milianw / springer_download Goto Github PK

(deleted)

After downloading all individual pdfs, chapters are not put in the right order. For example when running:

./springer_download.py -c 978-3-642-18101-6

The frontpage is succeeded by the 5th chapter.

http://springerlink.com/content/978-3-540-17069-3/contents/

    <h1 lang="en" class="title">
        A Course in H<sub>8</sub> Control Theory

Sometimes, the chapters of the books are sorted alphabetically on the contents page of springerlink, as the script only uses this information for its list order, the chapters are mixed up which isn't very nice.
Maybe there could be a sorting, based on the page numbers of the chapters. I think it should be possible, but I'm not very good on regex, so I can't present a solution myself.

Some books I've downloaded contained the back matter after the front matter and also after the other chapters.
This patch removes the first appearance after the front matter.

--- springer_download.py.ori    2010-09-29 00:59:40.000000000 +0100
+++ springer_download.py    2010-10-15 13:30:46.000000000 +0100
@@ -142,6 +155,9 @@
                     front_matter = True
             if re.search(r'back-matter.pdf', chapterLink) and re.search(r'<a href="([^"]+)">Next</a>', page):
                 continue
+            #skip backmatter if it is in list as second chapter - will be there at the end of the book also
+            if re.search(r'back-matter.pdf', chapterLink):
+                if len(chapters)<2:
+                    continue
 
             chapters.append(chapterLink)

Its helpful to include a howto for setting http_proxy environment variable, example (this commit depends on socks-related changes and is therefore not compatible to milanw/springer_download@master): yvesf/springer_download@75e4a086a0a16aa072b7

the python behavior regarding http_proxy is described in http://docs.python.org/library/urllib.html#urllib.urlopen

... because the link now has also a Title attribute, and the old regex doesn't work. Here is the new regex which worked for me:

<a href="([^"#]+)"[^>]*>Next</a>

(replaced the old one on lines 148 and 158)

Thanks for a great script!

my university requires to login to dowload the pdfs.

click on login
"Log In as an Institution (via Shibboleth or Athens)"
Select Country and University
Login

is there a way to pass the login credentials?

if i run the script as python2 springer-download.py it works just fine.
./springer-download.py gives me (with /usr/bin/env python defaulting to python3)

https://github.com/milianw/springer_download

File "./springer_download.py", line 92
print "fetching book information...\n\t%s" % link
^
SyntaxError: invalid syntax

Adding ISBN from Springer into filename helps to identify and to rename book with external scripts

Such form as (ISBN XXXXXXXXXX) at the end of the pdf file would be optimal

Seems that Springer has updated their website structure and URL space, for instance a book previously located at http://www.springerlink.com/content/978-94-007-4655-8 is now being redirected to http://link.springer.com/book/10.1007/978-94-007-4655-8/page/1, causing the script not to work.

Before this change, Springer, I guess, made some changes to the old site as well since the chapter ordering in the bound pdf is incorrect (see comments in http://milianw.de/code-snippets/take-2-download-script-for-springerlinkcom-ebooks)

Would be great if these issues were fixed. This is a great tool!

The script always stops at chapter 3:

ERROR: downloaded chapter http://springerlink.com/content/m65623732836hr43/fulltext.pdf has invalid mime type text/html - are you allowed to download XXX

The problem is, I’am 1) allowed and 2) it works via the browser.

How to fix this?

Nice to have:
It would be helpfull getting bookmarks to the merged file.

Unfortunately I'm not good at regexp but I think it could work like described here: http://blog.tremily.us/posts/PDF_bookmarks_with_Ghostscript/

…but not merge.