twitter_ebooks 2.0.0
Sign up to get free protection for your applications and to get access to all the features.
- data/.gitignore +1 -0
- data/Gemfile +4 -0
- data/Gemfile.lock +78 -0
- data/NOTES.md +4 -0
- data/README.md +20 -0
- data/bin/ebooks +83 -0
- data/data/adjectives.txt +1466 -0
- data/data/nouns.txt +2193 -0
- data/data/stopwords.txt +639 -0
- data/lib/twitter_ebooks/archiver.rb +86 -0
- data/lib/twitter_ebooks/bot.rb +145 -0
- data/lib/twitter_ebooks/markov.rb +89 -0
- data/lib/twitter_ebooks/model.rb +147 -0
- data/lib/twitter_ebooks/nlp.rb +142 -0
- data/lib/twitter_ebooks/version.rb +3 -0
- data/lib/twitter_ebooks.rb +20 -0
- data/skeleton/Procfile +1 -0
- data/skeleton/bots.rb +47 -0
- data/skeleton/corpus/README.md +1 -0
- data/skeleton/model/README.md +1 -0
- data/test/corpus/0xabad1dea.tweets +14696 -0
- data/test/tokenize.rb +18 -0
- data/twitter_ebooks.gemspec +30 -0
- metadata +247 -0
data/data/stopwords.txt
ADDED
@@ -0,0 +1,639 @@
|
|
1
|
+
a
|
2
|
+
able
|
3
|
+
about
|
4
|
+
above
|
5
|
+
abst
|
6
|
+
across
|
7
|
+
act
|
8
|
+
actually
|
9
|
+
added
|
10
|
+
adj
|
11
|
+
after
|
12
|
+
afterwards
|
13
|
+
again
|
14
|
+
against
|
15
|
+
ah
|
16
|
+
all
|
17
|
+
almost
|
18
|
+
alone
|
19
|
+
along
|
20
|
+
already
|
21
|
+
also
|
22
|
+
although
|
23
|
+
always
|
24
|
+
am
|
25
|
+
among
|
26
|
+
amongst
|
27
|
+
an
|
28
|
+
and
|
29
|
+
announce
|
30
|
+
another
|
31
|
+
any
|
32
|
+
anybody
|
33
|
+
anyhow
|
34
|
+
anymore
|
35
|
+
anyone
|
36
|
+
anything
|
37
|
+
anyway
|
38
|
+
anyways
|
39
|
+
anywhere
|
40
|
+
apparently
|
41
|
+
are
|
42
|
+
aren
|
43
|
+
arent
|
44
|
+
arise
|
45
|
+
around
|
46
|
+
as
|
47
|
+
aside
|
48
|
+
ask
|
49
|
+
asking
|
50
|
+
at
|
51
|
+
auth
|
52
|
+
away
|
53
|
+
b
|
54
|
+
back
|
55
|
+
be
|
56
|
+
became
|
57
|
+
because
|
58
|
+
become
|
59
|
+
becomes
|
60
|
+
becoming
|
61
|
+
been
|
62
|
+
before
|
63
|
+
beforehand
|
64
|
+
behind
|
65
|
+
being
|
66
|
+
believe
|
67
|
+
below
|
68
|
+
beside
|
69
|
+
besides
|
70
|
+
between
|
71
|
+
beyond
|
72
|
+
biol
|
73
|
+
both
|
74
|
+
brief
|
75
|
+
briefly
|
76
|
+
but
|
77
|
+
by
|
78
|
+
c
|
79
|
+
ca
|
80
|
+
came
|
81
|
+
can
|
82
|
+
cannot
|
83
|
+
can't
|
84
|
+
cause
|
85
|
+
causes
|
86
|
+
co
|
87
|
+
com
|
88
|
+
come
|
89
|
+
comes
|
90
|
+
contain
|
91
|
+
contains
|
92
|
+
could
|
93
|
+
couldnt
|
94
|
+
d
|
95
|
+
date
|
96
|
+
did
|
97
|
+
didn't
|
98
|
+
different
|
99
|
+
do
|
100
|
+
does
|
101
|
+
doesn't
|
102
|
+
doing
|
103
|
+
done
|
104
|
+
don't
|
105
|
+
down
|
106
|
+
downwards
|
107
|
+
due
|
108
|
+
during
|
109
|
+
e
|
110
|
+
each
|
111
|
+
ed
|
112
|
+
edu
|
113
|
+
effect
|
114
|
+
eg
|
115
|
+
eight
|
116
|
+
eighty
|
117
|
+
either
|
118
|
+
else
|
119
|
+
elsewhere
|
120
|
+
end
|
121
|
+
ending
|
122
|
+
enough
|
123
|
+
especially
|
124
|
+
et
|
125
|
+
et-al
|
126
|
+
etc
|
127
|
+
even
|
128
|
+
ever
|
129
|
+
ex
|
130
|
+
except
|
131
|
+
f
|
132
|
+
far
|
133
|
+
few
|
134
|
+
ff
|
135
|
+
fifth
|
136
|
+
first
|
137
|
+
five
|
138
|
+
fix
|
139
|
+
followed
|
140
|
+
following
|
141
|
+
follows
|
142
|
+
for
|
143
|
+
former
|
144
|
+
formerly
|
145
|
+
forth
|
146
|
+
found
|
147
|
+
four
|
148
|
+
from
|
149
|
+
further
|
150
|
+
furthermore
|
151
|
+
g
|
152
|
+
gave
|
153
|
+
get
|
154
|
+
gets
|
155
|
+
getting
|
156
|
+
give
|
157
|
+
given
|
158
|
+
gives
|
159
|
+
giving
|
160
|
+
go
|
161
|
+
goes
|
162
|
+
gone
|
163
|
+
got
|
164
|
+
gotten
|
165
|
+
h
|
166
|
+
had
|
167
|
+
happens
|
168
|
+
hardly
|
169
|
+
has
|
170
|
+
hasn't
|
171
|
+
have
|
172
|
+
haven't
|
173
|
+
having
|
174
|
+
he
|
175
|
+
hed
|
176
|
+
hence
|
177
|
+
her
|
178
|
+
here
|
179
|
+
hereafter
|
180
|
+
hereby
|
181
|
+
herein
|
182
|
+
heres
|
183
|
+
hereupon
|
184
|
+
hers
|
185
|
+
herself
|
186
|
+
hes
|
187
|
+
hi
|
188
|
+
hid
|
189
|
+
him
|
190
|
+
himself
|
191
|
+
his
|
192
|
+
hither
|
193
|
+
home
|
194
|
+
how
|
195
|
+
howbeit
|
196
|
+
however
|
197
|
+
hundred
|
198
|
+
i
|
199
|
+
id
|
200
|
+
ie
|
201
|
+
if
|
202
|
+
i'll
|
203
|
+
im
|
204
|
+
immediate
|
205
|
+
immediately
|
206
|
+
importance
|
207
|
+
important
|
208
|
+
in
|
209
|
+
inc
|
210
|
+
indeed
|
211
|
+
index
|
212
|
+
instead
|
213
|
+
into
|
214
|
+
inward
|
215
|
+
is
|
216
|
+
isn't
|
217
|
+
it
|
218
|
+
itd
|
219
|
+
it'll
|
220
|
+
its
|
221
|
+
itself
|
222
|
+
i've
|
223
|
+
j
|
224
|
+
just
|
225
|
+
k
|
226
|
+
keep
|
227
|
+
keeps
|
228
|
+
kept
|
229
|
+
kg
|
230
|
+
km
|
231
|
+
know
|
232
|
+
known
|
233
|
+
knows
|
234
|
+
l
|
235
|
+
largely
|
236
|
+
last
|
237
|
+
lately
|
238
|
+
later
|
239
|
+
latter
|
240
|
+
latterly
|
241
|
+
least
|
242
|
+
less
|
243
|
+
lest
|
244
|
+
let
|
245
|
+
lets
|
246
|
+
like
|
247
|
+
liked
|
248
|
+
likely
|
249
|
+
line
|
250
|
+
little
|
251
|
+
'll
|
252
|
+
look
|
253
|
+
looking
|
254
|
+
looks
|
255
|
+
ltd
|
256
|
+
m
|
257
|
+
made
|
258
|
+
mainly
|
259
|
+
make
|
260
|
+
makes
|
261
|
+
many
|
262
|
+
may
|
263
|
+
maybe
|
264
|
+
me
|
265
|
+
mean
|
266
|
+
means
|
267
|
+
meantime
|
268
|
+
meanwhile
|
269
|
+
merely
|
270
|
+
mg
|
271
|
+
might
|
272
|
+
miss
|
273
|
+
ml
|
274
|
+
more
|
275
|
+
moreover
|
276
|
+
most
|
277
|
+
mostly
|
278
|
+
mr
|
279
|
+
mrs
|
280
|
+
much
|
281
|
+
mug
|
282
|
+
must
|
283
|
+
my
|
284
|
+
myself
|
285
|
+
n
|
286
|
+
na
|
287
|
+
name
|
288
|
+
namely
|
289
|
+
nay
|
290
|
+
nd
|
291
|
+
near
|
292
|
+
nearly
|
293
|
+
necessarily
|
294
|
+
necessary
|
295
|
+
need
|
296
|
+
needs
|
297
|
+
neither
|
298
|
+
never
|
299
|
+
new
|
300
|
+
next
|
301
|
+
nine
|
302
|
+
ninety
|
303
|
+
no
|
304
|
+
nobody
|
305
|
+
non
|
306
|
+
none
|
307
|
+
nonetheless
|
308
|
+
noone
|
309
|
+
nor
|
310
|
+
normally
|
311
|
+
nos
|
312
|
+
not
|
313
|
+
noted
|
314
|
+
nothing
|
315
|
+
now
|
316
|
+
nowhere
|
317
|
+
o
|
318
|
+
obtain
|
319
|
+
obtained
|
320
|
+
obviously
|
321
|
+
of
|
322
|
+
off
|
323
|
+
often
|
324
|
+
oh
|
325
|
+
ok
|
326
|
+
okay
|
327
|
+
old
|
328
|
+
omitted
|
329
|
+
on
|
330
|
+
once
|
331
|
+
one
|
332
|
+
ones
|
333
|
+
only
|
334
|
+
onto
|
335
|
+
or
|
336
|
+
ord
|
337
|
+
other
|
338
|
+
others
|
339
|
+
otherwise
|
340
|
+
ought
|
341
|
+
our
|
342
|
+
ours
|
343
|
+
ourselves
|
344
|
+
out
|
345
|
+
outside
|
346
|
+
over
|
347
|
+
overall
|
348
|
+
owing
|
349
|
+
own
|
350
|
+
p
|
351
|
+
page
|
352
|
+
pages
|
353
|
+
part
|
354
|
+
particular
|
355
|
+
particularly
|
356
|
+
past
|
357
|
+
per
|
358
|
+
perhaps
|
359
|
+
placed
|
360
|
+
please
|
361
|
+
plus
|
362
|
+
poorly
|
363
|
+
possible
|
364
|
+
possibly
|
365
|
+
potentially
|
366
|
+
pp
|
367
|
+
predominantly
|
368
|
+
present
|
369
|
+
previously
|
370
|
+
primarily
|
371
|
+
probably
|
372
|
+
promptly
|
373
|
+
proud
|
374
|
+
provides
|
375
|
+
put
|
376
|
+
q
|
377
|
+
que
|
378
|
+
quickly
|
379
|
+
quite
|
380
|
+
qv
|
381
|
+
r
|
382
|
+
ran
|
383
|
+
rather
|
384
|
+
rd
|
385
|
+
re
|
386
|
+
readily
|
387
|
+
really
|
388
|
+
recent
|
389
|
+
recently
|
390
|
+
ref
|
391
|
+
refs
|
392
|
+
regarding
|
393
|
+
regardless
|
394
|
+
regards
|
395
|
+
related
|
396
|
+
relatively
|
397
|
+
research
|
398
|
+
respectively
|
399
|
+
resulted
|
400
|
+
resulting
|
401
|
+
results
|
402
|
+
right
|
403
|
+
run
|
404
|
+
s
|
405
|
+
said
|
406
|
+
same
|
407
|
+
saw
|
408
|
+
say
|
409
|
+
saying
|
410
|
+
says
|
411
|
+
sec
|
412
|
+
section
|
413
|
+
see
|
414
|
+
seeing
|
415
|
+
seem
|
416
|
+
seemed
|
417
|
+
seeming
|
418
|
+
seems
|
419
|
+
seen
|
420
|
+
self
|
421
|
+
selves
|
422
|
+
sent
|
423
|
+
seven
|
424
|
+
several
|
425
|
+
shall
|
426
|
+
she
|
427
|
+
shed
|
428
|
+
she'll
|
429
|
+
shes
|
430
|
+
should
|
431
|
+
shouldn't
|
432
|
+
show
|
433
|
+
showed
|
434
|
+
shown
|
435
|
+
showns
|
436
|
+
shows
|
437
|
+
similar
|
438
|
+
similarly
|
439
|
+
since
|
440
|
+
six
|
441
|
+
slightly
|
442
|
+
so
|
443
|
+
some
|
444
|
+
somebody
|
445
|
+
somehow
|
446
|
+
someone
|
447
|
+
somethan
|
448
|
+
something
|
449
|
+
sometime
|
450
|
+
sometimes
|
451
|
+
somewhat
|
452
|
+
somewhere
|
453
|
+
soon
|
454
|
+
sorry
|
455
|
+
specifically
|
456
|
+
specified
|
457
|
+
specify
|
458
|
+
specifying
|
459
|
+
still
|
460
|
+
stop
|
461
|
+
strongly
|
462
|
+
sub
|
463
|
+
substantially
|
464
|
+
successfully
|
465
|
+
such
|
466
|
+
sufficiently
|
467
|
+
suggest
|
468
|
+
sup
|
469
|
+
sure
|
470
|
+
t
|
471
|
+
take
|
472
|
+
taken
|
473
|
+
taking
|
474
|
+
tell
|
475
|
+
tends
|
476
|
+
th
|
477
|
+
than
|
478
|
+
thank
|
479
|
+
thanks
|
480
|
+
thanx
|
481
|
+
that
|
482
|
+
that'll
|
483
|
+
thats
|
484
|
+
that've
|
485
|
+
the
|
486
|
+
their
|
487
|
+
theirs
|
488
|
+
them
|
489
|
+
themselves
|
490
|
+
then
|
491
|
+
thence
|
492
|
+
there
|
493
|
+
thereafter
|
494
|
+
thereby
|
495
|
+
thered
|
496
|
+
therefore
|
497
|
+
therein
|
498
|
+
there'll
|
499
|
+
thereof
|
500
|
+
therere
|
501
|
+
theres
|
502
|
+
thereto
|
503
|
+
thereupon
|
504
|
+
there've
|
505
|
+
these
|
506
|
+
they
|
507
|
+
theyd
|
508
|
+
they'll
|
509
|
+
theyre
|
510
|
+
they've
|
511
|
+
think
|
512
|
+
this
|
513
|
+
those
|
514
|
+
thou
|
515
|
+
though
|
516
|
+
thoughh
|
517
|
+
thousand
|
518
|
+
throug
|
519
|
+
through
|
520
|
+
throughout
|
521
|
+
thru
|
522
|
+
thus
|
523
|
+
til
|
524
|
+
tip
|
525
|
+
to
|
526
|
+
together
|
527
|
+
too
|
528
|
+
took
|
529
|
+
toward
|
530
|
+
towards
|
531
|
+
tried
|
532
|
+
tries
|
533
|
+
truly
|
534
|
+
try
|
535
|
+
trying
|
536
|
+
ts
|
537
|
+
twice
|
538
|
+
two
|
539
|
+
u
|
540
|
+
un
|
541
|
+
under
|
542
|
+
unfortunately
|
543
|
+
unless
|
544
|
+
unlike
|
545
|
+
unlikely
|
546
|
+
until
|
547
|
+
unto
|
548
|
+
up
|
549
|
+
upon
|
550
|
+
ups
|
551
|
+
us
|
552
|
+
use
|
553
|
+
used
|
554
|
+
useful
|
555
|
+
usefully
|
556
|
+
usefulness
|
557
|
+
uses
|
558
|
+
using
|
559
|
+
usually
|
560
|
+
v
|
561
|
+
value
|
562
|
+
various
|
563
|
+
've
|
564
|
+
very
|
565
|
+
via
|
566
|
+
viz
|
567
|
+
vol
|
568
|
+
vols
|
569
|
+
vs
|
570
|
+
w
|
571
|
+
want
|
572
|
+
wants
|
573
|
+
was
|
574
|
+
wasn't
|
575
|
+
way
|
576
|
+
we
|
577
|
+
wed
|
578
|
+
welcome
|
579
|
+
we'll
|
580
|
+
went
|
581
|
+
were
|
582
|
+
weren't
|
583
|
+
we've
|
584
|
+
what
|
585
|
+
whatever
|
586
|
+
what'll
|
587
|
+
whats
|
588
|
+
when
|
589
|
+
whence
|
590
|
+
whenever
|
591
|
+
where
|
592
|
+
whereafter
|
593
|
+
whereas
|
594
|
+
whereby
|
595
|
+
wherein
|
596
|
+
wheres
|
597
|
+
whereupon
|
598
|
+
wherever
|
599
|
+
whether
|
600
|
+
which
|
601
|
+
while
|
602
|
+
whim
|
603
|
+
whither
|
604
|
+
who
|
605
|
+
whod
|
606
|
+
whoever
|
607
|
+
whole
|
608
|
+
who'll
|
609
|
+
whom
|
610
|
+
whomever
|
611
|
+
whos
|
612
|
+
whose
|
613
|
+
why
|
614
|
+
widely
|
615
|
+
willing
|
616
|
+
with
|
617
|
+
within
|
618
|
+
without
|
619
|
+
won't
|
620
|
+
words
|
621
|
+
would
|
622
|
+
wouldn't
|
623
|
+
www
|
624
|
+
x
|
625
|
+
y
|
626
|
+
yes
|
627
|
+
yet
|
628
|
+
you
|
629
|
+
youd
|
630
|
+
you'll
|
631
|
+
your
|
632
|
+
youre
|
633
|
+
yours
|
634
|
+
you've
|
635
|
+
z
|
636
|
+
zero
|
637
|
+
.
|
638
|
+
?
|
639
|
+
!
|
@@ -0,0 +1,86 @@
|
|
1
|
+
#!/usr/bin/env ruby
|
2
|
+
# encoding: utf-8
|
3
|
+
|
4
|
+
require 'twitter'
|
5
|
+
require 'tweetstream'
|
6
|
+
require 'rufus/scheduler'
|
7
|
+
require 'marky_markov'
|
8
|
+
require 'htmlentities'
|
9
|
+
|
10
|
+
module Ebooks
|
11
|
+
class Archiver
|
12
|
+
def initialize(username, outpath)
|
13
|
+
@username = username
|
14
|
+
@outpath = outpath
|
15
|
+
@client = Twitter::Client.new
|
16
|
+
end
|
17
|
+
|
18
|
+
# Read exiting corpus into memory.
|
19
|
+
# Return list of tweet lines and the last tweet id.
|
20
|
+
def read_corpus
|
21
|
+
lines = []
|
22
|
+
since_id = nil
|
23
|
+
|
24
|
+
if File.exists?(@outpath)
|
25
|
+
lines = File.read(@outpath).split("\n")
|
26
|
+
if lines[0].start_with?('#')
|
27
|
+
since_id = lines[0].split('# ').last
|
28
|
+
end
|
29
|
+
end
|
30
|
+
|
31
|
+
[lines, since_id]
|
32
|
+
end
|
33
|
+
|
34
|
+
# Retrieve all available tweets for a given user since the last tweet id
|
35
|
+
def tweets_since(since_id)
|
36
|
+
page = 1
|
37
|
+
retries = 0
|
38
|
+
tweets = []
|
39
|
+
max_id = nil
|
40
|
+
|
41
|
+
opts = {
|
42
|
+
count: 200,
|
43
|
+
include_rts: false,
|
44
|
+
trim_user: true
|
45
|
+
}
|
46
|
+
|
47
|
+
opts[:since_id] = since_id unless since_id.nil?
|
48
|
+
|
49
|
+
loop do
|
50
|
+
opts[:max_id] = max_id unless max_id.nil?
|
51
|
+
new = @client.user_timeline(@username, opts)
|
52
|
+
break if new.length <= 1
|
53
|
+
puts "Received #{new.length} tweets"
|
54
|
+
tweets += new
|
55
|
+
max_id = new.last.id
|
56
|
+
break
|
57
|
+
end
|
58
|
+
|
59
|
+
tweets
|
60
|
+
end
|
61
|
+
|
62
|
+
def fetch_tweets
|
63
|
+
lines, since_id = read_corpus
|
64
|
+
|
65
|
+
if since_id.nil?
|
66
|
+
puts "Retrieving tweets from @#{@username}"
|
67
|
+
else
|
68
|
+
puts "Retrieving tweets from @#{@username} since #{since_id}"
|
69
|
+
end
|
70
|
+
|
71
|
+
tweets = tweets_since(since_id)
|
72
|
+
|
73
|
+
if tweets.length == 0
|
74
|
+
puts "No new tweets"
|
75
|
+
return
|
76
|
+
end
|
77
|
+
|
78
|
+
new_lines = tweets.map { |tweet| tweet.text.gsub("\n", " ") }
|
79
|
+
new_since_id = tweets[0].id.to_s
|
80
|
+
lines = ["# " + new_since_id] + new_lines + lines
|
81
|
+
corpus = File.open(@outpath, 'w')
|
82
|
+
corpus.write(lines.join("\n"))
|
83
|
+
corpus.close
|
84
|
+
end
|
85
|
+
end
|
86
|
+
end
|