-/* $Id: mdoc_validate.c,v 1.342 2017/06/24 18:58:33 schwarze Exp $ */
+/* $Id: mdoc_validate.c,v 1.358 2018/04/11 17:11:13 schwarze Exp $ */
/*
* Copyright (c) 2008-2012 Kristaps Dzonsons <kristaps@bsd.lv>
- * Copyright (c) 2010-2017 Ingo Schwarze <schwarze@openbsd.org>
+ * Copyright (c) 2010-2018 Ingo Schwarze <schwarze@openbsd.org>
* Copyright (c) 2010 Joerg Sonnenberger <joerg@netbsd.org>
*
* Permission to use, copy, modify, and distribute this software for any
#include "mandoc_aux.h"
#include "mandoc.h"
+#include "mandoc_xr.h"
#include "roff.h"
#include "mdoc.h"
#include "libmandoc.h"
typedef void (*v_post)(POST_ARGS);
static int build_list(struct roff_man *, int);
-static void check_text(struct roff_man *, int, int, char *);
static void check_argv(struct roff_man *,
struct roff_node *, struct mdoc_argv *);
static void check_args(struct roff_man *, struct roff_node *);
+static void check_text(struct roff_man *, int, int, char *);
+static void check_text_em(struct roff_man *, int, int, char *);
static void check_toptext(struct roff_man *, int, int, const char *);
static int child_an(const struct roff_node *);
static size_t macro2len(enum roff_tok);
static void rewrite_macro2len(struct roff_man *, char **);
+static int similar(const char *, const char *);
static void post_an(POST_ARGS);
static void post_an_norm(POST_ARGS);
static void post_display(POST_ARGS);
static void post_dd(POST_ARGS);
static void post_delim(POST_ARGS);
+static void post_delim_nb(POST_ARGS);
static void post_dt(POST_ARGS);
static void post_en(POST_ARGS);
static void post_es(POST_ARGS);
static void post_sm(POST_ARGS);
static void post_st(POST_ARGS);
static void post_std(POST_ARGS);
+static void post_sx(POST_ARGS);
static void post_useless(POST_ARGS);
static void post_xr(POST_ARGS);
static void post_xx(POST_ARGS);
post_bl, /* Bl */
NULL, /* El */
post_it, /* It */
- post_delim, /* Ad */
+ post_delim_nb, /* Ad */
post_an, /* An */
NULL, /* Ap */
post_defaults, /* Ar */
NULL, /* Cd */
- post_delim, /* Cm */
- post_delim, /* Dv */
- post_delim, /* Er */
- post_delim, /* Ev */
+ post_delim_nb, /* Cm */
+ post_delim_nb, /* Dv */
+ post_delim_nb, /* Er */
+ post_delim_nb, /* Ev */
post_ex, /* Ex */
post_fa, /* Fa */
NULL, /* Fd */
- post_delim, /* Fl */
+ post_delim_nb, /* Fl */
post_fn, /* Fn */
- post_delim, /* Ft */
- post_delim, /* Ic */
- post_delim, /* In */
+ post_delim_nb, /* Ft */
+ post_delim_nb, /* Ic */
+ post_delim_nb, /* In */
post_defaults, /* Li */
post_nd, /* Nd */
post_nm, /* Nm */
- post_delim, /* Op */
+ post_delim_nb, /* Op */
post_obsolete, /* Ot */
post_defaults, /* Pa */
post_rv, /* Rv */
post_st, /* St */
- post_delim, /* Va */
- post_delim, /* Vt */
+ post_delim_nb, /* Va */
+ post_delim_nb, /* Vt */
post_xr, /* Xr */
NULL, /* %A */
post_hyph, /* %B */ /* FIXME: can be used outside Rs/Re. */
post_hyph, /* %T */ /* FIXME: can be used outside Rs/Re. */
NULL, /* %V */
NULL, /* Ac */
- post_delim, /* Ao */
- post_delim, /* Aq */
+ NULL, /* Ao */
+ post_delim_nb, /* Aq */
post_at, /* At */
NULL, /* Bc */
post_bf, /* Bf */
- post_delim, /* Bo */
+ NULL, /* Bo */
NULL, /* Bq */
post_xx, /* Bsx */
post_bx, /* Bx */
NULL, /* Dq */
NULL, /* Ec */
NULL, /* Ef */
- post_delim, /* Em */
+ post_delim_nb, /* Em */
NULL, /* Eo */
post_xx, /* Fx */
- post_delim, /* Ms */
+ post_delim_nb, /* Ms */
NULL, /* No */
post_ns, /* Ns */
post_xx, /* Nx */
post_xx, /* Ox */
NULL, /* Pc */
NULL, /* Pf */
- post_delim, /* Po */
- post_delim, /* Pq */
+ NULL, /* Po */
+ post_delim_nb, /* Pq */
NULL, /* Qc */
- post_delim, /* Ql */
- post_delim, /* Qo */
- post_delim, /* Qq */
+ post_delim_nb, /* Ql */
+ NULL, /* Qo */
+ post_delim_nb, /* Qq */
NULL, /* Re */
post_rs, /* Rs */
NULL, /* Sc */
- post_delim, /* So */
- post_delim, /* Sq */
+ NULL, /* So */
+ post_delim_nb, /* Sq */
post_sm, /* Sm */
- post_hyph, /* Sx */
- post_delim, /* Sy */
+ post_sx, /* Sx */
+ post_delim_nb, /* Sy */
post_useless, /* Tn */
post_xx, /* Ux */
NULL, /* Xc */
NULL, /* Xo */
post_fo, /* Fo */
NULL, /* Fc */
- post_delim, /* Oo */
+ NULL, /* Oo */
NULL, /* Oc */
post_bk, /* Bk */
NULL, /* Ek */
post_eoln, /* Ud */
post_lb, /* Lb */
post_par, /* Lp */
- post_delim, /* Lk */
+ post_delim_nb, /* Lk */
post_defaults, /* Mt */
- post_delim, /* Brq */
- post_delim, /* Bro */
+ post_delim_nb, /* Brq */
+ NULL, /* Bro */
NULL, /* Brc */
NULL, /* %C */
post_es, /* Es */
void
mdoc_node_validate(struct roff_man *mdoc)
{
- struct roff_node *n;
+ struct roff_node *n, *np;
const v_post *p;
n = mdoc->last;
mdoc->next = ROFF_NEXT_SIBLING;
switch (n->type) {
case ROFFT_TEXT:
+ np = n->parent;
if (n->sec != SEC_SYNOPSIS ||
- (n->parent->tok != MDOC_Cd && n->parent->tok != MDOC_Fd))
+ (np->tok != MDOC_Cd && np->tok != MDOC_Fd))
check_text(mdoc, n->line, n->pos, n->string);
- if (n->parent->tok == MDOC_It ||
- (n->parent->type == ROFFT_BODY &&
- (n->parent->tok == MDOC_Sh ||
- n->parent->tok == MDOC_Ss)))
+ if (np->tok != MDOC_Ql && np->tok != MDOC_Dl &&
+ (np->tok != MDOC_Bd ||
+ (mdoc->flags & MDOC_LITERAL) == 0) &&
+ (np->tok != MDOC_It || np->type != ROFFT_HEAD ||
+ np->parent->parent->norm->Bl.type != LIST_diag))
+ check_text_em(mdoc, n->line, n->pos, n->string);
+ if (np->tok == MDOC_It || (np->type == ROFFT_BODY &&
+ (np->tok == MDOC_Sh || np->tok == MDOC_Ss)))
check_toptext(mdoc, n->line, n->pos, n->string);
break;
+ case ROFFT_COMMENT:
case ROFFT_EQN:
case ROFFT_TBL:
break;
ln, pos + (int)(p - cp), NULL);
}
+static void
+check_text_em(struct roff_man *mdoc, int ln, int pos, char *p)
+{
+ const struct roff_node *np, *nn;
+ char *cp;
+
+ np = mdoc->last->prev;
+ nn = mdoc->last->next;
+
+ /* Look for em-dashes wrongly encoded as "--". */
+
+ for (cp = p; *cp != '\0'; cp++) {
+ if (cp[0] != '-' || cp[1] != '-')
+ continue;
+ cp++;
+
+ /* Skip input sequences of more than two '-'. */
+
+ if (cp[1] == '-') {
+ while (cp[1] == '-')
+ cp++;
+ continue;
+ }
+
+ /* Skip "--" directly attached to something else. */
+
+ if ((cp - p > 1 && cp[-2] != ' ') ||
+ (cp[1] != '\0' && cp[1] != ' '))
+ continue;
+
+ /* Require a letter right before or right afterwards. */
+
+ if ((cp - p > 2 ?
+ isalpha((unsigned char)cp[-3]) :
+ np != NULL &&
+ np->type == ROFFT_TEXT &&
+ np->string != '\0' &&
+ isalpha((unsigned char)np->string[
+ strlen(np->string) - 1])) ||
+ (cp[2] != '\0' ?
+ isalpha((unsigned char)cp[2]) :
+ nn != NULL &&
+ nn->type == ROFFT_TEXT &&
+ nn->string != '\0' &&
+ isalpha((unsigned char)*nn->string))) {
+ mandoc_msg(MANDOCERR_DASHDASH, mdoc->parse,
+ ln, pos + (int)(cp - p) - 1, NULL);
+ break;
+ }
+ }
+}
+
static void
check_toptext(struct roff_man *mdoc, int ln, int pos, const char *p)
{
static void
post_delim(POST_ARGS)
+{
+ const struct roff_node *nch;
+ const char *lc;
+ enum mdelim delim;
+ enum roff_tok tok;
+
+ tok = mdoc->last->tok;
+ nch = mdoc->last->last;
+ if (nch == NULL || nch->type != ROFFT_TEXT)
+ return;
+ lc = strchr(nch->string, '\0') - 1;
+ if (lc < nch->string)
+ return;
+ delim = mdoc_isdelim(lc);
+ if (delim == DELIM_NONE || delim == DELIM_OPEN)
+ return;
+ if (*lc == ')' && (tok == MDOC_Nd || tok == MDOC_Sh ||
+ tok == MDOC_Ss || tok == MDOC_Fo))
+ return;
+
+ mandoc_vmsg(MANDOCERR_DELIM, mdoc->parse,
+ nch->line, nch->pos + (lc - nch->string),
+ "%s%s %s", roff_name[tok],
+ nch == mdoc->last->child ? "" : " ...", nch->string);
+}
+
+static void
+post_delim_nb(POST_ARGS)
{
const struct roff_node *nch;
const char *lc, *cp;
/* At least three alphabetic words with a sentence ending. */
if (strchr("!.:?", *lc) != NULL && (tok == MDOC_Em ||
- tok == MDOC_Li || tok == MDOC_Po || tok == MDOC_Pq ||
- tok == MDOC_Sy)) {
+ tok == MDOC_Li || tok == MDOC_Pq || tok == MDOC_Sy)) {
nw = 0;
for (cp = lc - 1; cp >= nch->string; cp--) {
if (*cp == ' ') {
}
}
- mandoc_vmsg(MANDOCERR_DELIM, mdoc->parse,
+ mandoc_vmsg(MANDOCERR_DELIM_NB, mdoc->parse,
nch->line, nch->pos + (lc - nch->string),
"%s%s %s", roff_name[tok],
nch == mdoc->last->child ? "" : " ...", nch->string);
switch (n->norm->Bl.type) {
case LIST_tag:
- if (NULL == n->norm->Bl.width)
+ if (n->norm->Bl.width == NULL)
mandoc_msg(MANDOCERR_BL_NOWIDTH, mdoc->parse,
n->line, n->pos, "Bl -tag");
break;
case LIST_ohang:
case LIST_inset:
case LIST_item:
- if (n->norm->Bl.width)
+ if (n->norm->Bl.width != NULL)
mandoc_vmsg(MANDOCERR_BL_SKIPW, mdoc->parse,
wa->line, wa->pos, "Bl -%s",
mdoc_argnames[mdoclt]);
+ n->norm->Bl.width = NULL;
break;
case LIST_bullet:
case LIST_dash:
case LIST_hyphen:
- if (NULL == n->norm->Bl.width)
+ if (n->norm->Bl.width == NULL)
n->norm->Bl.width = "2n";
break;
case LIST_enum:
- if (NULL == n->norm->Bl.width)
+ if (n->norm->Bl.width == NULL)
n->norm->Bl.width = "3n";
break;
default:
struct roff_node *n;
const char *p;
- post_delim(mdoc);
+ post_delim_nb(mdoc);
n = mdoc->last;
assert(n->child->type == ROFFT_TEXT);
roff_word_alloc(mdoc, n->line, n->pos, "library");
mdoc->last->flags = NODE_NOSRC;
- roff_word_alloc(mdoc, n->line, n->pos, "\\(Lq");
+ roff_word_alloc(mdoc, n->line, n->pos, "\\(lq");
mdoc->last->flags = NODE_DELIMO | NODE_NOSRC;
mdoc->last = mdoc->last->next;
- roff_word_alloc(mdoc, n->line, n->pos, "\\(Rq");
+ roff_word_alloc(mdoc, n->line, n->pos, "\\(rq");
mdoc->last->flags = NODE_DELIMC | NODE_NOSRC;
mdoc->last = n;
}
{
struct roff_node *n;
+ post_delim(mdoc);
+
n = mdoc->last;
if (n->args && n->args->argc == 1)
if (n->args->argv[0].arg == MDOC_Std)
"Fo ... %s", n->child->next->string);
while (n->child != n->last)
roff_node_delete(mdoc, n->last);
- }
+ } else
+ post_delim(mdoc);
post_fname(mdoc);
}
break;
}
}
- post_delim(mdoc);
+ post_delim_nb(mdoc);
}
static void
n = mdoc->last;
+ if (n->sec == SEC_NAME && n->child != NULL &&
+ n->child->type == ROFFT_TEXT && mdoc->meta.msec != NULL)
+ mandoc_xr_add(mdoc->meta.msec, n->child->string, -1, -1);
+
if (n->last != NULL &&
(n->last->tok == MDOC_Pp ||
n->last->tok == MDOC_Lp))
mandoc_msg(MANDOCERR_NM_NONAME, mdoc->parse,
n->line, n->pos, "Nm");
- if (n->type == ROFFT_ELEM)
+ switch (n->type) {
+ case ROFFT_ELEM:
+ post_delim_nb(mdoc);
+ break;
+ case ROFFT_HEAD:
post_delim(mdoc);
+ break;
+ default:
+ return;
+ }
- if ((n->type != ROFFT_ELEM && n->type != ROFFT_HEAD) ||
- (n->child != NULL && n->child->type == ROFFT_TEXT) ||
+ if ((n->child != NULL && n->child->type == ROFFT_TEXT) ||
mdoc->meta.name == NULL)
return;
post_nd(POST_ARGS)
{
struct roff_node *n;
- size_t sz;
n = mdoc->last;
if (n->child == NULL)
mandoc_msg(MANDOCERR_ND_EMPTY, mdoc->parse,
n->line, n->pos, "Nd");
- else if (n->last->type == ROFFT_TEXT &&
- (sz = strlen(n->last->string)) != 0 &&
- n->last->string[sz - 1] == '.')
- mandoc_msg(MANDOCERR_ND_DOT, mdoc->parse,
- n->last->line, n->last->pos + sz - 1, NULL);
+ else
+ post_delim(mdoc);
post_hyph(mdoc);
}
struct roff_node *nn;
if (mdoc->last->child != NULL) {
- post_delim(mdoc);
+ post_delim_nb(mdoc);
return;
}
mandoc_msg(MANDOCERR_MACRO_EMPTY, mdoc->parse,
np->line, np->pos, "An");
else
- post_delim(mdoc);
+ post_delim_nb(mdoc);
} else if (nch != NULL)
mandoc_vmsg(MANDOCERR_ARG_EXCESS, mdoc->parse,
nch->line, nch->pos, "An ... %s", nch->string);
{
struct roff_node *n;
const char *os;
+ char *v;
- post_delim(mdoc);
+ post_delim_nb(mdoc);
n = mdoc->last;
switch (n->tok) {
break;
case MDOC_Nx:
os = "NetBSD";
+ if (n->child == NULL)
+ break;
+ v = n->child->string;
+ if ((v[0] != '0' && v[0] != '1') || v[1] != '.' ||
+ v[2] < '0' || v[2] > '9' ||
+ v[3] < 'a' || v[3] > 'z' || v[4] != '\0')
+ break;
+ n->child->flags |= NODE_NOPRT;
+ mdoc->next = ROFF_NEXT_CHILD;
+ roff_word_alloc(mdoc, n->child->line, n->child->pos, v);
+ v = mdoc->last->string;
+ v[3] = toupper((unsigned char)v[3]);
+ mdoc->last->flags |= NODE_NOSRC;
+ mdoc->last = n;
break;
case MDOC_Ox:
os = "OpenBSD";
assert(nit->head->child == NULL);
- i = 0;
- for (nch = nit->child; nch != NULL; nch = nch->next)
- if (nch->type == ROFFT_BODY)
- i++;
+ if (nit->head->next->child == NULL &&
+ nit->head->next->next == NULL) {
+ mandoc_msg(MANDOCERR_MACRO_EMPTY, mdoc->parse,
+ nit->line, nit->pos, "It");
+ roff_node_delete(mdoc, nit);
+ break;
+ }
+ i = 0;
+ for (nch = nit->child; nch != NULL; nch = nch->next) {
+ if (nch->type != ROFFT_BODY)
+ continue;
+ if (i++ && nch->flags & NODE_LINE)
+ mandoc_msg(MANDOCERR_TA_LINE, mdoc->parse,
+ nch->line, nch->pos, "Ta");
+ }
if (i < cols || i > cols + 1)
mandoc_vmsg(MANDOCERR_BL_COL,
mdoc->parse, nit->line, nit->pos,
"%d columns, %d cells", cols, i);
+ else if (nit->head->next->child != NULL &&
+ nit->head->next->child->line > nit->line)
+ mandoc_msg(MANDOCERR_IT_NOARG, mdoc->parse,
+ nit->line, nit->pos, "Bl -column It");
break;
default:
abort();
arch++;
if (*arch == NULL) {
n = mdoc->first->child;
- while (n->tok != MDOC_Dt)
+ while (n->tok != MDOC_Dt ||
+ n->child == NULL ||
+ n->child->next == NULL ||
+ n->child->next->next == NULL)
n = n->next;
n = n->child->next->next;
mandoc_vmsg(MANDOCERR_ARCH_BAD,
/* Check that we begin with a proper `Sh'. */
n = mdoc->first->child;
- while (n != NULL && n->tok != TOKEN_NONE &&
- mdoc_macros[n->tok].flags & MDOC_PROLOGUE)
+ while (n != NULL &&
+ (n->type == ROFFT_COMMENT ||
+ (n->tok >= MDOC_Dd &&
+ mdoc_macros[n->tok].flags & MDOC_PROLOGUE)))
n = n->next;
if (n == NULL)
static void
post_ns(POST_ARGS)
{
+ struct roff_node *n;
- if (mdoc->last->flags & NODE_LINE)
+ n = mdoc->last;
+ if (n->flags & NODE_LINE ||
+ (n->next != NULL && n->next->flags & NODE_DELIMC))
mandoc_msg(MANDOCERR_NS_SKIP, mdoc->parse,
- mdoc->last->line, mdoc->last->pos, NULL);
+ n->line, n->pos, NULL);
+}
+
+static void
+post_sx(POST_ARGS)
+{
+ post_delim(mdoc);
+ post_hyph(mdoc);
}
static void
mdoc->last->line, mdoc->last->pos, NULL);
}
+/*
+ * Return an upper bound for the string distance (allowing
+ * transpositions). Not a full Levenshtein implementation
+ * because Levenshtein is quadratic in the string length
+ * and this function is called for every standard name,
+ * so the check for each custom name would be cubic.
+ * The following crude heuristics is linear, resulting
+ * in quadratic behaviour for checking one custom name,
+ * which does not cause measurable slowdown.
+ */
+static int
+similar(const char *s1, const char *s2)
+{
+ const int maxdist = 3;
+ int dist = 0;
+
+ while (s1[0] != '\0' && s2[0] != '\0') {
+ if (s1[0] == s2[0]) {
+ s1++;
+ s2++;
+ continue;
+ }
+ if (++dist > maxdist)
+ return INT_MAX;
+ if (s1[1] == s2[1]) { /* replacement */
+ s1++;
+ s2++;
+ } else if (s1[0] == s2[1] && s1[1] == s2[0]) {
+ s1 += 2; /* transposition */
+ s2 += 2;
+ } else if (s1[0] == s2[1]) /* insertion */
+ s2++;
+ else if (s1[1] == s2[0]) /* deletion */
+ s1++;
+ else
+ return INT_MAX;
+ }
+ dist += strlen(s1) + strlen(s2);
+ return dist > maxdist ? INT_MAX : dist;
+}
+
static void
post_sh_head(POST_ARGS)
{
struct roff_node *nch;
const char *goodsec;
+ const char *const *testsec;
+ int dist, mindist;
enum roff_sec sec;
/*
/* We don't care about custom sections after this. */
- if (sec == SEC_CUSTOM)
+ if (sec == SEC_CUSTOM) {
+ if ((nch = mdoc->last->child) == NULL ||
+ nch->type != ROFFT_TEXT || nch->next != NULL)
+ return;
+ goodsec = NULL;
+ mindist = INT_MAX;
+ for (testsec = secnames + 1; *testsec != NULL; testsec++) {
+ dist = similar(nch->string, *testsec);
+ if (dist < mindist) {
+ goodsec = *testsec;
+ mindist = dist;
+ }
+ }
+ if (goodsec != NULL)
+ mandoc_vmsg(MANDOCERR_SEC_TYPO, mdoc->parse,
+ nch->line, nch->pos, "Sh %s instead of %s",
+ nch->string, goodsec);
return;
+ }
/*
* Check whether our non-custom section is being repeated or is
if (nch->next == NULL) {
mandoc_vmsg(MANDOCERR_XR_NOSEC, mdoc->parse,
n->line, n->pos, "Xr %s", nch->string);
- } else
+ } else {
assert(nch->next == n->last);
- post_delim(mdoc);
+ if(mandoc_xr_add(nch->next->string, nch->string,
+ nch->line, nch->pos))
+ mandoc_vmsg(MANDOCERR_XR_SELF, mdoc->parse,
+ nch->line, nch->pos, "Xr %s %s",
+ nch->string, nch->next->string);
+ }
+ post_delim_nb(mdoc);
}
static void
post_prevpar(mdoc);
return;
case ROFFT_HEAD:
+ post_delim(mdoc);
post_hyph(mdoc);
return;
case ROFFT_BODY:
struct roff_node *n, *nch;
const char *macro;
- post_delim(mdoc);
+ post_delim_nb(mdoc);
n = mdoc->last;
nch = n->child;
mandoc_msg(MANDOCERR_PROLOG_LATE, mdoc->parse,
n->line, n->pos, "Os");
+ post_delim(mdoc);
+
/*
* Set the operating system by way of the `Os' macro.
* The order of precedence is:
git.ckatri.com / mandoc.git / blobdiff