Use raw strings for less escaping

[paperdoorknob] / glowfic.py

diff --git a/glowfic.py b/glowfic.py
index c969ac055de6f96b921b9a6dc0b489173193cd90..bda6d1941c84eb1b578461439dfa87ada971469a 100644 (file)
--- a/glowfic.py
+++ b/glowfic.py
@@ -8,6 +8,7 @@
 from abc import ABC, abstractmethod
 from dataclasses import dataclass
 import itertools
+from urllib.parse import parse_qsl, urlencode, urlparse, urlunparse
 
 from typing import Iterable
 
@@ -18,12 +19,29 @@ from images import ImageStore
 from texify import Texifier
 
 
+def _removeViewFromURL(url: str) -> str:
+    u = urlparse(url)
+    old_qs = parse_qsl(u.query)
+    new_qs = [(k, v) for k, v in old_qs if k != 'view']
+    return urlunparse(u._replace(query=urlencode(new_qs)))
+
+
+def nonFlatURL(url: str) -> str:
+    return _removeViewFromURL(url)
+
+
+def flatURL(url: str) -> str:
+    u = urlparse(_removeViewFromURL(url))
+    qs = parse_qsl(u.query) + [('view', 'flat')]
+    return urlunparse(u._replace(query=urlencode(qs)))
+
+
 @dataclass(frozen=True)
 class Chunk:
     icon: str | None
-    character: str | None
-    screen_name: str | None
-    author: str | None
+    character: Tag | None
+    screen_name: Tag | None
+    author: Tag | None
     content: Tag
 
 # We avoid the name "post" because the Glowfic community uses the term
@@ -56,34 +74,48 @@ def chunkDOMs(html: BeautifulSoup) -> Iterable[Tag]:
 def makeChunk(chunk_dom: Tag, image_store: ImageStore) -> Chunk:
 
     def getIcon() -> str | None:
-        icon_div = chunk_dom.find_next('div', class_='post-icon')
+        icon_div = chunk_dom.findChild('div', class_='post-icon')
         if icon_div is None:
             return None
-        icon_img = icon_div.find_next('img')
+        assert isinstance(icon_div, Tag)
+        icon_img = icon_div.findChild('img')
         if icon_img is None:
             return None
         assert isinstance(icon_img, Tag)
         return image_store.get_image(icon_img.attrs['src'])
 
-    def getTextByClass(css_class: str) -> str | None:
-        div = chunk_dom.find_next('div', class_=css_class)
-        if div is None:
+    def getByClass(css_class: str) -> Tag | None:
+        tag = chunk_dom.findChild('div', class_=css_class)
+        assert tag is None or isinstance(tag, Tag)
+        return tag
+
+    def stripHREF(tag: Tag) -> None:
+        for c in tag.findChildren("a"):
+            if "href" in c.attrs:
+                del c.attrs["href"]
+
+    def getMeta(css_class: str) -> Tag | None:
+        tag = getByClass(css_class)
+        if tag is None:
             return None
-        return div.text.strip()
+        stripHREF(tag)
+        return tag
 
-    content = chunk_dom.find_next('div', class_='post-content')
+    content = chunk_dom.findChild('div', class_='post-content')
     assert isinstance(content, Tag)
 
     return Chunk(getIcon(),
-                 getTextByClass('post-character'),
-                 getTextByClass('post-screenname'),
-                 getTextByClass('post-author'),
+                 getMeta('post-character'),
+                 getMeta('post-screenname'),
+                 getMeta('post-author'),
                  content)
 
 
-def renderIcon(icon_path: str | None, image_size: float) -> bytes:
-    return b'\\includegraphics[width=%fmm,height=%fmm,keepaspectratio]{%s}' % (
-        image_size, image_size, icon_path.encode('UTF-8')) if icon_path else b''
+def renderIcon(icon_path: str | None, image_size: float) -> bytes | None:
+    if icon_path is None:
+        return None
+    return br'\includegraphics[width=%fmm,height=%fmm,keepaspectratio]{%s}' % (
+        image_size, image_size, icon_path.encode('UTF-8'))
 
 
 class Layout(ABC):
@@ -99,7 +131,7 @@ class ContentOnlyLayout(Layout):
         self._texifier = texifier
 
     def renderChunk(self, chunk: Chunk) -> bytes:
-        return self._texifier.texify(chunk.content)
+        return self._texifier.texify(chunk.content) + b'\n'
 
 
 class BelowIconLayout(Layout):
@@ -109,18 +141,68 @@ class BelowIconLayout(Layout):
         self._image_size = image_size
 
     def renderChunk(self, chunk: Chunk) -> bytes:
-        return b'''\\begin{wrapstuff}[l]\\oalign{
-%s\\cr
-%s\\cr
-%s\\cr
-%s\\cr
-}\\end{wrapstuff}
+        icon = renderIcon(chunk.icon, self._image_size)
+        meta = [icon] if icon else []
+        meta += [self._texifier.texify(x)
+                 for x in [chunk.character, chunk.screen_name, chunk.author]
+                 if x is not None]
+
+        return br'''\wrapstuffclear
+\begin{wrapstuff}[l]
+\fbox{
+\begin{varwidth}{0.5\textwidth}
+  \smash{\parbox[t][0pt]{0pt}{
+    \setlength{\fboxrule}{0.2pt}
+    \setlength{\fboxsep}{0pt}
+    \vspace{-3.4pt}
+    \fbox{\hspace{107mm}}
+  }\\*}
+  \vspace{-1em}
+\begin{center}
+%s
+\end{center}
+\end{varwidth}
+}
+\end{wrapstuff}
+
+\strut
+
+\noindent %s
+''' % (
+            br'\\*'.join(meta),
+            self._texifier.texify(chunk.content))
+
 
+class BesideIconLayout(Layout):
+
+    def __init__(self, texifier: Texifier, image_size: float) -> None:
+        self._texifier = texifier
+        self._image_size = image_size
+
+    def renderChunk(self, chunk: Chunk) -> bytes:
+        icon = renderIcon(chunk.icon, self._image_size)
+        meta = [
+            chunk.character,
+            chunk.screen_name,
+            chunk.author,
+        ]
+
+        # Why is \textwidth not the width of the text?
+        # Why is the width of the text .765\textwidth?
+        return br'''\noindent\fbox{
 %s
+\parbox[b]{.765\textwidth}{
+\begin{center}
+%s
+\end{center}
+}
+}\\*
+\vspace{-0.75em}\\*
+\noindent %s
+
+\strut
 
 ''' % (
-            renderIcon(chunk.icon, self._image_size),
-            chunk.character.encode('UTF-8') if chunk.character else b'',
-            chunk.screen_name.encode('UTF-8') if chunk.screen_name else b'',
-            chunk.author.encode('UTF-8') if chunk.author else b'',
+            icon if icon else b'',
+            br'\\*'.join(self._texifier.texify(x) for x in meta if x is not None),
             self._texifier.texify(chunk.content))