-
Notifications
You must be signed in to change notification settings - Fork 577
/
changelog.txt
225 lines (178 loc) · 8.33 KB
/
changelog.txt
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
# twitterscraper changelog
# 1.6.1 ( 2020-07-28 )
## Fixed
- Issue 330: Added KeyError to the try / except so that it no longer breaks when json_resp does not have this key.
# 1.6.0 ( 2020-07-22 )
## Added
- PR234: Adds command line argument -dp or --disableproxy to disable to use of proxy when querying.
## Improved
- PR261: Improve logging; there is no ts_logger file, logger is initiated in main.py and query.py, loglevel is set via CLI.
# 1.5.0 ( 2020-07-22 )
## Fixed
- PR304: Fixed query.py by adding 'X-Requested-With': 'XMLHttpRequest' to header value.
- PR253: Fixed Docker build
## Added
- PR313: Added example to README (section 2.3.1).
- PR277: Support emojis by adding the alt text of images to the tweet text.
# 1.4.0 ( 2019-11-03 )
## Fixed
- PR228: Fixed Typo in Readme
- PR224: Force CSV quoting for all non-numeric values
## Added
- PR213: Added Dockerfile for Docker support
- PR220: Passed timeout value of 60s from method to requests.get()
- PR231: Added a lot of tweet attributes to the output, regarding links, media and replies.
- PR233: Added support for searching for the '&' sign.
## Improved
- PR223: Pretty printing the output which is dumped
# 1.3.1 ( 2019-09-07 )
## Fixed
- Change two uses of f-strings to .format() since f-strings only work well with Python 3.6+
# 1.3.0 ( 2019-09-07 )
## Added
- Added the use of proxies while making an request.
- PR #204: Added a max timeout to twitterscraper requests which is set to 60s by default.
# 1.2.1 ( 2019-08-06 )
### Fixed
- PR #208: Fixed a type in a print statement which was breaking down twitterscraper
- Remove the use of fake_useragent library
# 1.2.0 ( 2019-06-22 )
### Added
- PR #186: adds the fields is_retweet, retweeter related information, and timestamp_epochs to the output.
- PR #184: use fake_useragent for generation of random user agent headers.
- Additionally scraper for 'is_verified' when scraping for user profile pages.
# 1.1.0 ( 2019-06-15 )
### Added
- PR #176: Using billiard library instead of multiprocessing to add the ability to use this library with Celery.
# 1.0.1 ( 2019-06-15 )
### Fixed
- PR #191: wrong argument was used in the method query_tweets_from_user()
- CSV output file has as default ";" as a separator.
- PR #173: Some small improvements on the profile page scraping.
### Added
- Command line argument -ow / --overwrite to indicate if an existing output file should be overwritten.
# 1.0.0 ( 2019-02-04 )
### Added
- PR #159: scrapes user profile pages for additional information.
### Fixed:
- Moved example scripts demonstrating use of get_user_info() functionality to examples folder
- removed screenshot demonstrating get_user_info() works
- Added command line argument to main.py which calls get_user_info() for all users in list of scraped tweets.
# 0.9.3 ( 2018-11-04 )
### Fixed
- PR #143: cancels query if end-date is earlier than begin-date.
- PR #151: returned json_resp['min_position] is parsed in order to quote special characters.
- PR #153: cast Tweets attributes to proper data types (int instead of str)
- Use codecs.open() to write to file. Should fix issues 144 and 147.
# 0.9.0 ( 2018-07-18 )
### Added
- Added -u / --user command line argument which can be used to scrape all
tweets from an users profile page.
## 0.8.1 ( 2018-07-18 )
- saving .csv files as an utf-8 encoded file. This fixes https://github.com/taspinar/twitterscraper/issues/138
## 0.8.0 ( 2018-07-17 )
### Fixed
- remove two headers which caused bad fetching results https://github.com/taspinar/twitterscraper/issues/126#issuecomment-405132147
- fix python2 logger bug https://github.com/taspinar/twitterscraper/issues/134 https://github.com/taspinar/twitterscraper/issues/132 https://github.com/taspinar/twitterscraper/issues/127
### Improved
- Use a generator to get tweets, but convert to list in `query_tweets_once`
- this is useful for low memory applications, like massively parallelizing twitter scraping through AWS Lambda (128MB RAM)
- use single quotes for all strings (it was inconsistent prior)
- pep8 compliance on L28
### Removed
- remove `eliminate_duplicates` dead code
# 0.7.2 ( 2018-07-09 )
### Fixed
- twitterscraper.logging is imported as logger instead of logging in order to
avoid a module name clash with Python2's logging module.
# 0.7.1 ( 2018-06-12 )
### Improved
- Give access to logger for scripts which import this module. Create the module,
`logging.py`, which contains the logger used by twitterscraper.
### Removed
- fake_useragent is removed as a dependency, since it has been giving
user-agent headers which keep being blocked by Twitter.
## 0.7.0 ( 2018-05-06 )
### Fixed
- By using linspace() instead of range() to divide the number of days into
the number of parallel processes, edge cases ( p = 1 ) now also work fine.
This fixes https://github.com/taspinar/twitterscraper/issues/108.
### Improved
- The default value of begindate is set to 2006-03-21. The previous value (2017-01-01)
was chosen arbitrarily and leaded to questions why not all tweets were retrieved.
This fixes https://github.com/taspinar/twitterscraper/issues/88.
### Added
- Users can now save the tweets in a csv-format, with the arguments "-c" or "--csv"
## 0.6.2 ( 2018-03-21 )
### Fixed
- Errors occuring during the serialization of a non-html response (everything after 1st request),
No longer crashes the program but is catched with a try / except.
- Fixes https://github.com/taspinar/twitterscraper/issues/93
- The '@' character in an username is now removed by the ".strip('\@')" method instead of "[1:]".
- This fixes issue https://github.com/taspinar/twitterscraper/issues/105
## 0.6.1 ( 2018-03-17 )
### Improved
- The way the number of days are divided over the number of parallel processes is improved.
- The maximum number of parallel processes is limited to the max no of days.
- Fixes https://github.com/taspinar/twitterscraper/issues/101
## 0.6.0 ( 2018-02-17 )
### Fixed
- PR #89: closed pools to prevent zombie processes.
## 0.5.1 ( 2018-02-17 )
### Fixed
- Fixed MaxRecursionError crashes which was introduced with version 0.5.0
## 0.5.0 ( 2018-01-11 )
### Added
- Added the html code of a tweet message to the Tweet class as one of its attributes
## 0.4.2 ( 2018-01-09 )
### Fixed
- Fixed backward compatability of the new --lang parameter by placing it at the end of all arguments.
## 0.4.1 ( 2018-01-07 )
### Fixed
- Fixed --lang functionality by passing the lang parameter from its CL argument form to the generater url.
## 0.4 ( 2017-12-19 )
-----------
### Added
- Added "-bd / --begindate" command line arguments to set the begin date of the query
- Added "-ed / --enddate" command line arguments to set the end date of the query.
- Added "-p / --poolsize" command line arguments which can change the number of parallel processes.
Default number of parallel processes is set to 20.
### Improved
- Outputfile is only created if tweets are actually retrieved.
### Removed
- The ´query_all_tweets' method in the Query module is removed. Since twitterscraper is starting parallel processes by default,
this method is no longer necessary.
### Changed
- The 'query_tweets' method now takes as arguments query, limit, begindate, enddate, poolsize.
- The 'query_tweets_once' no longer has the argument 'num_tweets'
- The default value of the 'retry' argument of the 'query_single_page' method has been increased from 3 to 10.
- The ´query_tweets_once' method does not log to screen at every single scrape, but at the end of a batch.
## 0.3.3 ( 2017-12-06 )
-----------
### Added
-PR #61: Adding --lang functionality which can retrieve tweets written in a specific language.
-PR #62: Tweet class now also contains the tweet url. This closes https://github.com/taspinar/twitterscraper/issues/59
## 0.3.2 ( 2017-11-12 )
-----------
### Improved
-PR #55: Adding --dump functionality which dumps the scraped tweets to screen, instead of an outputfile.
## 0.3.1 ( 2017-11-05 )
-----------
### Improved
-PR #49: scraping of replies, retweets and likes is improved.
## 0.3.0 ( 2017-08-01 )
-----------
### Added
- Tweet class now also includes 'replies', 'retweets' and 'likes'
## 0.2.7 ( 2017-01-10 )
-----------
### Improved
- PR #26: use ``requests`` library for HTTP requests. Makes the use of urllib2 / urllib redundant.
### Added:
- changelog.txt for GitHub
- HISTORY.rst for PyPi
- README.rst for PyPi
## 0.2.6 ( 2017-01-02 )
-----------
### Improved
- PR #25: convert date retrieved from timestamp to day precision