File: [cvsweb.bsd.lv] / docbook2mdoc / statistics.c (download)
Revision 1.1, Fri Mar 29 15:55:28 2019 UTC (5 years, 5 months ago) by schwarze
Branch: MAIN
Add a utility for docbook2mdoc developers
to collect element usage and parenting statistics,
to help decide which nodes should be most urgently worked on.
|
/* $Id: statistics.c,v 1.1 2019/03/29 15:55:28 schwarze Exp $ */
/*
* Copyright (c) 2019 Ingo Schwarze <schwarze@openbsd.org>
*
* Permission to use, copy, modify, and distribute this software for any
* purpose with or without fee is hereby granted, provided that the above
* copyright notice and this permission notice appear in all copies.
*
* THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHORS DISCLAIM ALL WARRANTIES
* WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
* MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR
* ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
* ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
*/
#include <assert.h>
#include <ctype.h>
#include <err.h>
#include <fcntl.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>
/*
* Count parent-child element relations in a corpus of DocBook documents.
*
* Read absolute or relative input file names from standard input,
* one per line.
* For each parent-child relation, print the total number of occurrences,
* the parent name, and the child name, separated by tab characters
* and followed by a newline character.
*
* Typical usage:
* statistics < filenames.txt | sort -n
* statistics < filenames.txt | grep '\<listitem\>' | sort -n
*/
struct entry {
char *parent;
char *child;
int count;
};
static struct entry *table;
static size_t tablesz;
static size_t tablei;
static char **stack;
static size_t stacksz;
static size_t stacki;
/*
* Count one instance of a parent-child relation.
*/
static void
table_add(const char *parent, const char *child)
{
size_t i;
/* If the table entry already exists, increment its count. */
for (i = 0; i < tablei; i++) {
if (strcmp(parent, table[i].parent) == 0 &&
strcmp(child, table[i].child) == 0) {
table[i].count++;
return;
}
}
/* If the table is full, make room. */
if (tablei == tablesz) {
tablesz += 64;
table = reallocarray(table, tablesz, sizeof(*table));
if (table == NULL)
err(1, NULL);
}
/* Add a new entry to the table. */
if ((table[tablei].parent = strdup(parent)) == NULL)
err(1, NULL);
if ((table[tablei].child = strdup(child)) == NULL)
err(1, NULL);
table[tablei++].count = 1;
}
/*
* Enter an element.
*/
static void
stack_push(const char *name)
{
if (stacki == stacksz) {
stacksz += 8;
stack = reallocarray(stack, stacksz, sizeof(*stack));
if (stack == NULL)
err(1, NULL);
}
if ((stack[stacki++] = strdup(name)) == NULL)
err(1, NULL);
}
/*
* Exit an element.
*/
static void
stack_pop(const char *name)
{
if (stacki > 0 && (name == NULL ||
strcmp(name, stack[stacki - 1]) == 0))
free(stack[--stacki]);
}
/*
* Simplified version from parse.c.
*/
static int
advance(char *b, size_t rlen, size_t *pend, const char *charset)
{
int space;
if (*charset == ' ') {
space = 1;
charset++;
} else
space = 0;
while (*pend < rlen) {
if (space && isspace((unsigned char)b[*pend]))
break;
if (strchr(charset, b[*pend]) != NULL)
break;
++*pend;
}
if (*pend == rlen) {
b[rlen] = '\0';
return 1;
} else
return 0;
}
/*
* Simplified version from parse.c.
*/
static void
parse_file(int fd, char *fname)
{
char b[4096];
ssize_t rsz; /* Return value from read(2). */
size_t rlen; /* Number of bytes in b[]. */
size_t poff; /* Parse offset in b[]. */
size_t pend; /* Offset of the end of the current word. */
int in_tag, in_arg, in_quotes, elem_end;
rlen = 0;
in_tag = in_arg = in_quotes = 0;
while ((rsz = read(fd, b + rlen, sizeof(b) - rlen - 1)) >= 0) {
if ((rlen += rsz) == 0)
break;
pend = 0;
for (;;) {
if ((poff = pend) == rlen)
break;
if (isspace((unsigned char)b[pend])) {
pend++;
continue;
}
if (in_arg) {
if (in_quotes == 0 && b[pend] == '"') {
in_quotes = 1;
pend++;
continue;
}
if (advance(b, rlen, &pend,
in_quotes ? "\"" : " >") && rsz > 0)
break;
in_arg = in_quotes = elem_end = 0;
if (b[pend] == '>') {
in_tag = 0;
if (pend > 0 && b[pend - 1] == '/') {
b[pend - 1] = '\0';
elem_end = 1;
}
}
b[pend] = '\0';
if (pend < rlen)
pend++;
if (elem_end)
stack_pop(NULL);
} else if (in_tag) {
if (advance(b, rlen, &pend, " =>") && rsz > 0)
break;
elem_end = 0;
switch (b[pend]) {
case '>':
in_tag = 0;
if (pend > 0 && b[pend - 1] == '/') {
b[pend - 1] = '\0';
elem_end = 1;
}
break;
case '=':
in_arg = 1;
break;
default:
break;
}
b[pend] = '\0';
if (pend < rlen)
pend++;
if (elem_end)
stack_pop(NULL);
} else if (b[poff] == '<') {
if (advance(b, rlen, &pend, " >") && rsz > 0)
break;
elem_end = 0;
if (b[pend] != '>')
in_tag = 1;
else if (pend > 0 && b[pend - 1] == '/') {
b[pend - 1] = '\0';
elem_end = 1;
}
b[pend] = '\0';
if (pend < rlen)
pend++;
if (b[++poff] == '/') {
elem_end = 1;
poff++;
} else if (b[poff] != '!' && b[poff] != '?') {
table_add(stacki > 0 ?
stack[stacki - 1] : "",
b + poff);
stack_push(b + poff);
}
if (elem_end)
stack_pop(b + poff);
} else {
advance(b, rlen, &pend, "<");
if (stacki > 0)
table_add(stack[stacki - 1], "TEXT");
}
}
assert(poff > 0);
memmove(b, b + poff, rlen - poff);
rlen -= poff;
}
if (rsz < 0)
perror(fname);
}
int
main(int argc, char *argv[])
{
char *fname;
size_t fsz, i;
ssize_t rsz;
int fd;
fd = -1;
fname = NULL;
/* Loop over input files. */
while ((rsz = getline(&fname, &fsz, stdin)) != -1) {
if (fname[rsz - 1] == '\n')
fname[--rsz] = '\0';
if ((fd = open(fname, O_RDONLY, 0)) == -1)
err(1, "%s", fname);
parse_file(fd, fname);
close(fd);
}
/* Cleanup and error handling. */
free(fname);
if (ferror(stdin))
err(1, "standard input");
if (fd == -1)
errx(1, "No input file names found on standard input");
/* Dump results. */
for (i = 0; i < tablei; i++)
printf("%d\t%s\t%s\n", table[i].count,
table[i].parent, table[i].child);
return 0;
}