to the travails below, I must add one more example -- yesterday I
asked BD to import stuff from a Library of Congress search from within
BD. It was amazing how easily I was able to search for a book and
import it into BD all from within BD. Well, even those entries were
running afoul of LyX... illegal characters in the Annote field. Again,
making BD strict from the get-go might be very useful. There could be
modes of strictness... sort of like IGNORE, WARN, or CROAK.
On 10/30/07, P Kishor <[EMAIL PROTECTED]> wrote:
> Dear Developers, (long-ish email... for the punchline, just read the
> last para, else read the entire thing),
>
> First, thanks to Adam M. I am on my way now happily adding and
> subtracting entries to groups. Indeed, the pasteboard server, whatever
> that is, had died, and rebooting brought everything back to normal.
>
> Now, a list of suggestions (mind it, this is NOT a list of complaints).
>
> We have had long, continuing discussions on what BD really is -- in
> the minds of many, dare I say all, it is the winningest bib manager.
> For those who work with TeX, it is also makes for a robust citation
> system. But, at least for me, who prefers to stay away from TeX,
> making the final product, a document with citations and
> bibliographies, is still quite difficult.
>
> First thing, perhaps the most important thing, I would suggest is to
> make BD as strict as a "strongly typed language." For now, I can stuff
> anything in BD, and it happily takes it. The problem comes when I try
> to interface BD with any other program. Some of you might recall that
> I was having the darndest time trying to import my bib into CiteULike
> or Connotea. There were many opaque errors (or no error messages at
> all), but the closest descriptive error I would get was that some of
> my entries were too long. Well, the culprits were the Annote and
> Abstract fields that I had merrily filled with my notes. In one case,
> I had stuffed the text of the entire paper in the Annote field... it
> was convenient for me to look up the plain text, and BD didn't
> complain when I stuffed the text in. Well, apparently it violated
> BibTeX's rules.
>
> Moving on, I gave up on CiteULike and Connotea, for the time being,
> and went back to BD. Until yesterday -- I tried using LyX, as much as
> I disliked using such an "anvil and hammer" looking system on my Mac.
> It seemed to work. Funnily, LyX's citation inserting workflow was
> really smooth and quick. Soon, I had my article text in LyX, with
> several citations scattered within. Then I decided to do a preview in
> PDF, and all hell broke loose. All kinds of errors that I couldn't
> even figure out where they were coming from. After much diagnosis, I
> realized that LyX was complaining about errors in my BD database.
> Because there were "illegal characters" (mostly quotation marks in
> Abstract and Annote fields), LyX was not even processing my document
> for replacing \cite{...} with citations. Seems like I can't type
> something "as simple as this" in the Abstract field without violating
> BibTex rules (Tex items at zero-depth... etc.)
>
> So, I tried my Perl script for doing a scan and replace. Well, I got
> the following --
>
> =================
> ~/bibliography.bib, line 69, warning: found " at brace-depth zero in
> string (TeX accents in BibTeX should be inside braces)
> ~/bibliography.bib, line 140, warning: found " at brace-depth zero in
> string (TeX accents in BibTeX should be inside braces)
> lexical buffer overflowed (reallocating to 4000 bytes)
> ..
> lexical buffer overflowed (reallocating to 10000 bytes)
> lexical buffer overflowed (reallocating to 12000 bytes)
> lex_auxiliary.c:162: failed assertion `(txt[0] == '{' && txt[len-1] ==
> '}') || (txt[0] == '\"' && txt[len-1] == '\"')'
> Abort trap
> =================
>
> Finally, getting desperate, I went into my bibliography using
> TextWrangler, cropped all text entries to really short (to avoid the
> "lexical buffer overflows") removed all double-quotes, and was finally
> able to import the entire db into RefWorks (web edition). Then I
> re-inserted the citations in my text using RefWorks schema, and was
> successfully able to have RefWorks scan and replace to create a decent
> looking manuscript in RTF.
>
> The cleaned up bib also imported successfully into CiteULike and Connotea.
>
> BD + some word processor of choice (my choice happens to be Apple's
> excellent Pages) workflow is still a long ways away. I can very nicely
> and smoothly insert \cite{...} keys into Scrivener, but then I can't
> do much with that. However, many of my problems would have been
> avoided if I had made a "valid" BD database in the first place.
>
> For that, I urge the developers (who, by the way, are compleatly
> wonderful folks) to make BD more strict by not allowing me, the user,
> enter into it anything that would be illegal downstream.
>
> Many thanks,
>
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Bibdesk-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bibdesk-users
