-
Notifications
You must be signed in to change notification settings - Fork 70
/
fetchpost.rs
1393 lines (1257 loc) · 56.1 KB
/
fetchpost.rs
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
#![allow(unused_assignments)]
static USAGE: &str = r#"
Fetchpost fetches data from web services for every row using HTTP Post.
As opposed to fetch, which uses HTTP Get.
Fetchpost is integrated with `jaq` (a jq clone) to directly parse out values from an API JSON response.
(See https://github.com/01mf02/jaq for more info on how to use the jaq JSON Query Language)
CACHE OPTIONS:
Fetchpost caches responses to minimize traffic and maximize performance. It has four
mutually-exclusive caching options:
1. In memory cache (the default)
2. Disk cache
3. Redis cache
4. No cache
In memory Cache:
In memory cache is the default and is used if no caching option is set.
It uses a non-persistent, in-memory, 2 million entry Least Recently Used (LRU)
cache for each fetch session. To change the maximum number of entries in the cache,
set the --mem-cache-size option.
Disk Cache:
For persistent, inter-session caching, a DiskCache can be enabled with the --disk-cache flag.
By default, it will store the cache in the directory ~/.qsv/cache/fetchpost, with a cache expiry
Time-to-Live (TTL) of 2,419,200 seconds (28 days), and cache hits NOT refreshing the TTL
of cached values.
Set the --disk-cache-dir option and the environment variables QSV_DISKCACHE_TTL_SECS and
QSV_DISKCACHE_TTL_REFRESH to change default DiskCache settings.
Redis Cache:
Another persistent, inter-session cache option is a Redis cache enabled with the --redis flag.
By default, it will connect to a local Redis instance at redis://127.0.0.1:6379/2,
with a cache expiry Time-to-Live (TTL) of 2,419,200 seconds (28 days),
and cache hits NOT refreshing the TTL of cached values.
Set the environment variables QSV_FP_REDIS_CONNSTR, QSV_REDIS_TTL_SECONDS and
QSV_REDIS_TTL_REFRESH to change default Redis settings.
Note that the default values are the same as the fetch command, except fetchpost creates the
cache at database 2, as opposed to database 1 with fetch.
If you don't want responses to be cached at all, use the --no-cache flag.
NETWORK OPTIONS:
Fetchpost recognizes RateLimit and Retry-After headers and dynamically throttles requests
to be as fast as allowed. The --rate-limit option sets the maximum number of queries per second
(QPS) to be made. The default is 0, which means to go as fast as possible,
automatically throttling as required.
To use a proxy, please set env vars HTTP_PROXY, HTTPS_PROXY or ALL_PROXY
(e.g. export HTTPS_PROXY=socks5://127.0.0.1:1086).
qsv fetchpost supports brotli, gzip and deflate automatic decompression for improved throughput
and performance, preferring brotli over gzip over deflate.
Gzip compression of requests bodies is supported with the --compress flag. Note that
public APIs typically do not support gzip compression of request bodies because of the
"zip bomb" vulnerability. This option should only be used with private APIs where this
is not a concern.
It automatically upgrades its connection to the much faster and more efficient HTTP/2 protocol
with adaptive flow control if the server supports it.
See https://www.cloudflare.com/learning/performance/http2-vs-http1.1/ and
https://medium.com/coderscorner/http-2-flow-control-77e54f7fd518 for more info.
URL OPTIONS:
<url-column> needs to be a fully qualified URL path. It can be specified as a column name
from which the URL value will be retrieved for each record, or as the URL literal itself.
EXAMPLES:
data.csv
URL, zipcode, country
https://httpbin.org/post, 90210, USA
https://httpbin.org/post, 94105, USA
https://httpbin.org/post, 92802, USA
Given the data.csv above, fetch the JSON response.
$ qsv fetchpost URL zipcode,country data.csv
Note the output will be a JSONL file - with a minified JSON response per line, not a CSV file.
Now, if we want to generate a CSV file with a parsed response - getting only the "form" property,
we use the new-column and jaq options.
$ qsv fetchpost URL zipcode,country --new-column form --jaq '."form"' data.csv > data_with_response.csv
data_with_response.csv
URL,zipcode,country,form
https://httpbin.org/post,90210,USA,"{""country"": String(""USA""), ""zipcode"": String(""90210"")}"
https://httpbin.org/post,94105,USA,"{""country"": String(""USA""), ""zipcode"": String(""94105"")}"
https://httpbin.org/post,92802,USA,"{""country"": String(""USA""), ""zipcode"": String(""92802"")}"
Alternatively, since we're using the same URL for all the rows, we can just pass the url directly on the command-line.
$ qsv fetchpost https://httpbin.org/post 2,3 --new-column form --jaqfile form.jaq data.csv > data_with_formdata.csv
Also note that for the column-list argument, we used the column index (2,3 for second & third column)
instead of using the column names, and we loaded the jaq selector from the form.jaq file.
The form.jaq file simply contains the string literal ".form", including the enclosing double quotes:
form.jaq
".form"
USING THE HTTP-HEADER OPTION:
The --http-header option allows you to append arbitrary key value pairs (a valid pair is a key and value
separated by a colon) to the HTTP header (to authenticate against an API, pass custom header fields, etc.).
Note that you can pass as many key-value pairs by using --http-header option repeatedly. For example:
$ qsv fetchpost https://httpbin.org/post col1-col3 data.csv -H "X-Api-Key:TEST_KEY" -H "X-Api-Secret:ABC123XYZ"
For more extensive examples, see https://github.com/jqnatividad/qsv/blob/master/tests/test_fetch.rs.
Usage:
qsv fetchpost (<url-column> <column-list>) [--jaq <selector> | --jaqfile <file>] [--http-header <k:v>...] [options] [<input>]
qsv fetchpost --help
Fetchpost options:
<url-column> Name of the column with the URL.
Otherwise, if the argument starts with `http`, the URL to use.
<column-list> Comma-delimited list of columns to insert into the HTTP Post body.
Uses `qsv select` syntax - i.e. Columns can be referenced by index or
by name if there is a header row (duplicate column names can be disambiguated
with more indexing). Column ranges can also be specified. Finally, columns
can be selected using regular expressions.
See 'qsv select --help' for examples.
-c, --new-column <name> Put the fetched values in a new column. Specifying this option
results in a CSV. Otherwise, the output is in JSONL format.
--jaq <selector> Apply jaq selector to API returned JSON response.
Mutually exclusive with --jaqfile.
--jaqfile <file> Load jaq selector from file instead.
Mutually exclusive with --jaq.
--pretty Prettify JSON responses. Otherwise, they're minified.
If the response is not in JSON format, it's passed through unchanged.
Note that --pretty requires the --new-column option.
--rate-limit <qps> Rate Limit in Queries Per Second (max: 1000). Note that fetch
dynamically throttles as well based on rate-limit and
retry-after response headers.
Set to 0 to go as fast as possible, automatically throttling as required.
CAUTION: Only use zero for APIs that use RateLimit and/or Retry-After headers,
otherwise your fetchpost job may look like a Denial Of Service attack.
Even though zero is the default, this is mitigated by --max-errors having a
default of 10.
[default: 0 ]
--timeout <seconds> Timeout for each URL request.
[default: 30 ]
-H, --http-header <k:v> Append custom header(s) to the HTTP header. Pass multiple key-value pairs
by adding this option multiple times, once for each pair. The key and value
should be separated by a colon.
--compress Compress the HTTP request body using gzip. Note that most servers do not support
compressed request bodies unless they are specifically configured to do so. This
should only be enabled for trusted scenarios where "zip bombs" are not a concern.
see https://github.com/postmanlabs/httpbin/issues/577#issuecomment-875814469
for more info.
--max-retries <count> Maximum number of retries per record before an error is raised.
[default: 5]
--max-errors <count> Maximum number of errors before aborting.
Set to zero (0) to continue despite errors.
[default: 10 ]
--store-error On error, store error code/message instead of blank value.
--cookies Allow cookies.
--user-agent <agent> Specify custom user agent. It supports the following variables -
$QSV_VERSION, $QSV_TARGET, $QSV_BIN_NAME, $QSV_KIND and $QSV_COMMAND.
Try to follow the syntax here -
https://developer.mozilla.org/en-US/docs/Web/HTTP/Headers/User-Agent
--report <d|s> Creates a report of the fetchpost job. The report has the same name as the
input file with the ".fetchpost-report" suffix.
There are two kinds of report - d for "detailed" & s for "short". The detailed
report has the same columns as the input CSV with seven additional columns -
qsv_fetchp_url, qsv_fetchp_form, qsv_fetchp_status, qsv_fetchp_cache_hit,
qsv_fetchp_retries, qsv_fetchp_elapsed_ms & qsv_fetchp_response.
The short report only has the seven columns without the "qsv_fetchp_" prefix.
[default: none]
CACHING OPTIONS:
--no-cache Do not cache responses.
--mem-cache-size <count> Maximum number of entries in the in-memory LRU cache.
[default: 2000000]
--disk-cache Use a persistent disk cache for responses. The cache is stored in the directory
specified by --disk-cache-dir. If the directory does not exist, it will be
created. If the directory exists, it will be used as is.
It has a default Time To Live (TTL)/lifespan of 28 days and cache hits do not
refresh the TTL of cached values.
Adjust the QSV_DISKCACHE_TTL_SECS & QSV_DISKCACHE_TTL_REFRESH env vars
to change DiskCache settings.
--disk-cache-dir <dir> The directory <dir> to store the disk cache. Note that if the directory
does not exist, it will be created. If the directory exists, it will be used as is,
and will not be flushed. This option allows you to maintain several disk caches
for different fetchpost jobs (e.g. one for geocoding, another for weather, etc.)
[default: ~/.qsv/cache/fetchpost]
--redis-cache Use Redis to cache responses. It connects to "redis://127.0.0.1:6379/2"
with a connection pool size of 20, with a TTL of 28 days, and a cache hit
NOT renewing an entry's TTL.
Adjust the QSV_FP_REDIS_CONNSTR, QSV_REDIS_MAX_POOL_SIZE, QSV_REDIS_TTL_SECONDS &
QSV_REDIS_TTL_REFRESH respectively to change Redis settings.
--cache-error Cache error responses even if a request fails. If an identical URL is requested,
the cached error is returned. Otherwise, the fetch is attempted again
for --max-retries.
--flush-cache Flush all the keys in the current cache on startup. This only applies to
Disk and Redis caches.
Common options:
-h, --help Display this message
-o, --output <file> Write output to <file> instead of stdout.
-n, --no-headers When set, the first row will not be interpreted
as headers. Namely, it will be sorted with the rest
of the rows. Otherwise, the first row will always
appear as the header row in the output.
-d, --delimiter <arg> The field delimiter for reading CSV data.
Must be a single character. (default: ,)
-p, --progressbar Show progress bars. Will also show the cache hit rate upon completion.
Not valid for stdin.
"#;
use std::{fs, io::Write, num::NonZeroU32, sync::OnceLock, thread, time};
use cached::{
proc_macro::{cached, io_cached},
stores::DiskCacheBuilder,
Cached, IOCached, RedisCache, Return, SizedCache,
};
use flate2::{write::GzEncoder, Compression};
use governor::{
clock::DefaultClock,
middleware::NoOpMiddleware,
state::{direct::NotKeyed, InMemoryState},
Quota, RateLimiter,
};
use indicatif::{HumanCount, MultiProgress, ProgressBar, ProgressDrawTarget};
use log::{
debug, error, info, log_enabled, warn,
Level::{Debug, Trace, Warn},
};
use rand::Rng;
use regex::Regex;
use reqwest::{
blocking::Client,
header::{HeaderMap, HeaderName, HeaderValue},
};
use serde::Deserialize;
use serde_json::{json, Value};
use simdutf8::basic::from_utf8;
use simple_expand_tilde::expand_tilde;
use url::Url;
use crate::{
cmd::fetch::{
get_ratelimit_header_value, parse_ratelimit_header_value, process_jaq, CacheType,
DiskCacheConfig, FetchResponse, RedisConfig, ReportKind, DEFAULT_ACCEPT_ENCODING,
},
config::{Config, Delimiter},
select::SelectColumns,
util, CliError, CliResult,
};
#[derive(Deserialize)]
struct Args {
flag_new_column: Option<String>,
flag_jaq: Option<String>,
flag_jaqfile: Option<String>,
flag_pretty: bool,
flag_rate_limit: u32,
flag_timeout: u16,
flag_http_header: Vec<String>,
flag_compress: bool,
flag_max_retries: u8,
flag_max_errors: u64,
flag_store_error: bool,
flag_cookies: bool,
flag_user_agent: Option<String>,
flag_report: String,
flag_no_cache: bool,
flag_mem_cache_size: usize,
flag_disk_cache: bool,
flag_disk_cache_dir: Option<String>,
flag_redis_cache: bool,
flag_cache_error: bool,
flag_flush_cache: bool,
flag_output: Option<String>,
flag_no_headers: bool,
flag_delimiter: Option<Delimiter>,
flag_progressbar: bool,
arg_url_column: SelectColumns,
arg_column_list: SelectColumns,
arg_input: Option<String>,
}
// set memcache size - the default is 2 million entries
// and is set through the docopt usage text
static MEM_CACHE_SIZE: OnceLock<usize> = OnceLock::new();
static DEFAULT_REDIS_CONN_STRING: OnceLock<String> = OnceLock::new();
static TIMEOUT_FP_SECS: OnceLock<u64> = OnceLock::new();
const FETCHPOST_REPORT_PREFIX: &str = "qsv_fetchp_";
const FETCHPOST_REPORT_SUFFIX: &str = ".fetchpost-report.tsv";
// for governor/ratelimiter
const MINIMUM_WAIT_MS: u64 = 10;
const MIN_WAIT: time::Duration = time::Duration::from_millis(MINIMUM_WAIT_MS);
static DISKCACHE_DIR: OnceLock<String> = OnceLock::new();
static REDISCONFIG: OnceLock<RedisConfig> = OnceLock::new();
static DISKCACHECONFIG: OnceLock<DiskCacheConfig> = OnceLock::new();
pub fn run(argv: &[&str]) -> CliResult<()> {
let args: Args = util::get_args(USAGE, argv)?;
// connect to Redis at localhost, using database 2 by default when --redis-cache is enabled
// fetchpost uses database 2 by default, as opposed to database 1 with fetch
let fp_redis_conn_str = std::env::var("QSV_FP_REDIS_CONNSTR")
.unwrap_or_else(|_| "redis://127.0.0.1:6379/2".to_string());
DEFAULT_REDIS_CONN_STRING.set(fp_redis_conn_str).unwrap();
// set memcache size
MEM_CACHE_SIZE.set(args.flag_mem_cache_size).unwrap();
TIMEOUT_FP_SECS
.set(util::timeout_secs(args.flag_timeout)?)
.unwrap();
// setup diskcache dir response caching
let diskcache_dir = if let Some(dir) = &args.flag_disk_cache_dir {
if dir.starts_with('~') {
// expand the tilde
let expanded_dir = expand_tilde(dir).unwrap();
expanded_dir.to_string_lossy().to_string()
} else {
dir.to_string()
}
} else {
String::new()
};
let cache_type = if args.flag_no_cache {
CacheType::None
} else if args.flag_disk_cache {
// if --flush-cache is set, flush the cache directory first if it exists
if args.flag_flush_cache
&& !diskcache_dir.is_empty()
&& fs::metadata(&diskcache_dir).is_ok()
{
if let Err(e) = fs::remove_dir_all(&diskcache_dir) {
return fail_clierror!(r#"Cannot remove cache directory "{diskcache_dir}": {e:?}"#);
}
info!("flushed DiskCache directory: {diskcache_dir}");
}
// check if the cache directory exists, if it doesn't, create it
if !diskcache_dir.is_empty() {
if let Err(e) = fs::create_dir_all(&diskcache_dir) {
return fail_clierror!(r#"Cannot create cache directory "{diskcache_dir}": {e:?}"#);
}
}
DISKCACHE_DIR.set(diskcache_dir).unwrap();
// initialize DiskCache Config
DISKCACHECONFIG.set(DiskCacheConfig::new()).unwrap();
CacheType::Disk
} else if args.flag_redis_cache {
// initialize Redis Config
REDISCONFIG.set(RedisConfig::new()).unwrap();
// check if redis connection is valid
let conn_str = &REDISCONFIG.get().unwrap().conn_str;
let redis_client = match redis::Client::open(conn_str.to_string()) {
Ok(rc) => rc,
Err(e) => {
return fail_incorrectusage_clierror!(
r#"Invalid Redis connection string "{conn_str}": {e:?}"#
)
},
};
let mut redis_conn;
match redis_client.get_connection() {
Err(e) => {
return fail_clierror!(r#"Cannot connect to Redis using "{conn_str}": {e:?}"#)
},
Ok(x) => redis_conn = x,
}
if args.flag_flush_cache {
redis::cmd("FLUSHDB")
.exec(&mut redis_conn)
.map_err(|_| "Cannot flush Redis cache")?;
info!("flushed Redis database.");
}
CacheType::Redis
} else {
CacheType::InMemory
};
log::info!("Cache Type: {cache_type:?}");
let mut rconfig = Config::new(args.arg_input.as_ref())
.delimiter(args.flag_delimiter)
.trim(csv::Trim::All)
.no_headers(args.flag_no_headers);
let mut rdr = rconfig.reader()?;
let mut wtr = if args.flag_new_column.is_some() {
// when adding a new column for the response, the output
// is a regular CSV file
Config::new(args.flag_output.as_ref()).writer()?
} else {
// otherwise, the output is a JSONL file. So we need to configure
// the csv writer so it doesn't double double quote the JSON response
// and its flexible (i.e. "column counts are different row to row")
Config::new(args.flag_output.as_ref())
.quote_style(csv::QuoteStyle::Never)
.flexible(true)
.writer()?
};
let mut headers = rdr.byte_headers()?.clone();
let include_existing_columns = if let Some(name) = args.flag_new_column {
// write header with new column
headers.push_field(name.as_bytes());
wtr.write_byte_record(&headers)?;
true
} else {
if args.flag_pretty {
return fail_incorrectusage_clierror!(
"The --pretty option requires the --new-column option."
);
}
false
};
// validate column-list is a list of valid column names
let cl_config = Config::new(args.arg_input.as_ref())
.delimiter(args.flag_delimiter)
.trim(csv::Trim::All)
.no_headers(args.flag_no_headers)
.select(args.arg_column_list.clone());
let col_list = cl_config.selection(&headers)?;
debug!("column-list: {col_list:?}");
// check if the url_column arg was passed as a URL literal
// or as a column selector
let url_column_str = format!("{:?}", args.arg_url_column);
let re = Regex::new(r"^IndexedName\((.*)\[0\]\)$").unwrap();
let literal_url = if let Some(caps) = re.captures(&url_column_str) {
caps[1].to_lowercase()
} else {
String::new()
};
let literal_url_used = literal_url.starts_with("http");
let mut column_index = 0;
if !literal_url_used {
rconfig = rconfig.select(args.arg_url_column);
let sel = rconfig.selection(&headers)?;
column_index = *sel.iter().next().unwrap();
if sel.len() != 1 {
return fail!("Only a single URL column may be selected.");
}
}
let rate_limit = match args.flag_rate_limit {
0 => NonZeroU32::new(u32::MAX).unwrap(),
1..=1000 => NonZeroU32::new(args.flag_rate_limit).unwrap(),
_ => {
return fail_incorrectusage_clierror!(
"Rate Limit should be between 0 to 1000 queries per second."
)
},
};
info!("RATE LIMIT: {rate_limit}");
let http_headers: HeaderMap = {
let mut map = HeaderMap::with_capacity(args.flag_http_header.len() + 1);
for header in args.flag_http_header {
let vals: Vec<&str> = header.split(':').collect();
if vals.len() != 2 {
return fail_incorrectusage_clierror!(
"{vals:?} is not a valid key-value pair. Expecting a key and a value \
separated by a colon."
);
}
// allocate new String for header key to put into map
let k: String = String::from(vals[0].trim());
let header_name: HeaderName =
match HeaderName::from_lowercase(k.to_lowercase().as_bytes()) {
Ok(h) => h,
Err(e) => return fail_incorrectusage_clierror!("Invalid header name: {e}"),
};
// allocate new String for header value to put into map
let v: String = String::from(vals[1].trim());
let header_val: HeaderValue = match HeaderValue::from_str(v.as_str()) {
Ok(v) => v,
Err(e) => return fail_incorrectusage_clierror!("Invalid header value: {e}"),
};
map.append(header_name, header_val);
}
map.append(
reqwest::header::ACCEPT_ENCODING,
HeaderValue::from_str(DEFAULT_ACCEPT_ENCODING).unwrap(),
);
map.append(
reqwest::header::CONTENT_TYPE,
HeaderValue::from_str("application/x-www-form-urlencoded").unwrap(),
);
if args.flag_compress {
map.append(
reqwest::header::CONTENT_ENCODING,
HeaderValue::from_str("gzip").unwrap(),
);
}
map
};
debug!("HTTP Header: {http_headers:?}");
let client_timeout = time::Duration::from_secs(*TIMEOUT_FP_SECS.get().unwrap_or(&30));
let client = Client::builder()
.user_agent(util::set_user_agent(args.flag_user_agent)?)
.default_headers(http_headers)
.cookie_store(args.flag_cookies)
.brotli(true)
.gzip(true)
.deflate(true)
.zstd(true)
.use_rustls_tls()
.http2_adaptive_window(true)
.connection_verbose(log_enabled!(Debug) || log_enabled!(Trace))
.timeout(client_timeout)
.build()?;
// set rate limiter with allow_burst set to 1 - see https://github.com/antifuchs/governor/issues/39
let limiter =
RateLimiter::direct(Quota::per_second(rate_limit).allow_burst(NonZeroU32::new(1).unwrap()));
// prep progress bars
let show_progress =
(args.flag_progressbar || util::get_envvar_flag("QSV_PROGRESSBAR")) && !rconfig.is_stdin();
// create multi_progress to stderr with a maximum refresh of 5 per second
let multi_progress = MultiProgress::with_draw_target(ProgressDrawTarget::stderr_with_hz(5));
let progress = multi_progress.add(ProgressBar::new(0));
let mut record_count = 0;
let error_progress = multi_progress.add(ProgressBar::new(args.flag_max_errors));
if args.flag_max_errors > 0 && show_progress {
console::set_colors_enabled(true); // as error progress bar is red
error_progress.set_style(
indicatif::ProgressStyle::default_bar()
.template("{bar:37.red/white} {percent}%{msg} ({per_sec:7})")
.unwrap(),
);
error_progress.set_message(format!(
" of {} max errors",
HumanCount(args.flag_max_errors)
));
} else {
error_progress.set_draw_target(ProgressDrawTarget::hidden());
}
if show_progress {
record_count = util::count_rows(&rconfig)?;
util::prep_progress(&progress, record_count);
} else {
multi_progress.set_draw_target(ProgressDrawTarget::hidden());
}
let jaq_selector: Option<String> = match args.flag_jaqfile {
Some(ref jaq_file) => Some(fs::read_to_string(jaq_file)?),
None => args.flag_jaq.as_ref().map(std::string::ToString::to_string),
};
// prepare report
let report = if args.flag_report.to_lowercase().starts_with('d') {
// if it starts with d, its a detailed report
ReportKind::Detailed
} else if args.flag_report.to_lowercase().starts_with('s') {
// if it starts with s, its a short report
ReportKind::Short
} else {
ReportKind::None
};
let mut report_wtr;
let report_path;
if report == ReportKind::None {
// no report, point report_wtr to /dev/null (AKA sink)
report_wtr = Config::new(Some("sink".to_string()).as_ref()).writer()?;
report_path = String::new();
} else {
report_path = args
.arg_input
.clone()
.unwrap_or_else(|| "stdin.csv".to_string());
report_wtr =
Config::new(Some(report_path.clone() + FETCHPOST_REPORT_SUFFIX).as_ref()).writer()?;
let mut report_headers = if report == ReportKind::Detailed {
headers.clone()
} else {
csv::ByteRecord::new()
};
let rptcol_prefix = if report == ReportKind::Detailed {
FETCHPOST_REPORT_PREFIX
} else {
""
};
// the fetchpost report has the following columns:
// url - URL used, form - form data sent, status - HTTP status code,
// cache_hit - cache hit flag, retries - retry attempts,
// elapsed - elapsed time (milliseconds) & response.
let report_header_fields = vec![
format!("{rptcol_prefix}url"),
format!("{rptcol_prefix}status"),
format!("{rptcol_prefix}cache_hit"),
format!("{rptcol_prefix}retries"),
format!("{rptcol_prefix}elapsed_ms"),
format!("{rptcol_prefix}response"),
];
report_headers = csv::ByteRecord::from(report_header_fields);
report_wtr.write_byte_record(&report_headers)?;
}
// amortize memory allocations
// why optimize for mem & speed, when we're just doing single-threaded, throttled URL fetches?
// we still optimize since fetch is backed by a memoized cache (in memory or Redis, when --redis
// is used), so we want to return responses as fast as possible as we bypass the network
// request with a cache hit
let mut record = csv::ByteRecord::new();
let mut jsonl_record = csv::ByteRecord::new();
let mut report_record = csv::ByteRecord::new();
let mut url = String::with_capacity(100);
let mut redis_cache_hits: u64 = 0;
let mut disk_cache_hits: u64 = 0;
let mut intermediate_redis_value: Return<String> = Return {
was_cached: false,
value: String::new(),
};
let mut intermediate_value: Return<FetchResponse> = Return {
was_cached: false,
value: FetchResponse {
response: String::new(),
status_code: 0_u16,
retries: 0_u8,
},
};
let mut final_value = String::with_capacity(150);
let mut final_response = FetchResponse {
response: String::new(),
status_code: 0_u16,
retries: 0_u8,
};
let empty_response = FetchResponse {
response: String::new(),
status_code: 0_u16,
retries: 0_u8,
};
let mut running_error_count = 0_u64;
let mut running_success_count = 0_u64;
let mut was_cached;
let mut now = time::Instant::now();
let mut form_body_jsonmap = serde_json::map::Map::with_capacity(col_list.len());
let header_key_vec: Vec<String> = headers
.iter()
.map(|x| String::from_utf8_lossy(x).to_string())
.collect();
let debug_flag = log_enabled!(Debug);
while rdr.read_byte_record(&mut record)? {
if show_progress {
progress.inc(1);
}
if report != ReportKind::None {
now = time::Instant::now();
};
// construct body per the column-list
form_body_jsonmap.clear();
for col_idx in &*col_list {
form_body_jsonmap.insert(
(header_key_vec[*col_idx]).to_string(),
serde_json::Value::String(
from_utf8(&record[*col_idx]).unwrap_or_default().to_owned(),
),
);
}
if debug_flag {
// deserializing the form_body_jsonmap to a string is expensive
// so we only do it when debug is enabled
debug!("{form_body_jsonmap:?}");
}
if literal_url_used {
url.clone_from(&literal_url);
} else if let Ok(s) = from_utf8(&record[column_index]) {
s.clone_into(&mut url);
} else {
url = String::new();
}
if url.is_empty() {
final_response.clone_from(&empty_response);
was_cached = false;
} else {
match cache_type {
CacheType::InMemory => {
intermediate_value = get_cached_response(
&url,
&form_body_jsonmap,
&client,
&limiter,
jaq_selector.as_ref(),
args.flag_store_error,
args.flag_pretty,
args.flag_compress,
include_existing_columns,
args.flag_max_retries,
);
final_response = intermediate_value.value;
was_cached = intermediate_value.was_cached;
if !args.flag_cache_error && final_response.status_code != 200 {
let mut cache = GET_CACHED_RESPONSE.lock().unwrap();
cache.cache_remove(&url);
}
},
CacheType::Disk => {
intermediate_value = get_diskcache_response(
&url,
&form_body_jsonmap,
&client,
&limiter,
jaq_selector.as_ref(),
args.flag_store_error,
args.flag_pretty,
args.flag_compress,
include_existing_columns,
args.flag_max_retries,
)?;
final_response = intermediate_value.value;
was_cached = intermediate_value.was_cached;
if was_cached {
disk_cache_hits += 1;
// log::debug!("Disk cache hit for {url} hit: {disk_cache_hits}");
}
if !args.flag_cache_error && final_response.status_code != 200 {
let _ = GET_DISKCACHE_RESPONSE.cache_remove(&url);
// log::debug!("Removed Disk cache for {url}");
}
},
CacheType::Redis => {
intermediate_redis_value = get_redis_response(
&url,
&form_body_jsonmap,
&client,
&limiter,
jaq_selector.as_ref(),
args.flag_store_error,
args.flag_pretty,
args.flag_compress,
include_existing_columns,
args.flag_max_retries,
)?;
was_cached = intermediate_redis_value.was_cached;
if was_cached {
redis_cache_hits += 1;
}
final_response = match serde_json::from_str(&intermediate_redis_value) {
Ok(r) => r,
Err(e) => {
return fail_clierror!(
"Cannot deserialize Redis cache value. Try flushing the Redis \
cache with --flushdb: {e}"
)
},
};
if !args.flag_cache_error && final_response.status_code != 200 {
let key = format!(
"{}{:?}{}{}{}",
url,
jaq_selector,
args.flag_store_error,
args.flag_pretty,
include_existing_columns
);
if GET_REDIS_RESPONSE.cache_remove(&key).is_err() && log_enabled!(Warn) {
// failure to remove cache keys is non-fatal. Continue, but log it.
wwarn!(r#"Cannot remove Redis key "{key}""#);
};
}
},
CacheType::None => {
final_response = get_response(
&url,
&form_body_jsonmap,
&client,
&limiter,
jaq_selector.as_ref(),
args.flag_store_error,
args.flag_pretty,
args.flag_compress,
include_existing_columns,
args.flag_max_retries,
);
was_cached = false;
},
}
};
if final_response.status_code == 200 {
running_success_count += 1;
} else {
running_error_count += 1;
error_progress.inc(1);
}
final_value.clone_from(&final_response.response);
if include_existing_columns {
record.push_field(final_value.as_bytes());
wtr.write_byte_record(&record)?;
} else {
jsonl_record.clear();
if final_value.is_empty() {
jsonl_record.push_field(b"{}");
} else {
jsonl_record.push_field(final_value.as_bytes());
}
wtr.write_byte_record(&jsonl_record)?;
}
if report != ReportKind::None {
if report == ReportKind::Detailed {
report_record.clone_from(&record);
} else {
report_record.clear();
}
report_record.push_field(url.as_bytes());
report_record.push_field(format!("{form_body_jsonmap:?}").as_bytes());
report_record.push_field(final_response.status_code.to_string().as_bytes());
report_record.push_field(if was_cached { b"1" } else { b"0" });
report_record.push_field(final_response.retries.to_string().as_bytes());
report_record.push_field(now.elapsed().as_millis().to_string().as_bytes());
if include_existing_columns {
report_record.push_field(final_value.as_bytes());
} else {
report_record.push_field(jsonl_record.as_slice());
}
report_wtr.write_byte_record(&report_record)?;
}
if args.flag_max_errors > 0 && running_error_count >= args.flag_max_errors {
break;
}
}
report_wtr.flush()?;
if show_progress {
match cache_type {
CacheType::InMemory => {
util::update_cache_info!(progress, GET_CACHED_RESPONSE);
},
CacheType::Disk => {
util::update_cache_info!(progress, disk_cache_hits, record_count);
},
CacheType::Redis => {
util::update_cache_info!(progress, redis_cache_hits, record_count);
},
CacheType::None => (),
}
util::finish_progress(&progress);
if running_error_count == 0 {
error_progress.finish_and_clear();
} else if running_error_count >= args.flag_max_errors {
error_progress.finish();
// sleep so we can dependably write eprintln without messing up progress bars
thread::sleep(time::Duration::from_nanos(10));
let abort_msg = format!(
"{} max errors. Fetchpost aborted.",
HumanCount(args.flag_max_errors)
);
winfo!("{abort_msg}");
} else {
error_progress.abandon();
}
}
let mut end_msg = format!(
"{} records successfully fetchposted as {}. {} errors.",
HumanCount(running_success_count),
if include_existing_columns {
"CSV"
} else {
"JSONL"
},
HumanCount(running_error_count)
);
if report != ReportKind::None {
use std::fmt::Write;
write!(
&mut end_msg,
" {} report created: \"{}{FETCHPOST_REPORT_SUFFIX}\"",
if report == ReportKind::Detailed {
"Detailed"
} else {
"Short"
},
report_path
)
.unwrap();
}
winfo!("{end_msg}");
// if using a Diskcache, explicitly flush it
// to ensure all entries are written to disk
if cache_type == CacheType::Disk {
GET_DISKCACHE_RESPONSE
.connection()
.flush()
.map_err(|e| CliError::Other(format!("Error flushing DiskCache: {e}")))?;
}
Ok(wtr.flush()?)
}
// we only need url in the cache key
// as this is an in-memory cache that is only used for one qsv session
#[cached(
ty = "SizedCache<String, Return<FetchResponse>>",
create = r##"{
let cache_size = MEM_CACHE_SIZE.get().unwrap();
let memcache = SizedCache::with_size(*cache_size);
log::info!("In Memory cache created - size: {cache_size} entries");
memcache
}"##,
key = "String",
convert = r#"{ format!("{:?}", form_body_jsonmap) }"#,
with_cached_flag = true
)]
fn get_cached_response(
url: &str,
form_body_jsonmap: &serde_json::Map<String, Value>,
client: &reqwest::blocking::Client,
limiter: &governor::RateLimiter<NotKeyed, InMemoryState, DefaultClock, NoOpMiddleware>,
flag_jaq: Option<&String>,
flag_store_error: bool,
flag_pretty: bool,
flag_compress: bool,
include_existing_columns: bool,
flag_max_retries: u8,
) -> cached::Return<FetchResponse> {
Return::new(get_response(
url,
form_body_jsonmap,
client,
limiter,
flag_jaq,
flag_store_error,
flag_pretty,
flag_compress,
include_existing_columns,
flag_max_retries,
))
}
// this is a disk cache that can be used across qsv sessions
// so we need to include the values of flag_jaq, flag_store_error, flag_pretty and
// include_existing_columns in the cache key
#[io_cached(
disk = true,
ty = "cached::DiskCache<String, FetchResponse>",
cache_prefix_block = r##"{ "dc_" }"##,
key = "String",
convert = r#"{ format!("{}{:?}{:?}{}{}{}{}", url, form_body_jsonmap, flag_jaq, flag_store_error, flag_pretty, flag_compress, include_existing_columns) }"#,
create = r##"{
let cache_dir = DISKCACHE_DIR.get().unwrap();
let diskcache_config = DISKCACHECONFIG.get().unwrap();
let diskcache = DiskCacheBuilder::new("fetchpost")
.set_disk_directory(cache_dir)
.set_lifespan(diskcache_config.ttl_secs)
.set_refresh(diskcache_config.ttl_refresh)