From: kristaps@mdocml.bsd.lv
To: source@mdocml.bsd.lv
Subject: mdocml: Give libroff an extremely simple table-lookup instead of
Date: Sat, 29 May 2010 15:41:47 -0400 (EDT) [thread overview]
Message-ID: <201005291941.o4TJflxU029996@krisdoz.my.domain> (raw)
Log Message:
-----------
Give libroff an extremely simple table-lookup instead of iterating over
each and every macro. The table keys on the first character then walks
a chain. Good enough for now.
Modified Files:
--------------
mdocml:
roff.c
Revision Data
-------------
Index: roff.c
===================================================================
RCS file: /usr/vhosts/mdocml.bsd.lv/cvs/mdocml/roff.c,v
retrieving revision 1.84
retrieving revision 1.85
diff -Lroff.c -Lroff.c -u -p -r1.84 -r1.85
--- roff.c
+++ roff.c
@@ -19,6 +19,7 @@
#endif
#include <assert.h>
+#include <ctype.h>
#include <stdlib.h>
#include <string.h>
#include <stdio.h>
@@ -91,6 +92,7 @@ struct roffmac {
roffproc sub; /* process as child of macro */
int flags;
#define ROFFMAC_STRUCT (1 << 0) /* always interpret */
+ struct roffmac *next;
};
static enum rofferr roff_block(ROFF_ARGS);
@@ -103,32 +105,65 @@ static enum rofferr roff_cond_text(ROFF
static enum rofferr roff_cond_sub(ROFF_ARGS);
static enum rofferr roff_line(ROFF_ARGS);
-const struct roffmac roffs[ROFF_MAX] = {
- { "am", roff_block, roff_block_text, roff_block_sub, 0 },
- { "ami", roff_block, roff_block_text, roff_block_sub, 0 },
- { "am1", roff_block, roff_block_text, roff_block_sub, 0 },
- { "de", roff_block, roff_block_text, roff_block_sub, 0 },
- { "dei", roff_block, roff_block_text, roff_block_sub, 0 },
- { "de1", roff_block, roff_block_text, roff_block_sub, 0 },
- { "ds", roff_line, NULL, NULL, 0 },
- { "el", roff_cond, roff_cond_text, roff_cond_sub, ROFFMAC_STRUCT },
- { "ie", roff_cond, roff_cond_text, roff_cond_sub, ROFFMAC_STRUCT },
- { "if", roff_cond, roff_cond_text, roff_cond_sub, ROFFMAC_STRUCT },
- { "ig", roff_block, roff_block_text, roff_block_sub, 0 },
- { "rm", roff_line, NULL, NULL, 0 },
- { "tr", roff_line, NULL, NULL, 0 },
- { ".", roff_cblock, NULL, NULL, 0 },
- { "\\}", roff_ccond, NULL, NULL, 0 },
+/* See roff_hash_find() */
+
+#define ASCII_HI 126
+#define ASCII_LO 33
+#define HASHWIDTH (ASCII_HI - ASCII_LO + 1)
+
+static struct roffmac *hash[HASHWIDTH];
+
+static struct roffmac roffs[ROFF_MAX] = {
+ { "am", roff_block, roff_block_text, roff_block_sub, 0, NULL },
+ { "ami", roff_block, roff_block_text, roff_block_sub, 0, NULL },
+ { "am1", roff_block, roff_block_text, roff_block_sub, 0, NULL },
+ { "de", roff_block, roff_block_text, roff_block_sub, 0, NULL },
+ { "dei", roff_block, roff_block_text, roff_block_sub, 0, NULL },
+ { "de1", roff_block, roff_block_text, roff_block_sub, 0, NULL },
+ { "ds", roff_line, NULL, NULL, 0, NULL },
+ { "el", roff_cond, roff_cond_text, roff_cond_sub, ROFFMAC_STRUCT, NULL },
+ { "ie", roff_cond, roff_cond_text, roff_cond_sub, ROFFMAC_STRUCT, NULL },
+ { "if", roff_cond, roff_cond_text, roff_cond_sub, ROFFMAC_STRUCT, NULL },
+ { "ig", roff_block, roff_block_text, roff_block_sub, 0, NULL },
+ { "rm", roff_line, NULL, NULL, 0, NULL },
+ { "tr", roff_line, NULL, NULL, 0, NULL },
+ { ".", roff_cblock, NULL, NULL, 0, NULL },
+ { "\\}", roff_ccond, NULL, NULL, 0, NULL },
};
static void roff_free1(struct roff *);
static enum rofft roff_hash_find(const char *);
+static void roff_hash_init(void);
static void roffnode_cleanscope(struct roff *);
static int roffnode_push(struct roff *,
enum rofft, int, int);
static void roffnode_pop(struct roff *);
static enum rofft roff_parse(const char *, int *);
+/* See roff_hash_find() */
+#define ROFF_HASH(p) (p[0] - ASCII_LO)
+
+static void
+roff_hash_init(void)
+{
+ struct roffmac *n;
+ int buc, i;
+
+ for (i = 0; i < (int)ROFF_MAX; i++) {
+ assert(roffs[i].name[0] >= ASCII_LO);
+ assert(roffs[i].name[0] <= ASCII_HI);
+
+ buc = ROFF_HASH(roffs[i].name);
+
+ if (NULL != (n = hash[buc])) {
+ for ( ; n->next; n = n->next)
+ /* Do nothing. */ ;
+ n->next = &roffs[i];
+ } else
+ hash[buc] = &roffs[i];
+ }
+}
+
/*
* Look up a roff token by its name. Returns ROFF_MAX if no macro by
@@ -137,13 +172,26 @@ static enum rofft roff_parse(const char
static enum rofft
roff_hash_find(const char *p)
{
- int i;
-
- /* FIXME: make this be fast and efficient. */
+ int buc;
+ struct roffmac *n;
- for (i = 0; i < (int)ROFF_MAX; i++)
- if (0 == strcmp(roffs[i].name, p))
- return((enum rofft)i);
+ /*
+ * libroff has an extremely simple hashtable, for the time
+ * being, which simply keys on the first character, which must
+ * be printable, then walks a chain. It works well enough until
+ * optimised.
+ */
+
+ if (p[0] < ASCII_LO || p[0] > ASCII_HI)
+ return(ROFF_MAX);
+
+ buc = ROFF_HASH(p);
+
+ if (NULL == (n = hash[buc]))
+ return(ROFF_MAX);
+ for ( ; n; n = n->next)
+ if (0 == strcmp(n->name, p))
+ return((enum rofft)(n - roffs));
return(ROFF_MAX);
}
@@ -236,6 +284,8 @@ roff_alloc(const mandocmsg msg, void *da
r->msg = msg;
r->data = data;
r->rstackpos = -1;
+
+ roff_hash_init();
return(r);
}
--
To unsubscribe send an email to source+unsubscribe@mdocml.bsd.lv
reply other threads:[~2010-05-29 19:41 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=201005291941.o4TJflxU029996@krisdoz.my.domain \
--to=kristaps@mdocml.bsd.lv \
--cc=source@mdocml.bsd.lv \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).