add a style message about overlong text lines,

trying very hard to avoid false positives,
not at all trying to catch as many cases as possible;

feature originally suggested by tb@,
OK tb@ kn@ jmc@

diff --git a/usr.bin/mandoc/libmandoc.h b/usr.bin/mandoc/libmandoc.h
index b291631..354845f 100644 (file)
--- a/usr.bin/mandoc/libmandoc.h
+++ b/usr.bin/mandoc/libmandoc.h
@@ -1,4 +1,4 @@
-/* $OpenBSD: libmandoc.h,v 1.64 2020/04/03 11:34:19 schwarze Exp $ */
+/* $OpenBSD: libmandoc.h,v 1.65 2021/06/27 17:57:13 schwarze Exp $ */
 /*
  * Copyright (c) 2013-2015,2017,2018,2020 Ingo Schwarze <schwarze@openbsd.org>
  * Copyright (c) 2009, 2010, 2011, 2012 Kristaps Dzonsons <kristaps@bsd.lv>
@@ -73,7 +73,7 @@ void           roff_reset(struct roff *);
 void            roff_man_free(struct roff_man *);
 struct roff_man        *roff_man_alloc(struct roff *, const char *, int);
 void            roff_man_reset(struct roff_man *);
-int             roff_parseln(struct roff *, int, struct buf *, int *);
+int             roff_parseln(struct roff *, int, struct buf *, int *, size_t);
 void            roff_userret(struct roff *);
 void            roff_endparse(struct roff *);
 void            roff_setreg(struct roff *, const char *, int, char);

diff --git a/usr.bin/mandoc/mandoc.1 b/usr.bin/mandoc/mandoc.1
index acb737c..cfa47b4 100644 (file)
--- a/usr.bin/mandoc/mandoc.1
+++ b/usr.bin/mandoc/mandoc.1
@@ -1,4 +1,4 @@
-.\" $OpenBSD: mandoc.1,v 1.175 2021/06/02 18:27:36 schwarze Exp $
+.\" $OpenBSD: mandoc.1,v 1.176 2021/06/27 17:57:13 schwarze Exp $
 .\"
 .\" Copyright (c) 2012, 2014-2021 Ingo Schwarze <schwarze@openbsd.org>
 .\" Copyright (c) 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
@@ -15,7 +15,7 @@
 .\" ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
 .\" OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
 .\"
-.Dd $Mdocdate: June 2 2021 $
+.Dd $Mdocdate: June 27 2021 $
 .Dt MANDOC 1
 .Os
 .Sh NAME
@@ -1066,6 +1066,9 @@ An
 request occurs even though the document already switched to no-fill mode
 and did not switch back to fill mode yet.
 It has no effect.
+.It Sy "input text line longer than 80 bytes"
+Consider breaking the input text line
+at one of the blank characters before column 80.
 .It Sy "verbatim \(dq--\(dq, maybe consider using \e(em"
 .Pq mdoc
 Even though the ASCII output device renders an em-dash as

diff --git a/usr.bin/mandoc/mandoc.h b/usr.bin/mandoc/mandoc.h
index 40f54b8..0bd681c 100644 (file)
--- a/usr.bin/mandoc/mandoc.h
+++ b/usr.bin/mandoc/mandoc.h
@@ -1,4 +1,4 @@
-/* $OpenBSD: mandoc.h,v 1.212 2021/06/02 18:27:37 schwarze Exp $ */
+/* $OpenBSD: mandoc.h,v 1.213 2021/06/27 17:57:13 schwarze Exp $ */
 /*
  * Copyright (c) 2012-2020 Ingo Schwarze <schwarze@openbsd.org>
  * Copyright (c) 2010, 2011, 2014 Kristaps Dzonsons <kristaps@bsd.lv>
@@ -72,6 +72,7 @@ enum  mandocerr {
        MANDOCERR_DELIM_NB, /* no blank before trailing delimiter: macro ... */
        MANDOCERR_FI_SKIP, /* fill mode already enabled, skipping: fi */
        MANDOCERR_NF_SKIP, /* fill mode already disabled, skipping: nf */
+       MANDOCERR_TEXT_LONG, /* input text line longer than 80 bytes */
        MANDOCERR_DASHDASH, /* verbatim "--", maybe consider using \(em */
        MANDOCERR_FUNC, /* function name without markup: name() */
        MANDOCERR_SPACE_EOL, /* whitespace at end of input line */

diff --git a/usr.bin/mandoc/mandoc_msg.c b/usr.bin/mandoc/mandoc_msg.c
index c6f2c72..1d20cfc 100644 (file)
--- a/usr.bin/mandoc/mandoc_msg.c
+++ b/usr.bin/mandoc/mandoc_msg.c
@@ -1,4 +1,4 @@
-/* $OpenBSD: mandoc_msg.c,v 1.11 2021/06/02 18:27:37 schwarze Exp $ */
+/* $OpenBSD: mandoc_msg.c,v 1.12 2021/06/27 17:57:13 schwarze Exp $ */
 /*
  * Copyright (c) 2014-2020 Ingo Schwarze <schwarze@openbsd.org>
  * Copyright (c) 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
@@ -71,6 +71,7 @@ static        const char *const type_message[MANDOCERR_MAX] = {
        "no blank before trailing delimiter",
        "fill mode already enabled, skipping",
        "fill mode already disabled, skipping",
+       "input text line longer than 80 bytes",
        "verbatim \"--\", maybe consider using \\(em",
        "function name without markup",
        "whitespace at end of input line",

diff --git a/usr.bin/mandoc/read.c b/usr.bin/mandoc/read.c
index a6a25de..223dc82 100644 (file)
--- a/usr.bin/mandoc/read.c
+++ b/usr.bin/mandoc/read.c
@@ -1,4 +1,4 @@
-/* $OpenBSD: read.c,v 1.190 2020/04/24 11:58:02 schwarze Exp $ */
+/* $OpenBSD: read.c,v 1.191 2021/06/27 17:57:13 schwarze Exp $ */
 /*
  * Copyright (c) 2010-2020 Ingo Schwarze <schwarze@openbsd.org>
  * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
@@ -152,6 +152,7 @@ mparse_buf_r(struct mparse *curp, struct buf blk, size_t i, int start)
        struct buf      *firstln, *lastln, *thisln, *loop;
        char            *cp;
        size_t           pos; /* byte number in the ln buffer */
+       size_t           spos; /* at the start of the current line parse */
        int              line_result, result;
        int              of;
        int              lnn; /* line number in the real file */
@@ -178,6 +179,7 @@ mparse_buf_r(struct mparse *curp, struct buf blk, size_t i, int start)
                            curp->filenc & MPARSE_LATIN1)
                                curp->filenc = preconv_cue(&blk, i);
                }
+               spos = pos;
 
                while (i < blk.sz && (start || blk.buf[i] != '\0')) {
 
@@ -277,7 +279,8 @@ mparse_buf_r(struct mparse *curp, struct buf blk, size_t i, int start)
 
                of = 0;
 rerun:
-               line_result = roff_parseln(curp->roff, curp->line, &ln, &of);
+               line_result = roff_parseln(curp->roff, curp->line,
+                   &ln, &of, start && spos == 0 ? pos : 0);
 
                /* Process options. */
 

diff --git a/usr.bin/mandoc/roff.c b/usr.bin/mandoc/roff.c
index 00db7b8..8518d19 100644 (file)
--- a/usr.bin/mandoc/roff.c
+++ b/usr.bin/mandoc/roff.c
@@ -1,4 +1,4 @@
-/* $OpenBSD: roff.c,v 1.248 2020/08/27 12:58:00 schwarze Exp $ */
+/* $OpenBSD: roff.c,v 1.249 2021/06/27 17:57:13 schwarze Exp $ */
 /*
  * Copyright (c) 2010-2015, 2017-2020 Ingo Schwarze <schwarze@openbsd.org>
  * Copyright (c) 2008-2012, 2014 Kristaps Dzonsons <kristaps@bsd.lv>
@@ -1821,7 +1821,7 @@ roff_parsetext(struct roff *r, struct buf *buf, int pos, int *offs)
 }
 
 int
-roff_parseln(struct roff *r, int ln, struct buf *buf, int *offs)
+roff_parseln(struct roff *r, int ln, struct buf *buf, int *offs, size_t len)
 {
        enum roff_tok    t;
        int              e;
@@ -1832,6 +1832,14 @@ roff_parseln(struct roff *r, int ln, struct buf *buf, int *offs)
 
        ppos = pos = *offs;
 
+       if (len > 80 && r->tbl == NULL && r->eqn == NULL &&
+           (r->man->flags & ROFF_NOFILL) == 0 &&
+           strchr(" .\\", buf->buf[pos]) == NULL &&
+           buf->buf[pos] != r->control &&
+           strcspn(buf->buf, " ") < 80)
+               mandoc_msg(MANDOCERR_TEXT_LONG, ln, (int)len - 1,
+                   "%.20s...", buf->buf + pos);
+
        /* Handle in-line equation delimiters. */
 
        if (r->tbl == NULL &&