Newer
Older
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
/* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/*
* mod_deflate.c: Perform deflate content-encoding on the fly
*
* Written by Ian Holsman, Justin Erenkrantz, and Nick Kew
*/
/*
* Portions of this software are based upon zlib code by Jean-loup Gailly
* (zlib functions gz_open and gzwrite, check_header)
*/
/* zlib flags */
#define ASCII_FLAG 0x01 /* bit 0 set: file probably ascii text */
#define HEAD_CRC 0x02 /* bit 1 set: header CRC present */
#define EXTRA_FIELD 0x04 /* bit 2 set: extra field present */
#define ORIG_NAME 0x08 /* bit 3 set: original file name present */
#define COMMENT 0x10 /* bit 4 set: file comment present */
#define RESERVED 0xE0 /* bits 5..7: reserved */
#include "httpd.h"
#include "http_config.h"
#include "http_log.h"
#include "http_core.h"
#include "apr_lib.h"
#include "apr_strings.h"
#include "apr_general.h"
#include "util_filter.h"
#include "apr_buckets.h"
#include "http_request.h"
#define APR_WANT_STRFUNC
#include "apr_want.h"
#include "mod_ssl.h"
#include "zlib.h"
static const char deflateFilterName[] = "DEFLATE";
module AP_MODULE_DECLARE_DATA deflate_module;
#define AP_INFLATE_RATIO_LIMIT 200
#define AP_INFLATE_RATIO_BURST 3
typedef struct deflate_filter_config_t
{
int windowSize;
int memlevel;
int compressionlevel;
apr_size_t bufferSize;
const char *note_ratio_name;
const char *note_input_name;
const char *note_output_name;
} deflate_filter_config;
typedef struct deflate_dirconf_t {
apr_off_t inflate_limit;
int ratio_limit,
ratio_burst;
} deflate_dirconf_t;
/* RFC 1952 Section 2.3 defines the gzip header:
*
* +---+---+---+---+---+---+---+---+---+---+
* |ID1|ID2|CM |FLG| MTIME |XFL|OS |
* +---+---+---+---+---+---+---+---+---+---+
*/
static const char gzip_header[10] =
{ '\037', '\213', Z_DEFLATED, 0,
0, 0, 0, 0, /* mtime */
0, 0x03 /* Unix OS_CODE */
};
/* magic header */
static const char deflate_magic[2] = { '\037', '\213' };
/* windowsize is negative to suppress Zlib header */
#define DEFAULT_COMPRESSION Z_DEFAULT_COMPRESSION
#define DEFAULT_WINDOWSIZE -15
#define DEFAULT_MEMLEVEL 9
#define DEFAULT_BUFFERSIZE 8096
static APR_OPTIONAL_FN_TYPE(ssl_var_lookup) *mod_deflate_ssl_var = NULL;
/* Check whether a request is gzipped, so we can un-gzip it.
* If a request has multiple encodings, we need the gzip
* to be the outermost non-identity encoding.
*/
static int check_gzip(request_rec *r, apr_table_t *hdrs1, apr_table_t *hdrs2)
{
int found = 0;
apr_table_t *hdrs = hdrs1;
const char *encoding = apr_table_get(hdrs, "Content-Encoding");
if (!encoding && (hdrs2 != NULL)) {
/* the output filter has two tables and a content_encoding to check */
encoding = apr_table_get(hdrs2, "Content-Encoding");
hdrs = hdrs2;
if (!encoding) {
encoding = r->content_encoding;
hdrs = NULL;
}
}
if (encoding && *encoding) {
/* check the usual/simple case first */
if (!strcasecmp(encoding, "gzip")
|| !strcasecmp(encoding, "x-gzip")) {
found = 1;
if (hdrs) {
apr_table_unset(hdrs, "Content-Encoding");
}
else {
r->content_encoding = NULL;
}
}
else if (ap_strchr_c(encoding, ',') != NULL) {
/* If the outermost encoding isn't gzip, there's nothing
* we can do. So only check the last non-identity token
*/
char *new_encoding = apr_pstrdup(r->pool, encoding);
char *ptr;
for(;;) {
char *token = ap_strrchr(new_encoding, ',');
if (!token) { /* gzip:identity or other:identity */
if (!strcasecmp(new_encoding, "gzip")
|| !strcasecmp(new_encoding, "x-gzip")) {
found = 1;
if (hdrs) {
apr_table_unset(hdrs, "Content-Encoding");
}
else {
r->content_encoding = NULL;
}
}
break; /* seen all tokens */
}
for (ptr=token+1; apr_isspace(*ptr); ++ptr);
if (!strcasecmp(ptr, "gzip")
|| !strcasecmp(ptr, "x-gzip")) {
*token = '\0';
if (hdrs) {
apr_table_setn(hdrs, "Content-Encoding", new_encoding);
}
else {
r->content_encoding = new_encoding;
}
found = 1;
}
else if (!ptr[0] || !strcasecmp(ptr, "identity")) {
*token = '\0';
continue; /* strip the token and find the next one */
}
break; /* found a non-identity token */
}
}
}
/*
* If we have dealt with the headers above but content_encoding was set
* before sync it with the new value in the hdrs table as
* r->content_encoding takes precedence later on in the http_header_filter
* and hence would destroy what we have just set in the hdrs table.
*/
if (hdrs && r->content_encoding) {
r->content_encoding = apr_table_get(hdrs, "Content-Encoding");
}
return found;
}
/* Outputs a long in LSB order to the given file
* only the bottom 4 bits are required for the deflate file format.
*/
static void putLong(unsigned char *string, unsigned long x)
{
string[0] = (unsigned char)(x & 0xff);
string[1] = (unsigned char)((x & 0xff00) >> 8);
string[2] = (unsigned char)((x & 0xff0000) >> 16);
string[3] = (unsigned char)((x & 0xff000000) >> 24);
}
/* Inputs a string and returns a long.
*/
static unsigned long getLong(unsigned char *string)
{
return ((unsigned long)string[0])
| (((unsigned long)string[1]) << 8)
| (((unsigned long)string[2]) << 16)
| (((unsigned long)string[3]) << 24);
}
static void *create_deflate_server_config(apr_pool_t *p, server_rec *s)
{
deflate_filter_config *c = apr_pcalloc(p, sizeof *c);
c->memlevel = DEFAULT_MEMLEVEL;
c->windowSize = DEFAULT_WINDOWSIZE;
c->bufferSize = DEFAULT_BUFFERSIZE;
c->compressionlevel = DEFAULT_COMPRESSION;
return c;
}
static void *create_deflate_dirconf(apr_pool_t *p, char *dummy)
{
deflate_dirconf_t *dc = apr_pcalloc(p, sizeof(*dc));
dc->ratio_limit = AP_INFLATE_RATIO_LIMIT;
dc->ratio_burst = AP_INFLATE_RATIO_BURST;
return dc;
}
static const char *deflate_set_window_size(cmd_parms *cmd, void *dummy,
const char *arg)
{
deflate_filter_config *c = ap_get_module_config(cmd->server->module_config,
&deflate_module);
int i;
i = atoi(arg);
if (i < 1 || i > 15)
return "DeflateWindowSize must be between 1 and 15";
c->windowSize = i * -1;
return NULL;
}
static const char *deflate_set_buffer_size(cmd_parms *cmd, void *dummy,
const char *arg)
{
deflate_filter_config *c = ap_get_module_config(cmd->server->module_config,
&deflate_module);
int n = atoi(arg);
if (n <= 0) {
return "DeflateBufferSize should be positive";
}
c->bufferSize = (apr_size_t)n;
return NULL;
}
static const char *deflate_set_note(cmd_parms *cmd, void *dummy,
const char *arg1, const char *arg2)
{
deflate_filter_config *c = ap_get_module_config(cmd->server->module_config,
&deflate_module);
if (arg2 == NULL) {
c->note_ratio_name = arg1;
}
else if (!strcasecmp(arg1, "ratio")) {
c->note_ratio_name = arg2;
}
else if (!strcasecmp(arg1, "input")) {
c->note_input_name = arg2;
}
else if (!strcasecmp(arg1, "output")) {
c->note_output_name = arg2;
}
else {
return apr_psprintf(cmd->pool, "Unknown note type %s", arg1);
}
return NULL;
}
static const char *deflate_set_memlevel(cmd_parms *cmd, void *dummy,
const char *arg)
{
deflate_filter_config *c = ap_get_module_config(cmd->server->module_config,
&deflate_module);
int i;
i = atoi(arg);
if (i < 1 || i > 9)
return "DeflateMemLevel must be between 1 and 9";
c->memlevel = i;
return NULL;
}
static const char *deflate_set_compressionlevel(cmd_parms *cmd, void *dummy,
const char *arg)
{
deflate_filter_config *c = ap_get_module_config(cmd->server->module_config,
&deflate_module);
int i;
i = atoi(arg);
if (i < 1 || i > 9)
return "Compression Level must be between 1 and 9";
c->compressionlevel = i;
return NULL;
}
static const char *deflate_set_inflate_limit(cmd_parms *cmd, void *dirconf,
const char *arg)
{
deflate_dirconf_t *dc = (deflate_dirconf_t*) dirconf;
char *errp;
if (APR_SUCCESS != apr_strtoff(&dc->inflate_limit, arg, &errp, 10)) {
return "DeflateInflateLimitRequestBody is not parsable.";
}
if (*errp || dc->inflate_limit < 0) {
return "DeflateInflateLimitRequestBody requires a non-negative integer.";
}
return NULL;
}
static const char *deflate_set_inflate_ratio_limit(cmd_parms *cmd,
void *dirconf,
const char *arg)
{
deflate_dirconf_t *dc = (deflate_dirconf_t*) dirconf;
int i;
i = atoi(arg);
if (i <= 0)
return "DeflateInflateRatioLimit must be positive";
dc->ratio_limit = i;
return NULL;
}
static const char *deflate_set_inflate_ratio_burst(cmd_parms *cmd,
void *dirconf,
const char *arg)
{
deflate_dirconf_t *dc = (deflate_dirconf_t*) dirconf;
int i;
i = atoi(arg);
if (i <= 0)
return "DeflateInflateRatioBurst must be positive";
dc->ratio_burst = i;
return NULL;
}
typedef struct deflate_ctx_t
{
z_stream stream;
unsigned char *buffer;
unsigned long crc;
apr_bucket_brigade *bb, *proc_bb;
int (*libz_end_func)(z_streamp);
unsigned char *validation_buffer;
apr_size_t validation_buffer_length;
char header[10]; /* sizeof(gzip_header) */
apr_size_t header_len;
int zlib_flags;
int ratio_hits;
apr_off_t inflate_total;
unsigned int consume_pos,
consume_len;
unsigned int filter_init:1;
unsigned int done:1;
} deflate_ctx;
/* Number of validation bytes (CRC and length) after the compressed data */
#define VALIDATION_SIZE 8
/* Do not update ctx->crc, see comment in flush_libz_buffer */
#define NO_UPDATE_CRC 0
/* Do update ctx->crc, see comment in flush_libz_buffer */
#define UPDATE_CRC 1
static int flush_libz_buffer(deflate_ctx *ctx, deflate_filter_config *c,
struct apr_bucket_alloc_t *bucket_alloc,
int (*libz_func)(z_streamp, int), int flush,
int crc)
{
int zRC = Z_OK;
int done = 0;
unsigned int deflate_len;
apr_bucket *b;
for (;;) {
deflate_len = c->bufferSize - ctx->stream.avail_out;
if (deflate_len != 0) {
/*
* Do we need to update ctx->crc? Usually this is the case for
* inflate action where we need to do a crc on the output, whereas
* in the deflate case we need to do a crc on the input
*/
if (crc) {
ctx->crc = crc32(ctx->crc, (const Bytef *)ctx->buffer,
deflate_len);
}
b = apr_bucket_heap_create((char *)ctx->buffer,
deflate_len, NULL,
bucket_alloc);
APR_BRIGADE_INSERT_TAIL(ctx->bb, b);
ctx->stream.next_out = ctx->buffer;
ctx->stream.avail_out = c->bufferSize;
}
if (done)
break;
zRC = libz_func(&ctx->stream, flush);
/*
* We can ignore Z_BUF_ERROR because:
* When we call libz_func we can assume that
*
* - avail_in is zero (due to the surrounding code that calls
* flush_libz_buffer)
* - avail_out is non zero due to our actions some lines above
*
* So the only reason for Z_BUF_ERROR is that the internal libz
* buffers are now empty and thus we called libz_func one time
* too often. This does not hurt. It simply says that we are done.
*/
if (zRC == Z_BUF_ERROR) {
zRC = Z_OK;
break;
}
done = (ctx->stream.avail_out != 0 || zRC == Z_STREAM_END);
if (zRC != Z_OK && zRC != Z_STREAM_END)
break;
}
return zRC;
}
static apr_status_t deflate_ctx_cleanup(void *data)
{
deflate_ctx *ctx = (deflate_ctx *)data;
if (ctx)
ctx->libz_end_func(&ctx->stream);
return APR_SUCCESS;
}
/* ETag must be unique among the possible representations, so a change
* to content-encoding requires a corresponding change to the ETag.
* This routine appends -transform (e.g., -gzip) to the entity-tag
* value inside the double-quotes if an ETag has already been set
* and its value already contains double-quotes. PR 39727
*/
static void deflate_check_etag(request_rec *r, const char *transform)
{
const char *etag = apr_table_get(r->headers_out, "ETag");
apr_size_t etaglen;
if ((etag && ((etaglen = strlen(etag)) > 2))) {
if (etag[etaglen - 1] == '"') {
apr_size_t transformlen = strlen(transform);
char *newtag = apr_palloc(r->pool, etaglen + transformlen + 2);
char *d = newtag;
char *e = d + etaglen - 1;
const char *s = etag;
for (; d < e; ++d, ++s) {
*d = *s; /* copy etag to newtag up to last quote */
}
*d++ = '-'; /* append dash to newtag */
s = transform;
e = d + transformlen;
for (; d < e; ++d, ++s) {
*d = *s; /* copy transform to newtag */
}
*d++ = '"'; /* append quote to newtag */
*d = '\0'; /* null terminate newtag */
apr_table_setn(r->headers_out, "ETag", newtag);
}
}
}
/* Check whether the (inflate) ratio exceeds the configured limit/burst. */
static int check_ratio(request_rec *r, deflate_ctx *ctx,
const deflate_dirconf_t *dc)
{
if (ctx->stream.total_in) {
int ratio = ctx->stream.total_out / ctx->stream.total_in;
if (ratio < dc->ratio_limit) {
ctx->ratio_hits = 0;
}
else if (++ctx->ratio_hits > dc->ratio_burst) {
return 0;
}
}
return 1;
}
static int have_ssl_compression(request_rec *r)
{
const char *comp;
if (mod_deflate_ssl_var == NULL)
return 0;
comp = mod_deflate_ssl_var(r->pool, r->server, r->connection, r,
"SSL_COMPRESS_METHOD");
if (comp == NULL || *comp == '\0' || strcmp(comp, "NULL") == 0)
return 0;
return 1;
}
static apr_status_t deflate_out_filter(ap_filter_t *f,
apr_bucket_brigade *bb)
{
apr_bucket *e;
request_rec *r = f->r;
deflate_ctx *ctx = f->ctx;
int zRC;
apr_size_t len = 0, blen;
const char *data;
deflate_filter_config *c;
/* Do nothing if asked to filter nothing. */
if (APR_BRIGADE_EMPTY(bb)) {
return APR_SUCCESS;
}
c = ap_get_module_config(r->server->module_config,
&deflate_module);
/* If we don't have a context, we need to ensure that it is okay to send
* the deflated content. If we have a context, that means we've done
* this before and we liked it.
* This could be not so nice if we always fail. But, if we succeed,
* we're in better shape.
*/
if (!ctx) {
char *token;
const char *encoding;
if (have_ssl_compression(r)) {
ap_log_rerror(APLOG_MARK, APLOG_TRACE1, 0, r,
"Compression enabled at SSL level; not compressing "
"at HTTP level.");
ap_remove_output_filter(f);
return ap_pass_brigade(f->next, bb);
}
/* We have checked above that bb is not empty */
e = APR_BRIGADE_LAST(bb);
if (APR_BUCKET_IS_EOS(e)) {
/*
* If we already know the size of the response, we can skip
* compression on responses smaller than the compression overhead.
* However, if we compress, we must initialize deflate_out before
* calling ap_pass_brigade() for the first time. Otherwise the
* headers will be sent to the client without
* "Content-Encoding: gzip".
*/
e = APR_BRIGADE_FIRST(bb);
while (1) {
apr_status_t rc;
if (APR_BUCKET_IS_EOS(e)) {
ap_log_rerror(APLOG_MARK, APLOG_TRACE1, 0, r,
"Not compressing very small response of %"
APR_SIZE_T_FMT " bytes", len);
ap_remove_output_filter(f);
return ap_pass_brigade(f->next, bb);
}
if (APR_BUCKET_IS_METADATA(e)) {
e = APR_BUCKET_NEXT(e);
continue;
}
rc = apr_bucket_read(e, &data, &blen, APR_BLOCK_READ);
if (rc != APR_SUCCESS)
return rc;
len += blen;
/* 50 is for Content-Encoding and Vary headers and ETag suffix */
if (len > sizeof(gzip_header) + VALIDATION_SIZE + 50)
break;
e = APR_BUCKET_NEXT(e);
}
}
ctx = f->ctx = apr_pcalloc(r->pool, sizeof(*ctx));
/*
* Only work on main request, not subrequests,
* that are not a 204 response with no content
* and are not tagged with the no-gzip env variable
* and not a partial response to a Range request.
*/
if ((r->main != NULL) || (r->status == HTTP_NO_CONTENT) ||
apr_table_get(r->subprocess_env, "no-gzip") ||
apr_table_get(r->headers_out, "Content-Range")
) {
if (APLOG_R_IS_LEVEL(r, APLOG_TRACE1)) {
const char *reason =
(r->main != NULL) ? "subrequest" :
(r->status == HTTP_NO_CONTENT) ? "no content" :
apr_table_get(r->subprocess_env, "no-gzip") ? "no-gzip" :
"content-range";
ap_log_rerror(APLOG_MARK, APLOG_TRACE1, 0, r,
"Not compressing (%s)", reason);
}
ap_remove_output_filter(f);
return ap_pass_brigade(f->next, bb);
}
/* Some browsers might have problems with content types
* other than text/html, so set gzip-only-text/html
* (with browsermatch) for them
*/
if (r->content_type == NULL
|| strncmp(r->content_type, "text/html", 9)) {
const char *env_value = apr_table_get(r->subprocess_env,
"gzip-only-text/html");
if ( env_value && (strcmp(env_value,"1") == 0) ) {
ap_log_rerror(APLOG_MARK, APLOG_TRACE1, 0, r,
"Not compressing, (gzip-only-text/html)");
ap_remove_output_filter(f);
return ap_pass_brigade(f->next, bb);
}
}
/* Let's see what our current Content-Encoding is.
* If it's already encoded, don't compress again.
* (We could, but let's not.)
*/
encoding = apr_table_get(r->headers_out, "Content-Encoding");
if (encoding) {
const char *err_enc;
err_enc = apr_table_get(r->err_headers_out, "Content-Encoding");
if (err_enc) {
encoding = apr_pstrcat(r->pool, encoding, ",", err_enc, NULL);
}
}
else {
encoding = apr_table_get(r->err_headers_out, "Content-Encoding");
}
if (r->content_encoding) {
encoding = encoding ? apr_pstrcat(r->pool, encoding, ",",
r->content_encoding, NULL)
: r->content_encoding;
}
if (encoding) {
const char *tmp = encoding;
token = ap_get_token(r->pool, &tmp, 0);
while (token && *token) {
/* stolen from mod_negotiation: */
if (strcmp(token, "identity") && strcmp(token, "7bit") &&
strcmp(token, "8bit") && strcmp(token, "binary")) {
ap_log_rerror(APLOG_MARK, APLOG_TRACE1, 0, r,
"Not compressing (content-encoding already "
" set: %s)", token);
ap_remove_output_filter(f);
return ap_pass_brigade(f->next, bb);
}
/* Otherwise, skip token */
if (*tmp) {
++tmp;
}
token = (*tmp) ? ap_get_token(r->pool, &tmp, 0) : NULL;
}
}
/* Even if we don't accept this request based on it not having
* the Accept-Encoding, we need to note that we were looking
* for this header and downstream proxies should be aware of that.
*/
apr_table_mergen(r->headers_out, "Vary", "Accept-Encoding");
/* force-gzip will just force it out regardless if the browser
* can actually do anything with it.
*/
if (!apr_table_get(r->subprocess_env, "force-gzip")) {
const char *accepts;
/* if they don't have the line, then they can't play */
accepts = apr_table_get(r->headers_in, "Accept-Encoding");
if (accepts == NULL) {
ap_remove_output_filter(f);
return ap_pass_brigade(f->next, bb);
}
token = ap_get_token(r->pool, &accepts, 0);
while (token && token[0] && strcasecmp(token, "gzip")) {
/* skip parameters, XXX: ;q=foo evaluation? */
while (*accepts == ';') {
++accepts;
ap_get_token(r->pool, &accepts, 1);
}
/* retrieve next token */
if (*accepts == ',') {
++accepts;
}
token = (*accepts) ? ap_get_token(r->pool, &accepts, 0) : NULL;
}
/* No acceptable token found. */
if (token == NULL || token[0] == '\0') {
ap_log_rerror(APLOG_MARK, APLOG_TRACE1, 0, r,
"Not compressing (no Accept-Encoding: gzip)");
ap_remove_output_filter(f);
return ap_pass_brigade(f->next, bb);
}
}
else {
ap_log_rerror(APLOG_MARK, APLOG_TRACE1, 0, r,
"Forcing compression (force-gzip set)");
}
/* At this point we have decided to filter the content. Let's try to
* to initialize zlib (except for 304 responses, where we will only
* send out the headers).
*/
if (r->status != HTTP_NOT_MODIFIED) {
ctx->bb = apr_brigade_create(r->pool, f->c->bucket_alloc);
ctx->buffer = apr_palloc(r->pool, c->bufferSize);
ctx->libz_end_func = deflateEnd;
zRC = deflateInit2(&ctx->stream, c->compressionlevel, Z_DEFLATED,
c->windowSize, c->memlevel,
Z_DEFAULT_STRATEGY);
if (zRC != Z_OK) {
deflateEnd(&ctx->stream);
ap_log_rerror(APLOG_MARK, APLOG_ERR, 0, r, APLOGNO(01383)
"unable to init Zlib: "
"deflateInit2 returned %d: URL %s",
zRC, r->uri);
/*
* Remove ourselves as it does not make sense to return:
* We are not able to init libz and pass data down the chain
* uncompressed.
*/
ap_remove_output_filter(f);
return ap_pass_brigade(f->next, bb);
}
/*
* Register a cleanup function to ensure that we cleanup the internal
* libz resources.
*/
apr_pool_cleanup_register(r->pool, ctx, deflate_ctx_cleanup,
apr_pool_cleanup_null);
/* Set the filter init flag so subsequent invocations know we are
* active.
*/
ctx->filter_init = 1;
}
/*
* Zlib initialization worked, so we can now change the important
* content metadata before sending the response out.
*/
/* If the entire Content-Encoding is "identity", we can replace it. */
if (!encoding || !strcasecmp(encoding, "identity")) {
apr_table_setn(r->headers_out, "Content-Encoding", "gzip");
}
else {
apr_table_mergen(r->headers_out, "Content-Encoding", "gzip");
}
/* Fix r->content_encoding if it was set before */
if (r->content_encoding) {
r->content_encoding = apr_table_get(r->headers_out,
"Content-Encoding");
}
apr_table_unset(r->headers_out, "Content-Length");
apr_table_unset(r->headers_out, "Content-MD5");
deflate_check_etag(r, "gzip");
/* For a 304 response, only change the headers */
if (r->status == HTTP_NOT_MODIFIED) {
ap_remove_output_filter(f);
return ap_pass_brigade(f->next, bb);
}
/* add immortal gzip header */
e = apr_bucket_immortal_create(gzip_header, sizeof gzip_header,
f->c->bucket_alloc);
APR_BRIGADE_INSERT_TAIL(ctx->bb, e);
/* initialize deflate output buffer */
ctx->stream.next_out = ctx->buffer;
ctx->stream.avail_out = c->bufferSize;
} else if (!ctx->filter_init) {
/* Hmm. We've run through the filter init before as we have a ctx,
* but we never initialized. We probably have a dangling ref. Bail.
*/
return ap_pass_brigade(f->next, bb);
}
while (!APR_BRIGADE_EMPTY(bb))
{
apr_bucket *b;
/*
* Optimization: If we are a HEAD request and bytes_sent is not zero
* it means that we have passed the content-length filter once and
* have more data to sent. This means that the content-length filter
* could not determine our content-length for the response to the
* HEAD request anyway (the associated GET request would deliver the
* body in chunked encoding) and we can stop compressing.
*/
if (r->header_only && r->bytes_sent) {
ap_remove_output_filter(f);
return ap_pass_brigade(f->next, bb);
}
e = APR_BRIGADE_FIRST(bb);
if (APR_BUCKET_IS_EOS(e)) {
char *buf;
ctx->stream.avail_in = 0; /* should be zero already anyway */
/* flush the remaining data from the zlib buffers */
flush_libz_buffer(ctx, c, f->c->bucket_alloc, deflate, Z_FINISH,
NO_UPDATE_CRC);
buf = apr_palloc(r->pool, VALIDATION_SIZE);
putLong((unsigned char *)&buf[0], ctx->crc);
putLong((unsigned char *)&buf[4], ctx->stream.total_in);
b = apr_bucket_pool_create(buf, VALIDATION_SIZE, r->pool,
f->c->bucket_alloc);
APR_BRIGADE_INSERT_TAIL(ctx->bb, b);
ap_log_rerror(APLOG_MARK, APLOG_DEBUG, 0, r, APLOGNO(01384)
"Zlib: Compressed %ld to %ld : URL %s",
ctx->stream.total_in, ctx->stream.total_out, r->uri);
/* leave notes for logging */
if (c->note_input_name) {
apr_table_setn(r->notes, c->note_input_name,
(ctx->stream.total_in > 0)
? apr_off_t_toa(r->pool,
ctx->stream.total_in)
: "-");
}
if (c->note_output_name) {
apr_table_setn(r->notes, c->note_output_name,
(ctx->stream.total_in > 0)
? apr_off_t_toa(r->pool,
ctx->stream.total_out)
: "-");
}
if (c->note_ratio_name) {
apr_table_setn(r->notes, c->note_ratio_name,
(ctx->stream.total_in > 0)
? apr_itoa(r->pool,
(int)(ctx->stream.total_out
* 100
/ ctx->stream.total_in))
: "-");
}
deflateEnd(&ctx->stream);
/* No need for cleanup any longer */
apr_pool_cleanup_kill(r->pool, ctx, deflate_ctx_cleanup);
/* Remove EOS from the old list, and insert into the new. */
APR_BUCKET_REMOVE(e);
APR_BRIGADE_INSERT_TAIL(ctx->bb, e);
/* Okay, we've seen the EOS.
* Time to pass it along down the chain.
*/
return ap_pass_brigade(f->next, ctx->bb);
}
if (APR_BUCKET_IS_FLUSH(e)) {
apr_status_t rv;
/* flush the remaining data from the zlib buffers */
zRC = flush_libz_buffer(ctx, c, f->c->bucket_alloc, deflate,
Z_SYNC_FLUSH, NO_UPDATE_CRC);
if (zRC != Z_OK) {
ap_log_rerror(APLOG_MARK, APLOG_ERR, 0, r, APLOGNO(01385)
"Zlib error %d flushing zlib output buffer (%s)",
zRC, ctx->stream.msg);
return APR_EGENERAL;
}
/* Remove flush bucket from old brigade anf insert into the new. */
APR_BUCKET_REMOVE(e);
APR_BRIGADE_INSERT_TAIL(ctx->bb, e);
rv = ap_pass_brigade(f->next, ctx->bb);
if (rv != APR_SUCCESS) {
return rv;
}
continue;
}
if (APR_BUCKET_IS_METADATA(e)) {
/*
* Remove meta data bucket from old brigade and insert into the
* new.
*/
APR_BUCKET_REMOVE(e);
APR_BRIGADE_INSERT_TAIL(ctx->bb, e);
continue;
}
/* read */
apr_bucket_read(e, &data, &len, APR_BLOCK_READ);
if (!len) {
apr_bucket_delete(e);
continue;
}
if (len > APR_INT32_MAX) {
apr_bucket_split(e, APR_INT32_MAX);
apr_bucket_read(e, &data, &len, APR_BLOCK_READ);
}
/* This crc32 function is from zlib. */
ctx->crc = crc32(ctx->crc, (const Bytef *)data, len);
/* write */
ctx->stream.next_in = (unsigned char *)data; /* We just lost const-ness,
* but we'll just have to
* trust zlib */
ctx->stream.avail_in = len;
while (ctx->stream.avail_in != 0) {
if (ctx->stream.avail_out == 0) {
apr_status_t rv;
ctx->stream.next_out = ctx->buffer;
len = c->bufferSize - ctx->stream.avail_out;
b = apr_bucket_heap_create((char *)ctx->buffer, len,
NULL, f->c->bucket_alloc);
APR_BRIGADE_INSERT_TAIL(ctx->bb, b);
ctx->stream.avail_out = c->bufferSize;
/* Send what we have right now to the next filter. */
rv = ap_pass_brigade(f->next, ctx->bb);
if (rv != APR_SUCCESS) {
return rv;
}
}
zRC = deflate(&(ctx->stream), Z_NO_FLUSH);
if (zRC != Z_OK) {
ap_log_rerror(APLOG_MARK, APLOG_ERR, 0, r, APLOGNO(01386)
"Zlib error %d deflating data (%s)", zRC,
ctx->stream.msg);
return APR_EGENERAL;
}
}
apr_bucket_delete(e);
}
return APR_SUCCESS;
}
static apr_status_t consume_zlib_flags(deflate_ctx *ctx,
const char **data, apr_size_t *len)
{
if ((ctx->zlib_flags & EXTRA_FIELD)) {
/* Consume 2 bytes length prefixed data. */
if (ctx->consume_pos == 0) {
if (!*len) {
return APR_INCOMPLETE;
}
ctx->consume_len = (unsigned int)**data;
ctx->consume_pos++;
++*data;
--*len;
}
if (ctx->consume_pos == 1) {
if (!*len) {
return APR_INCOMPLETE;
}