twitterscraper-ruby 0.14.0 → 0.17.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/Gemfile.lock +1 -1
- data/README.md +89 -56
- data/lib/twitterscraper.rb +1 -0
- data/lib/twitterscraper/cache.rb +7 -1
- data/lib/twitterscraper/cli.rb +17 -5
- data/lib/twitterscraper/query.rb +81 -45
- data/lib/twitterscraper/template.rb +23 -41
- data/lib/twitterscraper/template/tweets.html.erb +98 -0
- data/lib/twitterscraper/tweet.rb +9 -0
- data/lib/twitterscraper/type.rb +15 -0
- data/lib/version.rb +1 -1
- metadata +4 -2
checksums.yaml
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA256:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: ac0c10b18d836983cc6b73e25b9ed333af2f620106a07c6bc6a40058fb127895
|
|
4
|
+
data.tar.gz: e6fc18219d9127fb30ba57e39dc4656c0f0a3c108428d959de5bac9e7d317088
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: 90cbf06b606878dc36b4bba44669139c273bf03b08a777ad87036834841bcb4b052e0559813dc56e4be124442abfc5a7fc44c5c9524c74929ca02b1d287d346b
|
|
7
|
+
data.tar.gz: ada0b74ee42ff62964b73ad9b49358227cdaf4fc87420cf12cf65af95168ad9775615a504345ebc83d3b791e9c0d892691c55bc477eddd647b3e8934f752fb9c
|
data/Gemfile.lock
CHANGED
data/README.md
CHANGED
|
@@ -5,15 +5,17 @@
|
|
|
5
5
|
|
|
6
6
|
A gem to scrape https://twitter.com/search. This gem is inspired by [taspinar/twitterscraper](https://github.com/taspinar/twitterscraper).
|
|
7
7
|
|
|
8
|
+
Please feel free to ask [@ts_3156](https://twitter.com/ts_3156) if you have any questions.
|
|
9
|
+
|
|
8
10
|
|
|
9
11
|
## Twitter Search API vs. twitterscraper-ruby
|
|
10
12
|
|
|
11
|
-
|
|
13
|
+
#### Twitter Search API
|
|
12
14
|
|
|
13
15
|
- The number of tweets: 180 - 450 requests/15 minutes (18,000 - 45,000 tweets/15 minutes)
|
|
14
16
|
- The time window: the past 7 days
|
|
15
17
|
|
|
16
|
-
|
|
18
|
+
#### twitterscraper-ruby
|
|
17
19
|
|
|
18
20
|
- The number of tweets: Unlimited
|
|
19
21
|
- The time window: from 2006-3-21 to today
|
|
@@ -30,48 +32,98 @@ $ gem install twitterscraper-ruby
|
|
|
30
32
|
|
|
31
33
|
## Usage
|
|
32
34
|
|
|
33
|
-
Command-line interface:
|
|
35
|
+
#### Command-line interface:
|
|
36
|
+
|
|
37
|
+
Returns a collection of relevant tweets matching a specified query.
|
|
34
38
|
|
|
35
39
|
```shell script
|
|
36
|
-
$ twitterscraper --query KEYWORD --start_date 2020-06-01 --end_date 2020-06-30 --lang ja \
|
|
37
|
-
--limit 100 --threads 10 --output
|
|
40
|
+
$ twitterscraper --type search --query KEYWORD --start_date 2020-06-01 --end_date 2020-06-30 --lang ja \
|
|
41
|
+
--limit 100 --threads 10 --output tweets.json
|
|
38
42
|
```
|
|
39
43
|
|
|
40
|
-
|
|
44
|
+
Returns a collection of the most recent tweets posted by the user indicated by the screen_name
|
|
45
|
+
|
|
46
|
+
```shell script
|
|
47
|
+
$ twitterscraper --type user --query SCREEN_NAME --limit 100 --output tweets.json
|
|
48
|
+
```
|
|
49
|
+
|
|
50
|
+
#### From Within Ruby:
|
|
41
51
|
|
|
42
52
|
```ruby
|
|
43
53
|
require 'twitterscraper'
|
|
54
|
+
client = Twitterscraper::Client.new(cache: true, proxy: true)
|
|
55
|
+
```
|
|
44
56
|
|
|
45
|
-
|
|
46
|
-
start_date: '2020-06-01',
|
|
47
|
-
end_date: '2020-06-30',
|
|
48
|
-
lang: 'ja',
|
|
49
|
-
limit: 100,
|
|
50
|
-
threads: 10,
|
|
51
|
-
}
|
|
57
|
+
Returns a collection of relevant tweets matching a specified query.
|
|
52
58
|
|
|
53
|
-
|
|
54
|
-
tweets = client.
|
|
59
|
+
```ruby
|
|
60
|
+
tweets = client.search(KEYWORD, start_date: '2020-06-01', end_date: '2020-06-30', lang: 'ja', limit: 100, threads: 10)
|
|
61
|
+
```
|
|
62
|
+
|
|
63
|
+
Returns a collection of the most recent tweets posted by the user indicated by the screen_name
|
|
64
|
+
|
|
65
|
+
```ruby
|
|
66
|
+
tweets = client.user_timeline(SCREEN_NAME, limit: 100)
|
|
67
|
+
```
|
|
68
|
+
|
|
69
|
+
|
|
70
|
+
## Examples
|
|
71
|
+
|
|
72
|
+
```shell script
|
|
73
|
+
$ twitterscraper --query twitter --limit 1000
|
|
74
|
+
$ cat tweets.json | jq . | less
|
|
75
|
+
```
|
|
76
|
+
|
|
77
|
+
|
|
78
|
+
## Attributes
|
|
55
79
|
|
|
80
|
+
### Tweet
|
|
81
|
+
|
|
82
|
+
```ruby
|
|
56
83
|
tweets.each do |tweet|
|
|
57
84
|
puts tweet.tweet_id
|
|
58
85
|
puts tweet.text
|
|
59
86
|
puts tweet.tweet_url
|
|
60
87
|
puts tweet.created_at
|
|
61
88
|
|
|
89
|
+
attr_names = hash.keys
|
|
62
90
|
hash = tweet.attrs
|
|
63
|
-
|
|
91
|
+
json = tweet.to_json
|
|
64
92
|
end
|
|
65
93
|
```
|
|
66
94
|
|
|
67
|
-
|
|
68
|
-
|
|
69
|
-
|
|
70
|
-
|
|
95
|
+
```json
|
|
96
|
+
[
|
|
97
|
+
{
|
|
98
|
+
"screen_name": "@name",
|
|
99
|
+
"name": "Name",
|
|
100
|
+
"user_id": 12340000,
|
|
101
|
+
"profile_image_url": "https://pbs.twimg.com/profile_images/1826000000/0000.png",
|
|
102
|
+
"tweet_id": 1234000000000000,
|
|
103
|
+
"text": "Thanks Twitter!",
|
|
104
|
+
"links": [],
|
|
105
|
+
"hashtags": [],
|
|
106
|
+
"image_urls": [],
|
|
107
|
+
"video_url": null,
|
|
108
|
+
"has_media": null,
|
|
109
|
+
"likes": 10,
|
|
110
|
+
"retweets": 20,
|
|
111
|
+
"replies": 0,
|
|
112
|
+
"is_replied": false,
|
|
113
|
+
"is_reply_to": false,
|
|
114
|
+
"parent_tweet_id": null,
|
|
115
|
+
"reply_to_users": [],
|
|
116
|
+
"tweet_url": "https://twitter.com/name/status/1234000000000000",
|
|
117
|
+
"timestamp": 1594793000,
|
|
118
|
+
"created_at": "2020-07-15 00:00:00 +0000"
|
|
119
|
+
}
|
|
120
|
+
]
|
|
121
|
+
```
|
|
71
122
|
|
|
72
123
|
- screen_name
|
|
73
124
|
- name
|
|
74
125
|
- user_id
|
|
126
|
+
- profile_image_url
|
|
75
127
|
- tweet_id
|
|
76
128
|
- text
|
|
77
129
|
- links
|
|
@@ -110,44 +162,25 @@ end
|
|
|
110
162
|
Search operators documentation is in [Standard search operators](https://developer.twitter.com/en/docs/tweets/rules-and-filtering/overview/standard-operators).
|
|
111
163
|
|
|
112
164
|
|
|
113
|
-
## Examples
|
|
114
|
-
|
|
115
|
-
```shell script
|
|
116
|
-
$ twitterscraper --query twitter --limit 1000
|
|
117
|
-
$ cat tweets.json | jq . | less
|
|
118
|
-
```
|
|
119
|
-
|
|
120
|
-
```json
|
|
121
|
-
[
|
|
122
|
-
{
|
|
123
|
-
"screen_name": "@screenname",
|
|
124
|
-
"name": "name",
|
|
125
|
-
"user_id": 1194529546483000000,
|
|
126
|
-
"tweet_id": 1282659891992000000,
|
|
127
|
-
"tweet_url": "https://twitter.com/screenname/status/1282659891992000000",
|
|
128
|
-
"created_at": "2020-07-13 12:00:00 +0000",
|
|
129
|
-
"text": "Thanks Twitter!"
|
|
130
|
-
}
|
|
131
|
-
]
|
|
132
|
-
```
|
|
133
|
-
|
|
134
165
|
## CLI Options
|
|
135
166
|
|
|
136
|
-
| Option | Description |
|
|
137
|
-
| ------------- | ------------- | ------------- |
|
|
138
|
-
|
|
|
139
|
-
| `--
|
|
140
|
-
| `--
|
|
141
|
-
| `--
|
|
142
|
-
| `--
|
|
143
|
-
| `--
|
|
144
|
-
| `--
|
|
145
|
-
| `--
|
|
146
|
-
| `--
|
|
147
|
-
| `--
|
|
148
|
-
| `--
|
|
149
|
-
| `--
|
|
150
|
-
| `--
|
|
167
|
+
| Option | Type | Description | Value |
|
|
168
|
+
| ------------- | ------------- | ------------- | ------------- |
|
|
169
|
+
| `--help` | string | This option displays a summary of twitterscraper. | |
|
|
170
|
+
| `--type` | string | Specify a search type. | search(default) or user |
|
|
171
|
+
| `--query` | string | Specify a keyword used during the search. | |
|
|
172
|
+
| `--start_date` | string | Used as "since:yyyy-mm-dd for your query. This means "since the date". | |
|
|
173
|
+
| `--end_date` | string | Used as "until:yyyy-mm-dd for your query. This means "before the date". | |
|
|
174
|
+
| `--lang` | string | Retrieve tweets written in a specific language. | |
|
|
175
|
+
| `--limit` | integer | Stop scraping when *at least* the number of tweets indicated with --limit is scraped. | 100 |
|
|
176
|
+
| `--order` | string | Sort a order of the results. | desc(default) or asc |
|
|
177
|
+
| `--threads` | integer | Set the number of threads twitterscraper-ruby should initiate while scraping for your query. | 2 |
|
|
178
|
+
| `--threads_granularity` | string | | auto |
|
|
179
|
+
| `--proxy` | boolean | Scrape https://twitter.com/search via proxies. | true(default) or false |
|
|
180
|
+
| `--cache` | boolean | Enable caching. | true(default) or false |
|
|
181
|
+
| `--format` | string | The format of the output. | json(default) or html |
|
|
182
|
+
| `--output` | string | The name of the output file. | tweets.json |
|
|
183
|
+
| `--verbose` | | Print debug messages. | |
|
|
151
184
|
|
|
152
185
|
|
|
153
186
|
## Contributing
|
data/lib/twitterscraper.rb
CHANGED
data/lib/twitterscraper/cache.rb
CHANGED
|
@@ -4,7 +4,7 @@ require 'digest/md5'
|
|
|
4
4
|
module Twitterscraper
|
|
5
5
|
class Cache
|
|
6
6
|
def initialize()
|
|
7
|
-
@ttl =
|
|
7
|
+
@ttl = 86400 # 1 day
|
|
8
8
|
@dir = 'cache'
|
|
9
9
|
Dir.mkdir(@dir) unless File.exist?(@dir)
|
|
10
10
|
end
|
|
@@ -25,6 +25,12 @@ module Twitterscraper
|
|
|
25
25
|
File.write(file, entry.to_json)
|
|
26
26
|
end
|
|
27
27
|
|
|
28
|
+
def delete(key)
|
|
29
|
+
key = cache_key(key)
|
|
30
|
+
file = File.join(@dir, key)
|
|
31
|
+
File.delete(file) if File.exist?(file)
|
|
32
|
+
end
|
|
33
|
+
|
|
28
34
|
def fetch(key, &block)
|
|
29
35
|
if (value = read(key))
|
|
30
36
|
value
|
data/lib/twitterscraper/cli.rb
CHANGED
|
@@ -16,6 +16,7 @@ module Twitterscraper
|
|
|
16
16
|
print_version || return if print_version?
|
|
17
17
|
|
|
18
18
|
query_options = {
|
|
19
|
+
type: options['type'],
|
|
19
20
|
start_date: options['start_date'],
|
|
20
21
|
end_date: options['end_date'],
|
|
21
22
|
lang: options['lang'],
|
|
@@ -23,19 +24,20 @@ module Twitterscraper
|
|
|
23
24
|
daily_limit: options['daily_limit'],
|
|
24
25
|
order: options['order'],
|
|
25
26
|
threads: options['threads'],
|
|
27
|
+
threads_granularity: options['threads_granularity'],
|
|
26
28
|
}
|
|
27
29
|
client = Twitterscraper::Client.new(cache: options['cache'], proxy: options['proxy'])
|
|
28
30
|
tweets = client.query_tweets(options['query'], query_options)
|
|
29
|
-
export(tweets) unless tweets.empty?
|
|
31
|
+
export(options['query'], tweets) unless tweets.empty?
|
|
30
32
|
end
|
|
31
33
|
|
|
32
|
-
def export(tweets)
|
|
34
|
+
def export(name, tweets)
|
|
33
35
|
write_json = lambda { File.write(options['output'], generate_json(tweets)) }
|
|
34
36
|
|
|
35
37
|
if options['format'] == 'json'
|
|
36
38
|
write_json.call
|
|
37
39
|
elsif options['format'] == 'html'
|
|
38
|
-
File.write('
|
|
40
|
+
File.write(options['output'], Template.new.tweets_embedded_html(name, tweets, options))
|
|
39
41
|
else
|
|
40
42
|
write_json.call
|
|
41
43
|
end
|
|
@@ -59,6 +61,7 @@ module Twitterscraper
|
|
|
59
61
|
'help',
|
|
60
62
|
'v',
|
|
61
63
|
'version',
|
|
64
|
+
'type:',
|
|
62
65
|
'query:',
|
|
63
66
|
'start_date:',
|
|
64
67
|
'end_date:',
|
|
@@ -67,6 +70,7 @@ module Twitterscraper
|
|
|
67
70
|
'daily_limit:',
|
|
68
71
|
'order:',
|
|
69
72
|
'threads:',
|
|
73
|
+
'threads_granularity:',
|
|
70
74
|
'output:',
|
|
71
75
|
'format:',
|
|
72
76
|
'cache:',
|
|
@@ -75,14 +79,16 @@ module Twitterscraper
|
|
|
75
79
|
'verbose',
|
|
76
80
|
)
|
|
77
81
|
|
|
82
|
+
options['type'] ||= 'search'
|
|
78
83
|
options['start_date'] = Query::OLDEST_DATE if options['start_date'] == 'oldest'
|
|
79
84
|
options['lang'] ||= ''
|
|
80
85
|
options['limit'] = (options['limit'] || 100).to_i
|
|
81
86
|
options['daily_limit'] = options['daily_limit'].to_i if options['daily_limit']
|
|
82
|
-
options['threads'] = (options['threads'] ||
|
|
87
|
+
options['threads'] = (options['threads'] || 10).to_i
|
|
88
|
+
options['threads_granularity'] ||= 'auto'
|
|
83
89
|
options['format'] ||= 'json'
|
|
84
90
|
options['order'] ||= 'desc'
|
|
85
|
-
options['output'] ||=
|
|
91
|
+
options['output'] ||= build_output_name(options)
|
|
86
92
|
|
|
87
93
|
options['cache'] = options['cache'] != 'false'
|
|
88
94
|
options['proxy'] = options['proxy'] != 'false'
|
|
@@ -90,6 +96,12 @@ module Twitterscraper
|
|
|
90
96
|
options
|
|
91
97
|
end
|
|
92
98
|
|
|
99
|
+
def build_output_name(options)
|
|
100
|
+
query = ERB::Util.url_encode(options['query'])
|
|
101
|
+
date = [options['start_date'], options['end_date']].select { |val| val && !val.empty? }.join('_')
|
|
102
|
+
[options['type'], 'tweets', date, query].compact.join('_') + '.' + options['format']
|
|
103
|
+
end
|
|
104
|
+
|
|
93
105
|
def initialize_logger
|
|
94
106
|
Twitterscraper.logger.level = ::Logger::DEBUG if options['verbose']
|
|
95
107
|
end
|
data/lib/twitterscraper/query.rb
CHANGED
|
@@ -22,23 +22,24 @@ module Twitterscraper
|
|
|
22
22
|
RELOAD_URL = 'https://twitter.com/i/search/timeline?f=tweets&vertical=' +
|
|
23
23
|
'default&include_available_features=1&include_entities=1&' +
|
|
24
24
|
'reset_error_state=false&src=typd&max_position=__POS__&q=__QUERY__&l=__LANG__'
|
|
25
|
-
INIT_URL_USER = 'https://twitter.com/
|
|
26
|
-
RELOAD_URL_USER = 'https://twitter.com/i/profiles/show/
|
|
25
|
+
INIT_URL_USER = 'https://twitter.com/__USER__'
|
|
26
|
+
RELOAD_URL_USER = 'https://twitter.com/i/profiles/show/__USER__/timeline/tweets?' +
|
|
27
27
|
'include_available_features=1&include_entities=1&' +
|
|
28
|
-
'max_position=
|
|
29
|
-
|
|
30
|
-
def build_query_url(query, lang,
|
|
31
|
-
|
|
32
|
-
|
|
33
|
-
|
|
34
|
-
|
|
35
|
-
|
|
36
|
-
|
|
37
|
-
# end
|
|
38
|
-
if pos
|
|
39
|
-
RELOAD_URL.sub('__QUERY__', query).sub('__LANG__', lang.to_s).sub('__POS__', pos)
|
|
28
|
+
'max_position=__POS__&reset_error_state=false'
|
|
29
|
+
|
|
30
|
+
def build_query_url(query, lang, type, pos)
|
|
31
|
+
if type.user?
|
|
32
|
+
if pos
|
|
33
|
+
RELOAD_URL_USER.sub('__USER__', query).sub('__POS__', pos.to_s)
|
|
34
|
+
else
|
|
35
|
+
INIT_URL_USER.sub('__USER__', query)
|
|
36
|
+
end
|
|
40
37
|
else
|
|
41
|
-
|
|
38
|
+
if pos
|
|
39
|
+
RELOAD_URL.sub('__QUERY__', query).sub('__LANG__', lang.to_s).sub('__POS__', pos)
|
|
40
|
+
else
|
|
41
|
+
INIT_URL.sub('__QUERY__', query).sub('__LANG__', lang.to_s)
|
|
42
|
+
end
|
|
42
43
|
end
|
|
43
44
|
end
|
|
44
45
|
|
|
@@ -50,7 +51,7 @@ module Twitterscraper
|
|
|
50
51
|
end
|
|
51
52
|
Http.get(url, headers, proxy, timeout)
|
|
52
53
|
rescue => e
|
|
53
|
-
logger.debug "
|
|
54
|
+
logger.debug "get_single_page: #{e.inspect}"
|
|
54
55
|
if (retries -= 1) > 0
|
|
55
56
|
logger.info "Retrying... (Attempts left: #{retries - 1})"
|
|
56
57
|
retry
|
|
@@ -68,26 +69,25 @@ module Twitterscraper
|
|
|
68
69
|
else
|
|
69
70
|
json_resp = JSON.parse(text)
|
|
70
71
|
items_html = json_resp['items_html'] || ''
|
|
71
|
-
logger.warn json_resp['message'] if json_resp['message'] # Sorry, you are rate limited.
|
|
72
72
|
end
|
|
73
73
|
|
|
74
74
|
[items_html, json_resp]
|
|
75
75
|
end
|
|
76
76
|
|
|
77
|
-
def query_single_page(query, lang,
|
|
77
|
+
def query_single_page(query, lang, type, pos, headers: [], proxies: [])
|
|
78
78
|
logger.info "Querying #{query}"
|
|
79
79
|
query = ERB::Util.url_encode(query)
|
|
80
80
|
|
|
81
|
-
url = build_query_url(query, lang,
|
|
81
|
+
url = build_query_url(query, lang, type, pos)
|
|
82
82
|
http_request = lambda do
|
|
83
|
-
logger.debug "Scraping tweets from
|
|
83
|
+
logger.debug "Scraping tweets from url=#{url}"
|
|
84
84
|
get_single_page(url, headers, proxies)
|
|
85
85
|
end
|
|
86
86
|
|
|
87
87
|
if cache_enabled?
|
|
88
88
|
client = Cache.new
|
|
89
89
|
if (response = client.read(url))
|
|
90
|
-
logger.debug
|
|
90
|
+
logger.debug "Fetching tweets from cache url=#{url}"
|
|
91
91
|
else
|
|
92
92
|
response = http_request.call
|
|
93
93
|
client.write(url, response) unless stop_requested?
|
|
@@ -99,6 +99,12 @@ module Twitterscraper
|
|
|
99
99
|
|
|
100
100
|
html, json_resp = parse_single_page(response, pos.nil?)
|
|
101
101
|
|
|
102
|
+
if json_resp && json_resp['message']
|
|
103
|
+
logger.warn json_resp['message'] # Sorry, you are rate limited.
|
|
104
|
+
@stop_requested = true
|
|
105
|
+
Cache.new.delete(url) if cache_enabled?
|
|
106
|
+
end
|
|
107
|
+
|
|
102
108
|
tweets = Tweet.from_html(html)
|
|
103
109
|
|
|
104
110
|
if tweets.empty?
|
|
@@ -107,8 +113,8 @@ module Twitterscraper
|
|
|
107
113
|
|
|
108
114
|
if json_resp
|
|
109
115
|
[tweets, json_resp['min_position']]
|
|
110
|
-
elsif
|
|
111
|
-
|
|
116
|
+
elsif type.user?
|
|
117
|
+
[tweets, tweets[-1].tweet_id]
|
|
112
118
|
else
|
|
113
119
|
[tweets, "TWEET-#{tweets[-1].tweet_id}-#{tweets[0].tweet_id}"]
|
|
114
120
|
end
|
|
@@ -116,7 +122,7 @@ module Twitterscraper
|
|
|
116
122
|
|
|
117
123
|
OLDEST_DATE = Date.parse('2006-03-21')
|
|
118
124
|
|
|
119
|
-
def validate_options!(queries, start_date:, end_date:, lang:, limit:, threads:)
|
|
125
|
+
def validate_options!(queries, type:, start_date:, end_date:, lang:, limit:, threads:)
|
|
120
126
|
query = queries[0]
|
|
121
127
|
if query.nil? || query == ''
|
|
122
128
|
raise Error.new('Please specify a search query.')
|
|
@@ -139,19 +145,33 @@ module Twitterscraper
|
|
|
139
145
|
raise Error.new(":start_date must be greater than or equal to #{OLDEST_DATE}")
|
|
140
146
|
end
|
|
141
147
|
end
|
|
142
|
-
|
|
143
|
-
if end_date
|
|
144
|
-
today = Date.today
|
|
145
|
-
if end_date > Date.today
|
|
146
|
-
raise Error.new(":end_date must be less than or equal to today(#{today})")
|
|
147
|
-
end
|
|
148
|
-
end
|
|
149
148
|
end
|
|
150
149
|
|
|
151
|
-
def build_queries(query, start_date, end_date)
|
|
150
|
+
def build_queries(query, start_date, end_date, threads_granularity)
|
|
152
151
|
if start_date && end_date
|
|
153
|
-
|
|
154
|
-
|
|
152
|
+
if threads_granularity == 'auto'
|
|
153
|
+
threads_granularity = start_date.upto(end_date - 1).to_a.size >= 28 ? 'day' : 'hour'
|
|
154
|
+
end
|
|
155
|
+
|
|
156
|
+
if threads_granularity == 'day'
|
|
157
|
+
date_range = start_date.upto(end_date - 1)
|
|
158
|
+
queries = date_range.map { |date| query + " since:#{date} until:#{date + 1}" }
|
|
159
|
+
else
|
|
160
|
+
time = Time.utc(start_date.year, start_date.month, start_date.day, 0, 0, 0)
|
|
161
|
+
end_time = Time.utc(end_date.year, end_date.month, end_date.day, 0, 0, 0)
|
|
162
|
+
queries = []
|
|
163
|
+
|
|
164
|
+
while true
|
|
165
|
+
if time < Time.now.utc
|
|
166
|
+
queries << (query + " since:#{time.strftime('%Y-%m-%d_%H:00:00')}_UTC until:#{(time + 3600).strftime('%Y-%m-%d_%H:00:00')}_UTC")
|
|
167
|
+
end
|
|
168
|
+
time += 3600
|
|
169
|
+
break if time >= end_time
|
|
170
|
+
end
|
|
171
|
+
end
|
|
172
|
+
|
|
173
|
+
queries
|
|
174
|
+
|
|
155
175
|
elsif start_date
|
|
156
176
|
[query + " since:#{start_date}"]
|
|
157
177
|
elsif end_date
|
|
@@ -161,12 +181,12 @@ module Twitterscraper
|
|
|
161
181
|
end
|
|
162
182
|
end
|
|
163
183
|
|
|
164
|
-
def main_loop(query, lang, limit, daily_limit, headers, proxies)
|
|
184
|
+
def main_loop(query, lang, type, limit, daily_limit, headers, proxies)
|
|
165
185
|
pos = nil
|
|
166
186
|
daily_tweets = []
|
|
167
187
|
|
|
168
188
|
while true
|
|
169
|
-
new_tweets, new_pos = query_single_page(query, lang, pos, headers: headers, proxies: proxies)
|
|
189
|
+
new_tweets, new_pos = query_single_page(query, lang, type, pos, headers: headers, proxies: proxies)
|
|
170
190
|
unless new_tweets.empty?
|
|
171
191
|
daily_tweets.concat(new_tweets)
|
|
172
192
|
daily_tweets.uniq! { |t| t.tweet_id }
|
|
@@ -195,12 +215,18 @@ module Twitterscraper
|
|
|
195
215
|
@stop_requested
|
|
196
216
|
end
|
|
197
217
|
|
|
198
|
-
def query_tweets(query, start_date: nil, end_date: nil, lang:
|
|
199
|
-
|
|
200
|
-
|
|
201
|
-
|
|
218
|
+
def query_tweets(query, type: 'search', start_date: nil, end_date: nil, lang: nil, limit: 100, daily_limit: nil, order: 'desc', threads: 10, threads_granularity: 'auto')
|
|
219
|
+
type = Type.new(type)
|
|
220
|
+
if type.search?
|
|
221
|
+
start_date = Date.parse(start_date) if start_date && start_date.is_a?(String)
|
|
222
|
+
end_date = Date.parse(end_date) if end_date && end_date.is_a?(String)
|
|
223
|
+
elsif type.user?
|
|
224
|
+
start_date = nil
|
|
225
|
+
end_date = nil
|
|
226
|
+
end
|
|
227
|
+
|
|
228
|
+
queries = build_queries(query, start_date, end_date, threads_granularity)
|
|
202
229
|
if threads > queries.size
|
|
203
|
-
logger.warn 'The maximum number of :threads is the number of dates between :start_date and :end_date.'
|
|
204
230
|
threads = queries.size
|
|
205
231
|
end
|
|
206
232
|
if proxy_enabled?
|
|
@@ -212,9 +238,9 @@ module Twitterscraper
|
|
|
212
238
|
end
|
|
213
239
|
logger.debug "Cache #{cache_enabled? ? 'enabled' : 'disabled'}"
|
|
214
240
|
|
|
241
|
+
validate_options!(queries, type: type, start_date: start_date, end_date: end_date, lang: lang, limit: limit, threads: threads)
|
|
215
242
|
|
|
216
|
-
|
|
217
|
-
|
|
243
|
+
logger.info "The number of queries #{queries.size}"
|
|
218
244
|
logger.info "The number of threads #{threads}"
|
|
219
245
|
|
|
220
246
|
headers = {'User-Agent': USER_AGENT_LIST.sample, 'X-Requested-With': 'XMLHttpRequest'}
|
|
@@ -229,17 +255,27 @@ module Twitterscraper
|
|
|
229
255
|
logger.debug "Set 'Thread.abort_on_exception' to true"
|
|
230
256
|
|
|
231
257
|
Parallel.each(queries, in_threads: threads) do |query|
|
|
232
|
-
main_loop(query, lang, limit, daily_limit, headers, proxies)
|
|
258
|
+
main_loop(query, lang, type, limit, daily_limit, headers, proxies)
|
|
233
259
|
raise Parallel::Break if stop_requested?
|
|
234
260
|
end
|
|
235
261
|
else
|
|
236
262
|
queries.each do |query|
|
|
237
|
-
main_loop(query, lang, limit, daily_limit, headers, proxies)
|
|
263
|
+
main_loop(query, lang, type, limit, daily_limit, headers, proxies)
|
|
238
264
|
break if stop_requested?
|
|
239
265
|
end
|
|
240
266
|
end
|
|
241
267
|
|
|
268
|
+
logger.info "Return #{@all_tweets.size} tweets"
|
|
269
|
+
|
|
242
270
|
@all_tweets.sort_by { |tweet| (order == 'desc' ? -1 : 1) * tweet.created_at.to_i }
|
|
243
271
|
end
|
|
272
|
+
|
|
273
|
+
def search(query, start_date: nil, end_date: nil, lang: '', limit: 100, daily_limit: nil, order: 'desc', threads: 10, threads_granularity: 'auto')
|
|
274
|
+
query_tweets(query, type: 'search', start_date: start_date, end_date: end_date, lang: lang, limit: limit, daily_limit: daily_limit, order: order, threads: threads, threads_granularity: threads_granularity)
|
|
275
|
+
end
|
|
276
|
+
|
|
277
|
+
def user_timeline(screen_name, limit: 100, order: 'desc')
|
|
278
|
+
query_tweets(screen_name, type: 'user', start_date: nil, end_date: nil, lang: nil, limit: limit, daily_limit: nil, order: order, threads: 1, threads_granularity: nil)
|
|
279
|
+
end
|
|
244
280
|
end
|
|
245
281
|
end
|
|
@@ -1,48 +1,30 @@
|
|
|
1
1
|
module Twitterscraper
|
|
2
|
-
|
|
3
|
-
|
|
2
|
+
class Template
|
|
3
|
+
def tweets_embedded_html(name, tweets, options)
|
|
4
|
+
path = File.join(File.dirname(__FILE__), 'template/tweets.html.erb')
|
|
5
|
+
template = ERB.new(File.read(path))
|
|
4
6
|
|
|
5
|
-
|
|
6
|
-
|
|
7
|
-
|
|
7
|
+
template.result_with_hash(
|
|
8
|
+
chart_name: name,
|
|
9
|
+
chart_data: chart_data(tweets).to_json,
|
|
10
|
+
first_tweet: tweets.sort_by { |t| t.created_at.to_i }[0],
|
|
11
|
+
last_tweet: tweets.sort_by { |t| t.created_at.to_i }[-1],
|
|
12
|
+
tweets: tweets,
|
|
13
|
+
convert_limit: 30,
|
|
14
|
+
)
|
|
8
15
|
end
|
|
9
16
|
|
|
10
|
-
|
|
11
|
-
|
|
12
|
-
|
|
13
|
-
|
|
14
|
-
|
|
17
|
+
def chart_data(tweets)
|
|
18
|
+
data = tweets.each_with_object(Hash.new(0)) do |tweet, memo|
|
|
19
|
+
t = tweet.created_at
|
|
20
|
+
min = (t.min.to_f / 5).floor * 5
|
|
21
|
+
time = Time.new(t.year, t.month, t.day, t.hour, min, 0, '+00:00')
|
|
22
|
+
memo[time.to_i] += 1
|
|
23
|
+
end
|
|
15
24
|
|
|
16
|
-
|
|
17
|
-
|
|
18
|
-
|
|
19
|
-
|
|
20
|
-
.twitter-tweet {
|
|
21
|
-
margin: 30px auto 0 auto !important;
|
|
22
|
-
}
|
|
23
|
-
</style>
|
|
24
|
-
<script>
|
|
25
|
-
window.twttr = (function(d, s, id) {
|
|
26
|
-
var js, fjs = d.getElementsByTagName(s)[0], t = window.twttr || {};
|
|
27
|
-
if (d.getElementById(id)) return t;
|
|
28
|
-
js = d.createElement(s);
|
|
29
|
-
js.id = id;
|
|
30
|
-
js.src = "https://platform.twitter.com/widgets.js";
|
|
31
|
-
fjs.parentNode.insertBefore(js, fjs);
|
|
32
|
-
|
|
33
|
-
t._e = [];
|
|
34
|
-
t.ready = function(f) {
|
|
35
|
-
t._e.push(f);
|
|
36
|
-
};
|
|
37
|
-
|
|
38
|
-
return t;
|
|
39
|
-
}(document, "script", "twitter-wjs"));
|
|
40
|
-
</script>
|
|
41
|
-
</head>
|
|
42
|
-
<body>
|
|
43
|
-
__TWEETS__
|
|
44
|
-
</body>
|
|
45
|
-
</html>
|
|
46
|
-
HTML
|
|
25
|
+
data.sort_by { |k, v| k }.map do |timestamp, count|
|
|
26
|
+
[timestamp * 1000, count]
|
|
27
|
+
end
|
|
28
|
+
end
|
|
47
29
|
end
|
|
48
30
|
end
|
|
@@ -0,0 +1,98 @@
|
|
|
1
|
+
<html>
|
|
2
|
+
<head>
|
|
3
|
+
<script src="https://cdnjs.cloudflare.com/ajax/libs/moment.js/2.27.0/moment.min.js" integrity="sha512-rmZcZsyhe0/MAjquhTgiUcb4d9knaFc7b5xAfju483gbEXTkeJRUMIPk6s3ySZMYUHEcjKbjLjyddGWMrNEvZg==" crossorigin="anonymous"></script>
|
|
4
|
+
<script src="https://cdnjs.cloudflare.com/ajax/libs/moment-timezone/0.5.31/moment-timezone-with-data.min.js" integrity="sha512-HZcf3uHWA+Y2P5KNv+F/xa87/flKVP92kUTe/KXjU8URPshczF1Dx+cL5bw0VBGhmqWAK0UbhcqxBbyiNtAnWQ==" crossorigin="anonymous"></script>
|
|
5
|
+
<script src="https://code.highcharts.com/stock/highstock.js"></script>
|
|
6
|
+
<script>
|
|
7
|
+
function updateTweets() {
|
|
8
|
+
window.twttr = (function (d, s, id) {
|
|
9
|
+
var js, fjs = d.getElementsByTagName(s)[0], t = window.twttr || {};
|
|
10
|
+
if (d.getElementById(id)) return t;
|
|
11
|
+
js = d.createElement(s);
|
|
12
|
+
js.id = id;
|
|
13
|
+
js.src = "https://platform.twitter.com/widgets.js";
|
|
14
|
+
fjs.parentNode.insertBefore(js, fjs);
|
|
15
|
+
|
|
16
|
+
t._e = [];
|
|
17
|
+
t.ready = function (f) {
|
|
18
|
+
t._e.push(f);
|
|
19
|
+
};
|
|
20
|
+
|
|
21
|
+
return t;
|
|
22
|
+
}(document, "script", "twitter-wjs"));
|
|
23
|
+
}
|
|
24
|
+
|
|
25
|
+
function drawChart() {
|
|
26
|
+
var data = <%= chart_data %>;
|
|
27
|
+
Highcharts.setOptions({
|
|
28
|
+
time: {
|
|
29
|
+
timezone: moment.tz.guess()
|
|
30
|
+
}
|
|
31
|
+
});
|
|
32
|
+
|
|
33
|
+
Highcharts.stockChart('chart', {
|
|
34
|
+
title: {
|
|
35
|
+
text: '<%= tweets.size %> tweets of <%= chart_name %>'
|
|
36
|
+
},
|
|
37
|
+
subtitle: {
|
|
38
|
+
text: 'since:<%= first_tweet.created_at.localtime.strftime('%Y-%m-%d %H:%M') %> until:<%= last_tweet.created_at.localtime.strftime('%Y-%m-%d %H:%M') %>'
|
|
39
|
+
},
|
|
40
|
+
series: [{
|
|
41
|
+
data: data
|
|
42
|
+
}],
|
|
43
|
+
rangeSelector: {enabled: false},
|
|
44
|
+
scrollbar: {enabled: false},
|
|
45
|
+
navigator: {enabled: false},
|
|
46
|
+
exporting: {enabled: false},
|
|
47
|
+
credits: {enabled: false}
|
|
48
|
+
});
|
|
49
|
+
}
|
|
50
|
+
|
|
51
|
+
document.addEventListener("DOMContentLoaded", function () {
|
|
52
|
+
drawChart();
|
|
53
|
+
updateTweets();
|
|
54
|
+
});
|
|
55
|
+
</script>
|
|
56
|
+
|
|
57
|
+
<style type=text/css>
|
|
58
|
+
.tweets-container {
|
|
59
|
+
max-width: 550px;
|
|
60
|
+
margin: 0 auto 0 auto;
|
|
61
|
+
}
|
|
62
|
+
|
|
63
|
+
.twitter-tweet {
|
|
64
|
+
margin: 15px 0 15px 0 !important;
|
|
65
|
+
}
|
|
66
|
+
</style>
|
|
67
|
+
</head>
|
|
68
|
+
<body>
|
|
69
|
+
<div id="chart" style="width: 100vw; height: 400px;"></div>
|
|
70
|
+
|
|
71
|
+
<div class="tweets-container">
|
|
72
|
+
<% tweets.each.with_index do |tweet, i| %>
|
|
73
|
+
<% tweet_time = tweet.created_at.localtime.strftime('%Y-%m-%d %H:%M') %>
|
|
74
|
+
<% if i < convert_limit %>
|
|
75
|
+
<blockquote class="twitter-tweet">
|
|
76
|
+
<% else %>
|
|
77
|
+
<div class="twitter-tweet" style="border: 1px solid rgb(204, 214, 221);">
|
|
78
|
+
<% end %>
|
|
79
|
+
|
|
80
|
+
<div style="display: grid; grid-template-rows: 24px 24px; grid-template-columns: 48px 1fr;">
|
|
81
|
+
<div style="grid-row: 1/3; grid-column: 1/2;"><img src="<%= tweet.profile_image_url %>" width="48" height="48" loading="lazy"></div>
|
|
82
|
+
<div style="grid-row: 1/2; grid-column: 2/3;"><%= tweet.name %></div>
|
|
83
|
+
<div style="grid-row: 2/3; grid-column: 2/3;"><a href="https://twitter.com/<%= tweet.screen_name %>">@<%= tweet.screen_name %></a></div>
|
|
84
|
+
</div>
|
|
85
|
+
|
|
86
|
+
<div><%= tweet.text %></div>
|
|
87
|
+
<div><a href="<%= tweet.tweet_url %>"><small><%= tweet_time %></small></a></div>
|
|
88
|
+
|
|
89
|
+
<% if i < convert_limit %>
|
|
90
|
+
</blockquote>
|
|
91
|
+
<% else %>
|
|
92
|
+
</div>
|
|
93
|
+
<% end %>
|
|
94
|
+
<% end %>
|
|
95
|
+
</div>
|
|
96
|
+
|
|
97
|
+
</body>
|
|
98
|
+
</html>
|
data/lib/twitterscraper/tweet.rb
CHANGED
|
@@ -6,6 +6,7 @@ module Twitterscraper
|
|
|
6
6
|
:screen_name,
|
|
7
7
|
:name,
|
|
8
8
|
:user_id,
|
|
9
|
+
:profile_image_url,
|
|
9
10
|
:tweet_id,
|
|
10
11
|
:text,
|
|
11
12
|
:links,
|
|
@@ -51,6 +52,11 @@ module Twitterscraper
|
|
|
51
52
|
end
|
|
52
53
|
end
|
|
53
54
|
|
|
55
|
+
# .js-stream-item
|
|
56
|
+
# .js-stream-tweet{data: {screen-name:, tweet-id:}}
|
|
57
|
+
# .stream-item-header
|
|
58
|
+
# .js-tweet-text-container
|
|
59
|
+
# .stream-item-footer
|
|
54
60
|
def from_html(text)
|
|
55
61
|
html = Nokogiri::HTML(text)
|
|
56
62
|
from_tweets_html(html.xpath("//li[@class[contains(., 'js-stream-item')]]/div[@class[contains(., 'js-stream-tweet')]]"))
|
|
@@ -72,6 +78,8 @@ module Twitterscraper
|
|
|
72
78
|
end
|
|
73
79
|
|
|
74
80
|
inner_html = Nokogiri::HTML(html.inner_html)
|
|
81
|
+
|
|
82
|
+
profile_image_url = inner_html.xpath("//img[@class[contains(., 'js-action-profile-avatar')]]").first.attr('src').gsub(/_bigger/, '')
|
|
75
83
|
text = inner_html.xpath("//div[@class[contains(., 'js-tweet-text-container')]]/p[@class[contains(., 'js-tweet-text')]]").first.text
|
|
76
84
|
links = inner_html.xpath("//a[@class[contains(., 'twitter-timeline-link')]]").map { |elem| elem.attr('data-expanded-url') }.select { |link| link && !link.include?('pic.twitter') }
|
|
77
85
|
image_urls = inner_html.xpath("//div[@class[contains(., 'AdaptiveMedia-photoContainer')]]").map { |elem| elem.attr('data-image-url') }
|
|
@@ -99,6 +107,7 @@ module Twitterscraper
|
|
|
99
107
|
screen_name: screen_name,
|
|
100
108
|
name: html.attr('data-name'),
|
|
101
109
|
user_id: html.attr('data-user-id').to_i,
|
|
110
|
+
profile_image_url: profile_image_url,
|
|
102
111
|
tweet_id: tweet_id,
|
|
103
112
|
text: text,
|
|
104
113
|
links: links,
|
data/lib/version.rb
CHANGED
metadata
CHANGED
|
@@ -1,14 +1,14 @@
|
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
|
2
2
|
name: twitterscraper-ruby
|
|
3
3
|
version: !ruby/object:Gem::Version
|
|
4
|
-
version: 0.
|
|
4
|
+
version: 0.17.0
|
|
5
5
|
platform: ruby
|
|
6
6
|
authors:
|
|
7
7
|
- ts-3156
|
|
8
8
|
autorequire:
|
|
9
9
|
bindir: bin
|
|
10
10
|
cert_chain: []
|
|
11
|
-
date: 2020-07-
|
|
11
|
+
date: 2020-07-18 00:00:00.000000000 Z
|
|
12
12
|
dependencies:
|
|
13
13
|
- !ruby/object:Gem::Dependency
|
|
14
14
|
name: nokogiri
|
|
@@ -72,7 +72,9 @@ files:
|
|
|
72
72
|
- lib/twitterscraper/proxy.rb
|
|
73
73
|
- lib/twitterscraper/query.rb
|
|
74
74
|
- lib/twitterscraper/template.rb
|
|
75
|
+
- lib/twitterscraper/template/tweets.html.erb
|
|
75
76
|
- lib/twitterscraper/tweet.rb
|
|
77
|
+
- lib/twitterscraper/type.rb
|
|
76
78
|
- lib/version.rb
|
|
77
79
|
- twitterscraper-ruby.gemspec
|
|
78
80
|
homepage: https://github.com/ts-3156/twitterscraper-ruby
|