static void test_str(abts_case *tc, void *data)
{
apr_pool_t *pool = p;
const apr_strmatch_pattern *pattern;
const apr_strmatch_pattern *pattern_nocase;
const apr_strmatch_pattern *pattern_onechar;
const apr_strmatch_pattern *pattern_zero;
const char *match = NULL;
const char *input1 = "string that contains a patterN...";
const char *input2 = "string that contains a pattern...";
const char *input3 = "pattern at the start of a string";
const char *input4 = "string that ends with a pattern";
const char *input5 = "patter\200n not found, negative chars in input";
const char *input6 = "patter\200n, negative chars, contains pattern...";
pattern = apr_strmatch_precompile(pool, "pattern", 1);
ABTS_PTR_NOTNULL(tc, pattern);
pattern_nocase = apr_strmatch_precompile(pool, "pattern", 0);
ABTS_PTR_NOTNULL(tc, pattern_nocase);
pattern_onechar = apr_strmatch_precompile(pool, "g", 0);
ABTS_PTR_NOTNULL(tc, pattern_onechar);
pattern_zero = apr_strmatch_precompile(pool, "", 0);
ABTS_PTR_NOTNULL(tc, pattern_zero);
match = apr_strmatch(pattern, input1, strlen(input1));
ABTS_PTR_EQUAL(tc, match, NULL);
match = apr_strmatch(pattern, input2, strlen(input2));
ABTS_PTR_EQUAL(tc, match, input2 + 23);
match = apr_strmatch(pattern_onechar, input1, strlen(input1));
ABTS_PTR_EQUAL(tc, match, input1 + 5);
match = apr_strmatch(pattern_zero, input1, strlen(input1));
ABTS_PTR_EQUAL(tc, match, input1);
match = apr_strmatch(pattern_nocase, input1, strlen(input1));
ABTS_PTR_EQUAL(tc, match, input1 + 23);
match = apr_strmatch(pattern, input3, strlen(input3));
ABTS_PTR_EQUAL(tc, match, input3);
match = apr_strmatch(pattern, input4, strlen(input4));
ABTS_PTR_EQUAL(tc, match, input4 + 24);
match = apr_strmatch(pattern, input5, strlen(input5));
ABTS_PTR_EQUAL(tc, match, NULL);
match = apr_strmatch(pattern, input6, strlen(input6));
ABTS_PTR_EQUAL(tc, match, input6 + 35);
}
static apr_status_t do_pattmatch(ap_filter_t *f, apr_bucket *inb,
apr_bucket_brigade *mybb,
apr_pool_t *pool)
{
int i;
int force_quick = 0;
ap_regmatch_t regm[AP_MAX_REG_MATCH];
apr_size_t bytes;
apr_size_t len;
const char *buff;
struct ap_varbuf vb;
apr_bucket *b;
apr_bucket *tmp_b;
subst_dir_conf *cfg =
(subst_dir_conf *) ap_get_module_config(f->r->per_dir_config,
&substitute_module);
subst_pattern_t *script;
APR_BRIGADE_INSERT_TAIL(mybb, inb);
ap_varbuf_init(pool, &vb, 0);
script = (subst_pattern_t *) cfg->patterns->elts;
/*
* Simple optimization. If we only have one pattern, then
* we can safely avoid the overhead of flattening
*/
if (cfg->patterns->nelts == 1) {
force_quick = 1;
}
for (i = 0; i < cfg->patterns->nelts; i++) {
for (b = APR_BRIGADE_FIRST(mybb);
b != APR_BRIGADE_SENTINEL(mybb);
b = APR_BUCKET_NEXT(b)) {
if (APR_BUCKET_IS_METADATA(b)) {
/*
* we should NEVER see this, because we should never
* be passed any, but "handle" it just in case.
*/
continue;
}
if (apr_bucket_read(b, &buff, &bytes, APR_BLOCK_READ)
== APR_SUCCESS) {
int have_match = 0;
vb.strlen = 0;
if (script->pattern) {
const char *repl;
/*
* space_left counts how many bytes we have left until the
* line length reaches max_line_length.
*/
apr_size_t space_left = cfg->max_line_length;
apr_size_t repl_len = strlen(script->replacement);
while ((repl = apr_strmatch(script->pattern, buff, bytes)))
{
have_match = 1;
/* get offset into buff for pattern */
len = (apr_size_t) (repl - buff);
if (script->flatten && !force_quick) {
/*
* We are flattening the buckets here, meaning
* that we don't do the fast bucket splits.
* Instead we copy over what the buckets would
* contain and use them. This is slow, since we
* are constanting allocing space and copying
* strings.
*/
if (vb.strlen + len + repl_len > cfg->max_line_length)
return APR_ENOMEM;
ap_varbuf_strmemcat(&vb, buff, len);
ap_varbuf_strmemcat(&vb, script->replacement, repl_len);
}
else {
/*
* The string before the match but after the
* previous match (if any) has length 'len'.
* Check if we still have space for this string and
* the replacement string.
*/
if (space_left < len + repl_len)
return APR_ENOMEM;
space_left -= len + repl_len;
/*
* We now split off the string before the match
* as its own bucket, then isolate the matched
* string and delete it.
*/
SEDRMPATBCKT(b, len, tmp_b, script->patlen);
/*
* Finally, we create a bucket that contains the
* replacement...
*/
tmp_b = apr_bucket_transient_create(script->replacement,
script->replen,
f->r->connection->bucket_alloc);
/* ... and insert it */
APR_BUCKET_INSERT_BEFORE(b, tmp_b);
}
/* now we need to adjust buff for all these changes */
len += script->patlen;
bytes -= len;
buff += len;
}
if (have_match) {
if (script->flatten && !force_quick) {
/* XXX: we should check for AP_MAX_BUCKETS here and
* XXX: call ap_pass_brigade accordingly
*/
char *copy = ap_varbuf_pdup(pool, &vb, NULL, 0,
buff, bytes, &len);
tmp_b = apr_bucket_pool_create(copy, len, pool,
f->r->connection->bucket_alloc);
APR_BUCKET_INSERT_BEFORE(b, tmp_b);
apr_bucket_delete(b);
b = tmp_b;
}
else {
/*
* We want the behaviour to be predictable.
* Therefore we try to always error out if the
* line length is larger than the limit,
* regardless of the content of the line. So,
* let's check if the remaining non-matching
* string does not exceed the limit.
*/
if (space_left < b->length)
return APR_ENOMEM;
}
}
}
else if (script->regexp) {
int left = bytes;
const char *pos = buff;
char *repl;
apr_size_t space_left = cfg->max_line_length;
while (!ap_regexec_len(script->regexp, pos, left,
AP_MAX_REG_MATCH, regm, 0)) {
apr_status_t rv;
have_match = 1;
if (script->flatten && !force_quick) {
/* check remaining buffer size */
/* Note that the last param in ap_varbuf_regsub below
* must stay positive. If it gets 0, it would mean
* unlimited space available. */
if (vb.strlen + regm[0].rm_so >= cfg->max_line_length)
return APR_ENOMEM;
/* copy bytes before the match */
if (regm[0].rm_so > 0)
ap_varbuf_strmemcat(&vb, pos, regm[0].rm_so);
/* add replacement string, last argument is unsigned! */
rv = ap_varbuf_regsub(&vb, script->replacement, pos,
AP_MAX_REG_MATCH, regm,
cfg->max_line_length - vb.strlen);
if (rv != APR_SUCCESS)
return rv;
}
else {
apr_size_t repl_len;
/* acount for string before the match */
if (space_left <= regm[0].rm_so)
return APR_ENOMEM;
space_left -= regm[0].rm_so;
rv = ap_pregsub_ex(pool, &repl,
script->replacement, pos,
AP_MAX_REG_MATCH, regm,
space_left);
if (rv != APR_SUCCESS)
return rv;
repl_len = strlen(repl);
space_left -= repl_len;
len = (apr_size_t) (regm[0].rm_eo - regm[0].rm_so);
SEDRMPATBCKT(b, regm[0].rm_so, tmp_b, len);
tmp_b = apr_bucket_transient_create(repl, repl_len,
f->r->connection->bucket_alloc);
APR_BUCKET_INSERT_BEFORE(b, tmp_b);
}
/*
* reset to past what we just did. pos now maps to b
* again
*/
pos += regm[0].rm_eo;
left -= regm[0].rm_eo;
}
if (have_match && script->flatten && !force_quick) {
char *copy;
/* Copy result plus the part after the last match into
* a bucket.
*/
copy = ap_varbuf_pdup(pool, &vb, NULL, 0, pos, left,
&len);
tmp_b = apr_bucket_pool_create(copy, len, pool,
f->r->connection->bucket_alloc);
APR_BUCKET_INSERT_BEFORE(b, tmp_b);
apr_bucket_delete(b);
b = tmp_b;
}
}
else {
ap_assert(0);
continue;
}
}
}
script++;
}
ap_varbuf_free(&vb);
return APR_SUCCESS;
}
static void do_pattmatch(ap_filter_t *f, apr_bucket *inb,
apr_bucket_brigade *mybb,
apr_pool_t *tmp_pool)
{
int i;
int force_quick = 0;
ap_regmatch_t regm[AP_MAX_REG_MATCH];
apr_size_t bytes;
apr_size_t len;
apr_size_t fbytes;
const char *buff;
const char *repl;
char *scratch;
char *p;
char *s1;
char *s2;
apr_bucket *b;
apr_bucket *tmp_b;
apr_pool_t *tpool;
subst_dir_conf *cfg =
(subst_dir_conf *) ap_get_module_config(f->r->per_dir_config,
&substitute_module);
subst_pattern_t *script;
APR_BRIGADE_INSERT_TAIL(mybb, inb);
script = (subst_pattern_t *) cfg->patterns->elts;
apr_pool_create(&tpool, tmp_pool);
scratch = NULL;
fbytes = 0;
/*
* Simple optimization. If we only have one pattern, then
* we can safely avoid the overhead of flattening
*/
if (cfg->patterns->nelts == 1) {
force_quick = 1;
}
for (i = 0; i < cfg->patterns->nelts; i++) {
for (b = APR_BRIGADE_FIRST(mybb);
b != APR_BRIGADE_SENTINEL(mybb);
b = APR_BUCKET_NEXT(b)) {
if (APR_BUCKET_IS_METADATA(b)) {
/*
* we should NEVER see this, because we should never
* be passed any, but "handle" it just in case.
*/
continue;
}
if (apr_bucket_read(b, &buff, &bytes, APR_BLOCK_READ)
== APR_SUCCESS) {
s1 = NULL;
if (script->pattern) {
while ((repl = apr_strmatch(script->pattern, buff, bytes)))
{
/* get offset into buff for pattern */
len = (apr_size_t) (repl - buff);
if (script->flatten && !force_quick) {
/*
* We are flattening the buckets here, meaning
* that we don't do the fast bucket splits.
* Instead we copy over what the buckets would
* contain and use them. This is slow, since we
* are constanting allocing space and copying
* strings.
*/
SEDSCAT(s1, s2, tmp_pool, buff, len,
script->replacement);
}
else {
/*
* We now split off the stuff before the regex
* as its own bucket, then isolate the pattern
* and delete it.
*/
SEDRMPATBCKT(b, len, tmp_b, script->patlen);
/*
* Finally, we create a bucket that contains the
* replacement...
*/
tmp_b = apr_bucket_transient_create(script->replacement,
script->replen,
f->r->connection->bucket_alloc);
/* ... and insert it */
APR_BUCKET_INSERT_BEFORE(b, tmp_b);
}
/* now we need to adjust buff for all these changes */
len += script->patlen;
bytes -= len;
buff += len;
}
if (script->flatten && s1 && !force_quick) {
/*
* we've finished looking at the bucket, so remove the
* old one and add in our new one
*/
s2 = apr_pstrmemdup(tmp_pool, buff, bytes);
s1 = apr_pstrcat(tmp_pool, s1, s2, NULL);
tmp_b = apr_bucket_transient_create(s1, strlen(s1),
f->r->connection->bucket_alloc);
APR_BUCKET_INSERT_BEFORE(b, tmp_b);
apr_bucket_delete(b);
b = tmp_b;
}
}
else if (script->regexp) {
/*
* we need a null terminated string here :(. To hopefully
* save time and memory, we don't alloc for each run
* through, but only if we need to have a larger chunk
* to save the string to. So we keep track of how much
* we've allocated and only re-alloc when we need it.
* NOTE: this screams for a macro.
*/
if (!scratch || (bytes > (fbytes + 1))) {
fbytes = bytes + 1;
scratch = apr_palloc(tpool, fbytes);
}
/* reset pointer to the scratch space */
p = scratch;
memcpy(p, buff, bytes);
p[bytes] = '\0';
while (!ap_regexec(script->regexp, p,
AP_MAX_REG_MATCH, regm, 0)) {
/* first, grab the replacement string */
repl = ap_pregsub(tmp_pool, script->replacement, p,
AP_MAX_REG_MATCH, regm);
if (script->flatten && !force_quick) {
SEDSCAT(s1, s2, tmp_pool, p, regm[0].rm_so, repl);
}
else {
len = (apr_size_t) (regm[0].rm_eo - regm[0].rm_so);
SEDRMPATBCKT(b, regm[0].rm_so, tmp_b, len);
tmp_b = apr_bucket_transient_create(repl,
strlen(repl),
f->r->connection->bucket_alloc);
APR_BUCKET_INSERT_BEFORE(b, tmp_b);
}
/*
* reset to past what we just did. buff now maps to b
* again
*/
p += regm[0].rm_eo;
}
if (script->flatten && s1 && !force_quick) {
s1 = apr_pstrcat(tmp_pool, s1, p, NULL);
tmp_b = apr_bucket_transient_create(s1, strlen(s1),
f->r->connection->bucket_alloc);
APR_BUCKET_INSERT_BEFORE(b, tmp_b);
apr_bucket_delete(b);
b = tmp_b;
}
}
else {
/* huh? */
continue;
}
}
}
script++;
}
apr_pool_destroy(tpool);
return;
}
apr_status_t dav_svn__location_body_filter(ap_filter_t *f,
apr_bucket_brigade *bb)
{
request_rec *r = f->r;
locate_ctx_t *ctx = f->ctx;
apr_bucket *bkt;
const char *master_uri, *root_dir, *canonicalized_uri;
apr_uri_t uri;
/* Don't filter if we're in a subrequest or we aren't setup to
proxy anything. */
master_uri = dav_svn__get_master_uri(r);
if (r->main || !master_uri) {
ap_remove_output_filter(f);
return ap_pass_brigade(f->next, bb);
}
/* And don't filter if our search-n-replace would be a noop anyway
(that is, if our root path matches that of the master server). */
apr_uri_parse(r->pool, master_uri, &uri);
root_dir = dav_svn__get_root_dir(r);
canonicalized_uri = svn_urlpath__canonicalize(uri.path, r->pool);
if (strcmp(canonicalized_uri, root_dir) == 0) {
ap_remove_output_filter(f);
return ap_pass_brigade(f->next, bb);
}
/* ### FIXME: GET and PROPFIND requests that make it here must be
### referring to data inside commit transactions-in-progress.
### We've got to be careful not to munge the versioned data
### they return in the process of trying to do URI fix-ups.
### See issue #3445 for details. */
/* We are url encoding the current url and the master url
as incoming(from master) request body has it encoded already. */
canonicalized_uri = svn_path_uri_encode(canonicalized_uri, r->pool);
root_dir = svn_path_uri_encode(root_dir, r->pool);
if (!f->ctx) {
ctx = f->ctx = apr_pcalloc(r->pool, sizeof(*ctx));
ctx->remotepath = canonicalized_uri;
ctx->remotepath_len = strlen(ctx->remotepath);
ctx->localpath = root_dir;
ctx->localpath_len = strlen(ctx->localpath);
ctx->pattern = apr_strmatch_precompile(r->pool, ctx->remotepath, 1);
ctx->pattern_len = ctx->remotepath_len;
}
bkt = APR_BRIGADE_FIRST(bb);
while (bkt != APR_BRIGADE_SENTINEL(bb)) {
const char *data, *match;
apr_size_t len;
/* read */
apr_bucket_read(bkt, &data, &len, APR_BLOCK_READ);
match = apr_strmatch(ctx->pattern, data, len);
if (match) {
apr_bucket *next_bucket;
apr_bucket_split(bkt, match - data);
next_bucket = APR_BUCKET_NEXT(bkt);
apr_bucket_split(next_bucket, ctx->pattern_len);
bkt = APR_BUCKET_NEXT(next_bucket);
apr_bucket_delete(next_bucket);
next_bucket = apr_bucket_pool_create(ctx->localpath,
ctx->localpath_len,
r->pool, bb->bucket_alloc);
APR_BUCKET_INSERT_BEFORE(bkt, next_bucket);
}
else {
bkt = APR_BUCKET_NEXT(bkt);
}
}
return ap_pass_brigade(f->next, bb);
}
apr_status_t dav_svn__location_in_filter(ap_filter_t *f,
apr_bucket_brigade *bb,
ap_input_mode_t mode,
apr_read_type_e block,
apr_off_t readbytes)
{
request_rec *r = f->r;
locate_ctx_t *ctx = f->ctx;
apr_status_t rv;
apr_bucket *bkt;
const char *master_uri, *root_dir, *canonicalized_uri;
apr_uri_t uri;
/* Don't filter if we're in a subrequest or we aren't setup to
proxy anything. */
master_uri = dav_svn__get_master_uri(r);
if (r->main || !master_uri) {
ap_remove_input_filter(f);
return ap_get_brigade(f->next, bb, mode, block, readbytes);
}
/* And don't filter if our search-n-replace would be a noop anyway
(that is, if our root path matches that of the master server). */
apr_uri_parse(r->pool, master_uri, &uri);
root_dir = dav_svn__get_root_dir(r);
canonicalized_uri = svn_urlpath__canonicalize(uri.path, r->pool);
if (strcmp(canonicalized_uri, root_dir) == 0) {
ap_remove_input_filter(f);
return ap_get_brigade(f->next, bb, mode, block, readbytes);
}
/* ### FIXME: While we want to fix up any locations in proxied XML
### requests, we do *not* want to be futzing with versioned (or
### to-be-versioned) data, such as the file contents present in
### PUT requests and properties in PROPPATCH requests.
### See issue #3445 for details. */
/* We are url encoding the current url and the master url
as incoming(from client) request body has it encoded already. */
canonicalized_uri = svn_path_uri_encode(canonicalized_uri, r->pool);
root_dir = svn_path_uri_encode(root_dir, r->pool);
if (!f->ctx) {
ctx = f->ctx = apr_pcalloc(r->pool, sizeof(*ctx));
ctx->remotepath = canonicalized_uri;
ctx->remotepath_len = strlen(ctx->remotepath);
ctx->localpath = root_dir;
ctx->localpath_len = strlen(ctx->localpath);
ctx->pattern = apr_strmatch_precompile(r->pool, ctx->localpath, 1);
ctx->pattern_len = ctx->localpath_len;
}
rv = ap_get_brigade(f->next, bb, mode, block, readbytes);
if (rv) {
return rv;
}
bkt = APR_BRIGADE_FIRST(bb);
while (bkt != APR_BRIGADE_SENTINEL(bb)) {
const char *data, *match;
apr_size_t len;
if (APR_BUCKET_IS_METADATA(bkt)) {
bkt = APR_BUCKET_NEXT(bkt);
continue;
}
/* read */
apr_bucket_read(bkt, &data, &len, APR_BLOCK_READ);
match = apr_strmatch(ctx->pattern, data, len);
if (match) {
apr_bucket *next_bucket;
apr_bucket_split(bkt, match - data);
next_bucket = APR_BUCKET_NEXT(bkt);
apr_bucket_split(next_bucket, ctx->pattern_len);
bkt = APR_BUCKET_NEXT(next_bucket);
apr_bucket_delete(next_bucket);
next_bucket = apr_bucket_pool_create(ctx->remotepath,
ctx->remotepath_len,
r->pool, bb->bucket_alloc);
APR_BUCKET_INSERT_BEFORE(bkt, next_bucket);
}
else {
bkt = APR_BUCKET_NEXT(bkt);
}
}
return APR_SUCCESS;
}
static apr_status_t line_edit_filter(ap_filter_t* f, apr_bucket_brigade* bb) {
int i, j;
unsigned int match ;
unsigned int nmatch = 10 ;
ap_regmatch_t pmatch[10] ;
const char* bufp;
const char* subs ;
apr_size_t bytes ;
apr_size_t fbytes ;
apr_size_t offs ;
const char* buf ;
const char* le = NULL ;
const char* le_n ;
const char* le_r ;
char* fbuf ;
apr_bucket* b = APR_BRIGADE_FIRST(bb) ;
apr_bucket* b1 ;
int found = 0 ;
apr_status_t rv ;
apr_bucket_brigade* bbline ;
line_edit_cfg* cfg
= ap_get_module_config(f->r->per_dir_config, &line_edit_module) ;
rewriterule* rules = (rewriterule*) cfg->rewriterules->elts ;
rewriterule* newrule;
line_edit_ctx* ctx = f->ctx ;
if (ctx == NULL) {
/* check env to see if we're wanted, to give basic control with 2.0 */
buf = apr_table_get(f->r->subprocess_env, "LineEdit");
if (buf && f->r->content_type) {
char* lcbuf = apr_pstrdup(f->r->pool, buf) ;
char* lctype = apr_pstrdup(f->r->pool, f->r->content_type) ;
char* c ;
for (c = lcbuf; *c; ++c)
if (isupper(*c))
*c = tolower(*c) ;
for (c = lctype; *c; ++c)
if (isupper(*c))
*c = tolower(*c) ;
else if (*c == ';') {
*c = 0 ;
break ;
}
if (!strstr(lcbuf, lctype)) {
/* don't filter this content type */
ap_filter_t* fnext = f->next ;
ap_remove_output_filter(f) ;
return ap_pass_brigade(fnext, bb) ;
}
}
ctx = f->ctx = apr_palloc(f->r->pool, sizeof(line_edit_ctx)) ;
ctx->bbsave = apr_brigade_create(f->r->pool, f->c->bucket_alloc) ;
/* If we have any regex matches, we'll need to copy everything, so we
* have null-terminated strings to parse. That's a lot of memory if
* we're streaming anything big. So we'll use (and reuse) a local
* subpool. Fall back to the request pool if anything bad happens.
*/
ctx->lpool = f->r->pool ;
for (i = 0; i < cfg->rewriterules->nelts; ++i) {
if ( rules[i].flags & M_REGEX ) {
if (apr_pool_create(&ctx->lpool, f->r->pool) != APR_SUCCESS) {
ctx->lpool = f->r->pool ;
}
break ;
}
}
/* If we have env interpolation, we'll need a private copy of
* our rewrite rules with this requests env. Otherwise we can
* save processing time by using the original.
*
* If one ENV is found, we also have to copy all previous and
* subsequent rules, even those with no interpolation.
*/
ctx->rewriterules = cfg->rewriterules;
for (i = 0; i < cfg->rewriterules->nelts; ++i) {
found |= (rules[i].flags & M_ENV) ;
if ( found ) {
if (ctx->rewriterules == cfg->rewriterules) {
ctx->rewriterules = apr_array_make(f->r->pool,
cfg->rewriterules->nelts, sizeof(rewriterule));
for (j = 0; j < i; ++j) {
newrule = apr_array_push (((line_edit_ctx*)ctx)->rewriterules) ;
newrule->from = rules[j].from;
newrule->to = rules[j].to;
newrule->flags = rules[j].flags;
newrule->length = rules[j].length;
}
}
/* this rule needs to be interpolated */
newrule = apr_array_push (((line_edit_ctx*)ctx)->rewriterules) ;
newrule->from = rules[i].from;
if (rules[i].flags & M_ENV) {
newrule->to = interpolate_env(f->r, rules[i].to);
} else {
newrule->to = rules[i].to ;
}
newrule->flags = rules[i].flags;
newrule->length = rules[i].length;
}
}
/* for back-compatibility with Apache 2.0, set some protocol stuff */
apr_table_unset(f->r->headers_out, "Content-Length") ;
apr_table_unset(f->r->headers_out, "Content-MD5") ;
apr_table_unset(f->r->headers_out, "Accept-Ranges") ;
}
/* by now our rules are in ctx->rewriterules */
rules = (rewriterule*) ctx->rewriterules->elts ;
/* bbline is what goes to the next filter,
* so we (can) have a new one each time.
*/
bbline = apr_brigade_create(f->r->pool, f->c->bucket_alloc) ;
/* first ensure we have no mid-line breaks that might be in the
* middle of a search string causing us to miss it! At the same
* time we split into lines to avoid pattern-matching over big
* chunks of memory.
*/
while ( b != APR_BRIGADE_SENTINEL(bb) ) {
if ( !APR_BUCKET_IS_METADATA(b) ) {
if ( apr_bucket_read(b, &buf, &bytes, APR_BLOCK_READ) == APR_SUCCESS ) {
if ( bytes == 0 ) {
APR_BUCKET_REMOVE(b) ;
} else while ( bytes > 0 ) {
switch (cfg->lineend) {
case LINEEND_UNIX:
le = memchr(buf, '\n', bytes) ;
break ;
case LINEEND_MAC:
le = memchr(buf, '\r', bytes) ;
break ;
case LINEEND_DOS:
/* Edge-case issue: if a \r\n spans buckets it'll get missed.
* Not a problem for present purposes, but would be an issue
* if we claimed to support pattern matching on the lineends.
*/
found = 0 ;
le = memchr(buf+1, '\n', bytes-1) ;
while ( le && !found ) {
if ( le[-1] == '\r' ) {
found = 1 ;
} else {
le = memchr(le+1, '\n', bytes-1 - (le+1 - buf)) ;
}
}
if ( !found )
le = 0 ;
break;
case LINEEND_ANY:
case LINEEND_UNSET:
/* Edge-case notabug: if a \r\n spans buckets it'll get seen as
* two line-ends. It'll insert the \n as a one-byte bucket.
*/
le_n = memchr(buf, '\n', bytes) ;
le_r = memchr(buf, '\r', bytes) ;
if ( le_n != NULL )
if ( le_n == le_r + sizeof(char))
le = le_n ;
else if ( (le_r < le_n) && (le_r != NULL) )
le = le_r ;
else
le = le_n ;
else
le = le_r ;
break;
case LINEEND_NONE:
le = 0 ;
break;
case LINEEND_CUSTOM:
le = memchr(buf, cfg->lechar, bytes) ;
break;
}
if ( le ) {
/* found a lineend in this bucket. */
offs = 1 + ((unsigned int)le-(unsigned int)buf) / sizeof(char) ;
apr_bucket_split(b, offs) ;
bytes -= offs ;
buf += offs ;
b1 = APR_BUCKET_NEXT(b) ;
APR_BUCKET_REMOVE(b);
/* Is there any previous unterminated content ? */
if ( !APR_BRIGADE_EMPTY(ctx->bbsave) ) {
/* append this to any content waiting for a lineend */
APR_BRIGADE_INSERT_TAIL(ctx->bbsave, b) ;
rv = apr_brigade_pflatten(ctx->bbsave, &fbuf, &fbytes, f->r->pool) ;
/* make b a new bucket of the flattened stuff */
b = apr_bucket_pool_create(fbuf, fbytes, f->r->pool,
f->r->connection->bucket_alloc) ;
/* bbsave has been consumed, so clear it */
apr_brigade_cleanup(ctx->bbsave) ;
}
/* b now contains exactly one line */
APR_BRIGADE_INSERT_TAIL(bbline, b);
b = b1 ;
} else {
/* no lineend found. Remember the dangling content */
APR_BUCKET_REMOVE(b);
APR_BRIGADE_INSERT_TAIL(ctx->bbsave, b);
bytes = 0 ;
}
} /* while bytes > 0 */
} else {
/* bucket read failed - oops ! Let's remove it. */
APR_BUCKET_REMOVE(b);
}
} else if ( APR_BUCKET_IS_EOS(b) ) {
/* If there's data to pass, send it in one bucket */
if ( !APR_BRIGADE_EMPTY(ctx->bbsave) ) {
rv = apr_brigade_pflatten(ctx->bbsave, &fbuf, &fbytes, f->r->pool) ;
b1 = apr_bucket_pool_create(fbuf, fbytes, f->r->pool,
f->r->connection->bucket_alloc) ;
APR_BRIGADE_INSERT_TAIL(bbline, b1);
}
apr_brigade_cleanup(ctx->bbsave) ;
/* start again rather than segfault if a seriously buggy
* filter in front of us sent a bogus EOS
*/
f->ctx = NULL ;
/* move the EOS to the new brigade */
APR_BUCKET_REMOVE(b);
APR_BRIGADE_INSERT_TAIL(bbline, b);
} else {
/* chop flush or unknown metadata bucket types */
apr_bucket_delete(b);
}
/* OK, reset pointer to what's left (since we're not in a for-loop) */
b = APR_BRIGADE_FIRST(bb) ;
}
/* OK, now we have a bunch of complete lines in bbline,
* so we can apply our edit rules
*/
/* When we get a match, we split the line into before+match+after.
* To flatten that back into one buf every time would be inefficient.
* So we treat it as three separate bufs to apply future rules.
*
* We can only reasonably do that by looping over buckets *inside*
* the loop over rules.
*
* That means concepts like one-match-per-line or start-of-line-only
* won't work, except for the first rule. So we won't pretend.
*/
for (i = 0; i < ctx->rewriterules->nelts; ++i) {
for ( b = APR_BRIGADE_FIRST(bbline) ;
b != APR_BRIGADE_SENTINEL(bbline) ;
b = APR_BUCKET_NEXT(b) ) {
if ( !APR_BUCKET_IS_METADATA(b)
&& (apr_bucket_read(b, &buf, &bytes, APR_BLOCK_READ) == APR_SUCCESS)) {
if ( rules[i].flags & M_REGEX ) {
bufp = apr_pstrmemdup(ctx->lpool, buf, bytes) ;
while ( ! ap_regexec(rules[i].from.r, bufp, nmatch, pmatch, 0) ) {
match = pmatch[0].rm_so ;
subs = ap_pregsub(f->r->pool, rules[i].to, bufp, nmatch, pmatch) ;
apr_bucket_split(b, match) ;
b1 = APR_BUCKET_NEXT(b) ;
apr_bucket_split(b1, pmatch[0].rm_eo - match) ;
b = APR_BUCKET_NEXT(b1) ;
apr_bucket_delete(b1) ;
b1 = apr_bucket_pool_create(subs, strlen(subs), f->r->pool,
f->r->connection->bucket_alloc) ;
APR_BUCKET_INSERT_BEFORE(b, b1) ;
bufp += pmatch[0].rm_eo ;
}
} else {
bufp = buf ;
while (subs = apr_strmatch(rules[i].from.s, bufp, bytes),
subs != NULL) {
match = ((unsigned int)subs - (unsigned int)bufp) / sizeof(char) ;
bytes -= match ;
bufp += match ;
apr_bucket_split(b, match) ;
b1 = APR_BUCKET_NEXT(b) ;
apr_bucket_split(b1, rules[i].length) ;
b = APR_BUCKET_NEXT(b1) ;
apr_bucket_delete(b1) ;
bytes -= rules[i].length ;
bufp += rules[i].length ;
b1 = apr_bucket_immortal_create(rules[i].to, strlen(rules[i].to),
f->r->connection->bucket_alloc) ;
APR_BUCKET_INSERT_BEFORE(b, b1) ;
}
}
}
}
/* If we used a local pool, clear it now */
if ( (ctx->lpool != f->r->pool) && (rules[i].flags & M_REGEX) ) {
apr_pool_clear(ctx->lpool) ;
}
}
/* now pass it down the chain */
rv = ap_pass_brigade(f->next, bbline) ;
/* if we have leftover data, don't risk it going out of scope */
for ( b = APR_BRIGADE_FIRST(ctx->bbsave) ;
b != APR_BRIGADE_SENTINEL(ctx->bbsave) ;
b = APR_BUCKET_NEXT(b)) {
apr_bucket_setaside(b, f->r->pool) ;
}
return rv ;
}
int main (void)
{
apr_pool_t *pool;
const apr_strmatch_pattern *pattern;
const apr_strmatch_pattern *pattern_nocase;
const apr_strmatch_pattern *pattern_onechar;
const apr_strmatch_pattern *pattern_zero;
const char *input1 = "string that contains a patterN...";
const char *input2 = "string that contains a pattern...";
const char *input3 = "pattern at the start of a string";
const char *input4 = "string that ends with a pattern";
const char *input5 = "patter\200n not found, negative chars in input";
const char *input6 = "patter\200n, negative chars, contains pattern...";
(void) apr_initialize();
apr_pool_create(&pool, NULL);
printf("testing pattern precompilation...");
pattern = apr_strmatch_precompile(pool, "pattern", 1);
if (!pattern) {
printf("FAILED\n");
exit(1);
}
pattern_nocase = apr_strmatch_precompile(pool, "pattern", 0);
if (!pattern_nocase) {
printf("FAILED\n");
exit(1);
}
pattern_onechar = apr_strmatch_precompile(pool, "g", 0);
if (!pattern_onechar) {
printf("FAILED\n");
exit(1);
}
pattern_zero = apr_strmatch_precompile(pool, "", 0);
if (!pattern_zero) {
printf("FAILED\n");
exit(1);
}
printf("OK\n");
printf("testing invalid match...");
if (apr_strmatch(pattern, input1, strlen(input1)) != NULL) {
printf("FAILED\n");
exit(1);
}
printf("OK\n");
printf("testing valid match...");
if (apr_strmatch(pattern, input2, strlen(input2)) != input2 + 23) {
printf("FAILED\n");
exit(1);
}
printf("OK\n");
printf("testing single-character match...");
if (apr_strmatch(pattern_onechar, input1, strlen(input1)) != input1 + 5) {
printf("FAILED\n");
exit(1);
}
printf("OK\n");
printf("testing zero-length pattern...");
if (apr_strmatch(pattern_zero, input1, strlen(input1)) != input1) {
printf("FAILED\n");
exit(1);
}
printf("OK\n");
printf("testing inexact-case match...");
if (apr_strmatch(pattern_nocase, input1, strlen(input1)) != input1 + 23) {
printf("FAILED\n");
exit(1);
}
printf("OK\n");
printf("testing match at start of string...");
if (apr_strmatch(pattern, input3, strlen(input3)) != input3) {
printf("FAILED\n");
exit(1);
}
printf("OK\n");
printf("testing match at end of string...");
if (apr_strmatch(pattern, input4, strlen(input4)) != input4 + 24) {
printf("FAILED\n");
exit(1);
}
printf("OK\n");
printf("testing invalid match with negative chars in input string...");
if (apr_strmatch(pattern, input5, strlen(input5)) != NULL) {
printf("FAILED\n");
exit(1);
}
printf("OK\n");
printf("testing valid match with negative chars in input string...");
if (apr_strmatch(pattern, input6, strlen(input6)) != input6 + 35) {
printf("FAILED\n");
exit(1);
}
printf("OK\n");
exit(0);
}
static apr_status_t google_analytics_out_filter(ap_filter_t *f, apr_bucket_brigade *bb)
{
request_rec *r = f->r;
google_analytics_filter_ctx *ctx = f->ctx;
google_analytics_filter_config *c;
apr_bucket *b = APR_BRIGADE_FIRST(bb);
apr_size_t bytes;
apr_size_t fbytes;
apr_size_t offs;
const char *buf;
const char *le = NULL;
const char *le_n;
const char *le_r;
const char *bufp;
const char *subs;
unsigned int match;
apr_bucket *b1;
char *fbuf;
int found = 0;
apr_status_t rv;
apr_bucket_brigade *bbline;
// サブリクエストならなにもしない
if (r->main) {
ap_remove_output_filter(f);
return ap_pass_brigade(f->next, bb);
}
c = ap_get_module_config(r->per_dir_config, &google_analytics_module);
if (ctx == NULL) {
ctx = f->ctx = apr_pcalloc(r->pool, sizeof(google_analytics_filter_ctx));
ctx->bbsave = apr_brigade_create(r->pool, f->c->bucket_alloc);
}
// length かわってしまうので unset で OK?
apr_table_unset(r->headers_out, "Content-Length");
apr_table_unset(r->headers_out, "Content-MD5");
apr_table_unset(r->headers_out, "Accept-Ranges");
apr_table_unset(r->headers_out, "ETag");
bbline = apr_brigade_create(r->pool, f->c->bucket_alloc);
// 改行毎なbucketに編成しなおす
while ( b != APR_BRIGADE_SENTINEL(bb) ) {
if ( !APR_BUCKET_IS_METADATA(b) ) {
if ( apr_bucket_read(b, &buf, &bytes, APR_BLOCK_READ) == APR_SUCCESS ) {
if ( bytes == 0 ) {
APR_BUCKET_REMOVE(b);
} else {
while ( bytes > 0 ) {
le_n = memchr(buf, '\n', bytes);
le_r = memchr(buf, '\r', bytes);
if ( le_n != NULL ) {
if ( le_n == le_r + sizeof(char)) {
le = le_n;
}
else if ( (le_r < le_n) && (le_r != NULL) ) {
le = le_r;
}
else {
le = le_n;
}
}
else {
le = le_r;
}
if ( le ) {
offs = 1 + ((unsigned int)le-(unsigned int)buf) / sizeof(char);
apr_bucket_split(b, offs);
bytes -= offs;
buf += offs;
b1 = APR_BUCKET_NEXT(b);
APR_BUCKET_REMOVE(b);
if ( !APR_BRIGADE_EMPTY(ctx->bbsave) ) {
APR_BRIGADE_INSERT_TAIL(ctx->bbsave, b);
rv = apr_brigade_pflatten(ctx->bbsave, &fbuf, &fbytes, r->pool);
b = apr_bucket_pool_create(fbuf, fbytes, r->pool,
r->connection->bucket_alloc);
apr_brigade_cleanup(ctx->bbsave);
}
APR_BRIGADE_INSERT_TAIL(bbline, b);
b = b1;
} else {
APR_BUCKET_REMOVE(b);
APR_BRIGADE_INSERT_TAIL(ctx->bbsave, b);
bytes = 0;
}
} /* while bytes > 0 */
}
} else {
APR_BUCKET_REMOVE(b);
}
} else if ( APR_BUCKET_IS_EOS(b) ) {
if ( !APR_BRIGADE_EMPTY(ctx->bbsave) ) {
rv = apr_brigade_pflatten(ctx->bbsave, &fbuf, &fbytes, r->pool);
b1 = apr_bucket_pool_create(fbuf, fbytes, r->pool,
r->connection->bucket_alloc);
APR_BRIGADE_INSERT_TAIL(bbline, b1);
}
apr_brigade_cleanup(ctx->bbsave);
f->ctx = NULL;
APR_BUCKET_REMOVE(b);
APR_BRIGADE_INSERT_TAIL(bbline, b);
} else {
apr_bucket_delete(b);
}
b = APR_BRIGADE_FIRST(bb);
}
// 改行毎なbucketをまわす
for ( b = APR_BRIGADE_FIRST(bbline);
b != APR_BRIGADE_SENTINEL(bbline);
b = APR_BUCKET_NEXT(b) ) {
if ( !APR_BUCKET_IS_METADATA(b)
&& (apr_bucket_read(b, &buf, &bytes, APR_BLOCK_READ) == APR_SUCCESS)) {
bufp = buf;
if (ap_regexec(regex_tag_exists, bufp, 0, NULL, 0) == 0) {
break;
}
subs = apr_strmatch(pattern_body_end_tag, bufp, bytes);
if (subs != NULL) {
match = ((unsigned int)subs - (unsigned int)bufp) / sizeof(char);
bytes -= match;
bufp += match;
apr_bucket_split(b, match);
b1 = APR_BUCKET_NEXT(b);
apr_bucket_split(b1, body_end_tag_length);
b = APR_BUCKET_NEXT(b1);
apr_bucket_delete(b1);
bytes -= body_end_tag_length;
bufp += body_end_tag_length;
b1 = apr_bucket_immortal_create(c->replace, strlen(c->replace),
r->connection->bucket_alloc);
APR_BUCKET_INSERT_BEFORE(b, b1);
}
}
}
rv = ap_pass_brigade(f->next, bbline);
for ( b = APR_BRIGADE_FIRST(ctx->bbsave);
b != APR_BRIGADE_SENTINEL(ctx->bbsave);
b = APR_BUCKET_NEXT(b)) {
apr_bucket_setaside(b, r->pool);
}
return rv;
}
