These are dev questions, so moved to koha-devel. On 2003-09-19 08:34:10 +0100 paul POULAIN <paul.poulain@free.fr> wrote:
To support any MARC flavour, NO MARC TAG/SUBFIELDS ARE HARDCODED into Koha.
The MARC export appears to call as_usmarc(), which needs MARC21 to be used to get valid results, doesn't it? I spoke on Wednesday's IRC with jferraro (excuse misspelling) who was having some trouble with MARC export. Any progress? -- MJR/slef My Opinion Only and possibly not of any group I know. http://mjr.towers.org.uk/ gopher://g.towers.org.uk/ slef@jabber.at Creative copyleft computing services via http://www.ttllp.co.uk/
MJ Ray wrote:
These are dev questions, so moved to koha-devel.
On 2003-09-19 08:34:10 +0100 paul POULAIN <paul.poulain@free.fr> wrote:
To support any MARC flavour, NO MARC TAG/SUBFIELDS ARE HARDCODED into Koha.
The MARC export appears to call as_usmarc(), which needs MARC21 to be used to get valid results, doesn't it?
I spoke on Wednesday's IRC with jferraro (excuse misspelling) who was having some trouble with MARC export. Any progress?
Hi Mark, joshua tested the exported records with screen copy/paste. As there are some non-printable characters in iso2709, the result was wrong, of course. The as_usmarc() creates iso2709 records, not us_marc specifically ones. Let me explain how it's used in Koha : * the MARCgetbiblio() sub (a Koha one) builds a MARC::Record from Koha-DB. The MARC::Record in independant of the MARC flavour. * the as_usmarc() is applied on the MARC::Record object. It creates a valid iso2709 data. * some MARC::Record methods like $record->title() won't work. It returns the title that is located in 100$a in usmarc. In 100$a in UNIMARC is ... something else (don't remember). so Koha NEVER uses MARC::Record shortcuts like ->title or ->author. It uses an internal sub (MARCmarc2koha) that transforms a MARC::Record into an hash, having {title} being the title {author} being the author... Koha uses MARC parameter table to do this, so it's MARC flavour independant. * I think/am afraid extended char (é,è,à,...) are wrong. they are a pain in iso2709. HTH -- Paul POULAIN Consultant indépendant en logiciels libres responsable francophone de koha (SIGB libre http://www.koha-fr.org)
On 2003-09-29 08:30:09 +0100 paul POULAIN <paul.poulain@free.fr> wrote:
The as_usmarc() creates iso2709 records, not us_marc specifically ones.
Isn't that a bug in MARC::Record? Can other things read iso2709? They seem to want MARC21 or USMARC.
* I think/am afraid extended char (é,è,à,...) are wrong. they are a pain in iso2709.
MARC21 has a leader byte that indicates some form of unicode to be used, IIRC. -- MJR/slef My Opinion Only and possibly not of any group I know. http://mjr.towers.org.uk/ gopher://g.towers.org.uk/ slef@jabber.at Creative copyleft computing services via http://www.ttllp.co.uk/
MJ Ray wrote:
* I think/am afraid extended char (é,è,à,...) are wrong. they are a pain in iso2709.
MARC21 has a leader byte that indicates some form of unicode to be used, IIRC.
which one pls ? i can't find it in UNIMARC, so i'm waiting your answer anxiously... -- Paul POULAIN Consultant indépendant en logiciels libres responsable francophone de koha (SIGB libre http://www.koha-fr.org)
On 2003-09-30 15:19:18 +0100 paul POULAIN <paul.poulain@free.fr> wrote:
MARC21 has a leader byte that indicates some form of unicode to be used, which one pls ? i can't find it in UNIMARC, so i'm waiting your answer anxiously...
http://www.loc.gov/marc/umb/um07to10.html#part9 says that an "a" is position 9 signifies UCS/Unicode, but I don't know if that's UTF-8 or otherwise. If someone could find out from the MARC::Record and MARC21 experts how to use this properly, I'd be very happy. -- MJR/slef My Opinion Only and possibly not of any group I know. http://mjr.towers.org.uk/ gopher://g.towers.org.uk/ slef@jabber.at Creative copyleft computing services via http://www.ttllp.co.uk/
MJ Ray wrote:
On 2003-09-30 15:19:18 +0100 paul POULAIN <paul.poulain@free.fr> wrote:
MARC21 has a leader byte that indicates some form of unicode to be used,
which one pls ? i can't find it in UNIMARC, so i'm waiting your answer anxiously...
http://www.loc.gov/marc/umb/um07to10.html#part9 says that an "a" is position 9 signifies UCS/Unicode, but I don't know if that's UTF-8 or otherwise. If someone could find out from the MARC::Record and MARC21 experts how to use this properly, I'd be very happy.
UNIMARC => leader => digit 9 => undefined Not sure wether it's good or bad for us. -- Paul POULAIN Consultant indépendant en logiciels libres responsable francophone de koha (SIGB libre http://www.koha-fr.org)
On 2003-09-30 15:19:18 +0100 paul POULAIN <paul.poulain@free.fr> wrote:
which one pls ? i can't find it in UNIMARC, so i'm waiting your answer anxiously...
Further to my last message, http://www.loc.gov/marc/specifications/speccharucs.html seems to say that it is the popular UTF-8 encoding. -- MJR/slef My Opinion Only and possibly not of any group I know. http://mjr.towers.org.uk/ gopher://g.towers.org.uk/ slef@jabber.at Creative copyleft computing services via http://www.ttllp.co.uk/
participants (2)
-
MJ Ray -
paul POULAIN