zenfyr.dev
social media crossposting tool. 3rd time's the charm
mastodon
misskey
crossposting
bluesky
mastodon input
+258
-5
+3
-1
bluesky/input.py
+3
-1
bluesky/input.py
···
167
167
reposted = self._get_post(self.url, self.did, reposted_uri)
168
168
if not reposted:
169
169
self.log.info(
170
-
"Skipping repost '%s' as reposted post '%s' was not found in the db."
170
+
"Skipping repost '%s' as reposted post '%s' was not found in the db.",
171
+
post_uri,
172
+
reposted_uri,
171
173
)
172
174
return
173
175
+132
-4
mastodon/input.py
+132
-4
mastodon/input.py
···
6
6
7
7
import websockets
8
8
9
+
from cross.attachments import (
10
+
LabelsAttachment,
11
+
LanguagesAttachment,
12
+
MediaAttachment,
13
+
RemoteUrlAttachment,
14
+
SensitiveAttachment,
15
+
)
16
+
from cross.media import Blob, download_blob
17
+
from cross.post import Post
9
18
from cross.service import InputService
10
19
from database.connection import DatabasePool
11
20
from mastodon.info import MastodonService, validate_and_transform
21
+
from mastodon.parser import StatusParser
12
22
13
23
ALLOWED_VISIBILITY: list[str] = ["public", "unlisted"]
14
24
···
55
65
return self.options.token
56
66
57
67
def _on_create_post(self, status: dict[str, Any]):
58
-
self.log.info(status) # TODO
68
+
if status["account"]["id"] != self.user_id:
69
+
return
70
+
71
+
if status["visibility"] not in self.options.allowed_visibility:
72
+
return
73
+
74
+
reblog: dict[str, Any] | None = status.get("reblog")
75
+
if reblog:
76
+
if reblog["account"]["id"] != self.user_id:
77
+
return
78
+
self._on_reblog(status, reblog)
79
+
return
80
+
81
+
if status.get("poll"):
82
+
self.log.info("Skipping '%s'! Contains a poll..", status["id"])
83
+
return
84
+
85
+
if status.get("quote"):
86
+
self.log.info("Skipping '%s'! Quote..", status["id"])
87
+
return
88
+
89
+
in_reply: str | None = status.get("in_reply_to_id")
90
+
in_reply_to: str | None = status.get("in_reply_to_account_id")
91
+
if in_reply_to and in_reply_to != self.user_id:
92
+
return
93
+
94
+
parent = None
95
+
if in_reply:
96
+
parent = self._get_post(self.url, self.user_id, in_reply)
97
+
if not parent:
98
+
self.log.info(
99
+
"Skipping %s, parent %s not found in db", status["id"], in_reply
100
+
)
101
+
return
102
+
parser = StatusParser()
103
+
parser.feed(status["content"])
104
+
text, fragments = parser.get_result()
105
+
106
+
post = Post(id=status["id"], parent_id=in_reply, text=text)
107
+
post.fragments.extend(fragments)
108
+
109
+
if status.get("url"):
110
+
post.attachments.put(RemoteUrlAttachment(url=status["url"]))
111
+
if status.get("sensitive"):
112
+
post.attachments.put(SensitiveAttachment(sensitive=True))
113
+
if status.get("language"):
114
+
post.attachments.put(LanguagesAttachment(langs=[status["language"]]))
115
+
if status.get("spoiler"):
116
+
post.attachments.put(LabelsAttachment(labels=[status["spoiler"]]))
117
+
118
+
blobs: list[Blob] = []
119
+
for media in status.get("media_attachments", []):
120
+
self.log.info("Downloading %s...", media["url"])
121
+
blob: Blob | None = download_blob(media["url"], media.get("alt"))
122
+
if not blob:
123
+
self.log.error(
124
+
"Skipping %s! Failed to download media %s.",
125
+
status["id"],
126
+
media["url"],
127
+
)
128
+
return
129
+
blobs.append(blob)
130
+
131
+
if blobs:
132
+
post.attachments.put(MediaAttachment(blobs=blobs))
133
+
134
+
if parent:
135
+
self._insert_post(
136
+
{
137
+
"user": self.user_id,
138
+
"service": self.url,
139
+
"identifier": status["id"],
140
+
"parent": parent["id"],
141
+
"root": parent["id"] if not parent["root"] else parent["root"],
142
+
}
143
+
)
144
+
else:
145
+
self._insert_post(
146
+
{
147
+
"user": self.user_id,
148
+
"service": self.url,
149
+
"identifier": status["id"],
150
+
}
151
+
)
152
+
153
+
for out in self.outputs:
154
+
self.submitter(lambda: out.accept_post(post))
155
+
156
+
def _on_reblog(self, status: dict[str, Any], reblog: dict[str, Any]):
157
+
reposted = self._get_post(self.url, self.user_id, reblog["id"])
158
+
if not reposted:
159
+
self.log.info(
160
+
"Skipping repost '%s' as reposted post '%s' was not found in the db.",
161
+
status["id"],
162
+
reblog["id"],
163
+
)
164
+
return
165
+
166
+
self._insert_post(
167
+
{
168
+
"user": self.user_id,
169
+
"service": self.url,
170
+
"identifier": status["id"],
171
+
"reposted": reposted["id"],
172
+
}
173
+
)
174
+
175
+
for out in self.outputs:
176
+
self.submitter(lambda: out.accept_repost(status["id"], reposted["id"]))
59
177
60
178
def _on_delete_post(self, status_id: str):
61
-
self.log.info(status_id) # TODO
179
+
post = self._get_post(self.url, self.user_id, status_id)
180
+
if not post:
181
+
return
182
+
183
+
if post["reposted_id"]:
184
+
for output in self.outputs:
185
+
self.submitter(lambda: output.delete_repost(status_id))
186
+
else:
187
+
for output in self.outputs:
188
+
self.submitter(lambda: output.delete_post(status_id))
189
+
self._delete_post_by_id(post["id"])
62
190
63
191
def _accept_msg(self, msg: websockets.Data) -> None:
64
192
data: dict[str, Any] = cast(dict[str, Any], json.loads(msg))
65
-
event: str = cast(str, data['event'])
66
-
payload: str = cast(str, data['payload'])
193
+
event: str = cast(str, data["event"])
194
+
payload: str = cast(str, data["payload"])
67
195
68
196
if event == "update":
69
197
self._on_create_post(json.loads(payload))
+123
mastodon/parser.py
+123
mastodon/parser.py
···
1
+
from html.parser import HTMLParser
2
+
from typing import override
3
+
import cross.fragments as f
4
+
5
+
6
+
class StatusParser(HTMLParser):
7
+
def __init__(self) -> None:
8
+
super().__init__()
9
+
self.text: str = ""
10
+
self.fragments: list[f.Fragment] = []
11
+
12
+
self._tag_stack: dict[str, tuple[int, dict[str, str | None]]] = {}
13
+
self.in_pre: bool = False
14
+
self.in_code: bool = False
15
+
16
+
self.invisible: bool = False
17
+
18
+
@override
19
+
def handle_starttag(self, tag: str, attrs: list[tuple[str, str | None]]) -> None:
20
+
_attr = dict(attrs)
21
+
22
+
def append_newline():
23
+
if self.text and not self.text.endswith("\n"):
24
+
self.text += "\n"
25
+
26
+
if self.invisible:
27
+
return
28
+
29
+
match tag:
30
+
case "p":
31
+
cls = _attr.get('class', '')
32
+
if cls and 'quote-inline' in cls:
33
+
self.invisible = True
34
+
case "a":
35
+
self._tag_stack["a"] = (len(self.text), _attr)
36
+
case "code":
37
+
if not self.in_pre:
38
+
self.text += "`"
39
+
self.in_code = True
40
+
case "pre":
41
+
append_newline()
42
+
self.text += "```\n"
43
+
self.in_pre = True
44
+
case "blockquote":
45
+
append_newline()
46
+
self.text += "> "
47
+
case "strong" | "b":
48
+
self.text += "**"
49
+
case "em" | "i":
50
+
self.text += "*"
51
+
case "del" | "s":
52
+
self.text += "~~"
53
+
case "br":
54
+
self.text += "\n"
55
+
case _:
56
+
if tag in {"h1", "h2", "h3", "h4", "h5", "h6"}:
57
+
level = int(tag[1])
58
+
self.text += "\n" + "#" * level + " "
59
+
60
+
@override
61
+
def handle_endtag(self, tag: str) -> None:
62
+
if self.invisible:
63
+
if tag == "p":
64
+
self.invisible = False
65
+
return
66
+
67
+
current_end = len(self.text)
68
+
match tag:
69
+
case "a":
70
+
if "a" in self._tag_stack:
71
+
start, _attr = self._tag_stack.pop("a")
72
+
73
+
href = _attr.get('href')
74
+
if href and current_end > start:
75
+
cls = _attr.get('class', '')
76
+
if cls:
77
+
if 'hashtag' in cls:
78
+
tag = self.text[start:current_end]
79
+
tag = tag[1:] if tag.startswith('#') else tag
80
+
81
+
self.fragments.append(
82
+
f.TagFragment(start=start, end=current_end, tag=tag)
83
+
)
84
+
return
85
+
if 'mention' in cls: # TODO put the full acct in the fragment
86
+
mention = self.text[start:current_end]
87
+
self.fragments.append(
88
+
f.MentionFragment(start=start, end=current_end, uri=mention)
89
+
)
90
+
return
91
+
self.fragments.append(
92
+
f.LinkFragment(start=start, end=current_end, url=href)
93
+
)
94
+
case "code":
95
+
if not self.in_pre and self.in_code:
96
+
self.text += "`"
97
+
self.in_code = False
98
+
case "pre":
99
+
self.text += "\n```\n"
100
+
self.in_pre = False
101
+
case "blockquote":
102
+
self.text += "\n"
103
+
case "strong" | "b":
104
+
self.text += "**"
105
+
case "em" | "i":
106
+
self.text += "*"
107
+
case "del" | "s":
108
+
self.text += "~~"
109
+
case "p":
110
+
self.text += "\n\n"
111
+
case _:
112
+
if tag in ["h1", "h2", "h3", "h4", "h5", "h6"]:
113
+
self.text += '\n'
114
+
115
+
@override
116
+
def handle_data(self, data: str) -> None:
117
+
if not self.invisible:
118
+
self.text += data
119
+
120
+
def get_result(self) -> tuple[str, list[f.Fragment]]:
121
+
if self.text.endswith('\n\n'):
122
+
return self.text[:-2], self.fragments
123
+
return self.text, self.fragments