diff options
Diffstat (limited to '')
| -rwxr-xr-x | pyblog | 30 | ||||
| -rw-r--r-- | requirements.txt | 2 |
2 files changed, 20 insertions, 12 deletions
@@ -381,7 +381,7 @@ def generate_sitemap(feed): # try to extract updated time updated = None with open(fullpath, encoding="utf-8") as htmlobj: - soup = bs4.BeautifulSoup(htmlobj.read()) + soup = bs4.BeautifulSoup(htmlobj.read(), "lxml") if soup.article.footer is not None: updated_tag = soup.article.footer.find(attrs={"class": "updated"}) if updated_tag is not None: @@ -480,7 +480,7 @@ def generate_index_and_feed(): item = RssItem() try: with open(htmlpath, encoding="utf-8") as htmlfile: - soup = bs4.BeautifulSoup(htmlfile.read()) + soup = bs4.BeautifulSoup(htmlfile.read(), "lxml") # generate atom entry entry.author = copy.deepcopy(feed.author) # assume it's always the same author @@ -500,21 +500,25 @@ def generate_index_and_feed(): entry.updated.text = entry.updated_datetime.isoformat() # process content - # extract the article content without header and footer + tags_to_remove = [] + # mark header and footer for removal article = soup.article if article.header is not None: - article.header.extract() + tags_to_remove.append(article.header) if article.footer is not None: - article.footer.extract() - # remove line numbers + tags_to_remove.append(article.footer) + # mark line numbers for removal for line_number_span in article.find_all("span", attrs={"class": "line-number"}): - line_number_span.extract() - # remove script tags + tags_to_remove.append(line_number_span) + # mark script tags for removal for script_tag in article.find_all("script"): - script_tag.extract() + tags_to_remove.append(script_tag) # make internal links absolute abosolutify_links(article, entry_url) + # remove marked tags + for tag in tags_to_remove: + tag.extract() entry.content_html = ''.join([str(content) for content in article.contents]) @@ -540,7 +544,7 @@ def generate_index_and_feed(): item.assemble_item() rss.items.append(item) except Exception: - sys.stderr.write("failed to generate feed entry from %s" % name) + sys.stderr.write("error: failed to generate feed entry from %s\n" % name) with open(htmlpath, encoding="utf-8") as htmlfile: sys.stderr.write("dumping HTML:%s\n\n" % htmlfile.read()) raise @@ -591,7 +595,7 @@ def _pre_tag_insert_line_numbers(soup, pre_tag): def number_code_lines(htmlfilepath): """Insert line numbers to preformatted code blocks.""" with open(htmlfilepath, "r+", encoding="utf-8") as htmlfileobj: - soup = bs4.BeautifulSoup(htmlfileobj.read()) + soup = bs4.BeautifulSoup(htmlfileobj.read(), "lxml") for pre_tag in soup.find_all("pre"): if ((pre_tag.code is None or "class" not in pre_tag.attrs or not "sourceCode" in pre_tag["class"])): @@ -650,6 +654,10 @@ def generate_blog(fresh=False, report_total_errors=True): else: os.remove(obj) + # nojekyll: https://help.github.com/articles/files-that-start-with-an-underscore-are-missing/ + with open(os.path.join(BUILDDIR, ".nojekyll"), "w") as fileobj: + pass + failed_builds = 0 template_mtime = os.path.getmtime(HTMLTEMPLATE) anything_modified = False diff --git a/requirements.txt b/requirements.txt index 4e0569be..577d17cf 100644 --- a/requirements.txt +++ b/requirements.txt @@ -1,4 +1,4 @@ -beautifulsoup4 +beautifulsoup4>=4.4.0 colorama lxml python-dateutil |