Sophie

Sophie

distrib > Mandriva > 2010.0 > i586 > media > contrib-release > by-pkgid > 2c1283c538215030a47ce161ff7fc751 > files > 120

alexandria-0.6.4.1-2mdv2010.0.noarch.rpm

Copyright (C) 2007 Marco Costantini

########################################################################

Question: I have a ( Bug | Support Request | Patch | Feature Request ).

Answer: You are welcome to report it. Please use the tracker at
http://rubyforge.org/tracker/?group_id=205

It may be helpful to read the text "How to Report Bugs Effectively", at
http://www.chiark.greenend.org.uk/~sgtatham/bugs.html

Please, one bug per report, and one report per bug.

Please use a meaningful and specific summary.

In order to let us investigate the bug and fix it for you, we need a clear
and specific information about how to replicate it. For instance: report
the ISBN of the book for which the bug happens; report the book provider
used if relevant; copy and paste the error message. You may also start
Alexandria in terminal window, and copy and paste the output on that
window.

In the past we got the bug report [#6767] saying:
"Entering a valid ISBN number (many different) is rejected by the
program. Stating to be an invalid ISBN."
This user didn't report any single non-working ISBN; and didn't explain
"rejected" enough. Hence, even if further work has been done on this
topic, we can't know whether this bug was a real one, and whether it has
been already solved or not.

If you report only the backtrace without saying how to replicate the 
bug, it is usually impossible to understand and fix the bug.

Please have a look at the already reported bugs or feature request:
it is useless to report again an already reported one. 
If you have more information, please add it at the already reported bug,
instead of opening a new one.

Your bug could be a bug already solved in the development version of
Alexandria. You can get the development version at
http://rubyforge.org/scm/?group_id=205


########################################################################


Question: Where are the Z39.50 stuff, the providers US Library of 
Congress and British Library, the advanced options, and the window for 
adding new providers? I can't install Ruby/ZOOM.

Answer: All these things require that the library Ruby/ZOOM,
version >= 0.2.0 is installed and working. You can get it at
http://ruby-zoom.rubyforge.org/ Install it according to its
documentation.

Ruby/ZOOM requires YAZ, http://www.indexdata.dk/yaz/ , compiled with 
passing the --enable-shared option to the configure script:

./configure --enable-shared

before building YAZ with make.

The following works for me with Ubuntu

sudo apt-get -y install libyaz-dev ruby ruby1.8-dev make gcc
wget http://rubyforge.org/frs/download.php/6940/ruby-zoom-0.2.2.tar.gz
tar xfz ruby-zoom-0.2.2.tar.gz
cd ruby-zoom-0.2.2/
ruby extconf.rb
make
sudo make install



List of Z39.50 server are at
http://www.webclarity.info/registry/browse
http://targettest.indexdata.com/


When adding new Z39.50 providers, you must choose the record syntax.
Currently (March 2007), these are offered: "USMARC", "SUTRS", "UNIMARC".

"USMARC" works, "UNIMARC" works very limited, "SUTRS" doesn't work. 
Choose a provider that supports "USMARC", or if none is available, one 
that supports "UNIMARC".
MARC21 is the same as USMARC (at least for Alexandria).

You must choose also the charset encoding. Some are 

UTF-8
ISO-8859-1 (also known as Latin-1, Western Europe)
ISO-8859-15
ISO_6937-2 (also known as Ala, Ansel)

The list of all possible charset encoding could be at
/usr/lib/gconv/gconv-modules


There is a plan to replace in future the provider British Library with 
Copac, which includes British Library and others; see
http://copac.ac.uk/libraries/
http://en.wikipedia.org/wiki/Copac


########################################################################


Question: A provider is not working, or only partial information it 
retrieved from it, or there are books on the providers for which 
Alexandria fails to retrieve information.

Answer: Yes, the provider may have changed the structure of its
web pages, and hence it is necessary to update the file for that
provider; this file is in the directory
/<something>/alexandria/book_providers/

It is also possible that the file handles only the "simple" cases, and the 
case of your book is a "complex" one.

Try to get the development version at http://rubyforge.org/scm/?group_id=205 
because the problem could be already fixed there. 

Try to fix by yourself, and thereafter send us a patch. 
Otherwise, report the bug, saying for which ISBN this problem happens.


########################################################################


Question: I know another book provider

Answer: There are many other providers, see 
http://en.wikipedia.org/wiki/Wikipedia:Book_sources
That page contains also links to the pages for book providers in other
languages.
If you now other providers, please add them to that lists.


########################################################################


Question: Please add a new provider

Answer: First, choose wisely which provider(s) to ask. Do a 
search on Internet of the various existing providers for your language 
(see also, but not only, the previous question). 

You may prefer providers that list also out-of-prints, instead of 
providers that list only available books. It is also preferable a 
provider that lists not only the books of your country, but also the 
books of the neighbor countries.

Check whether the provider supports the Z39.50 protocol. This is probable
for public libraries, and very unlikely for commercial booksellers. 
(If the Z39.50 protocol is supported, just go to "Preferences", 
"Providers", "Advanced Settings", and add the provider there.)

It is useful if you choose a provider that allows to specify the ISBN in 
the address, such as 
http://www.internetbookshop.it/ame/ser/serdsp.asp?e=0060748222&shop=1
or
http://www.deastore.com/product.asp?isbn=9780060748227

It is necessary that the HTML page of each provider is easily parsable, 
that is, that the computer can easily find title, author(s), ISBN, 
publisher, year of publication, binding.
For instance, the following piece of HTML page is easily parsable.

<span class="BDtitoloLibro">The Krusty Book</span>
<span class="BDauthLibro">by: Matt 
Groening </span><span class="BDformatoLibro"><span 
class="BDEticLibro">Format: </span>paperback</span><span 
class="BDEticLibro">Publication date: </span><span 
class="BDdataPubbLibro">10/23/2006</span><span 
class="BDEticLibro">Publisher: </span><span 
class="BDeditoreLibro">HarperCollins Publishers (Australia) Pty 
Ltd</span><br /><span class="BDEticLibro">ISBN: </span><span 
class="isbn">0-06-074822-2</span><br /><span class="BDEticLibro">ISBN 
13: </span><span class="isbn">978-0-06-074822-7</span></div>

It is not necessary that the provider reports for each book all the
information: title, author(s), ISBN, publisher, year of publication,
binding. In case that one of these is missing, Alexandria will not get 
it.

It is easier, if the provider doesn't use cookies (however, provider
Deastore uses cookies and has been implemented).

After the choice of the provider, try by yourself to modify the file for 
an already implemented provider. Probably it may be only required that 
you modify some regular expressions. Please send us the patch.
If you can program already, learning Ruby may require 20 minutes, see 
http://www.ruby-lang.org/en/documentation/quickstart/

If you can't implement the new provider by yourself, write a 'feature 
request' at http://rubyforge.org/tracker/?group_id=205
Please report some ISBNs of books that your provider knows and no other 
provider implemented by Alexandria knows. Report also how are the words 
title, author(s), ISBN, publisher, year of publication, binding in the 
language used by the site. Report any other relevant information.


########################################################################


Question: What are the ISBN, the ISBN-13, and the EAN? 

Answer: For more information about ISBN, ISBN-13, EAN, see 
http://en.wikipedia.org/wiki/International_Standard_Book_Number


########################################################################


Question: Does Alexandria support the ISBN-13? Do I need to enter the 
ISBN with or without dashes?

Answer: Alexandria support both the 10-digit and 13-digit ISBN, and the 
dashes entered with the ISBN are ignored. Hence you can enter the book 
in any of these equivalent forms:

88-06-14930-X
880614930X
88-06-14930-x
880614930x
978-88-06-14930-7
9788806149307


########################################################################


Question: I entered the ISBN exactly as it shows on the book, and 
Alexandria says that it is invalid.

Answer: You can check whether an ISBN is valid with
http://www.isbn-check.com/ 
Sometimes the ISBN reported on the book is wrong. Try to use the
13-digit one, or to use the ISBN reported on the back of the title page,
or to search by title/author on a book provider, in order to find the
correct ISBN.


########################################################################


Question: Where is stored the data about the books?

Answer: In the files ~/.alexandria/<library_name>/<ISBN>.yaml
These files are in YAML ( http://www.yaml.org/ ) format, and you can 
edit them manually. The covers are in the files 
~/.alexandria/<library_name>/<ISBN>.cover
These files persist if you install or deinstall Alexandria.


########################################################################


Question: I was enjoying Alexandria a lot and now it doesn't work any 
more, I can't enter other books, or crashes when it starts.

Answer: This usually happens when there is some invalid data in one of 
the .yaml files (see the question above). For one or more of the 
latest books that you have entered, a bug happened, and now the data 
stored is invalid and causes trouble. Try to find which book it was, 
(please send also a bug report), and to delete or to move to another 
directory that .yaml file or the whole ~/.alexandria directory.

This kind of problem may show only with the manual entered books, and 
not with the books searched by ISBN.


########################################################################


Question: can I catalog other objects?

Answer: Yes. Some kinds of object, such as illustrated calendars, or 
geographical maps, have the ISBN, and can be treated as books.

For other objects, insert the EAN or bar-code number as ISBN. Provider 
Thalia knows about music CDs, video DVDs, games and software, and can be 
used to retrieve information about them. Other providers may be added in 
future (possibly by you). Otherwise, the other objects may be added 
manually.


########################################################################


Question: I prefer to have the field "Author" as "Surname, Name" instead 
of "Name Surname".

Answer: Alexandria relies on what the providers say: 
some providers report "Surname, Name": Deastore, Renaud, Mcu,
some "Surname Name": Webster.it, Ibs.it,
some "Name Surname": Bn, Thalia, Bol.it, Adlibris, L Siciliano, Worldcat,
and some without any rule: Amazon, Proxis.
You can change, in the preferences, the order in which the providers are 
queried.

If Alexandria reports "Name Surname" with one provider, and in the HTML 
page of that provider there is also (in a parsable way) "Surname, Name", 
then this is a bug to be reported. In some of these cases, it would be 
possible to get "Surname, Name", but the code would be complicated and 
unreliable. 

There is no algorithm that can decide, given the Author as a string, 
which part of the string is the name and which part is the surname.
For instance, the name or the surname can be composed by more than one 
word, or the author may be a collective author, or be a title or a 
pseudonym, or in the author's country they may use that the surname is 
the first word and the name is the second, or there may be the 
patronymic, and so on.

For one provider, for books with more that one author, it's non-trivial 
even to split the string of all the authors into strings "Name Surname" for 
each author.

The only possibility to convert "Name Surname" to "Surname, Name" is to do
that by hand.

If you still believe that it is possible for a software to convert "Name 
Surname" into "Surname, Name" in a reliable way, here are some cases of 
real authors for you to try. (Some accented letters have been replaced by 
the non accented ones.)

"Agente Italiano"
"Albert Sydney Hornby"
"Alexandre Dumas fils"
"Antoine de Saint-Exupery"
"Boston Women's Health Book Collective"
"Camilla"
"Central Intelligence Agency"
"Christiane F."
"Dario De Toffoli"
"Elaine St. James"
"Ernesto Che Guevara"
"Francois-Marie Arouet de Voltaire"
"Gaius Caesar Octavianus Augustus"
"Geographers A-Z Map Company"
"Gian Antonio Stella"
"H G Wells"
"John Paul II" (the surname is "Wojtyla")
"Jose Ortega y Gasset"
"Joseph K. von Eichendorff"
"Leonardo da Vinci" (he had no surname, "da Vinci" means "from the village
'Vinci'")
"Leone Stella" (this only is fictional, from a comic movie, whose main
characters are some "Leone Stella" and a "Stella Leone"
"Levi Strauss"
"Lewis Carroll" ("Lewis" may be both name and surname, and here the 
surname is "Dodgson")
"Lorenzo il Magnifico"
"Lucio Lombado Radice"
"Luigi Luca Cavalli-Sforza"
"Malcolm X"
"Mao Zedong"
"Martin Luther"
"Martin Luther King Jr."
"Mihail Sergeevic Gorbacev"
"Robert Van der Plas"
"Saint Augustine"
"Scuola di Barbiana"
"Subcomandante Insurgente Marcos"
"Vladimir Lenin" (the surname is "Ulyanov")

########################################################################