Problem is with MARC records. Same MARC record has shown in the
Catalog but if the record title has turkish character then has not at
all. Search result says "nothing found".
On 10 Nisan, 22:12, "don.mo...@gmail.com" <don.mo...@gmail.com> wrote:
Is there any idea about this problem? Casey, Lichen? Do you think any
solution for this? First, i thought there is a problem about our MARC
records. But when i have tried import the same record with turkish
character and without turkish character i saw that there is not a
problem with our records. (unimarc) This is a Scriblio bug. If your
records have just english character you can easily import the
Scriblio. But what will we do? Do you think just english people use
Scriblio? Is there anybody use scriblio with korean, chinise,
japanese, arabic marc records? What are you then? Why somebody doesn't
help? Where is the project managers?
We at Hong Kong also encountered various problems on CJK (Chinese, Japanese, Korean) characters during initial installation of Scriblio back in December last year. We did some customization to enable the importing, display, indexing and searching of CJK characters, such as:
1. Inspect php codes that involve string conversion functions. For example, ucwords() works for ASCII characters only, and not for multibyte characters such as UTF-8. And mb_convert_encoding(..., 'UTF-8') is not needed if the source string is already UTF-8. And htmlentities(...,UTF-8) should be htmlentities(...,'UTF-8') etc. etc.
2. Enhance importer to support capturing of vernacular characters, which usually appear in tag 880 of MARC21 records.
3. Turn on Sphinx for CJK character searching. And add TSVCC (Traditional, Simplified and Variant Chinese Characters) linking as a "word form file" in Sphinx to enable searching of various forms of Chinese characters.
I will work with Casey to see how we can contribute our customization to the Scriblio code base.
Zebil, I think your problem is very likely related to Item 1 above. It may be useful if you can show us an URL that will bring up a record on your ILS system so that we can take a look.
Regards, K.T. Lam Head of Library Systems The Hong Kong University of Science and Technology Library
Reply-To: scriblio@googlegroups.com To: Scriblio <scriblio@googlegroups.com> References: <3F6EB7BC-A27E-4413-B4EC-B9BB9F1EC643@gmail.com> <d432d0d5-e4b7-4653-9d85-1790e3fa5...@f63g2000hsf.googlegroups.com> <999d73a5-3e47-4ec7-a14f-672458852...@v26g2000prm.googlegroups.com>
Is there any idea about this problem? Casey, Lichen? Do you think any solution for this? First, i thought there is a problem about our MARC records. But when i have tried import the same record with turkish character and without turkish character i saw that there is not a problem with our records. (unimarc) This is a Scriblio bug. If your records have just english character you can easily import the Scriblio. But what will we do? Do you think just english people use Scriblio? Is there anybody use scriblio with korean, chinise, japanese, arabic marc records? What are you then? Why somebody doesn't help? Where is the project managers?
Thank you for sharing your experience with this. I'd very much like to incorporate both the extended character set support and the many other fixes and enhancements you've made into Scriblio.
HKUST's efforts are really important, especially at this early stage of the project, and demonstrate how OSS is built upon the contributions from those involved with a project.
> We at Hong Kong also encountered various problems on CJK > (Chinese, Japanese, Korean) characters during initial > installation of Scriblio back in December last year. We > did some customization to enable the importing, display, > indexing and searching of CJK characters, such as:
> 1. Inspect php codes that involve string conversion > functions. For example, ucwords() works for ASCII > characters only, and not for multibyte characters > such as UTF-8. And mb_convert_encoding(..., 'UTF-8') > is not needed if the source string is already UTF-8. > And htmlentities(...,UTF-8) should be > htmlentities(...,'UTF-8') > etc. etc.
> 2. Enhance importer to support capturing of vernacular > characters, which usually appear in tag 880 of MARC21 > records.
> 3. Turn on Sphinx for CJK character searching. And add > TSVCC (Traditional, Simplified and Variant Chinese > Characters) linking as a "word form file" in Sphinx > to enable searching of various forms of Chinese > characters.
> I will work with Casey to see how we can contribute our > customization to the Scriblio code base.
> Zebil, I think your problem is very likely related to > Item 1 above. It may be useful if you can show us an URL > that will bring up a record on your ILS system so that > we can take a look.
> Regards, > K.T. Lam > Head of Library Systems > The Hong Kong University of Science and Technology Library
> Is there any idea about this problem? Casey, Lichen? Do you think any > solution for this? First, i thought there is a problem about our MARC > records. But when i have tried import the same record with turkish > character and without turkish character i saw that there is not a > problem with our records. (unimarc) This is a Scriblio bug. If your > records have just english character you can easily import the > Scriblio. But what will we do? Do you think just english people use > Scriblio? Is there anybody use scriblio with korean, chinise, > japanese, arabic marc records? What are you then? Why somebody doesn't > help? Where is the project managers?
