docs2epub 0.1.7__py3-none-any.whl → 0.1.8__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (5) hide show
  1. docs2epub/docusaurus_next.py +20 -0
  2. {docs2epub-0.1.7.dist-info → docs2epub-0.1.8.dist-info}/METADATA +1 -1
  3. {docs2epub-0.1.7.dist-info → docs2epub-0.1.8.dist-info}/RECORD +5 -5
  4. {docs2epub-0.1.7.dist-info → docs2epub-0.1.8.dist-info}/WHEEL +0 -0
  5. {docs2epub-0.1.7.dist-info → docs2epub-0.1.8.dist-info}/entry_points.txt +0 -0
docs2epub/docusaurus_next.py CHANGED
@@ -85,6 +85,22 @@ def _extract_article(soup: BeautifulSoup) -> Tag:
85
85
  role_main = soup.find(attrs={"role": "main"})
86
86
  if role_main:
87
87
  return role_main
88
+ for selector in [
89
+ "div#content",
90
+ "div.content",
91
+ "div#main",
92
+ "div.main",
93
+ "div#page",
94
+ "div.page",
95
+ "div.document",
96
+ "div#document",
97
+ ]:
98
+ candidate = soup.select_one(selector)
99
+ if candidate:
100
+ return candidate
101
+ body = soup.find("body")
102
+ if body:
103
+ return body
88
104
  raise RuntimeError("Could not find <article> in page HTML")
89
105
 
90
106
 
@@ -386,6 +402,10 @@ def iter_docusaurus_next(options: DocusaurusNextOptions) -> list[Chapter]:
386
402
  if title_el
387
403
  else f"Chapter {len(chapters) + 1}"
388
404
  )
405
+ if title_el is None and article.name == "body":
406
+ body_text = " ".join(article.get_text(" ", strip=True).split())
407
+ if len(body_text) < 200:
408
+ return None
389
409
 
390
410
  _remove_unwanted(article)
391
411
  _absolutize_urls(article, base_url=target_url)
{docs2epub-0.1.7.dist-info → docs2epub-0.1.8.dist-info}/METADATA RENAMED
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: docs2epub
3
- Version: 0.1.7
3
+ Version: 0.1.8
4
4
  Summary: Turn documentation sites into an EPUB (Kindle-friendly).
5
5
  Author: Breno Brito
6
6
  License: MIT
{docs2epub-0.1.7.dist-info → docs2epub-0.1.8.dist-info}/RECORD RENAMED
@@ -1,11 +1,11 @@
1
1
  docs2epub/__init__.py,sha256=iccyEu4zlubhvd6pM7Z2Gjwn8tPw9IhZ4ABKhbiFjUY,54
2
2
  docs2epub/cli.py,sha256=pt1crvrkr2k1ybf_p0m4xSYyoZVluFsDNGuwJ7CykYM,3863
3
- docs2epub/docusaurus_next.py,sha256=_PTPnObjYrR6Yxz9rFxooiJhzYlBpRutLMTG0SYshUg,12030
3
+ docs2epub/docusaurus_next.py,sha256=nQYkNecXgh4TsxaTydoiC1tVmIqjYiLiyEtYlpXGmXg,12507
4
4
  docs2epub/epub.py,sha256=OsPWcPGTgazAeNpWASIE6e4HQ5ILQr2VFO1-Aj3y1kg,2986
5
5
  docs2epub/kindle_html.py,sha256=LN0CGj9ap9b8iC_MlZcQLuhJ7FehZr_VbIfMOz78E5c,2297
6
6
  docs2epub/model.py,sha256=uL7uwbG6yU0bEGpSFxxIv2pcZHQR9cs2prfqk5iNQwc,160
7
7
  docs2epub/pandoc_epub2.py,sha256=l22-QAQcCgJyl7HF0_b5weC3qEGVQLwOhxdbAvd8C2o,3610
8
- docs2epub-0.1.7.dist-info/METADATA,sha256=m8ea2WWxNYzeCXHnab2P19oHCQIULTtrTdYMmtnLES8,1886
9
- docs2epub-0.1.7.dist-info/WHEEL,sha256=WLgqFyCfm_KASv4WHyYy0P3pM_m7J5L9k2skdKLirC8,87
10
- docs2epub-0.1.7.dist-info/entry_points.txt,sha256=DHK4mzthrIXUvM8Y8Vo_3jG2IhegEDDM7T9CvCkUtvw,49
11
- docs2epub-0.1.7.dist-info/RECORD,,
8
+ docs2epub-0.1.8.dist-info/METADATA,sha256=KdwbHGiBRLuXLQKlTypnDH8eOogD5bDoSGDIJgNriZs,1886
9
+ docs2epub-0.1.8.dist-info/WHEEL,sha256=WLgqFyCfm_KASv4WHyYy0P3pM_m7J5L9k2skdKLirC8,87
10
+ docs2epub-0.1.8.dist-info/entry_points.txt,sha256=DHK4mzthrIXUvM8Y8Vo_3jG2IhegEDDM7T9CvCkUtvw,49
11
+ docs2epub-0.1.8.dist-info/RECORD,,