Sonny Software

The newest version of Bookends now supports unicode text in import filter tags, however I'm still having trouble writing one for the national central library web site. Anyone get this to work?

Just wondering... are you trying to set up an internet-enabled import filter, or an import filter that only works on text files/clipboard text? If the former, what server information, etc. are you working with? My quick searching didn't turn up obvious candidates, just the following:

It appears that NBINET is the name of the NCL/pan-Taiwan z39.50 server: http://nbinet1.ncl.edu.tw/screens/opac_menu.html has a link to http://nbinet.ncl.edu.tw:211/z39m

So I duplicated the Hong Kong U of Science and Technology import filter (arbitrary choice), and under the Internet tab entered nbinet.ncl.edu.tw for the Server Address and 211 for the Server Port... beyond this I'm clueless.

Also, I recall from an e-mail exchange with Jon that the Library of Congress 'Gateway' that Bookends is using is *not* unicode compatible, so I'm doubtful (but please correct me if I'm wrong) that an internet-enabled import filter would not work (on entries containing unicode, such as Chinese entries) until that's available.

p.s. When I tried mimicking the HKU entry by entering z39m,nbinet.ncl.edu.tw as the Server Address and ran a search, I got an error message that didn't fit in the error dialog (it was cut off at the end). It said:

Your gateway doesn't include a search form named <B>/prod/www/data/z3950/innopac</B>. Please read the lsite documentation for

The Bookends documentation PDF didn't list any results when I searched for lsite.

[Edit: added the following]

I found some information (dated May 2004) about NBINET in http://hkcan.ln.edu.hk/papers/asaihl.pdf :

Taiwan
In Taiwan, the National Bibliographic Information Network (NBINet) is the leading bibliographic utility under the auspice of the National Central Library (NCL). In 1998, to meet the needs of the Taiwan region for Chinese name authority control, the NCL cooperated with the National Taiwan University Library (NTUL) to build the Chinese name authority file. The initial authority records database was built by merging records from the two libraries. At present, the Chinese Name Authority Database (CNAD) 中文名稱權威資料庫 contains over 300,000 records. The authority records are in Chinese MARC (CMARC format). Sample authority record from CNAD (CMARC) :

Control # 001 31834
100 $a19950621achiy09 ea
Fixed-length data 152 $aCCR
Heading (Personal Name) 200 1 $a賈$b恩紱$f1866-1945? 400
0 $aP’ei-ch’ing,$f1866-1945? 400
0 $aSsu-i-ts’ao-lu,$f1866-1945? 400
0 $aSsu-i-ts’ao-t’ang,$f1866-1945? 400
0 $a思易草堂$f1866-1945? 400
0 $a思易草廬$f1866-1945? 400
0 $a佩卿$f1866-1945? 400
1 $aChia,$bEn-fu, $f1866-1945? 400
1 $aChia,$bP’ei-ch’ing,$f1866-1945? 400
1 $aJia,$bEnfu, $f1866-1945?
See From 400 1 $a賈$b佩卿$f1866-1945?
Cat. Source 801 0 $acw$b中圖$c199506 810
$a中國近現代人物名號大辭典, 民82 :$b面 718 810
$a定縣志. 台北市: 成文, 民58
Source Data found 810
$aHis Tsao-Ch’ang hsien chih, 1987 :$bcaption (Chia En-fu)
General Cataloguer’s note 830 $a天津人, 字佩卿; 室名思易草堂, 一作思易草廬

If there is a Z50 import that could import Chinese language text records, that would be great - but I don't know of any. So I'm just looking to copy the record displayed on the web page of the National Central Library and paste it into Bookends.

You mean import it into Bookends from the clipboard, I imagine. If it looks like the example that Joe posted (CMARC), I don't think you can get a terribly satisfactory import, but it should be possible to import chunks of data to individual fields.

Jon
Sonny Software

Great, I'm glad to hear there's hope. Here's a URL with a MARC record (at least it says MARC, not CMARC) that I got by searching for a book on the library's website:

http://lib.ncl.edu.tw/cgi-bin/uz1?routi ... =101321044

If you switch your encoding to "Traditional Chinese (BIG5)", you should see something like this.

MARC FORMAT


系統號:101321044
000 cam2
001 101321044
0100 $a 9579405999 $b 平裝 $d NT$330
100 $a 20021113d1996 k y0chib09 e
1010 $a chi
102 $a cw
105 $a a ai 000yy
2001 $a 光復初期臺灣經濟政策的檢討 $f / 劉士永著
205 $a 初版
210 $a 臺北縣板橋市 $c : 稻鄉 $d , 民85
2150 $a 308面 $c : 圖 $d ; 21公分
2252 $a 臺灣文化系列 $v ; 10
320 $a 參考書目: 面275-302
410 0$1 2001 $a臺灣文化系列$v10
606 $a 經濟 $y -臺灣 $z -光復以後(1945- ) $2 csh
681 $a 552.2832 $b 8746 $v 增訂七版
700 1$a 劉 $b 士永
801 0$a cw $b ISBN $c 19960111 $g CCR
801 1$a cw $b ISBN $c 19960111 $g CCR
801 2$a cw $b 國圖 $c 20031016 $g CCR
002542233中文書庫 552.2832 8746 限館內閱覽

Maybe there's a detailed description of how the site constructs/defines its MARC code? Or, at worst, example searches could be used to create an import filter.

This looks pretty much like standard MARC21, so it should be importable if you turn on MARC in the filter (I'd just start with an existing MARC importer, make a New filter based on that, remove the internet search info, and use it for importing from the clipboard).

Jon
Sonny Software

Thanks, Jon. I tried making a new filter based on the Milwaukee Public Library (my hometown), copied the text starting with line 000 into the clipboard, and tried to import, but I got: "Done. Number Imported: 0". Under File > Import Text Encoding, I had chosen UTF-8 - thinking that this is what the clipboard should be using (even though the website is BIG5 encoded, as I mentioned above). I also pasted the clipboard into TextWrangler and saved the file as UTF-8, and tried to run it through the importer. No luck.

Do you have any suggestions? I'll e-mail you the text file.

Thanks again,
Joe

Make sure to change "reference starts with" form an * to a real tag (like 000).

I'll take a look at the filter you sent me and get back to you.

Jon
Sonny Software

Thanks again, Jon. FYI, Luhmann, here was Jon's response:

Jon wrote:Sorry, Joe, but I can't get this to work. It looks like it would require a lot of work to look into this, so for now at least it's not possible.

Hi Joe,

I think the Chinese (UTF-8 ) MARC import will work in the next update. Please try it when it's released.

Jon
Sonny Software

Great! Thanks, Jon.

I just successfully imported a Chinese language MARC record! Great job with the latest update!

Luhmann wrote:I just successfully imported a Chinese language MARC record! Great job with the latest update!

Same here - thanks so much, Jon!

I'm still sorting out a few details on the filter I'm working on: a Harvard library filter. The nice thing about their entries for CJK titles is that they give romanization too. Once polished and tested, I'll share it here.

Has anyone gotten a Taiwan Central Library filter together (or any other filters from Taiwanese or Mainland libraries for that matter?) If so, I'd love to get ahold of a copy.

I found this thread and I wonder if anyone has succeeded in importing from Taiwan ......

Thanks............

YingLam