bibtex8 error, capacity exceeded

[nutsmuggler]

Hello folks.
I am using biblatex to manage my bibliography. I have plenty of entries
begining with non ascii characters (Ó, Á, etc); my final biblography
reported all these entries after the Z.
So I thought I would use bibtex8; yet for some reason I cannot compile
and get an error message, saying that I have exceeded the memomry
capacity of bibtex (sorry, I can't report it verbatim, I am using a
different workstation now). I have tried the bibtex 8 --big --huge --
and --wolfgang option, but get the same error message.
My .bib file is codified in utf8, contains 274 entries, and compiles
perfectly with bibtex. Why doesn't it compile with bibtex8? Any
suggestion?
Cheers,
Davide

[Philipp Lehman]

nutsmuggler wrote:

> I am using biblatex to manage my bibliography. I have plenty of
> entries begining with non ascii characters (Ó, Á, etc); my final
> biblography reported all these entries after the Z.

Yes, that's what is to be expected.

> So I thought I would use bibtex8; yet for some reason I cannot
> compile and get an error message, saying that I have exceeded the
> memomry capacity of bibtex (sorry, I can't report it verbatim, I am
> using a different workstation now). I have tried the bibtex 8 --big
> --huge -- and --wolfgang option, but get the same error message.

The manual of the latest biblatex version includes a section (2.4.1)
which discusses this problem.

> My .bib file is codified in utf8, contains 274 entries, and compiles
> perfectly with bibtex.

Neither legacy bibtex nor bibtex8 support UTF-8 (or even have any
concept of a multibyte encoding). If you think it compiles fine,
you're probably overlooking some of the finer sorting errors (apart
from the obvious ones you mentioned).

You have two options:

1) Stay with UTF-8 as the main encoding on the Latex side. In this
case your bib file must be pure Ascii! That is, instead of things
like:

author = {José Vázques},
title = {?uvres completès},
location = {München},

you'd have:

author = {Jos{\'e} V{\'a}zques},
title = {{\OE}uvres complet{\`e}s},
location = {M{\"u}nchen},

There is absolutely no way to make bibtex or bibtex8 handle UTF-8
properly.

2) Switch to a suitable 8-bit encoding such as Latin1 or Latin9 on the
Latex side and use the same encoding in your bib file. Invoke bibtex8
like this:

bibtex8 --csfile <encoding>.csf --wolfgang --mcites 30000 --mentints
30000 --mentstrs 40000 --mfields 250000 --mstrings 35000 --mpool
750000 --mwizfuns 20000

This will solve the capacity issues.

--
Sender address blackholed; do not reply to From: address.
You can still reach me by email at: plehman gmx net.

[Philipp Lehman]

nutsmuggler wrote:

> I am using biblatex to manage my bibliography. I have plenty of
> entries begining with non ascii characters (Ó, Á, etc); my final
> biblography reported all these entries after the Z.

Yes, that's what is to be expected.

> So I thought I would use bibtex8; yet for some reason I cannot

> compile and get an error message, saying that I have exceeded the
> memomry capacity of bibtex (sorry, I can't report it verbatim, I am
> using a different workstation now). I have tried the bibtex 8 --big
> --huge -- and --wolfgang option, but get the same error message.

The manual of the latest biblatex version includes a section (2.4.1)
which discusses this problem.

> My .bib file is codified in utf8, contains 274 entries, and compiles
> perfectly with bibtex.

Neither legacy bibtex nor bibtex8 support UTF-8 (or even have any

concept of a multibyte encoding). If you think it compiles fine,
you're probably overlooking some of the finer sorting errors (apart
from the obvious ones you mentioned).

You have two options:

1) Stay with UTF-8 as the main encoding on the Latex side. In this
case your bib file must be pure Ascii! That is, instead of things
like:

author = {José Vázques},

title = {Œuvres completès},

[nutsmuggler]

The problem is still there...

> You have two options:
>
> 1) Stay with UTF-8 as the main encoding on the Latex side. In this
> case your bib file must be pure Ascii!

This would drive me mad: yes, I can use scripts, serch and replace and
stuff, but I am working on a multilingual thesis; I don't think I want
to convert all non-ascii characters in the .bib file

> 2) Switch to a suitable 8-bit encoding such as Latin1 or Latin9 on the
> Latex side and use the same encoding in your bib file. Invoke bibtex8
> like this:
>
> bibtex8 --csfile <encoding>.csf --wolfgang --mcites 30000 --mentints
> 30000 --mentstrs 40000 --mfields 250000 --mstrings 35000 --mpool
> 750000 --mwizfuns 20000
>

I've done this. Yet I get the same error; I can now post the log:

bibtex8 --csfile latin1.csf --wolfgang --mcites 30000 --mentints 30000

--mentstrs 40000 --mfields 250000 --mstrings 35000 --mpool 750000

--mwizfuns 20000 tesi

Error: cannot open CS file: latin1.csf
The top-level auxiliary file: tesi.aux
The style file: biblatex.bst
A level-1 auxilliary file: introduzione.aux
A level-1 auxilliary file: introduction.aux
A level-1 auxilliary file: orality.aux
Database file #1: blx-tesi.bib
Database file #2: bibLatin.bib
Sorry---you've exceeded BibTeX's buffer size 3000

Aborted at line 0 of file bibLatin.bib
(That was a fatal error)

Any suggestion?

[nutsmuggler]

Update: I get the same mistake with

bibtex8 --csfile 88591lat.csf00 --mentints 30000 --mentstrs 40000

--mfields 250000 --mstrings 35000 --mpool 750000 --mwizfuns 20000 tesi

(latin1.csf doesn't exist; I think 88591lat is the right file)

Still, nothing solved.
Cheers,
Davide

[nutsmuggler]

Update again: solved.
The problem derives from my bib file, which is converted from a
macroman encoded file. bibtex8 does not recognize the macroman line
endings and thus reads all the entries as one big line; thus the buffer
is exceed.
Thanks for your previous feedback,
Davide

On 24 Gen, 16:37, "nutsmuggler" <benini.dav...@gmail.com> wrote:
> Update: I get the same mistake with
>

> bibtex8--csfile 88591lat.csf00 --mentints 30000 --mentstrs 40000