diff options
| -rw-r--r-- | autodownload.py | 66 |
1 files changed, 66 insertions, 0 deletions
diff --git a/autodownload.py b/autodownload.py new file mode 100644 index 00000000..e77288df --- /dev/null +++ b/autodownload.py @@ -0,0 +1,66 @@ +#/usr/bin/env python3 +""" Automate the download of availlable manual. + + This script automates the download of manual for every languages availlable, + from the specifyed server (see at the bottom of the script) as long as the + server follows this tree: + server_path/language/manual.zip + + To use it, you have to modify the server URL at the bottom of the script + (variable "base_url"), + and modify the last line of the script to tell which kind of manual it is + ("installer" or "mcc") + +""" +from urllib.request import urlopen, urlretrieve +from urllib.error import HTTPError +from html.parser import HTMLParser + + +manual_name_prefix = {"installer": "DrakX-", + "mcc": "MCC-"} +manual_name_suffix = "WebHelp-zipped.zip" + +class ArchivePageHTMLParser(HTMLParser): + def __init__(self): + HTMLParser.__init__(self) + self.recording = 0 + self.data = [] + + def handle_data(self, data): + if data.endswith('/'): + self.data.append(data[:-1]) + +def getPage(url): + parseur = ArchivePageHTMLParser() + for word in urlopen(url).readlines(): + parseur.feed(word.strip().decode('ascii')) + return parseur.data + +def getZip(url, manual, lang): + if lang == 'en' and manual == "mcc": + filename = manual_name_prefix[manual] + \ + manual_name_suffix + else: + filename = manual_name_prefix[manual] + \ + lang.upper() + '-' + \ + manual_name_suffix + + language_url = "/".join([url, lang, filename]) + print("\n%s" % lang) + print(filename) + print(language_url) + try: + urlretrieve(language_url, filename) + except HTTPError as e: + print(e) + + + +if __name__ == "__main__": + base_url = "http://waesvanm.home.xs4all.nl/Mageia_4_documentation/zipped/installer/" + language_list = getPage(base_url) + for elem in language_list: + getZip(base_url, "installer", elem) + + |