def convert_blog_parts(self, str_line):
"""Convert blog parts to reST.
Argument:
str_line: text string of blog entry.
"""
# remove hatena internal link
str_line = convert.remove_internal_link(str_line)
# for ditto
str_line = convert.ditto(str_line)
# remove span element
str_line = convert.remove_span(str_line)
# remove del element
str_line = convert.remove_del(str_line)
# for google maps
str_line = convert.google_maps(str_line)
# for gmodules
str_line = convert.gmodules(str_line)
# for img element
str_line = convert.img2image(str_line)
# for amazlet
pat_amazon, m = utils.regex_search('amazlet', str_line)
if not m:
pat_image, m = utils.regex_search(
'(<a href="(.+?)" .+?><img src="(.+?)".*?/?></.+?>)', str_line)
if m:
img_path = utils.retrieve_image(m.group(3),
self.dstdir + __imgdir__,
self.retrieve_image_flag)
str_line = pat_image.sub(
'\n.. image:: ' + __imgdir__ + img_path + '\n :target: '
+ m.group(2) + '\n\n', str_line)
# for youtube
str_line = convert.youtube(str_line)
# for tweet
str_line = convert.tweet(str_line)
# for blogparts
str_line = convert.extract_blog_parts(str_line)
return str_line
def fotolife2rest(self, str_line):
"""Convert fotolife to image directive.
Argument:
str_line: text string of blog entry.
convert is
from: hatena; [f:id:imageid:image]
to: reST ; .. image:: imgsrc
:target: uri
"""
r, m = utils.regex_search(
'\[f:id:(.*):([0-9]*)[a-z]:image(|:.+?)\]', str_line)
if m:
img_uri_partial = ('http://cdn-ak.f.st-hatena.com/images/fotolife/'
+ m.group(1)[0] + '/' + m.group(1) + '/'
+ m.group(2)[0:8] + '/' + m.group(2))
# get image file
img_src = utils.retrieve_image(img_uri_partial,
self.dstdir + __imgdir__,
self.retrieve_image_flag)
repl_str = ('\n.. image:: ' + __imgdir__ + img_src)
str_line = r.sub(repl_str, str_line)
return str_line
def listing2rest(self, str_line):
"""Convert hatena syntax to reST of list.
Argument:
str_line: text string of blog entry.
"""
for i in range(1, 4)[::-1]:
"""list lv is indent depth
order is 3,2,1 why short matche is stronger than long.
3 is --- or +++
2 is -- or ++
1 is - or +
"""
r, m = utils.regex_search(
'(^(-{%d}))|(^(\+{%d}))' % (i, i), str_line)
if m:
item = (' ' * (i - 1) + '* ' if m.group(1)
else ' ' * (i - 1) + '#. ')
if self.list_lv == i:
repl = item
else:
repl = '\n' + item
self.list_lv = i
str_line = r.sub(repl, str_line)
str_line += '\n'
return str_line
def extract_tables(string_line, table, tables):
"""Extract tables
Argument:
string_line: parsing target string.
table: parsing target table
tables: parsing target tables
"""
pat_table, match_obj = utils.regex_search(
'^\|(.+?)\|$', string_line)
if match_obj:
row_data = (match_obj.group(0),
match_obj.groups()[0].split('|'))
if not self.table_flag:
# table start
self.table_flag = True
table.append(row_data)
else:
if self.table_flag:
# table close
tables.append(table)
table = []
self.table_flag = False
return table, tables
def tweet(string):
"""Convert blog parts of twitter to reST hyperlink.
Argument:
string: blog entry body string.
"""
pat_comment, m = utils.regex_search(
'((<!-- (.+?) -->) (<.+?>(.+?)</.+?> )(<!-- (.+) -->))', string)
if m:
str_tmp = string.replace(m.group(2), '')
str_tmp = str_tmp.replace(m.group(6), '')
pat_style, m2 = utils.regex_search(
' <style .+?>(.+?)</style> ', str_tmp)
if m2:
str_tmp = str_tmp.replace(m2.group(0), '')
str_tmp = str_tmp.replace('><', '>\n<')
str_tmp = str_tmp.replace('> <', '>\n<')
str_tmp = str_tmp.replace('</span>\n', '')
pat_tweet = re.compile(
'((<.+?>(.+?)</.+?>)(.+?)(<.+?>(.+?)</.+?>))')
m3 = pat_tweet.search(str_tmp)
if m3:
pat_anchor = re.compile('<a.+?>')
tweet_msg = (pat_anchor.sub('', m3.group(3)) +
pat_anchor.sub('', m3.group(4))
+ pat_anchor.sub('', m3.group(5))
).replace('</a>', '')
if parse_blog_parts(str_tmp.encode('utf-8')):
uri = parse_blog_parts(str_tmp.encode('utf-8'))
repl_str = '\n' + uri + '::\n\n ' + tweet_msg + '\n\n'
else:
repl_str = ''
string = pat_comment.sub(repl_str, string)
return string
def img2image(string):
"""Convert html img element to reST image directive.
Argument:
string: blog entry body string.
"""
pat_img, m = utils.regex_search('^<img src="(.+?)" .+?(/?)>', string)
if m:
string = pat_img.sub('\n.. image:: ' + m.group(1)
+ '\n\n', string)
return string
def remove_del(string):
"""Remove del element.
Argument:
string: blog entry body string.
"""
pat_del, m = utils.regex_search(
'(<del( .+?|)>(.+?)</del>)', string)
if m:
string = pat_del.sub('', string)
return string
def remove_span(string):
"""Remove span element.
Argument:
string: blog entry body string.
"""
pat_span, m = utils.regex_search(
'(<span .+?>(.+?)</span>)', string)
if m:
string = pat_span.sub(m.group(2), string)
return string
def replace_shell_variable(string):
"""Replace shell variable
Argument:
string: text string of blog entry.
"""
pat_shell_var, match_obj = utils.regex_search(
'(\${.+?}[a-zA-Z0-9/_\\\*]+)', string)
if match_obj:
string = pat_shell_var.sub(
' :command:`' + match_obj.group() + '` ', string)
return string
def remove_internal_link(string):
"""Remove hatena internal link.
Argument:
string: blog entry body string.
"""
pat_hatena_internal_link, m = utils.regex_search(
'(\[\[|\]\])', string)
if m:
string = pat_hatena_internal_link.sub('', string)
return string
def parse_end_ref(string_line):
"""Parse ending of reference block
Argument:
string: convert target string.
"""
pat_end_ref, match_obj = utils.regex_search(
'^<<', string_line)
if match_obj:
string_line = pat_end_ref.sub('\n\n', string_line)
self.ref_flag = False
else:
string_line = re.sub('^', ' ', string_line)
return string_line
def parse_end_codeblock(string_line):
"""Parse ending of codeblock.
Argument:
string_line: parsing target string.
"""
pat_code_close, match_obj = utils.regex_search(
'^\|\|<|^\|<$', string_line)
if match_obj:
string_line = pat_code_close.sub('\n', string_line)
# code block closing
self.code_flag = False
else:
string_line = re.sub('^', ' ', string_line)
return string_line
def gmodules(string):
"""Convert blog parts of gmodules to reST raw directive.
Argument:
string: blog entry body string.
"""
if (string.find('http://gmodules.com') > 0 or
string.find('https://gist.github.com') > 0):
pat_gmodules, m = utils.regex_search(
'^<script .+?></script>', string)
if m:
string = pat_gmodules.sub(
'\n.. raw:: html\n\n ' + m.group(0) + '\n', string)
return string
def google_maps(string):
"""Convert blog parts of google maps to reST raw directive.
Argument:
string: blog entry body string.
"""
if (string.find('http://maps.google.com/') > 0 or
string.find('http://maps.google.co.jp/') > 0):
pat_google_maps, m = utils.regex_search(
'(<iframe .+?></iframe><br />(<.+?>.+?</.+?>)(.*?)</.+?>)',
string)
if m:
string = pat_google_maps.sub(
'\n.. raw:: html\n\n ' + m.group(0) + '\n', string)
return string
def youtube(string):
"""Convert blog parts of YouTube to reST raw directive.
Argument:
string: blog entry body string.
"""
if string.find('http://www.youtube.com') > 0:
pat_youtube, m = utils.regex_search(
'(<object .+?>(.*?)</.+?>)', string)
if m:
string = pat_youtube.sub(m.group(0), string)
string = string.replace('\n', '')
string = string.replace('&hl=ja', '')
string = string.replace('&fs=1', '')
string = '\n.. raw:: html\n\n ' + string + '\n'
return string
def section2rest(string):
"""Convert hatena syntax to reST of section.
Argument:
string: text string of blog entry.
"""
for i in range(2, 4)[::-1]:
"""2:section, 3:subsection"""
sep = '-' if i == 2 else '^'
r, m = utils.regex_search('^(\*){%d}(.*)' % i, string)
if m:
pat_space = re.compile('^\s+')
section_str = pat_space.sub('', m.group(2))
string = r.sub(
'\n' + section_str + '\n'
+ sep * utils.length_str(section_str) + '\n', string)
return string
def parse_begin_ref(string_line):
"""Parse begining of reference block
Argument:
string: convert target string.
"""
pat_start_ref, match_obj = utils.regex_search(
'^>((http|https)://(.+?)|)>$', string_line)
if match_obj:
self.ref_flag = True
if match_obj.group(1):
repl_str = match_obj.group(1)
else:
repl_str = ''
string_line = pat_start_ref.sub(
repl_str,
string_line)
return string_line
def parse_begin_codeblock(string_line):
"""Parse begining of code block
Argument:
string: convert target string.
"""
pat_code_open, match_obj = utils.regex_search(
'>\|([a-zA-Z0-9]*)\|$|>\|()$', string_line)
if match_obj:
# code block opening
self.code_flag = True
if match_obj.group(1):
lexer_str = convert.replace_lexer(match_obj.group(1))
string_line = pat_code_open.sub(
'\n.. code-block:: ' + lexer_str + '\n',
string_line)
else:
string_line = pat_code_open.sub(
'\n.. code-block:: sh\n', string_line)
return string_line
def ditto(string):
"""Convert blog parts of twitter with ditto to reST hyperlink.
Argument:
string: blog entry body string.
"""
pat_ditto, m = utils.regex_search(
'(<style .+?>.+?</style>)(<div .+?>.+?</div>)', string)
if m:
ex_ref_char = re.compile('\&(?!amp;)', flags=re.U)
string = ex_ref_char.sub('&', m.group(2))
# get uri
uri = ''
xmltree = xml.etree.ElementTree.fromstring(string.encode('utf-8'))
for p_child in xmltree.find('p').getchildren():
for i, p_child_child in enumerate(p_child.getchildren()):
if i == 1 and p_child_child.get('href'):
uri = p_child_child.get('href')
# get tweet message
tweet_msg = ''
if xmltree.get('class').find('ditto') == 0:
span_element = xmltree.find('p').find('span').find('span')
for i, v in enumerate(xmltree.itertext()):
if i > 1:
pat = re.compile(' |via', flags=re.U)
if pat.search(v) > 0:
break
else:
tweet_msg += str(v.encode('utf-8'))
repl_str = '\n' + uri + '::\n\n ' + tweet_msg + '\n\n'
string = pat_ditto.sub(m.group(), repl_str).decode('utf-8')
return string
