diff --git a/recipes/ming_pao.recipe b/recipes/ming_pao.recipe index 9e9522f26e..d79125edee 100644 --- a/recipes/ming_pao.recipe +++ b/recipes/ming_pao.recipe @@ -24,6 +24,7 @@ __Date__ = '' ''' Change Log: +2011/12/01: take care of situation that in txt source parsing, the article content does start with special character u'\u3010' 2011/10/21: fix a bug that hi-res img is unavailable in pages parsed from source txt 2011/10/19: fix a bug in txt source parsing 2011/10/17: disable fetching of premium content, also improved txt source parsing @@ -533,12 +534,22 @@ class MPRecipe(BasicNewsRecipe): new_raw_html = 'Untitled
' next_is_img_txt = False title_started = False + title_break_reached = False met_article_start_char = False for item in splitter.split(raw_html): item = item.strip() - if item.startswith(u'\u3010'): - met_article_start_char = True - new_raw_html = new_raw_html + '

' + item + '

\n' + # if title already reached but break between title and content not yet found, record title_break_reached + if title_started == True and title_break_reached == False and item == '': + title_break_reached = True + # if title reached and title_break_reached and met_article_start_char == False and item is not empty + # start content + elif title_started == True and title_break_reached == True and met_article_start_char == False: + if item <> '': + met_article_start_char = True + new_raw_html = new_raw_html + '

' + item + '

\n' + #if item.startswith(u'\u3010'): + # met_article_start_char = True + # new_raw_html = new_raw_html + '

' + item + '

\n' else: if next_is_img_txt == False: if item.startswith("=@"): @@ -787,3 +798,4 @@ class MPRecipe(BasicNewsRecipe): with nested(open(opf_path, 'wb'), open(ncx_path, 'wb')) as (opf_file, ncx_file): opf.render(opf_file, ncx_file) +