; |
mlink->name = p + 1; |
*p = '\0'; |
*p = '\0'; |
} |
} |
mlink->name = mandoc_strdup(mlink->name); |
|
|
|
mlink_add(mlink, &st); |
mlink_add(mlink, &st); |
} |
} |
|
|
Line 770 mlink_add(struct mlink *mlink, const struct stat *st) |
|
Line 775 mlink_add(struct mlink *mlink, const struct stat *st) |
|
|
|
assert(NULL != mlink->file); |
assert(NULL != mlink->file); |
|
|
if (NULL == mlink->dsec) |
mlink->dsec = mandoc_strdup(mlink->dsec ? mlink->dsec : ""); |
mlink->dsec = mandoc_strdup(""); |
mlink->arch = mandoc_strdup(mlink->arch ? mlink->arch : ""); |
if (NULL == mlink->arch) |
mlink->name = mandoc_strdup(mlink->name ? mlink->name : ""); |
mlink->arch = mandoc_strdup(""); |
mlink->fsec = mandoc_strdup(mlink->fsec ? mlink->fsec : ""); |
if (NULL == mlink->name) |
|
mlink->name = mandoc_strdup(""); |
|
if (NULL == mlink->fsec) |
|
mlink->fsec = mandoc_strdup(""); |
|
|
|
if ('0' == *mlink->fsec) { |
if ('0' == *mlink->fsec) { |
free(mlink->fsec); |
free(mlink->fsec); |
|
|
} |
} |
} |
} |
|
|
|
static int |
|
mlink_check(struct mpage *mpage, struct mlink *mlink) |
|
{ |
|
int match; |
|
|
|
match = 1; |
|
|
|
/* |
|
* Check whether the manual section given in a file |
|
* agrees with the directory where the file is located. |
|
* Some manuals have suffixes like (3p) on their |
|
* section number either inside the file or in the |
|
* directory name, some are linked into more than one |
|
* section, like encrypt(1) = makekey(8). |
|
*/ |
|
|
|
if (FORM_SRC == mpage->form && |
|
strcasecmp(mpage->sec, mlink->dsec)) { |
|
match = 0; |
|
say(mlink->file, "Section \"%s\" manual in %s directory", |
|
mpage->sec, mlink->dsec); |
|
} |
|
|
|
/* |
|
* Manual page directories exist for each kernel |
|
* architecture as returned by machine(1). |
|
* However, many manuals only depend on the |
|
* application architecture as returned by arch(1). |
|
* For example, some (2/ARM) manuals are shared |
|
* across the "armish" and "zaurus" kernel |
|
* architectures. |
|
* A few manuals are even shared across completely |
|
* different architectures, for example fdformat(1) |
|
* on amd64, i386, sparc, and sparc64. |
|
*/ |
|
|
|
if (strcasecmp(mpage->arch, mlink->arch)) { |
|
match = 0; |
|
say(mlink->file, "Architecture \"%s\" manual in " |
|
"\"%s\" directory", mpage->arch, mlink->arch); |
|
} |
|
|
|
if (strcasecmp(mpage->title, mlink->name)) |
|
match = 0; |
|
|
|
return(match); |
|
} |
|
|
/* |
/* |
* Run through the files in the global vector "mpages" |
* Run through the files in the global vector "mpages" |
* and add them to the database specified in "basedir". |
* and add them to the database specified in "basedir". |
|
|
* and filename to determine whether the file is parsable or not. |
* and filename to determine whether the file is parsable or not. |
*/ |
*/ |
static void |
static void |
mpages_merge(struct mchars *mc, struct mparse *mp, int check_reachable) |
mpages_merge(struct mchars *mc, struct mparse *mp) |
{ |
{ |
struct ohash title_table; |
struct ohash_info str_info; |
struct ohash_info title_info, str_info; |
|
struct mpage *mpage; |
struct mpage *mpage; |
|
struct mlink *mlink; |
struct mdoc *mdoc; |
struct mdoc *mdoc; |
struct man *man; |
struct man *man; |
struct title *title_entry; |
|
char *title_str; |
|
const char *cp; |
const char *cp; |
int match; |
int match; |
unsigned int pslot, tslot; |
unsigned int pslot; |
enum mandoclevel lvl; |
enum mandoclevel lvl; |
|
|
str_info.alloc = hash_alloc; |
str_info.alloc = hash_alloc; |
Line 914 mpages_merge(struct mchars *mc, struct mparse *mp, int |
|
Line 961 mpages_merge(struct mchars *mc, struct mparse *mp, int |
|
str_info.hfree = hash_free; |
str_info.hfree = hash_free; |
str_info.key_offset = offsetof(struct str, key); |
str_info.key_offset = offsetof(struct str, key); |
|
|
if (check_reachable) { |
|
title_info.alloc = hash_alloc; |
|
title_info.halloc = hash_halloc; |
|
title_info.hfree = hash_free; |
|
title_info.key_offset = offsetof(struct title, title); |
|
ohash_init(&title_table, 6, &title_info); |
|
} |
|
|
|
mpage = ohash_first(&mpages, &pslot); |
mpage = ohash_first(&mpages, &pslot); |
while (NULL != mpage) { |
while (NULL != mpage) { |
mlinks_undupe(mpage); |
mlinks_undupe(mpage); |
Line 934 mpages_merge(struct mchars *mc, struct mparse *mp, int |
|
Line 973 mpages_merge(struct mchars *mc, struct mparse *mp, int |
|
mparse_reset(mp); |
mparse_reset(mp); |
mdoc = NULL; |
mdoc = NULL; |
man = NULL; |
man = NULL; |
match = 1; |
|
|
|
/* |
/* |
* Try interpreting the file as mdoc(7) or man(7) |
* Try interpreting the file as mdoc(7) or man(7) |
Line 974 mpages_merge(struct mchars *mc, struct mparse *mp, int |
|
Line 1012 mpages_merge(struct mchars *mc, struct mparse *mp, int |
|
mpage->title = |
mpage->title = |
mandoc_strdup(mpage->mlinks->name); |
mandoc_strdup(mpage->mlinks->name); |
} |
} |
|
putkey(mpage, mpage->sec, TYPE_sec); |
|
putkey(mpage, '\0' == *mpage->arch ? |
|
"any" : mpage->arch, TYPE_arch); |
|
|
/* |
for (mlink = mpage->mlinks; mlink; mlink = mlink->next) { |
* Check whether the manual section given in a file |
if ('\0' != *mlink->dsec) |
* agrees with the directory where the file is located. |
putkey(mpage, mlink->dsec, TYPE_sec); |
* Some manuals have suffixes like (3p) on their |
if ('\0' != *mlink->fsec) |
* section number either inside the file or in the |
putkey(mpage, mlink->fsec, TYPE_sec); |
* directory name, some are linked into more than one |
putkey(mpage, '\0' == *mlink->arch ? |
* section, like encrypt(1) = makekey(8). Do not skip |
"any" : mlink->arch, TYPE_arch); |
* manuals for such reasons. |
putkey(mpage, mlink->name, TYPE_Nm); |
*/ |
|
if (warnings && !use_all && FORM_SRC == mpage->form && |
|
strcasecmp(mpage->sec, mpage->mlinks->dsec)) { |
|
match = 0; |
|
say(mpage->mlinks->file, "Section \"%s\" " |
|
"manual in %s directory", |
|
mpage->sec, mpage->mlinks->dsec); |
|
} |
} |
|
|
/* |
if (warnings && !use_all) { |
* Manual page directories exist for each kernel |
|
* architecture as returned by machine(1). |
|
* However, many manuals only depend on the |
|
* application architecture as returned by arch(1). |
|
* For example, some (2/ARM) manuals are shared |
|
* across the "armish" and "zaurus" kernel |
|
* architectures. |
|
* A few manuals are even shared across completely |
|
* different architectures, for example fdformat(1) |
|
* on amd64, i386, sparc, and sparc64. |
|
* Thus, warn about architecture mismatches, |
|
* but don't skip manuals for this reason. |
|
*/ |
|
if (warnings && !use_all && |
|
strcasecmp(mpage->arch, mpage->mlinks->arch)) { |
|
match = 0; |
match = 0; |
say(mpage->mlinks->file, "Architecture \"%s\" " |
for (mlink = mpage->mlinks; mlink; |
"manual in \"%s\" directory", |
mlink = mlink->next) |
mpage->arch, mpage->mlinks->arch); |
if (mlink_check(mpage, mlink)) |
} |
match = 1; |
if (warnings && !use_all && |
} else |
strcasecmp(mpage->title, mpage->mlinks->name)) |
match = 1; |
match = 0; |
|
|
|
putkey(mpage, mpage->mlinks->name, TYPE_Nm); |
|
|
|
if (NULL != mdoc) { |
if (NULL != mdoc) { |
if (NULL != (cp = mdoc_meta(mdoc)->name)) |
if (NULL != (cp = mdoc_meta(mdoc)->name)) |
putkey(mpage, cp, TYPE_Nm); |
putkey(mpage, cp, TYPE_Nm); |
Line 1031 mpages_merge(struct mchars *mc, struct mparse *mp, int |
|
Line 1047 mpages_merge(struct mchars *mc, struct mparse *mp, int |
|
else |
else |
parse_cat(mpage); |
parse_cat(mpage); |
|
|
/* |
|
* Build a title string for the file. If it matches |
|
* the location of the file, remember the title as |
|
* found; else, remember it as missing. |
|
*/ |
|
|
|
if (check_reachable) { |
|
if (-1 == asprintf(&title_str, "%s(%s%s%s)", |
|
mpage->title, mpage->sec, |
|
'\0' == *mpage->arch ? "" : "/", |
|
mpage->arch)) { |
|
perror(NULL); |
|
exit((int)MANDOCLEVEL_SYSERR); |
|
} |
|
tslot = ohash_qlookup(&title_table, title_str); |
|
title_entry = ohash_find(&title_table, tslot); |
|
if (NULL == title_entry) { |
|
title_entry = mandoc_malloc( |
|
sizeof(struct title)); |
|
title_entry->title = title_str; |
|
title_entry->file = mandoc_strdup( |
|
match ? "" : mpage->mlinks->file); |
|
ohash_insert(&title_table, tslot, |
|
title_entry); |
|
} else { |
|
if (match) |
|
*title_entry->file = '\0'; |
|
free(title_str); |
|
} |
|
} |
|
|
|
dbindex(mpage, mc); |
dbindex(mpage, mc); |
ohash_delete(&strings); |
ohash_delete(&strings); |
mpage = ohash_next(&mpages, &pslot); |
mpage = ohash_next(&mpages, &pslot); |
} |
} |
|
|
if (check_reachable) { |
|
title_entry = ohash_first(&title_table, &tslot); |
|
while (NULL != title_entry) { |
|
if ('\0' != *title_entry->file) |
|
say(title_entry->file, |
|
"Probably unreachable, title is %s", |
|
title_entry->title); |
|
free(title_entry->title); |
|
free(title_entry->file); |
|
free(title_entry); |
|
title_entry = ohash_next(&title_table, &tslot); |
|
} |
|
ohash_delete(&title_table); |
|
} |
|
} |
} |
|
|
static void |
static void |
Line 1441 parse_mdoc_Fd(struct mpage *mpage, const struct mdoc_n |
|
Line 1411 parse_mdoc_Fd(struct mpage *mpage, const struct mdoc_n |
|
|
|
if (end > start) |
if (end > start) |
putkeys(mpage, start, end - start + 1, TYPE_In); |
putkeys(mpage, start, end - start + 1, TYPE_In); |
return(1); |
return(0); |
} |
} |
|
|
static int |
static int |
parse_mdoc_In(struct mpage *mpage, const struct mdoc_node *n) |
|
{ |
|
|
|
if (NULL != n->child && MDOC_TEXT == n->child->type) |
|
return(0); |
|
|
|
putkey(mpage, n->child->string, TYPE_In); |
|
return(1); |
|
} |
|
|
|
static int |
|
parse_mdoc_Fn(struct mpage *mpage, const struct mdoc_node *n) |
parse_mdoc_Fn(struct mpage *mpage, const struct mdoc_node *n) |
{ |
{ |
const char *cp; |
const char *cp; |
Line 1489 parse_mdoc_Fn(struct mpage *mpage, const struct mdoc_n |
|
Line 1448 parse_mdoc_Fn(struct mpage *mpage, const struct mdoc_n |
|
} |
} |
|
|
static int |
static int |
parse_mdoc_St(struct mpage *mpage, const struct mdoc_node *n) |
|
{ |
|
|
|
if (NULL == n->child || MDOC_TEXT != n->child->type) |
|
return(0); |
|
|
|
putkey(mpage, n->child->string, TYPE_St); |
|
return(1); |
|
} |
|
|
|
static int |
|
parse_mdoc_Xr(struct mpage *mpage, const struct mdoc_node *n) |
parse_mdoc_Xr(struct mpage *mpage, const struct mdoc_node *n) |
{ |
{ |
char *cp; |
char *cp; |
|
|
parse_mdoc_Nm(struct mpage *mpage, const struct mdoc_node *n) |
parse_mdoc_Nm(struct mpage *mpage, const struct mdoc_node *n) |
{ |
{ |
|
|
if (SEC_NAME == n->sec) |
return(SEC_NAME == n->sec || |
return(1); |
(SEC_SYNOPSIS == n->sec && MDOC_HEAD == n->type)); |
else if (SEC_SYNOPSIS != n->sec || MDOC_HEAD != n->type) |
|
return(0); |
|
|
|
return(1); |
|
} |
} |
|
|
static int |
static int |
Line 1670 utf8(unsigned int cp, char out[7]) |
|
Line 1614 utf8(unsigned int cp, char out[7]) |
|
} |
} |
|
|
/* |
/* |
* Store the UTF-8 version of a key, or alias the pointer if the key has |
* Store the rendered version of a key, or alias the pointer |
* no UTF-8 transcription marks in it. |
* if the key contains no escape sequences. |
*/ |
*/ |
static void |
static void |
utf8key(struct mchars *mc, struct str *key) |
render_key(struct mchars *mc, struct str *key) |
{ |
{ |
size_t sz, bsz, pos; |
size_t sz, bsz, pos; |
char utfbuf[7], res[5]; |
char utfbuf[7], res[5]; |
Line 1683 utf8key(struct mchars *mc, struct str *key) |
|
Line 1627 utf8key(struct mchars *mc, struct str *key) |
|
int len, u; |
int len, u; |
enum mandoc_esc esc; |
enum mandoc_esc esc; |
|
|
assert(NULL == key->utf8); |
assert(NULL == key->rendered); |
|
|
res[0] = '\\'; |
res[0] = '\\'; |
res[1] = '\t'; |
res[1] = '\t'; |
Line 1699 utf8key(struct mchars *mc, struct str *key) |
|
Line 1643 utf8key(struct mchars *mc, struct str *key) |
|
* pointer as ourselvse and get out of here. |
* pointer as ourselvse and get out of here. |
*/ |
*/ |
if (strcspn(val, res) == bsz) { |
if (strcspn(val, res) == bsz) { |
key->utf8 = key->key; |
key->rendered = key->key; |
return; |
return; |
} |
} |
|
|
Line 1734 utf8key(struct mchars *mc, struct str *key) |
|
Line 1678 utf8key(struct mchars *mc, struct str *key) |
|
/* Read past the slash. */ |
/* Read past the slash. */ |
|
|
val++; |
val++; |
u = 0; |
|
|
|
/* |
/* |
* Parse the escape sequence and see if it's a |
* Parse the escape sequence and see if it's a |
* predefined character or special character. |
* predefined character or special character. |
*/ |
*/ |
|
|
esc = mandoc_escape |
esc = mandoc_escape |
((const char **)&val, &seq, &len); |
((const char **)&val, &seq, &len); |
if (ESCAPE_ERROR == esc) |
if (ESCAPE_ERROR == esc) |
break; |
break; |
|
|
if (ESCAPE_SPECIAL != esc) |
if (ESCAPE_SPECIAL != esc) |
continue; |
continue; |
if (0 == (u = mchars_spec2cp(mc, seq, len))) |
|
continue; |
|
|
|
/* |
/* |
* If we have a Unicode codepoint, try to convert that |
* Render the special character |
* to a UTF-8 byte string. |
* as either UTF-8 or ASCII. |
*/ |
*/ |
cpp = utfbuf; |
|
if (0 == (sz = utf8(u, utfbuf))) |
|
continue; |
|
|
|
|
if (write_utf8) { |
|
if (0 == (u = mchars_spec2cp(mc, seq, len))) |
|
continue; |
|
cpp = utfbuf; |
|
if (0 == (sz = utf8(u, utfbuf))) |
|
continue; |
|
sz = strlen(cpp); |
|
} else { |
|
cpp = mchars_spec2str(mc, seq, len, &sz); |
|
if (NULL == cpp) |
|
continue; |
|
if (ASCII_NBRSP == *cpp) { |
|
cpp = " "; |
|
sz = 1; |
|
} |
|
} |
|
|
/* Copy the rendered glyph into the stream. */ |
/* Copy the rendered glyph into the stream. */ |
|
|
sz = strlen(cpp); |
|
bsz += sz; |
bsz += sz; |
|
|
buf = mandoc_realloc(buf, bsz); |
buf = mandoc_realloc(buf, bsz); |
|
|
memcpy(&buf[pos], cpp, sz); |
memcpy(&buf[pos], cpp, sz); |
pos += sz; |
pos += sz; |
} |
} |
|
|
buf[pos] = '\0'; |
buf[pos] = '\0'; |
key->utf8 = buf; |
key->rendered = buf; |
} |
} |
|
|
/* |
/* |
* Flush the current page's terms (and their bits) into the database. |
* Flush the current page's terms (and their bits) into the database. |
* Wrap the entire set of additions in a transaction to make sqlite be a |
* Wrap the entire set of additions in a transaction to make sqlite be a |
* little faster. |
* little faster. |
* Also, UTF-8-encode the description at the last possible moment. |
* Also, handle escape sequences at the last possible moment. |
*/ |
*/ |
static void |
static void |
dbindex(const struct mpage *mpage, struct mchars *mc) |
dbindex(const struct mpage *mpage, struct mchars *mc) |
{ |
{ |
struct mlink *mlink; |
struct mlink *mlink; |
struct str *key; |
struct str *key; |
const char *desc; |
|
int64_t recno; |
int64_t recno; |
size_t i; |
size_t i; |
unsigned int slot; |
unsigned int slot; |
Line 1795 dbindex(const struct mpage *mpage, struct mchars *mc) |
|
Line 1746 dbindex(const struct mpage *mpage, struct mchars *mc) |
|
if (nodb) |
if (nodb) |
return; |
return; |
|
|
desc = ""; |
|
if (NULL != mpage->desc && '\0' != *mpage->desc) { |
|
key = ohash_find(&strings, |
|
ohash_qlookup(&strings, mpage->desc)); |
|
assert(NULL != key); |
|
if (NULL == key->utf8) |
|
utf8key(mc, key); |
|
desc = key->utf8; |
|
} |
|
|
|
SQL_EXEC("BEGIN TRANSACTION"); |
SQL_EXEC("BEGIN TRANSACTION"); |
|
|
i = 1; |
i = 1; |
/* |
|
* XXX The following three lines are obsolete |
|
* and only kept for backward compatibility |
|
* until apropos(1) and friends have caught up. |
|
*/ |
|
SQL_BIND_TEXT(stmts[STMT_INSERT_PAGE], i, mpage->mlinks->file); |
|
SQL_BIND_TEXT(stmts[STMT_INSERT_PAGE], i, mpage->mlinks->dsec); |
|
SQL_BIND_TEXT(stmts[STMT_INSERT_PAGE], i, mpage->mlinks->arch); |
|
SQL_BIND_TEXT(stmts[STMT_INSERT_PAGE], i, desc); |
|
SQL_BIND_INT(stmts[STMT_INSERT_PAGE], i, FORM_SRC == mpage->form); |
SQL_BIND_INT(stmts[STMT_INSERT_PAGE], i, FORM_SRC == mpage->form); |
SQL_STEP(stmts[STMT_INSERT_PAGE]); |
SQL_STEP(stmts[STMT_INSERT_PAGE]); |
recno = sqlite3_last_insert_rowid(db); |
recno = sqlite3_last_insert_rowid(db); |
Line 1836 dbindex(const struct mpage *mpage, struct mchars *mc) |
|
Line 1768 dbindex(const struct mpage *mpage, struct mchars *mc) |
|
for (key = ohash_first(&strings, &slot); NULL != key; |
for (key = ohash_first(&strings, &slot); NULL != key; |
key = ohash_next(&strings, &slot)) { |
key = ohash_next(&strings, &slot)) { |
assert(key->mpage == mpage); |
assert(key->mpage == mpage); |
if (NULL == key->utf8) |
if (NULL == key->rendered) |
utf8key(mc, key); |
render_key(mc, key); |
i = 1; |
i = 1; |
SQL_BIND_INT64(stmts[STMT_INSERT_KEY], i, key->mask); |
SQL_BIND_INT64(stmts[STMT_INSERT_KEY], i, key->mask); |
SQL_BIND_TEXT(stmts[STMT_INSERT_KEY], i, key->utf8); |
SQL_BIND_TEXT(stmts[STMT_INSERT_KEY], i, key->rendered); |
SQL_BIND_INT64(stmts[STMT_INSERT_KEY], i, recno); |
SQL_BIND_INT64(stmts[STMT_INSERT_KEY], i, recno); |
SQL_STEP(stmts[STMT_INSERT_KEY]); |
SQL_STEP(stmts[STMT_INSERT_KEY]); |
sqlite3_reset(stmts[STMT_INSERT_KEY]); |
sqlite3_reset(stmts[STMT_INSERT_KEY]); |
if (key->utf8 != key->key) |
if (key->rendered != key->key) |
free(key->utf8); |
free(key->rendered); |
free(key); |
free(key); |
} |
} |
|
|
Line 1952 dbopen(int real) |
|
Line 1884 dbopen(int real) |
|
return(0); |
return(0); |
} |
} |
|
|
/* |
|
* XXX The first three columns in table mpages are obsolete |
|
* and only kept for backward compatibility |
|
* until apropos(1) and friends have caught up. |
|
*/ |
|
sql = "CREATE TABLE \"mpages\" (\n" |
sql = "CREATE TABLE \"mpages\" (\n" |
" \"file\" TEXT NOT NULL,\n" |
|
" \"sec\" TEXT NOT NULL,\n" |
|
" \"arch\" TEXT NOT NULL,\n" |
|
" \"desc\" TEXT NOT NULL,\n" |
|
" \"form\" INTEGER NOT NULL,\n" |
" \"form\" INTEGER NOT NULL,\n" |
" \"id\" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL\n" |
" \"id\" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL\n" |
");\n" |
");\n" |
Line 1997 prepare_statements: |
|
Line 1920 prepare_statements: |
|
sql = "DELETE FROM mpages where file=?"; |
sql = "DELETE FROM mpages where file=?"; |
sqlite3_prepare_v2(db, sql, -1, &stmts[STMT_DELETE_PAGE], NULL); |
sqlite3_prepare_v2(db, sql, -1, &stmts[STMT_DELETE_PAGE], NULL); |
sql = "INSERT INTO mpages " |
sql = "INSERT INTO mpages " |
"(file,sec,arch,desc,form) VALUES (?,?,?,?,?)"; |
"(form) VALUES (?)"; |
sqlite3_prepare_v2(db, sql, -1, &stmts[STMT_INSERT_PAGE], NULL); |
sqlite3_prepare_v2(db, sql, -1, &stmts[STMT_INSERT_PAGE], NULL); |
sql = "INSERT INTO mlinks " |
sql = "INSERT INTO mlinks " |
"(file,sec,arch,name,pageid) VALUES (?,?,?,?,?)"; |
"(file,sec,arch,name,pageid) VALUES (?,?,?,?,?)"; |