]>
Commit | Line | Data |
---|---|---|
1 | # paperdoorknob: Print glowfic | |
2 | # | |
3 | # This program is free software: you can redistribute it and/or modify it | |
4 | # under the terms of the GNU General Public License as published by the | |
5 | # Free Software Foundation, version 3. | |
6 | ||
7 | ||
8 | from dataclasses import dataclass | |
9 | import itertools | |
10 | from urllib.parse import parse_qsl, urlencode, urlparse, urlunparse | |
11 | ||
12 | from typing import Iterable | |
13 | ||
14 | from bs4 import BeautifulSoup | |
15 | from bs4.element import Tag | |
16 | ||
17 | from images import ImageStore | |
18 | from texify import Texifier | |
19 | ||
20 | ||
21 | def _removeViewFromURL(url: str) -> str: | |
22 | u = urlparse(url) | |
23 | old_qs = parse_qsl(u.query) | |
24 | new_qs = [(k, v) for k, v in old_qs if k != 'view'] | |
25 | return urlunparse(u._replace(query=urlencode(new_qs))) | |
26 | ||
27 | ||
28 | def nonFlatURL(url: str) -> str: | |
29 | return _removeViewFromURL(url) | |
30 | ||
31 | ||
32 | def flatURL(url: str) -> str: | |
33 | u = urlparse(_removeViewFromURL(url)) | |
34 | qs = parse_qsl(u.query) + [('view', 'flat')] | |
35 | return urlunparse(u._replace(query=urlencode(qs))) | |
36 | ||
37 | ||
38 | @dataclass(frozen=True) | |
39 | class Chunk: | |
40 | icon: str | None | |
41 | character: Tag | None | |
42 | screen_name: Tag | None | |
43 | author: Tag | None | |
44 | content: Tag | |
45 | ||
46 | # We avoid the name "post" because the Glowfic community uses the term | |
47 | # inconsistently: | |
48 | # * The Glowfic software sometimes uses "post" to refer to a whole thread | |
49 | # (in the URL), sometimes uses "post" to refer to chunks (in the CSS), | |
50 | # but mostly uses "post" to refer to just the first chunk in a thread | |
51 | # (in the HTML and UI). The non-first chunks are "replies". | |
52 | # * Readers and this software don't need to distinguish first-chunks and | |
53 | # non-first-chunks. | |
54 | # * Humans in the community tend to use "posts" to mean chunks. | |
55 | ||
56 | ||
57 | class Thread: | |
58 | ||
59 | def __init__(self, dom: BeautifulSoup) -> None: | |
60 | self._dom = dom | |
61 | ||
62 | def title(self) -> str | None: | |
63 | span = self._dom.findChild("span", id="post-title") | |
64 | if not isinstance(span, Tag): | |
65 | return None | |
66 | return span.text.strip() | |
67 | ||
68 | def chunkDOMs(self) -> Iterable[Tag]: | |
69 | def text() -> Tag: | |
70 | body = self._dom.body | |
71 | assert body | |
72 | text = body.find_next("div", class_="post-post") | |
73 | assert isinstance(text, Tag) | |
74 | return text | |
75 | ||
76 | def the_replies() -> Iterable[Tag]: | |
77 | rs = self._dom.find_all("div", class_="post-reply") | |
78 | assert all(isinstance(r, Tag) for r in rs) | |
79 | return rs | |
80 | ||
81 | return itertools.chain([text()], the_replies()) | |
82 | ||
83 | ||
84 | def makeChunk(chunk_dom: Tag, image_store: ImageStore) -> Chunk: | |
85 | ||
86 | def getIcon() -> str | None: | |
87 | icon_div = chunk_dom.findChild('div', class_='post-icon') | |
88 | if icon_div is None: | |
89 | return None | |
90 | assert isinstance(icon_div, Tag) | |
91 | icon_img = icon_div.findChild('img') | |
92 | if icon_img is None: | |
93 | return None | |
94 | assert isinstance(icon_img, Tag) | |
95 | return image_store.get_image(icon_img.attrs['src']) | |
96 | ||
97 | def getByClass(css_class: str) -> Tag | None: | |
98 | tag = chunk_dom.findChild('div', class_=css_class) | |
99 | assert tag is None or isinstance(tag, Tag) | |
100 | return tag | |
101 | ||
102 | def stripHREF(tag: Tag) -> None: | |
103 | for c in tag.findChildren("a"): | |
104 | if "href" in c.attrs: | |
105 | del c.attrs["href"] | |
106 | ||
107 | def getMeta(css_class: str) -> Tag | None: | |
108 | tag = getByClass(css_class) | |
109 | if tag is None: | |
110 | return None | |
111 | stripHREF(tag) | |
112 | return tag | |
113 | ||
114 | content = chunk_dom.findChild('div', class_='post-content') | |
115 | assert isinstance(content, Tag) | |
116 | ||
117 | return Chunk(getIcon(), | |
118 | getMeta('post-character'), | |
119 | getMeta('post-screenname'), | |
120 | getMeta('post-author'), | |
121 | content) | |
122 | ||
123 | ||
124 | def renderChunk(texifier: Texifier, chunk: Chunk) -> bytes: | |
125 | return b''.join([ | |
126 | br'\glowhead{', | |
127 | br'\glowicon{%s}' % chunk.icon.encode('UTF-8') if chunk.icon else b'', | |
128 | b'}{', | |
129 | texifier.texify(chunk.character) if chunk.character else b'', | |
130 | b'}{', | |
131 | texifier.texify(chunk.screen_name) if chunk.screen_name else b'', | |
132 | b'}{', | |
133 | texifier.texify(chunk.author) if chunk.author else b'', | |
134 | b'}', | |
135 | texifier.texify(chunk.content)]) | |
136 | ||
137 | ||
138 | ContentOnlyLayout = br''' | |
139 | \newcommand{\glowhead}[4]{} | |
140 | ''' | |
141 | ||
142 | ||
143 | BelowIconLayout = br''' | |
144 | \newcommand{\glowhead}[4]{\wrapstuffclear | |
145 | \begin{wrapstuff}[l] | |
146 | \fbox{ | |
147 | \begin{varwidth}{0.5\textwidth} | |
148 | \smash{\parbox[t][0pt]{0pt}{ | |
149 | \setlength{\fboxrule}{0.2pt} | |
150 | \setlength{\fboxsep}{0pt} | |
151 | \vspace{-3.4pt} | |
152 | \fbox{\hspace{107mm}} | |
153 | }\\*} | |
154 | \vspace{-1em} | |
155 | \begin{center} | |
156 | #1\ifnotempty | |
157 | {#1}{\\*}#2\ifnotempty | |
158 | {#2}{\\*}#3\ifnotempty | |
159 | {#3}{\\*}#4 | |
160 | \end{center} | |
161 | \end{varwidth} | |
162 | } | |
163 | \end{wrapstuff} | |
164 | ||
165 | \strut | |
166 | ||
167 | \noindent}''' | |
168 | ||
169 | ||
170 | # Why is \textwidth not the width of the text? | |
171 | # Why is the width of the text .765\textwidth? | |
172 | BesideIconLayout = br''' | |
173 | \newcommand{\glowhead}[4]{ | |
174 | ||
175 | \strut | |
176 | ||
177 | \noindent\fbox{ | |
178 | #1 | |
179 | \parbox[b]{.765\textwidth}{ | |
180 | \begin{center} | |
181 | #2\ifnotempty | |
182 | {#2}{\\*}#3\ifnotempty | |
183 | {#3}{\\*}#4 | |
184 | \end{center} | |
185 | } | |
186 | }\\* | |
187 | \vspace{-0.75em}\\* | |
188 | \noindent}''' |