Test bare chunks

[paperdoorknob] / glowfic.py

diff --git a/glowfic.py b/glowfic.py
index 012bd069aca5a3066b443640bc66904d796e44ee..ef71eef019e8005b4c2bbf8a34ab49b0362f3e85 100644 (file)
--- a/glowfic.py
+++ b/glowfic.py
@@ -7,17 +7,22 @@
 
 from dataclasses import dataclass
 import itertools
-from urllib.parse import parse_qsl, urlencode, urlparse, urlunparse
+from urllib.parse import parse_qsl, urlencode, urljoin, urlparse, urlunparse
 
-from typing import Iterable
+from typing import Any, Iterable
 
 from bs4 import BeautifulSoup
 from bs4.element import Tag
 
 from images import ImageStore
+from spec import Spec
 from texify import Texifier
 
 
+def ilen(it: Iterable[Any]) -> int:
+    return sum(1 for _ in it)
+
+
 def _removeViewFromURL(url: str) -> str:
     u = urlparse(url)
     old_qs = parse_qsl(u.query)
@@ -54,20 +59,63 @@ class Chunk:
 #  * Humans in the community tend to use "posts" to mean chunks.
 
 
-def chunkDOMs(html: BeautifulSoup) -> Iterable[Tag]:
-    def text() -> Tag:
-        body = html.body
-        assert body
-        text = body.find_next("div", class_="post-post")
-        assert isinstance(text, Tag)
-        return text
+class Thread:
 
-    def the_replies() -> Iterable[Tag]:
-        rs = html.find_all("div", class_="post-reply")
-        assert all(isinstance(r, Tag) for r in rs)
-        return rs
+    def __init__(self, spec: Spec) -> None:
+        def find_next_thread(dom: BeautifulSoup) -> str | None:
+            for c in dom.findChildren('div', class_='post-navheader'):
+                for a in c.findChildren('a'):
+                    if 'Next Post' in a.text and 'href' in a.attrs and isinstance(
+                            a.attrs['href'], str):
+                        return urljoin(spec.url, a.attrs['href'])
+            return None
 
-    return itertools.chain([text()], the_replies())
+        spec.log('Fetching HTML...\r')
+        html = spec.fetcher.fetch(spec.url)
+        flat_html = spec.fetcher.fetch(flatURL(spec.url))
+        spec.log('Parsing HTML...\r')
+        self._next_thread = find_next_thread(
+            BeautifulSoup(spec.htmlfilter(html), 'html.parser'))
+        self._dom = BeautifulSoup(spec.htmlfilter(flat_html), 'html.parser')
+        self._spec = spec
+
+    def title(self) -> str | None:
+        span = self._dom.findChild("span", id="post-title")
+        if not isinstance(span, Tag):
+            return None
+        return span.text.strip()
+
+    def next_thread(self) -> str | None:
+        return self._next_thread
+
+    def chunkDOMs(self) -> Iterable[Tag]:
+        def text() -> Tag:
+            body = self._dom.body
+            assert body
+            text = body.find_next("div", class_="post-post")
+            assert isinstance(text, Tag)
+            return text
+
+        def the_replies() -> Iterable[Tag]:
+            rs = self._dom.find_all("div", class_="post-reply")
+            assert all(isinstance(r, Tag) for r in rs)
+            return rs
+
+        return itertools.chain([text()], the_replies())
+
+    def emit(self) -> None:
+        self._spec.log('Counting chunks...\r')
+        num_chunks = ilen(self.chunkDOMs())
+        title = self.title() or "chunk"
+        for i, r in enumerate(self.chunkDOMs()):
+            percent = 100.0 * i / num_chunks
+            self._spec.log(
+                f'Processing {title} {i} of {num_chunks} ({percent:.1f}%)\r')
+            self._spec.domfilter(r)
+            chunk = makeChunk(r, self._spec.images)
+            self._spec.texout.write(
+                self._spec.texfilter(renderChunk(self._spec.texifier, chunk)))
+        self._spec.log('')
 
 
 def makeChunk(chunk_dom: Tag, image_store: ImageStore) -> Chunk: