Commit | Line | Data |
---|---|---|
cd541405 | 1 | #lang typed/racket/no-check |
4764ff89 | 2 | |
1d753430 | 3 | (require openssl/sha1) |
4764ff89 | 4 | (require racket/date) |
9c464d95 | 5 | (require |
d718efc4 | 6 | net/head |
edadb804 | 7 | net/uri-codec |
8da029e4 | 8 | net/url) |
4764ff89 | 9 | |
3c9c8266 SK |
10 | (require (prefix-in info: "info.rkt")) |
11 | ||
78398948 | 12 | (module+ test |
de3ff448 | 13 | (require rackunit)) |
78398948 | 14 | |
98529d3d SK |
15 | (define-type Url |
16 | net/url-structs:url) | |
17 | ||
18 | (define-type Out-Format | |
19 | (U 'single-line | |
20 | 'multi-line)) | |
21 | ||
22 | (define-type Timeline-Order | |
23 | (U 'old->new | |
24 | 'new->old)) | |
25 | ||
b0ff061a | 26 | (struct Msg |
78142acb | 27 | ([ts-epoch : Integer] |
3877a0c4 | 28 | [ts-orig : String] |
13c11724 | 29 | [nick : (Option String)] |
98529d3d | 30 | [uri : Url] |
13c11724 | 31 | [text : String] |
b0ff061a | 32 | [mentions : (Listof Peer)])) |
9c464d95 | 33 | |
dbc26280 | 34 | (struct Peer |
13c11724 | 35 | ([nick : (Option String)] |
d0a0e073 SK |
36 | [uri : Url]) |
37 | #:transparent) | |
4764ff89 | 38 | |
f65d6338 SK |
39 | (struct Resp |
40 | ([status-line : String] | |
41 | [headers : (Listof Bytes)] | |
42 | [body-input : Input-Port])) | |
43 | ||
edadb804 SK |
44 | (: tt-home-dir Path-String) |
45 | (define tt-home-dir (build-path (expand-user-path "~") ".tt")) | |
46 | ||
9a346534 | 47 | (: concurrent-filter-map (∀ (α β) (-> Natural (-> α β) (Listof α) (Listof β)))) |
78142acb | 48 | (define (concurrent-filter-map num-workers f xs) |
dad4504d | 49 | ; TODO preserve order of elements OR communicate that reorder is expected |
a239a233 | 50 | ; TODO switch from mailboxes to channels |
895a32cf SK |
51 | (define (make-worker id f) |
52 | (define parent (current-thread)) | |
53 | (λ () | |
a9511f7c SK |
54 | (define self : Thread (current-thread)) |
55 | (: work (∀ (α) (-> α))) | |
895a32cf SK |
56 | (define (work) |
57 | (thread-send parent (cons 'next self)) | |
58 | (match (thread-receive) | |
c562bea3 SK |
59 | ['done (thread-send parent (cons 'exit id))] |
60 | [(cons 'unit x) (begin | |
61 | (define y (f x)) | |
62 | (when y (thread-send parent (cons 'result y))) | |
63 | (work))])) | |
895a32cf | 64 | (work))) |
a9511f7c | 65 | (: dispatch (∀ (α β) (-> (Listof Nonnegative-Integer) (Listof α) (Listof β)))) |
895a32cf SK |
66 | (define (dispatch ws xs ys) |
67 | (if (empty? ws) | |
f1493e49 SK |
68 | ys |
69 | (match (thread-receive) | |
c562bea3 SK |
70 | [(cons 'exit w) (dispatch (remove w ws =) xs ys)] |
71 | [(cons 'result y) (dispatch ws xs (cons y ys))] | |
72 | [(cons 'next thd) (match xs | |
73 | ['() (begin | |
74 | (thread-send thd 'done) | |
75 | (dispatch ws xs ys))] | |
76 | [(cons x xs) (begin | |
77 | (thread-send thd (cons 'unit x)) | |
78 | (dispatch ws xs ys))])]))) | |
78142acb | 79 | (define workers (range num-workers)) |
9926c9a9 SK |
80 | (define threads (map (λ (id) (thread (make-worker id f))) workers)) |
81 | (define results (dispatch workers xs '())) | |
895a32cf SK |
82 | (for-each thread-wait threads) |
83 | results) | |
84 | ||
dad4504d | 85 | (module+ test |
de3ff448 SK |
86 | (let* ([f (λ (x) (if (even? x) x #f))] |
87 | [xs (range 11)] | |
88 | [actual (sort (concurrent-filter-map 10 f xs) <)] | |
89 | [expected (sort ( filter-map f xs) <)]) | |
c562bea3 | 90 | (check-equal? actual expected "concurrent-filter-map"))) |
dad4504d | 91 | |
98529d3d | 92 | (: msg-print (-> Out-Format Integer Msg Void)) |
3d042e75 SK |
93 | (define msg-print |
94 | (let* ([colors (vector 36 33)] | |
95 | [n (vector-length colors)]) | |
96 | (λ (out-format color-i msg) | |
01e4c499 | 97 | (let ([color (vector-ref colors (modulo color-i n))] |
b0ff061a SK |
98 | [nick (Msg-nick msg)] |
99 | [uri (url->string (Msg-uri msg))] | |
100 | [text (Msg-text msg)] | |
101 | [mentions (Msg-mentions msg)]) | |
3d042e75 | 102 | (match out-format |
01e4c499 | 103 | ['single-line |
13c11724 SK |
104 | (let ([nick (if nick nick uri)]) |
105 | (printf "~a \033[1;37m<~a>\033[0m \033[0;~am~a\033[0m~n" | |
106 | (parameterize | |
107 | ([date-display-format 'iso-8601]) | |
b0ff061a | 108 | (date->string (seconds->date (Msg-ts-epoch msg)) #t)) |
13c11724 | 109 | nick color text))] |
01e4c499 | 110 | ['multi-line |
13c11724 SK |
111 | (let ([nick (if nick (string-append nick " ") "")]) |
112 | (printf "~a (~a)~n\033[1;37m<~a~a>\033[0m~n\033[0;~am~a\033[0m~n~n" | |
113 | (parameterize | |
114 | ([date-display-format 'rfc2822]) | |
b0ff061a SK |
115 | (date->string (seconds->date (Msg-ts-epoch msg)) #t)) |
116 | (Msg-ts-orig msg) | |
13c11724 | 117 | nick uri color text))]))))) |
e96264cc | 118 | |
3877a0c4 SK |
119 | (: rfc3339->epoch (-> String (Option Nonnegative-Integer))) |
120 | (define rfc3339->epoch | |
121 | (let ([re (pregexp "^([0-9]{4})-([0-9]{2})-([0-9]{2})T([0-9]{2}):([0-9]{2})(:([0-9]{2}))?(\\.[0-9]+)?(Z|([+-])([0-9]{1,2}):?([0-9]{2}))?$")]) | |
122 | (λ (ts) | |
123 | (match (regexp-match re ts) | |
124 | [(list _wholething yyyy mm dd HH MM _:SS SS _fractional tz-whole tz-sign tz-HH tz-MM) | |
125 | (let* | |
126 | ([tz-offset | |
127 | (match* (tz-whole tz-sign tz-HH tz-MM) | |
128 | [("Z" #f #f #f) | |
129 | 0] | |
130 | [(_ (or "-" "+") (? identity h) (? identity m)) | |
131 | (let ([h (string->number h)] | |
132 | [m (string->number m)] | |
133 | ; Reverse to get back to UTC: | |
134 | [op (match tz-sign ["+" -] ["-" +])]) | |
135 | (op 0 (+ (* 60 m) (* 60 (* 60 h)))))] | |
136 | [(a b c d) | |
137 | (log-warning "Impossible TZ string: ~v, components: ~v ~v ~v ~v" tz-whole a b c d) | |
138 | 0])] | |
139 | [ts-orig ts] | |
140 | [local-time? #f] | |
141 | [ts-epoch (find-seconds (if SS (string->number SS) 0) | |
142 | (string->number MM) | |
143 | (string->number HH) | |
144 | (string->number dd) | |
145 | (string->number mm) | |
146 | (string->number yyyy) | |
147 | local-time?)]) | |
148 | (+ ts-epoch tz-offset))] | |
149 | [_ | |
150 | (log-error "Invalid timestamp: ~v" ts) | |
151 | #f])))) | |
152 | ||
13c11724 | 153 | (: str->msg (-> (Option String) Url String (Option Msg))) |
b4689464 | 154 | (define str->msg |
3877a0c4 | 155 | (let ([re (pregexp "^([^\\s\t]+)[\\s\t]+(.*)$")]) |
b4689464 | 156 | (λ (nick uri str) |
d3ac9e11 | 157 | (define str-head (substring str 0 (min 100 (string-length str)))) |
b4689464 SK |
158 | (with-handlers* |
159 | ([exn:fail? | |
160 | (λ (e) | |
9c464d95 SK |
161 | (log-error |
162 | "Failed to parse msg: ~v, from: ~v, at: ~v, because: ~v" | |
d3ac9e11 | 163 | str-head nick (url->string uri) e) |
b4689464 SK |
164 | #f)]) |
165 | (match (regexp-match re str) | |
3877a0c4 SK |
166 | [(list _wholething ts-orig text) |
167 | (let ([ts-epoch (rfc3339->epoch ts-orig)]) | |
168 | (if ts-epoch | |
13c11724 SK |
169 | (let ([mentions |
170 | (filter-map | |
171 | (λ (m) (match (regexp-match #px"@<([^>]+)>" m) | |
172 | [(list _wholething nick-uri) | |
dbc26280 | 173 | (str->peer nick-uri)])) |
13c11724 | 174 | (regexp-match* #px"@<[^\\s]+([\\s]+)?[^>]+>" text))]) |
b0ff061a | 175 | (Msg ts-epoch ts-orig nick uri text mentions)) |
3877a0c4 SK |
176 | (begin |
177 | (log-error | |
178 | "Msg rejected due to invalid timestamp: ~v, nick:~v, uri:~v" | |
d3ac9e11 | 179 | str-head nick (url->string uri)) |
3877a0c4 | 180 | #f)))] |
b4689464 | 181 | [_ |
d3ac9e11 | 182 | (log-debug "Non-msg line from nick:~v, line:~a" nick str-head) |
b4689464 | 183 | #f]))))) |
88d50b3e | 184 | |
63afa259 | 185 | (module+ test |
13c11724 | 186 | ; TODO Test for when missing-nick case |
b4689464 SK |
187 | (let* ([tzs (for*/list ([d '("-" "+")] |
188 | [h '("5" "05")] | |
189 | [m '("00" ":00" "57" ":57")]) | |
190 | (string-append d h m))] | |
191 | [tzs (list* "" "Z" tzs)]) | |
192 | (for* ([n '("fake-nick")] | |
193 | [u '("fake-uri")] | |
194 | [s '("" ":10")] | |
195 | [f '("" ".1337")] | |
196 | [z tzs] | |
197 | [sep (list "\t" " ")] | |
198 | [txt '("foo bar baz" "'jaz poop bear giraffe / tea" "@*\"``")]) | |
199 | (let* ([ts (string-append "2020-11-18T22:22" | |
200 | (if (non-empty-string? s) s ":00") | |
201 | z)] | |
202 | [m (str->msg n u (string-append ts sep txt))]) | |
203 | (check-not-false m) | |
b0ff061a SK |
204 | (check-equal? (Msg-nick m) n) |
205 | (check-equal? (Msg-uri m) u) | |
206 | (check-equal? (Msg-text m) txt) | |
207 | (check-equal? (Msg-ts-orig m) ts (format "Given: ~v" ts)) | |
b4689464 SK |
208 | ))) |
209 | ||
de3ff448 SK |
210 | (let* ([ts "2020-11-18T22:22:09-0500"] |
211 | [tab " "] | |
212 | [text "Lorem ipsum"] | |
213 | [nick "foo"] | |
214 | [uri "bar"] | |
215 | [actual (str->msg nick uri (string-append ts tab text))] | |
a139076c | 216 | [expected (Msg 1605756129 ts nick uri text '())]) |
c562bea3 | 217 | (check-equal? |
b0ff061a SK |
218 | (Msg-ts-epoch actual) |
219 | (Msg-ts-epoch expected) | |
78142acb | 220 | "str->msg ts-epoch") |
3877a0c4 | 221 | (check-equal? |
b0ff061a SK |
222 | (Msg-ts-orig actual) |
223 | (Msg-ts-orig expected) | |
3877a0c4 | 224 | "str->msg ts-orig") |
c562bea3 | 225 | (check-equal? |
b0ff061a SK |
226 | (Msg-nick actual) |
227 | (Msg-nick expected) | |
c562bea3 SK |
228 | "str->msg nick") |
229 | (check-equal? | |
b0ff061a SK |
230 | (Msg-uri actual) |
231 | (Msg-uri expected) | |
c562bea3 SK |
232 | "str->msg uri") |
233 | (check-equal? | |
b0ff061a SK |
234 | (Msg-text actual) |
235 | (Msg-text expected) | |
c562bea3 | 236 | "str->msg text"))) |
63afa259 | 237 | |
98529d3d | 238 | (: str->lines (-> String (Listof String))) |
e96264cc SK |
239 | (define (str->lines str) |
240 | (string-split str (regexp "[\r\n]+"))) | |
241 | ||
63afa259 | 242 | (module+ test |
de3ff448 | 243 | (check-equal? (str->lines "abc\ndef\n\nghi") '("abc" "def" "ghi"))) |
63afa259 | 244 | |
13c11724 | 245 | (: str->msgs (-> (Option String) Url String (Listof Msg))) |
b201e854 | 246 | (define (str->msgs nick uri str) |
3877a0c4 | 247 | (filter-map (λ (line) (str->msg nick uri line)) (filter-comments (str->lines str)))) |
4764ff89 | 248 | |
edadb804 SK |
249 | (: cache-dir Path-String) |
250 | (define cache-dir (build-path tt-home-dir "cache")) | |
251 | ||
d718efc4 SK |
252 | (define cache-object-dir (build-path cache-dir "objects")) |
253 | ||
edadb804 SK |
254 | (: url->cache-file-path-v1 (-> Url Path-String)) |
255 | (define (url->cache-file-path-v1 uri) | |
256 | (define (hash-sha1 str) : (-> String String) | |
257 | (define in (open-input-string str)) | |
258 | (define digest (sha1 in)) | |
259 | (close-input-port in) | |
260 | digest) | |
d718efc4 | 261 | (build-path cache-object-dir (hash-sha1 (url->string uri)))) |
edadb804 SK |
262 | |
263 | (: url->cache-file-path-v2 (-> Url Path-String)) | |
264 | (define (url->cache-file-path-v2 uri) | |
d718efc4 SK |
265 | (build-path cache-object-dir (uri-encode (url->string uri)))) |
266 | ||
267 | (define url->cache-object-path url->cache-file-path-v2) | |
1d753430 | 268 | |
d3ac9e11 SK |
269 | (: cache-object-filename->url (-> Path-String Url)) |
270 | (define (cache-object-filename->url name) | |
271 | (string->url (uri-decode (path->string name)))) | |
272 | ||
d718efc4 SK |
273 | (define (url->cache-etag-path uri) |
274 | (build-path cache-dir "etags" (uri-encode (url->string uri)))) | |
275 | ||
276 | (define (url->cache-lmod-path uri) | |
277 | (build-path cache-dir "lmods" (uri-encode (url->string uri)))) | |
9c464d95 | 278 | |
98529d3d SK |
279 | ; TODO Return Option |
280 | (: uri-read-cached (-> Url String)) | |
4214c0f3 | 281 | (define (uri-read-cached uri) |
edadb804 SK |
282 | (define path-v1 (url->cache-file-path-v1 uri)) |
283 | (define path-v2 (url->cache-file-path-v2 uri)) | |
284 | (when (file-exists? path-v1) | |
285 | (rename-file-or-directory path-v1 path-v2 #t)) | |
286 | (if (file-exists? path-v2) | |
287 | (file->string path-v2) | |
0e16a46c | 288 | (begin |
9c464d95 | 289 | (log-warning "Cache file not found for URI: ~a" (url->string uri)) |
4214c0f3 SK |
290 | ""))) |
291 | ||
a60c484e SK |
292 | (: uri? (-> String Boolean)) |
293 | (define (uri? str) | |
294 | (regexp-match? #rx"^[a-z]+://.*" (string-downcase str))) | |
295 | ||
dbc26280 SK |
296 | (: str->peer (String (Option Peer))) |
297 | (define (str->peer str) | |
298 | (log-debug "Parsing peer string: ~v" str) | |
13c11724 SK |
299 | (with-handlers* |
300 | ([exn:fail? | |
301 | (λ (e) | |
302 | (log-error "Invalid URI in string: ~v, exn: ~v" str e) | |
303 | #f)]) | |
304 | (match (string-split str) | |
a60c484e SK |
305 | [(list u) #:when (uri? u) (Peer #f (string->url u))] |
306 | [(list n u) #:when (uri? u) (Peer n (string->url u))] | |
13c11724 | 307 | [_ |
dbc26280 | 308 | (log-error "Invalid peer string: ~v" str) |
13c11724 SK |
309 | #f]))) |
310 | ||
9c464d95 | 311 | |
98529d3d | 312 | (: filter-comments (-> (Listof String) (Listof String))) |
9c464d95 SK |
313 | (define (filter-comments lines) |
314 | (filter-not (λ (line) (string-prefix? line "#")) lines)) | |
315 | ||
dbc26280 SK |
316 | (: str->peers (-> String (Listof Peer))) |
317 | (define (str->peers str) | |
318 | (filter-map str->peer (filter-comments (str->lines str)))) | |
9c464d95 | 319 | |
a60c484e SK |
320 | (: peers->file (-> (Listof Peers) Path-String Void)) |
321 | (define (peers->file peers path) | |
322 | (display-lines-to-file | |
323 | (map (match-lambda | |
324 | [(Peer n u) | |
325 | (format "~a~a" (if n (format "~a " n) "") (url->string u))]) | |
326 | peers) | |
327 | path | |
328 | #:exists 'replace)) | |
329 | ||
dbc26280 | 330 | (: file->peers (-> Path-String (Listof Peer))) |
d0a0e073 SK |
331 | (define (file->peers file-path) |
332 | (if (file-exists? file-path) | |
333 | (str->peers (file->string file-path)) | |
334 | (begin | |
a60c484e | 335 | (log-warning "File does not exist: ~v" (path->string file-path)) |
d0a0e073 | 336 | '()))) |
9c464d95 | 337 | |
9c5e4499 SK |
338 | (define re-rfc2822 |
339 | #px"^(Mon|Tue|Wed|Thu|Fri|Sat|Sun), ([0-9]{2}) (Jan|Feb|Mar|Apr|May|Jun|Jul|Aug|Sep|Oct|Nov|Dec) ([0-9]{4}) ([0-2][0-9]):([0-6][0-9]):([0-6][0-9]) GMT") | |
340 | ||
341 | (: b->n (-> Bytes (Option Number))) | |
342 | (define (b->n b) | |
343 | (string->number (bytes->string/utf-8 b))) | |
344 | ||
345 | (: mon->num (-> Bytes Natural)) | |
346 | (define/match (mon->num mon) | |
347 | [(#"Jan") 1] | |
348 | [(#"Feb") 2] | |
349 | [(#"Mar") 3] | |
350 | [(#"Apr") 4] | |
351 | [(#"May") 5] | |
352 | [(#"Jun") 6] | |
353 | [(#"Jul") 7] | |
354 | [(#"Aug") 8] | |
355 | [(#"Sep") 9] | |
356 | [(#"Oct") 10] | |
357 | [(#"Nov") 11] | |
358 | [(#"Dec") 12]) | |
359 | ||
360 | (: rfc2822->epoch (-> Bytes (Option Nonnegative-Integer))) | |
361 | (define (rfc2822->epoch timestamp) | |
362 | (match (regexp-match re-rfc2822 timestamp) | |
363 | [(list _ _ dd mo yyyy HH MM SS) | |
364 | #:when (and dd mo yyyy HH MM SS) | |
365 | (find-seconds (b->n SS) | |
366 | (b->n MM) | |
367 | (b->n HH) | |
368 | (b->n dd) | |
369 | (mon->num mo) | |
370 | (b->n yyyy) | |
371 | #f)] | |
372 | [_ | |
373 | #f])) | |
374 | ||
98529d3d | 375 | (: user-agent String) |
9c464d95 SK |
376 | (define user-agent |
377 | (let* | |
378 | ([prog-name "tt"] | |
3c9c8266 | 379 | [prog-version (info:#%info-lookup 'version)] |
9c464d95 | 380 | [prog-uri "https://github.com/xandkar/tt"] |
dbc26280 | 381 | [user-peer-file (build-path tt-home-dir "me")] |
9c464d95 | 382 | [user |
dbc26280 SK |
383 | (if (file-exists? user-peer-file) |
384 | (match (first (file->peers user-peer-file)) | |
385 | [(Peer #f u) (format "+~a" (url->string u) )] | |
386 | [(Peer n u) (format "+~a; @~a" (url->string u) n)]) | |
9c464d95 SK |
387 | (format "+~a" prog-uri))]) |
388 | (format "~a/~a (~a)" prog-name prog-version user))) | |
389 | ||
d718efc4 SK |
390 | (: header-get (-> (Listof Bytes) Bytes (Option Bytes))) |
391 | (define (header-get headers name) | |
392 | (match (filter-map (curry extract-field name) headers) | |
393 | [(list val) val] | |
394 | [_ #f])) | |
395 | ||
f65d6338 SK |
396 | (: uri-download (-> Positive-Float Url Void)) |
397 | (define (uri-download timeout u) | |
d718efc4 SK |
398 | (define cached-object-path (url->cache-object-path u)) |
399 | (define cached-etag-path (url->cache-etag-path u)) | |
400 | (define cached-lmod-path (url->cache-lmod-path u)) | |
f65d6338 SK |
401 | (define u-str (url->string u)) |
402 | (log-debug "uri-download ~v into ~v" u-str cached-object-path) | |
403 | (define timeout-chan (make-channel)) | |
404 | (define result-chan (make-channel)) | |
405 | (define timeout-thread | |
406 | (thread (λ () | |
407 | ; Doing this instead of sync/timeout to distinguish error values, | |
408 | ; rather than just have #f to work with. | |
409 | (sleep timeout) | |
410 | (channel-put timeout-chan (cons 'error 'timeout))))) | |
411 | (define result-thread | |
412 | (thread (λ () | |
413 | ; XXX We timeout getting a response, but body download could | |
414 | ; also take a long time and we might want to time that out as | |
415 | ; well, but then we may end-up with partially downloaded | |
416 | ; objects. But that could happen anyway if the server drops the | |
417 | ; connection for whatever reason. | |
418 | ; | |
419 | ; Maybe that is OK once we start treating the | |
420 | ; downloaded object as an addition to the stored set of | |
421 | ; messages, rather than the final set of messages. | |
422 | ||
423 | ; TODO message db | |
424 | ; - 1st try can just be an in-memory set that gets written-to | |
425 | ; and read-from disk as a whole. | |
426 | (define result | |
427 | (with-handlers | |
428 | ([exn:fail? (λ (e) (cons 'error (cons 'net-error e)))]) | |
429 | (define-values (status-line headers body-input) | |
430 | (http-sendrecv/url | |
431 | u | |
432 | #:headers (list (format "User-Agent: ~a" user-agent)))) | |
433 | (cons 'ok (Resp status-line headers body-input)))) | |
434 | (channel-put result-chan result)))) | |
435 | (define result | |
436 | (sync timeout-chan | |
437 | result-chan)) | |
438 | (kill-thread result-thread) | |
439 | (kill-thread timeout-thread) | |
440 | (match result | |
441 | [(cons 'error 'timeout) | |
442 | (log-error "Download failed: timeout. URL:~v" u-str)] | |
443 | [(cons 'error (cons 'net-error e)) | |
444 | (log-error "Download failed. Network error. URL:~v EXN:~v" u-str e)] | |
445 | [(cons 'ok (Resp status-line headers body-input)) | |
446 | (log-debug "headers: ~v" headers) | |
447 | (log-debug "status-line: ~v" status-line) | |
448 | (define status | |
449 | (string->number (second (string-split (bytes->string/utf-8 status-line))))) | |
450 | (log-debug "status: ~v" status) | |
451 | ; TODO Handle redirects | |
452 | (match status | |
453 | [200 | |
454 | (let* ([etag (header-get headers #"ETag")] | |
455 | [lmod (header-get headers #"Last-Modified")] | |
456 | [lmod-curr (if lmod (rfc2822->epoch lmod) #f)] | |
457 | [lmod-prev (if (file-exists? cached-lmod-path) | |
458 | (rfc2822->epoch (file->bytes cached-lmod-path)) | |
459 | #f)]) | |
460 | (log-debug "lmod-curr:~v lmod-prev:~v" lmod-curr lmod-prev) | |
461 | (unless (or (and etag | |
462 | (file-exists? cached-etag-path) | |
463 | (bytes=? etag (file->bytes cached-etag-path)) | |
464 | (begin | |
465 | (log-info "ETags match, skipping the rest of ~v" u-str) | |
466 | #t)) | |
467 | (and lmod-curr | |
468 | lmod-prev | |
469 | (<= lmod-curr lmod-prev) | |
470 | (begin | |
471 | (log-info "Last-Modified <= current skipping the rest of ~v" u-str) | |
472 | #t))) | |
473 | (begin | |
474 | (log-info | |
475 | "Downloading the rest of ~v. ETag: ~a, Last-Modified: ~v" | |
476 | u-str etag lmod) | |
477 | (make-parent-directory* cached-object-path) | |
478 | (make-parent-directory* cached-etag-path) | |
479 | (make-parent-directory* cached-lmod-path) | |
480 | (call-with-output-file cached-object-path | |
481 | (curry copy-port body-input) | |
482 | #:exists 'replace) | |
483 | (when etag | |
484 | (display-to-file etag cached-etag-path #:exists 'replace)) | |
485 | (when lmod | |
1864c06e | 486 | (display-to-file lmod cached-lmod-path #:exists 'replace)))))] |
f65d6338 | 487 | [_ |
1864c06e SK |
488 | (log-error "HTTP error URL:~a status:~a" u-str status)]) |
489 | (close-input-port body-input)])) | |
4764ff89 | 490 | |
98529d3d | 491 | (: timeline-print (-> Out-Format (Listof Msg) Void)) |
b201e854 | 492 | (define (timeline-print out-format timeline) |
3d042e75 | 493 | (void (foldl (match-lambda** |
b0ff061a | 494 | [((and m (Msg _ _ nick _ _ _)) (cons prev-nick i)) |
13c11724 | 495 | (let ([i (if (equal? prev-nick nick) i (+ 1 i))]) |
3d042e75 SK |
496 | (msg-print out-format i m) |
497 | (cons nick i))]) | |
498 | (cons "" 0) | |
499 | timeline))) | |
4764ff89 | 500 | |
dbc26280 SK |
501 | (: peer->msgs (-> Peer (Listof Msg))) |
502 | (define (peer->msgs f) | |
503 | (match-define (Peer nick uri) f) | |
504 | (log-info "Reading peer nick:~v uri:~v" nick (url->string uri)) | |
9c464d95 | 505 | (str->msgs nick uri (uri-read-cached uri))) |
4214c0f3 | 506 | |
f65d6338 SK |
507 | (: peer-download (-> Positive-Float Peer Void)) |
508 | (define (peer-download timeout peer) | |
509 | (match-define (Peer nick uri) peer) | |
2db1b40f | 510 | (define u (url->string uri)) |
f65d6338 SK |
511 | (log-info "Download BEGIN uri:~a" u) |
512 | (define-values (_result _tm-cpu-ms tm-real-ms _tm-gc-ms) | |
513 | (time-apply uri-download (list timeout uri))) | |
514 | (log-info "Download END in ~a seconds, uri: ~a" (/ tm-real-ms 1000.0) u)) | |
4214c0f3 | 515 | |
f65d6338 SK |
516 | (: timeline-download (-> Integer Positive-Float (Listof Peer) Void)) |
517 | (define (timeline-download num-workers timeout peers) | |
4214c0f3 | 518 | ; TODO No need for map - can just iter |
f65d6338 | 519 | (void (concurrent-filter-map num-workers (λ (p) (peer-download timeout p)) peers))) |
9a6a9f9a | 520 | |
38c9ecd5 | 521 | (: uniq (∀ (α) (-> (Listof α) (Listof α)))) |
a60c484e SK |
522 | (define (uniq xs) |
523 | (set->list (list->set xs))) | |
524 | ||
525 | (: peers->timeline (-> (listof Peer) (listof Msg))) | |
526 | (define (peers->timeline peers) | |
527 | (append* (filter-map peer->msgs peers))) | |
528 | ||
529 | (: timeline-sort (-> (listof Msg) timeline-order (Listof Msgs))) | |
530 | (define (timeline-sort msgs order) | |
a4899240 SK |
531 | (define cmp (match order |
532 | ['old->new <] | |
533 | ['new->old >])) | |
a60c484e SK |
534 | (sort msgs (λ (a b) (cmp (Msg-ts-epoch a) |
535 | (Msg-ts-epoch b))))) | |
4764ff89 | 536 | |
d0a0e073 SK |
537 | (: paths->peers (-> (Listof String) (Listof Peer))) |
538 | (define (paths->peers paths) | |
539 | (let* ([paths (match paths | |
540 | ['() | |
541 | (let ([peer-refs-file (build-path tt-home-dir "peers")]) | |
542 | (log-debug | |
543 | "No peer ref file paths provided, defaulting to ~v" | |
544 | (path->string peer-refs-file)) | |
545 | (list peer-refs-file))] | |
546 | [paths | |
547 | (log-debug "Peer ref file paths provided: ~v" paths) | |
548 | (map string->path paths)])] | |
549 | [peers (append* (map file->peers paths))]) | |
550 | (log-info "Read-in ~a peers." (length peers)) | |
38c9ecd5 | 551 | (uniq peers))) |
d0a0e073 | 552 | |
d3ac9e11 SK |
553 | (: mentioned-peers-in-cache (-> (Listof Peer))) |
554 | (define (mentioned-peers-in-cache) | |
555 | (define msgs | |
556 | (append* (map (λ (filename) | |
557 | (define path (build-path cache-object-dir filename)) | |
558 | (define size (/ (file-size path) 1000000.0)) | |
559 | (log-info "BEGIN parsing ~a MB from file: ~v" | |
560 | size | |
561 | (path->string path)) | |
562 | (define t0 (current-inexact-milliseconds)) | |
563 | (define m (filter-map | |
564 | (λ (line) | |
565 | (str->msg #f (cache-object-filename->url filename) line)) | |
566 | (filter-comments | |
567 | (file->lines path)))) | |
568 | (define t1 (current-inexact-milliseconds)) | |
569 | (log-info "END parsing ~a MB in ~a seconds from file: ~v." | |
570 | size | |
571 | (* 0.001 (- t1 t0)) | |
572 | (path->string path)) | |
573 | (when (empty? m) | |
574 | (log-warning "No messages found in ~a" (path->string path))) | |
575 | m) | |
576 | (directory-list cache-object-dir)))) | |
577 | (uniq (append* (map Msg-mentions msgs)))) | |
578 | ||
56de6228 SK |
579 | (: log-writer-stop (-> Thread Void)) |
580 | (define (log-writer-stop log-writer) | |
581 | (log-message (current-logger) 'fatal 'stop "Exiting." #f) | |
582 | (thread-wait log-writer)) | |
583 | ||
0d3f753c SK |
584 | (: log-writer-start (-> Log-Level Thread)) |
585 | (define (log-writer-start level) | |
56de6228 SK |
586 | (let* ([logger |
587 | (make-logger #f #f level #f)] | |
588 | [log-receiver | |
589 | (make-log-receiver logger level)] | |
590 | [log-writer | |
591 | (thread | |
592 | (λ () | |
593 | (parameterize | |
594 | ([date-display-format 'iso-8601]) | |
595 | (let loop () | |
596 | (match-define (vector level msg _ topic) (sync log-receiver)) | |
597 | (unless (equal? topic 'stop) | |
598 | (eprintf "~a [~a] ~a~n" (date->string (current-date) #t) level msg) | |
599 | (loop))))))]) | |
600 | (current-logger logger) | |
601 | log-writer)) | |
01e4c499 | 602 | |
24c6a76b | 603 | (module+ main |
24f1f64b | 604 | (let ([log-level 'info]) |
c562bea3 | 605 | (command-line |
24f1f64b SK |
606 | #:program |
607 | "tt" | |
c562bea3 | 608 | #:once-each |
01e4c499 SK |
609 | [("-d" "--debug") |
610 | "Enable debug log level." | |
611 | (set! log-level 'debug)] | |
24f1f64b SK |
612 | #:help-labels |
613 | "" | |
614 | "and <command> is one of" | |
a60c484e | 615 | "r, read : Read the timeline (offline operation)." |
4214c0f3 | 616 | "d, download : Download the timeline." |
edadb804 | 617 | ; TODO Add path dynamically |
3a4b2233 | 618 | "u, upload : Upload your twtxt file (alias to execute ~/.tt/upload)." |
a60c484e | 619 | "c, crawl : Discover new peers mentioned by known peers (offline operation)." |
24f1f64b SK |
620 | "" |
621 | #:args (command . args) | |
0d3f753c | 622 | (define log-writer (log-writer-start log-level)) |
4214c0f3 | 623 | (current-command-line-arguments (list->vector args)) |
24f1f64b | 624 | (match command |
4214c0f3 | 625 | [(or "d" "download") |
2db1b40f | 626 | ; Initially, 15 was fastest out of the tried: 1, 5, 10, 20. Then I |
edadb804 | 627 | ; started noticing significant slowdowns. Reducing to 5 seems to help. |
f65d6338 SK |
628 | (let ([num-workers 5] |
629 | [timeout 10.0]) | |
24f1f64b SK |
630 | (command-line |
631 | #:program | |
4214c0f3 | 632 | "tt download" |
24f1f64b SK |
633 | #:once-each |
634 | [("-j" "--jobs") | |
635 | njobs "Number of concurrent jobs." | |
78142acb | 636 | (set! num-workers (string->number njobs))] |
f65d6338 SK |
637 | [("-t" "--timeout") |
638 | seconds "Timeout seconds per request." | |
639 | (set! timeout (string->number seconds))] | |
d0a0e073 | 640 | #:args file-paths |
a60c484e SK |
641 | (let ([peers (paths->peers file-paths)]) |
642 | (define-values (_res _cpu real-ms _gc) | |
f65d6338 | 643 | (time-apply timeline-download (list num-workers timeout peers))) |
a60c484e SK |
644 | (log-info "Downloaded timelines from ~a peers in ~a seconds." |
645 | (length peers) | |
646 | (/ real-ms 1000.0)))))] | |
3a4b2233 SK |
647 | [(or "u" "upload") |
648 | (command-line | |
3877a0c4 SK |
649 | #:program |
650 | "tt upload" | |
651 | #:args () | |
edadb804 | 652 | (if (system (path->string (build-path tt-home-dir "upload"))) |
3a4b2233 SK |
653 | (exit 0) |
654 | (exit 1)))] | |
4214c0f3 | 655 | [(or "r" "read") |
a4899240 | 656 | (let ([out-format 'multi-line] |
a993cb85 SK |
657 | [order 'old->new] |
658 | [ts-min #f] | |
659 | [ts-max #f]) | |
4214c0f3 SK |
660 | (command-line |
661 | #:program | |
662 | "tt read" | |
a4899240 SK |
663 | #:once-each |
664 | [("-r" "--rev") | |
665 | "Reverse displayed timeline order." | |
666 | (set! order 'new->old)] | |
a993cb85 SK |
667 | [("-m" "--min") |
668 | m "Earliest time to display (ignore anything before it)." | |
669 | (set! ts-min (rfc3339->epoch m))] | |
670 | [("-x" "--max") | |
671 | x "Latest time to display (ignore anything after it)." | |
672 | (set! ts-max (rfc3339->epoch x))] | |
24f1f64b SK |
673 | #:once-any |
674 | [("-s" "--short") | |
675 | "Short output format" | |
676 | (set! out-format 'single-line)] | |
677 | [("-l" "--long") | |
678 | "Long output format" | |
679 | (set! out-format 'multi-line)] | |
d0a0e073 | 680 | #:args file-paths |
a60c484e SK |
681 | (let* ([peers |
682 | (paths->peers file-paths)] | |
683 | [timeline | |
a993cb85 SK |
684 | (timeline-sort (peers->timeline peers) order)] |
685 | [timeline | |
686 | (filter (λ (m) (and (if ts-min (>= (Msg-ts-epoch m) | |
687 | ts-min) | |
688 | #t) | |
689 | (if ts-max (<= (Msg-ts-epoch m) | |
690 | ts-max) | |
691 | #t))) | |
692 | timeline)]) | |
a60c484e SK |
693 | (timeline-print out-format timeline))))] |
694 | [(or "c" "crawl") | |
695 | (command-line | |
696 | #:program | |
697 | "tt crawl" | |
d3ac9e11 | 698 | #:args () |
ef07b6ed SK |
699 | (let* ([peers-sort |
700 | (λ (peers) (sort peers (match-lambda** | |
701 | [((Peer n1 _) (Peer n2 _)) | |
702 | (string<? (if n1 n1 "") | |
703 | (if n2 n2 ""))])))] | |
704 | [peers-all-file | |
a60c484e SK |
705 | (build-path tt-home-dir "peers-all")] |
706 | [peers-mentioned-file | |
707 | (build-path tt-home-dir "peers-mentioned")] | |
a60c484e | 708 | [peers-mentioned-curr |
d3ac9e11 | 709 | (mentioned-peers-in-cache)] |
a60c484e SK |
710 | [peers-mentioned-prev |
711 | (file->peers peers-mentioned-file)] | |
712 | [peers-mentioned | |
ef07b6ed SK |
713 | (peers-sort (uniq (append peers-mentioned-prev |
714 | peers-mentioned-curr)))] | |
a60c484e SK |
715 | [peers-all-prev |
716 | (file->peers peers-all-file)] | |
717 | [peers-all | |
d3ac9e11 | 718 | (list->set (append peers-mentioned |
018b2abb | 719 | peers-all-prev))] |
d3ac9e11 SK |
720 | [peers-discovered |
721 | (set-subtract peers-all (list->set peers-all-prev))] | |
018b2abb SK |
722 | [peers-all |
723 | (peers-sort (set->list peers-all))]) | |
d3ac9e11 SK |
724 | (log-info "Known peers mentioned: ~a" (length peers-mentioned)) |
725 | (log-info "Known peers total: ~a" (length peers-all)) | |
726 | (log-info "Discovered ~a new peers:~n~a" | |
727 | (set-count peers-discovered) | |
728 | (pretty-format (map | |
729 | (λ (p) (cons (Peer-nick p) | |
730 | (url->string (Peer-uri p)))) | |
731 | (set->list peers-discovered)))) | |
a60c484e SK |
732 | (peers->file peers-mentioned |
733 | peers-mentioned-file) | |
734 | (peers->file peers-all | |
735 | peers-all-file)))] | |
5db4881e SK |
736 | [command |
737 | (eprintf "Error: invalid command: ~v\n" command) | |
738 | (eprintf "Please use the \"--help\" option to see a list of available commands.\n") | |
739 | (exit 1)]) | |
0d3f753c | 740 | (log-writer-stop log-writer)))) |