Re: [Bibdesk-users] Two columns template for the preview pane
On Sat, Oct 2, 2010 at 11:55, Christiaan Hofman cmhof...@gmail.com wrote: Yes, I did find a workaround for the issue! Anyway, bug filed with Apple. Yay, thanks a lot, as always. JiHO --- http://maururu.net -- Virtualization is moving to the mainstream and overtaking non-virtualized environment for deploying applications. Does it make network security easier or more difficult to achieve? Read this whitepaper to separate the two and get a better understanding. http://p.sf.net/sfu/hp-phase2-d2d ___ Bibdesk-users mailing list Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] Two columns template for the preview pane
On Fri, Oct 1, 2010 at 12:09, Christiaan Hofman cmhof...@gmail.com wrote:
Everything works perfectly except that changes to an external
CSS file are not reflected in the preview pane after a refresh (as are
changes to the HTML files, including to inline CSS declarations in an
HTML file). A restart of BibDesk is required.
That makes no sense, as there's absolutely nothing cached, only the file
location is remembered.
I have just retested it with Bibdesk 2010-09-30 and the behaviour is
the same. Here are some steps to reproduce it:
- create a new HTML template with
test.html: Main Page
!DOCTYPE HTML PUBLIC -//W3C//DTD HTML 4.01 TRANSITIONAL//EN
http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd;
html xmlns=http://www.w3.org/1999/xhtml;
head
meta http-equiv=content-type
content=text/html; charset=utf-8 /
link rel=stylesheet href=test.css
type=text/css
media=screen title=no title charset=utf-8/
/head
body
$publicationsUsingTemplate/
/body
/html
testItem.html: Default Item
$fields.Title/
test.css: Accessory file
body { color: black; }
- select this new HTML template in the preview
- change the color from black to red in the css and save the file
- add something to testItem.html, like: Title: $fields.Title/
- go back to BibDesk and choose a new bibliography record, or change
the template in the preview and change it back to the test one. Both
these things refresh the preview pane and change the text according to
the change in testItem.html but do not change the color of the text
- add some inline CSS in test.html
style type=text/css media=screen
body{ color: blue; }
/style
- now refresh the view (as above) and it changes the color to blue.
- remove the inline css and refresh; the text is back to black
- restart BibDesk; the text is now red.
I hope that helps pinpoint the issue.
JiHO
---
http://maururu.net
--
Start uncovering the many advantages of virtual appliances
and start using them to simplify application deployment and
accelerate your shift to cloud computing.
http://p.sf.net/sfu/novell-sfdev2dev
___
Bibdesk-users mailing list
Bibdesk-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] Two columns template for the preview pane
The build failed last night, and I forgot to report that; there were errors building the QL plugin and some frameworks, IIRC. OK, I downloaded the build from the 28th and tested some modifications with it. Everything works perfectly except that changes to an external CSS file are not reflected in the preview pane after a refresh (as are changes to the HTML files, including to inline CSS declarations in an HTML file). A restart of BibDesk is required. I updated my styles, in RTFD and HTML, according to Christiaan's advice. I added some icons, a little functionality and some color (since the links force some blue). The result is there: http://maururu.net/2010/two-columns-bibdesk-template/ as before The remaining issues (which are not BibDesk related) are: - when using HTML tables, the maximum width is fixed to something less than the full window width (~800px). So the two columns won't extend further. It could be considered a feature though: it enhances the readability of the abstract by ensuring a decent number of words per line ;) - in RTF, I cannot find a way to have part of a line aligned to the left and another part aligned to the right. In HTML I use a table with two columns and text-align: left in the left column, text-align:right in the right column. Doing so in RTF works but the content of the right column can actually bleed out from the column (i.e. the minimum column width is not fixed by the content). In consequence, If I put an image in the right column (the magnifying glass icon in my case) it bleeds out of the window and its right half is cut. As a consequence I placed this icon immediately after the title in the RTF template, which is less efficient UI wise, because it moves around and is therefore less reliable to click. I hope this template can be useful to others now. JiHO --- http://maururu.net -- Start uncovering the many advantages of virtual appliances and start using them to simplify application deployment and accelerate your shift to cloud computing. http://p.sf.net/sfu/novell-sfdev2dev ___ Bibdesk-users mailing list Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] Two columns template for the preview pane
On Fri, Sep 24, 2010 at 11:37, Christiaan Hofman cmhof...@gmail.com wrote: The raw RTF data is just a proprietary format used to save the data of the rich text. It could also have been some other format (html, doc, docx, odt.) In code, we don't see any of these details. What we have in code is a string-with-attributes (NSAttributedString). We parse the string (looking for template tags), trying to keep the attributes intact, and possibly adding attributes (depending on the replacement.) So the template tags are in the text, not in the attributes, and certainly not in some underlying raw data representation that was used to save the template. This also makes it possible to edit in the template tags, because the text is what the person editing the template actually sees. I see. I thought this data was actually in the RTF format (which, even though proprietary and gibberish-looking is a public standard, for which a parser could be built) while it looks more like it is Rich Text data with no assumption regarding the underlying format. In tomorrow's nightly we will also parse the link attributes in an extra parser run. Thats very cool! I'll get the nightly and modify the two templates to work with it, in the exact same way (it should to make them equivalent now). Thanks again for your work and reactivity. JiHO --- http://maururu.net -- Start uncovering the many advantages of virtual appliances and start using them to simplify application deployment and accelerate your shift to cloud computing. http://p.sf.net/sfu/novell-sfdev2dev ___ Bibdesk-users mailing list Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] Two columns template for the preview pane
On Mon, Sep 27, 2010 at 00:58, Christiaan Hofman cmhof...@gmail.com wrote: In tomorrow's nightly we will also parse the link attributes in an extra parser run. Oops, it seems the latest nightly is from the 26th. I'll wait for this tomorrow then. JiHO --- http://maururu.net -- Start uncovering the many advantages of virtual appliances and start using them to simplify application deployment and accelerate your shift to cloud computing. http://p.sf.net/sfu/novell-sfdev2dev ___ Bibdesk-users mailing list Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] Two columns template for the preview pane
On Wed, Sep 22, 2010 at 18:51, Christiaan Hofman cmhof...@gmail.com wrote:
I have few questions:
- what are stringByDeTeXifyingString and stringByTeXifyingString
supposed to do? Neither of those does anything: when I have latex
commands (\textit, \circ etc.) they do not replace, apply, or suppress
them.
They replace tex commands that are also used as escapes, such as {\'e}, and
they also clean some braces. They certainly do not convert all tex commands,
because we're not a tex parser.
Ok. An I guess DeTeXifying means removing those commands without
affecting the output while TeXifying is converting those commands into
their RTF equivalent. i.e.:
{\'e} -- DeTeX -- e
{\'e} -- TeX -- é
Is that right?
If it is, what is the difference between DeTeX and stringByRevingTeX?
- I am using a table with one line and two columns. Whether its width
is specified as 100% or unspecified, it seems limited to a maximum
with (it does not take the whole width when bibdesk is maximized on a
large screen). Is that expected also?
- More generally, I find that BD renders the HTML differently from
Safari (of CSS Edit). For example, it does not seem to honor the
padding settings on p elements.
It's converted to rich text (NSAttributedString), and rich text support for
html is limited wrt the webview.
OK, so I am actually better off doing everything in RTF directly then.
My only big remaining issue is that BibDesk template values are not
expanded within links. For example if I have an RTF file with a only a
link to the DOI:
{\rtf1\ansi\ansicpg1252\cocoartf1038\cocoasubrtf320
{\fonttbl\f0\froman\fcharset0 Palatino-Roman;}
{\colortbl;\red255\green255\blue255;}
{\info
{\author
Foo}}\paperw11900\paperh16840\margl1440\margr1440\vieww9000\viewh8400\viewkind0
\pard\tx566\tx1133\tx1700\tx2267\tx2834\tx3401\tx3968\tx4535\tx5102\tx5669\tx6236\tx6803\ql\qnatural\pardirnatural
{\field{\*\fldinst{HYPERLINK http://dx.doi.org/%3C$string.Doi/%3E}}{\fldrslt
\f0\fs22 \cf0 $string.Doi/}}}
The $string.Doi/ element of the link is converted by TextEdit in
HTML entities: %3C$string.Doi/%3E. Even if I replace the %3*
commands by and with a plain text editor, the link still points to
http://dx.doi.org/$string.Doi/ and not to the expanded form (i.e.,
not the actual value of the DOI). Is there any way that BibDesk can
expand template values in links.
As a subsidiary question: I can't seem to be able to edit the
appearance of links in RTF. They always are underlined and blue. Is it
impossible to have them look like regular text?
Thanks in advance,
JiHO
---
http://maururu.net
--
Start uncovering the many advantages of virtual appliances
and start using them to simplify application deployment and
accelerate your shift to cloud computing.
http://p.sf.net/sfu/novell-sfdev2dev
___
Bibdesk-users mailing list
Bibdesk-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] Two columns template for the preview pane
Thanks for the help Christiaan.
On Thu, Sep 23, 2010 at 11:38, Christiaan Hofman cmhof...@gmail.com wrote:
Ok. An I guess DeTeXifying means removing those commands without
affecting the output while TeXifying is converting those commands into
their RTF equivalent. i.e.:
{\'e} -- DeTeX -- e
{\'e} -- TeX -- é
Is that right?
That's correct.
If it is, what is the difference between DeTeX and stringByRevingTeX?
I assume you mean stringByRemovingTeX. That's very different, it removes tex
commands.
Indeed that was a spelling mistake on my end. However, I don't see how
stringByRemovingTeX is different from stringByDeTeXifying. For the
simple example above it seems to do the same:
{\'e} e
Could you give me an example of where they are different, so that I
can use the correct one.
It's converted to rich text (NSAttributedString), and rich text support for
html is limited wrt the webview.
OK, so I am actually better off doing everything in RTF directly then.
I think so.
I actually went ahead on the HTML end and I think I got pretty good
results. I'll try to keep the HTML and RTF versions on par once I
understand the odds and ends of both.
My only big remaining issue is that BibDesk template values are not
expanded within links. For example if I have an RTF file with a only a
link to the DOI:
{\rtf1\ansi\ansicpg1252\cocoartf1038\cocoasubrtf320
{\fonttbl\f0\froman\fcharset0 Palatino-Roman;}
{\colortbl;\red255\green255\blue255;}
{\info
{\author
Foo}}\paperw11900\paperh16840\margl1440\margr1440\vieww9000\viewh8400\viewkind0
\pard\tx566\tx1133\tx1700\tx2267\tx2834\tx3401\tx3968\tx4535\tx5102\tx5669\tx6236\tx6803\ql\qnatural\pardirnatural
{\field{\*\fldinst{HYPERLINK
http://dx.doi.org/%3C$string.Doi/%3E}}{\fldrslt
\f0\fs22 \cf0 $string.Doi/}}}
The $string.Doi/ element of the link is converted by TextEdit in
HTML entities: %3C$string.Doi/%3E. Even if I replace the %3*
commands by and with a plain text editor, the link still points to
http://dx.doi.org/$string.Doi/ and not to the expanded form (i.e.,
not the actual value of the DOI).
I am not sure what your'e doing here, but you're not doing something you're
supposed to do. Don't mess with the RTF data directly, that won't be
interpreted as the template, the rich text you see in a rich text editor will
be.
Also, I don't know what $string.Doi/ should stand for? I think you want
something like $urls.Doi/.
Is there any way that BibDesk can
expand template values in links.
Use the linkedText key, e.g. $urls.Doi.linkedText/.
Sorry, this was not very clear. I want the value of the DOI (which was
$fields.Doi/, not $string.Doi/ that was another mistake from me)
to be a link to http://dx.doi.org/$fields.Doi/ so that clicking on
the DOI resolves the DOI on doi.org servers and reroutes the user to
the page of the publication, online.
$urls.Doi.linkedText/ does that but instead of displaying only the
DOI number it displays the full http:// url, which takes a little more
room.
Similarly, I want to make the a link that searches for the current
paper on Google Scholar. In HTML I can do this:
a
href=http://scholar.google.com/scholar?as_q=$fields.Title.stringByRemovingTeX.htmlString/as_sauthors=$authors.lastna...@firstobject/Foo/a
and Foo is a link to a Google Scholar search which has the first
author and the title as arguments. This works well but I am not sure
if/how I can do this in RTF.
Finally, in RTFD I can easily include images. In HTML I can easily
link to online images. However, adding an image in the templates
folder and using a relative link does not work. I would prefer to use
a local image rather than an online one so that it works offline too.
How can I include local images in HTML?
FYI, the point of all this is to display a magnifying glass as the
link to the Scholar search for example.
Thanks in advance,
JiHO
---
http://maururu.net
--
Start uncovering the many advantages of virtual appliances
and start using them to simplify application deployment and
accelerate your shift to cloud computing.
http://p.sf.net/sfu/novell-sfdev2dev
___
Bibdesk-users mailing list
Bibdesk-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] Two columns template for the preview pane
On Thu, Sep 23, 2010 at 18:55, Christiaan Hofman cmhof...@gmail.com wrote:
Ok. An I guess DeTeXifying means removing those commands without
affecting the output while TeXifying is converting those commands into
their RTF equivalent. i.e.:
{\'e} -- DeTeX -- e
{\'e} -- TeX -- é
Is that right?
That's correct.
Sorry, that's NOT correct.
DeTeX: {\'e} - é
TeX: é - {\'e}
OK, that makes more sense indeed.
They're not the same at all, they don't even overlap, certainly not in this
particular example. The first gives é, the second \'e. In general,
stringByRemovingTeX removes TeX commands and curly braces, while
stringByDeTeXifyingString converts TeX accents and TeX specials to
accented/special characters.
Thanks to the new explanation. I understand now that they are
completely different indeed. It was the previous explanation that
confused me.
Values you get through the fields key in a template are already DeTeXified.
We use stringByRemovingTeX for display values of strings like the title in
the main table as well as in the default RTF template.
So I don't need $fields.Title.stringByDeTeXifyingString/ for
example, $fields.Title/ is enough. Is that right?
Now, even with fields. (or with .stringByDeTeXifyingString) there is
some TeX left (because, as you say, you would not want to build a
complete TeX parser). Now, if I do:
$fields.Title.stringByRemovingTeX/
Is the title is *first* filtered to convert known tex values into
unicode and *then* the remaining tex is removed, or is all tex removed
before having a chance to be converted?
I know, but with RTF it is not possible to do exactly what you want, because
there's no way to pass an extra argument (the URL or displayed string) to a
template key. What I said is the closest you can get.
OK. I'll just use .linkedText and be happy with it in RTF then. But
overall, I think I'll move to HTML.
I know nothing of the implementation of course, but I imagined that
the RTF file was going through a parser of some sort which replaced
the $???/ bits by the appropriate values of the fields. I was
thinking that maybe this parser could be made to also replace those
strings within a HYPERLINK construct (where they are delimited by %3C
and %3E). This way, all imaginable link would be possible, as they are
in HTML. One would only have to create the link manually in the RTF
editor (rather than have the templating system create it for him with
.linkedText).
Also, for HTML you should $urls.Doi/ for the linked URL, not
http://dx.doi.org/$fields.Doi/, because we do some more validation and
cleaning that you may otherwise not get.
OK. I'll correct that.
If you use some value as part of a URL you should also always use
stringByAddingPercentEscapes or stringByAddingPercentEscapesIncludingReserved
(in this case the latter, because you want to escape everything including
characters like /=). And as noted above, you can only add links in RTF with
the text equal to the full URL or path.
I'll correct that also.
You cannot include images in a HTML preview. As I noted before, you cannot
have external files (like css or images), as it is based on data, not on
files. You can also not link to remote images, because a text view also does
not download anything.
Actually, remote images worked today, on the latest stable version of BibDesk.
In tomorrow's nightly you can add images and external .css files as accessory
files, and they will be used for the template based preview. Refer to them by
the filename (relative path).
That's very cool thanks! It helps to have the CSS separate, and the
possibility of adding pictures makes for more creative templates.
Thanks a lot,
JiHO
---
http://maururu.net
--
Nokia and ATT present the 2010 Calling All Innovators-North America contest
Create new apps games for the Nokia N8 for consumers in U.S. and Canada
$10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing
Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store
http://p.sf.net/sfu/nokia-dev2dev
___
Bibdesk-users mailing list
Bibdesk-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] Two columns template for the preview pane
On Fri, Sep 24, 2010 at 00:36, Christiaan Hofman cmhof...@gmail.com wrote:
I know nothing of the implementation of course, but I imagined that
the RTF file was going through a parser of some sort which replaced
the $???/ bits by the appropriate values of the fields. I was
thinking that maybe this parser could be made to also replace those
strings within a HYPERLINK construct (where they are delimited by %3C
and %3E).
Again, no, the hyperlinks are in the underlying /data/, while the parser
replaces template tags in the /text/. The hyperlinks are not text, they're
attributes.
I understand that they are not the same thing conceptually. But
practically they look almost the same in the file: they are just ASCII
characters. Maybe I know too little about the parser to really
understand, but, since a RTF file is actually just a collection of
ASCII character, I imagined that this parser just does some
find-and-replace operations on the template tags, probably based on a
regular expression that looks something like:
\\$.*?\\\
So I imagined that it would be equally able to perform these
operations on other regular expressions too.
To reuse my example above, a minimalist RTF file with just the DOI as
a link to the resolved DOI online would be:
{\rtf1\ansi\ansicpg1252\cocoartf1038\cocoasubrtf320
{\fonttbl\f0\froman\fcharset0 Palatino-Roman;}
{\colortbl;\red255\green255\blue255;}
{\info
{\author
Foo}}\paperw11900\paperh16840\margl1440\margr1440\vieww9000\viewh8400\viewkind0
\pard\tx566\tx1133\tx1700\tx2267\tx2834\tx3401\tx3968\tx4535\tx5102\tx5669\tx6236\tx6803\ql\qnatural\pardirnatural
{\field{\*\fldinst{HYPERLINK %3C$urls.Doi/%3E}}{\fldrslt
\f0\fs22 \cf0 $fields.Doi/}}}
The first template tag is %3C$urls.Doi/%3E is the destination of the
link, which should be expanded to http://dx.doi.org/value-of-doi, and
the second is $fields.Doi/, which should be expanded to
value-of-doi.
So, in the end, what I am saying is: can the parser recognise both
$***/ and %3C$***/%3E as tag delimiters?
This way, all imaginable link would be possible, as they are
in HTML. One would only have to create the link manually in the RTF
editor (rather than have the templating system create it for him with
.linkedText).
But it wouldn't be manageable for most, because it would require editing
template tags into the underlying RTF data, which most people (including me)
would not be able to do. The point is that now you just include the tags in
the rich text as text, and you don't bother with the underlying RTF data
format. It's either one or the other, you can't mix things up.
I am displaying the raw RTF data above for the sole purpose of showing
that a tag in the text and a tag within the hyperlink are not so
different. But I actually created everything in TextEdit. Just select
a piece of text, select Edit Add Link..., and, in the pane that
opens, write the destination address, possibly including template
tags.
I acknowledge that it would be more difficult that letting .linkedText
do the work but it would allow for finer configuration where needed
(the custom google scholar search I did in HTML turns out to be quite
useful).
Anyhow, thanks for all your work on the rest. I'm just trying to
better understand how things are working under the hood but maybe I
just shouldn't ;)
JiHO
---
http://maururu.net
--
Nokia and ATT present the 2010 Calling All Innovators-North America contest
Create new apps games for the Nokia N8 for consumers in U.S. and Canada
$10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing
Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store
http://p.sf.net/sfu/nokia-dev2dev
___
Bibdesk-users mailing list
Bibdesk-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] Two columns template for the preview pane
I'm trying to fix some of the hiccups of this template and convert it to html too, in case people prefer that (and because it allows better configuration of links). The files are now on github: http://github.com/jiho/bd-two_columns I have few questions: - what are stringByDeTeXifyingString and stringByTeXifyingString supposed to do? Neither of those does anything: when I have latex commands (\textit, \circ etc.) they do not replace, apply, or suppress them. - the preview pane does not seem to support linked CSS, only embedded. Is that expected? - I am using a table with one line and two columns. Whether its width is specified as 100% or unspecified, it seems limited to a maximum with (it does not take the whole width when bibdesk is maximized on a large screen). Is that expected also? - More generally, I find that BD renders the HTML differently from Safari (of CSS Edit). For example, it does not seem to honor the padding settings on p elements. Thanks in advance, JiHO --- http://maururu.net -- Start uncovering the many advantages of virtual appliances and start using them to simplify application deployment and accelerate your shift to cloud computing. http://p.sf.net/sfu/novell-sfdev2dev ___ Bibdesk-users mailing list Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] Two columns template for the preview pane
Your template is really cool. I am having difficulty copying text within the abstract, when using the template in Bibdesk. Any suggestions? Indeed, I had problems clicking links in the left column and this seems related: by default, the cursor is placed in the right column and there is no way to select text only from the left column. You actually have to start from the top. I guess this is a bug in Cocoa's RTF view that's used in the bottom pane of BibDesk. Would there be any way around it? Would and HTML template be likely to display the same problem? (I actually started this as an HTML template but got lazy with the hand coding of the style -- I could not get 50% width divs to show up side by side -- and went for an RTF template instead). This is great! Isn't the wiki open to changes by anyone, so long as he or she registers? Yes, but you need to have editor privileges, which I can give you when you give me your sourceforge user name. I probably have a sourceforge account but it's also probably lost within my numerous online service subscription. I would prefer if someone more active on the BibDesk/Sourceforge front could edit the wiki. To everyone who wants to download this (see link in the email quoted above)---the links can be hard to spot because they match the page regular text---the link is in the next to last paragraph of the blog entry. I guess bolding the word download was not enough. I updated the post and made the download link more prominent. I think it is unmissable now ;) Great. I wish there was a version of this for PDF files. What do you mean exactly? JiHO --- http://maururu.net -- Start uncovering the many advantages of virtual appliances and start using them to simplify application deployment and accelerate your shift to cloud computing. http://p.sf.net/sfu/novell-sfdev2dev ___ Bibdesk-users mailing list Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] url problem
On Tue, Sep 21, 2010 at 16:39, Themis Matsoukas tmatsou...@me.com wrote: This is not a bibdesk problem (I think) but it happens in citations: if a url is hyphenated, then clinking on it does not work because it is interpreted to end at the hyphenated point. I am attaching a bibdesk record and a screenshot that shows that the url (on the TeX preview) is highlighted only up to the hyphen. The same happens when the url appears in a regular latex document. It seems to be a conflict int he url package, but I though tI'd run it by this list first. I use the package hyperref with the option breaklinks which, I think, solves this problem. JiHO --- http://maururu.net -- Start uncovering the many advantages of virtual appliances and start using them to simplify application deployment and accelerate your shift to cloud computing. http://p.sf.net/sfu/novell-sfdev2dev ___ Bibdesk-users mailing list Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
[Bibdesk-users] Two columns template for the preview pane
Hi everyone, For a while, I have wanted a more compact template for the preview pane at the bottom of BibDesk's main window, to fit more information on my 13 MacBook screen. I created a two column RTF layout today that might be of interest to other bibdesk users. I just posted a quick write up about it here: http://maururu.net/2010/two-columns-bibdesk-template/ where you will find screenshots and a link to download the template. Of cours, I would see no objection to have this linked from the Template page of BibDesk's wiki. I just don't know if I am allowed to do it. All improvements are welcome. JiHO --- http://maururu.net -- Start uncovering the many advantages of virtual appliances and start using them to simplify application deployment and accelerate your shift to cloud computing. http://p.sf.net/sfu/novell-sfdev2dev ___ Bibdesk-users mailing list Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] BibDesk for iPad?
On Thu, Jan 28, 2010 at 12:25, Christiaan Hofman cmhof...@gmail.com wrote: I am pretty sure that a combo of citation management, PDF management, PDF viewing, and PDF annotating is really not possible on a device like iPad (and certainly not iPhone). We came to the conclusion that this was not even feasible on a normal computer or laptop without compromising too much. There's a limited number of keyboard shortcuts and menu items that you can offer, and the choices for those are very different for a citation manager and a PDF viewer/annotator. This is the reason we went for separate apps. With much less interaction you're much more restricted in what you can do, so you need to be much more focused on a single feature, you can't work with menu items and keyboard shortcuts. PDF organize + viewer/annotator can be combined, but citation manager would be a different app. And lots of functions of BibDesk are really not appropriate for iPads at all. I think the situation on the iPad vs. a laptop/desktop computer is made very different by: - the kind of interface you have - the lack of file system (which, as a side note, drives me nuts: why is it impossible to save the PDFs I view on my iPhone through Safari? Anyhow, back to the point:). So on OS X you have menus and keyboard shortcuts as the main means of interaction; and those are unique but also (mostly) static for each app. In this situation the solution of having separate apps for each task and switching between them through the file system (save a file, open it with another app) is how you do things. On iPad/iPhone, there is no menu and the UI can therefore completely change inside the same app. Once example is the new book reader: there is the shelf, aimed at choosing your book; there is the reader, aimed at flicking through the pages; and there is the bookstore, aimed at buying books. All those are three completely separate tasks, with completely different interfaces. Essentially they are three different apps, but the lack of access to the file system prevents you to have a bookstore app which would store an eBook that you could then double click to file it in your library or open it in the reader. It has to be all integrated in one app. Whether you think this is a good thing or not is probably a matter of taste and proficiency with computers, but in any case that's the way it is. A BibDesk + Skim equivalent for the iPad (and I think it could even also work in the iPhone, UI wise) could work in a very similar way: - opening the app shows your library with a list of records - taping an item allows to edit its details (similarly to an address book card): author, title, but also group etc. - taping a PDF icon either in the list or in each items details opens the PDF reader/annotator - taping a search icon brings you to the web / a search interface i.e. three different apps in one. This, of course, is wishful thinking. I would love and definitively have a use for such an app but I understand that it represents a tremendous amount of work and that it basically means restarting from scratch code-wise. Given the business model of the iPhone/iPad, charging for it wouldn't make as much difference on it does on the Mac: most good apps are shareware on the iPhone while many apps are excellent and free on the Mac (BibDesk being a prime example). JiHO --- http://maururu.net -- The Planet: dedicated and managed hosting, cloud storage, colocation Stay online with enterprise data centers and the best network in the business Choose flexible plans and management services without long-term contracts Personal 24x7 support from experience hosting pros just a phone call away. http://p.sf.net/sfu/theplanet-com ___ Bibdesk-users mailing list Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
[Bibdesk-users] Toggle between abbreviated and full journal names
Hi everyone, It has been a while since I last posted on this list. I guess Bibdesk has fulfilled all my needs for all this time, thanks for that! There's something that has been annoying me for a while though: I prefer having full journal names in my bibliography (some abbreviations are unclear, some journals --although rare-- request full names, etc.) but very often journals prefer abbreviated forms (with or without dots). I have been toggling between the long and abbreviated forms using a home made perl script... which is quite ugly and inefficient since I don't know anything about perl! I only recently noticed that Jabref is able to toggle between those since 2006, using an external list of abbreviations. I was wondering if there was any plan to build similar functionality into BibDesk. The list of abbreviation could be homemade or even better, provided by Bibdesk (and possibly completed in community way: every user submits its abbreviations). To get the ball rolling, I cleaned my scripts a little and here is what I have: http://github.com/jiho/ISI-Bib-Toggle get_isi_abbreviations.sh shell/perl/sed/awk script to get the list of journal abbreviations from ISI and construct a text version of the list in a format similar to the one used by JabRef ( full name = ISO abbreviation). I creates the file isi_journal_abbreviations.txt and appends personal_abbreviations.txt to it. This script would need to be improved to deal better with edge cases but it might be a good start to get a comprehensive list of abbreviations. abbreviate_journals.pl the perl script I currently use. It is very inefficient in that is loops over all possible journal names (and there are a lot of them) instad of first searching through my list of references and replacing only those names. But as I said, I was and still am a perl newbie. In any case, this part would be rebuilt within bibdesk in Obj-C (I still welcome suggestions for improvements in the meantime ;) ). Thanks in advance for any help/suggestions/comments. JiHO --- http://maururu.net -- Join us December 9, 2009 for the Red Hat Virtual Experience, a free event focused on virtualization and cloud computing. Attend in-depth sessions from your desk. Your couch. Anywhere. http://p.sf.net/sfu/redhat-sfdev2dev ___ Bibdesk-users mailing list Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] Toggle between abbreviated and full journal names
On Sun, Dec 6, 2009 at 21:17, Adam R. Maxwell amaxw...@mac.com wrote:
The standard way to do this is by using BibTeX macros, so you never have to
edit the .bib file. You can add global macros in BibDesk for
display/searching, and keep separate abbreviated/full name files for
processing with BibTeX.
Getting the abbreviated vs. full names is then as simple as using
\bibliography{abbreviatednames, actualdatabase}
Thanks for the reply. I've tried this but it means that I have to
manually change the journal name to the macro for every new entry in
my bib file, while 95% of those come from RIS/Bibtex imported in
Bibdesk (thanks to the excellent integration with Google Scholar for
example). So it just puts the strain of having to maintain something
manually on another part of the workflow (and, if I have to choose,
I'd rather do one long conversion once in a while that repeated manual
edits). Or maybe I am missing something.
In addition, LaTeX is not really standard in my field
(biology/ecology) so there is no standard list of abbreviations. I
have 120 different journals in my bib file currently. This means that
I'll probably need to create ~100 abbreviations and that, once this is
done, my bib file+macros would become useful only to myself (I know I
could probably work around that by exporting/drag and dropping it with
the correct options, but accumulating work arounds is what I am trying
to avoid). Guided by some online advice, I tried to define the
abbreviated/full form as @string macros but this does not work because
the shortcut of the macro just contain no space.
I guess, the real issue comes from Bibtex itself not handling it
better, but since this is unlikely to change in the near future
(ever?), I figured that Bibdesk might be able to help, the way Jabref
did.
Thanks in advance,
JiHO
---
http://maururu.net
--
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
___
Bibdesk-users mailing list
Bibdesk-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] Toggle between abbreviated and full journal names
On Mon, Dec 7, 2009 at 01:11, Adam R. Maxwell amaxw...@mac.com wrote: Thanks for the reply. I've tried this but it means that I have to manually change the journal name to the macro for every new entry in my bib file, while 95% of those come from RIS/Bibtex imported in Bibdesk (thanks to the excellent integration with Google Scholar for example). So it just puts the strain of having to maintain something manually on another part of the workflow (and, if I have to choose, I'd rather do one long conversion once in a while that repeated manual edits). Or maybe I am missing something. How about setting up a script hook that converts the full title to the macro value automatically? That would avoid the manual conversion entirely, and be less work than what you're doing now. The script could even have an alert that says Missing @string for title foo. Thanks for the suggestion. I've never used script hooks before but it would be a nice application. Guided by some online advice, I tried to define the abbreviated/full form as @string macros but this does not work because the shortcut of the macro just contain no space. Do you mean that you tried to use a title with spaces as a macro key? That definitely won't work, unfortunately :). Yes this is what I tried. Bad online advice ;) I guess, the real issue comes from Bibtex itself not handling it better, but since this is unlikely to change in the near future (ever?), I figured that Bibdesk might be able to help, the way Jabref did. It could do something similar, but I think the right way to do this is by using BibTeX features, rather than rewriting your entire database all the time. Well, if I wanted to be picky, I would argue that this looks more like a band-aid fix using the macro feature of bibtex than a true journal name toggling feature. A band-aid that has become standard, but a band-aid nonetheless. All other software don't have you bother with this. Another more constructive argument is that collaborators finally got the better of my motivation to use latex and that I'll probably switch to Pages/Word/Google Docs soon. Fortunately, thanks to CiteInPages, I don't have to give up BibDesk. And to use BibDesk with CiteInPages, I need the journal names a single bib database to be the correct ones (either all abbreviated or all full). Off the top of my head, here are a few of the issues you'd encounter adding this to BibDesk. How would you detect and handle errors? You'd have to have a UI to manage your full-abbreviated name mapping, and import someone else's map...how would you deal with conflicts? How would you know which entries don't have a definition? Indeed that requires UI. Probably a new preference pane or a new sheet within an existing one. You can have a quick look at JabRef to see how they did it.The UI could be quite close to the one for custom character conversion: Bibdesk would have an internal list, as comprehensive as possible and the user would only add its own journals (and be warned if he tries to enter an existing journal). As for which entries don't have a definition you could either be lazy and consider that the user will notice when converting the database, or match the names against the existing database and pre-fill the custom list with those that are missing. Practically, how would you handle a case where you're using the same database to draft one paper that requires a full title and another that requires abbreviated titles? I think this would be an edge case: at the limit, one needs the form appropriate for the journal (abbrv or full) only for the last compilation, before submission. I wish I often had two papers at this stage at the same time but this unfortunately proves to be rare ;) JiHO --- http://maururu.net -- Join us December 9, 2009 for the Red Hat Virtual Experience, a free event focused on virtualization and cloud computing. Attend in-depth sessions from your desk. Your couch. Anywhere. http://p.sf.net/sfu/redhat-sfdev2dev ___ Bibdesk-users mailing list Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
[Bibdesk-users] BibDesk completions in TextMate
Hello everyone, I have been using the BibDesk completions in TextMate program found on this page: http://bibdesk.sourceforge.net/wiki/index.php/Other_Applications with TextMate last month and everything worked. Now when I use the command Bibdesk Do Completion I get: /tmp/temp_textmate.UNeFiD:18: undefined method `[]' for nil:NilClass (NoMethodError) I don't remember a TextMate update in the meantime, however I did update Bibdesk and modified my bib file. Currently the problem occurs with the latest versions of TM, BibDesk, the completion program and an up to date OS X. I am willing to do what's necessary to help debug that, I just don't know where to start. My bib file is there in case it helps: http://jo.irisson.free.fr/dropbox/irisson_bib.bib.zip Thanks in advance for your help. JiHO --- http://jo.irisson.free.fr/ - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100url=/ ___ Bibdesk-users mailing list Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] BibDesk completions in TextMate
On 2008-August-06 , at 14:58 , Maxwell, Adam R wrote: It's working fine for me with latest TextMate and svn checkout of bundles. same error with a fresh svn checkout of the bundles, the LaTeX one in particular. JiHO --- http://jo.irisson.free.fr/ - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100url=/ ___ Bibdesk-users mailing list Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] BibDesk completions in TextMate
On 2008-August-06 , at 15:26 , Maxwell, Adam R wrote: On 08/06/08 12:16, jiho [EMAIL PROTECTED] wrote: On 2008-August-06 , at 14:58 , Maxwell, Adam R wrote: On 2008-August-06 , at 13:59 , jiho wrote: Now when I use the command Bibdesk Do Completion I get: /tmp/temp_textmate.UNeFiD:18: undefined method `[]' for nil:NilClass (NoMethodError) However, using Build cite with Bibdesk, the completion works. If that helps. That's odd, since they should be doing the same thing. Can you post the BibDesk DO Completion .tmCommand script? It's working fine for me with latest TextMate and svn checkout of bundles. I'll try updating my bundles and support directory to svn versions too. I just reinstalled the tmCommand too so it should be the same version you have apart from the location of the binary. Here it is: Looks like it's using a different method to get the latex master file. If what you posted is currently on my site, I need to update it; sorry about that. Here's my current one (note the added 'require' line and latexMaster variable): yay it works. I reinstalled the command by double clicking it and telling TM to use the new one. So what I posted is probably what is currently online indeed. thanks a lot. JiHO --- http://jo.irisson.free.fr/ - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100url=/ ___ Bibdesk-users mailing list Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] BibDesk completions in TextMate
On 2008-August-07 , at 00:34 , Adam R. Maxwell wrote: On Aug 6, 2008, at 1:24 PM, jiho wrote: On 2008-August-06 , at 15:26 , Maxwell, Adam R wrote: Looks like it's using a different method to get the latex master file. If what you posted is currently on my site, I need to update it; sorry about that. Here's my current one (note the added 'require' line and latexMaster variable): yay it works. I reinstalled the command by double clicking it and telling TM to use the new one. So what I posted is probably what is currently online indeed. The one online is actually up to date, but I suspect that your local modification of the path is causing TextMate to keep the old version of the ruby script instead (I see the same problem on my home computer). I think you can fix it by messing around with the LaTeX.tmbundle innards in ~/Library/Application Support/TextMate/ Bundles (and maybe Pristine Copy?), but it's probably easier just to copy the script. well I removed the .tmDelta in the LaTeX.tmbundle of my home directory. But maybe there was something else funky going on. ISTR making these fixes a few months ago when it broke due to changes in the LaTeX bundle, but I didn't realize installing it by double-clicking would be a problem. Sorry about that, but I'm actually not sure if I can fix it since TM seems to use a diff mechanism for bundle changes. I indeed find this part of bundle handling to be one where TM is less than ideal. For small modifications by the user this delta system is a good idea and works surprisingly well. But as soon as the modifications are more extensive or that the bundle gets a major update, then it does not scale. There was a discussion on TM's list regarding the possible use of a distributed version control system to replace this ad-hoc one i.e. each bundle would be a git repository and git itself would handle the merges between the user's changes and upstream updates. With a decent UI for conflict resolution this would probably be excellent. We'll see what the future gives us. Sorry to have bothered you with the issue then, since it was specific to my system, and thanks for the work on BibDesk and on that. JiHO --- http://jo.irisson.free.fr/ - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100url=/ ___ Bibdesk-users mailing list Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] How to use Web Groups (Google Scholar)?
On 2007-November-09 , at 10:24 , Hendrik wrote: Hmm. Nevermind. After quitting, restarting and trying again it magically works now. Same here. Very very cool. ... a minute later ... But then after switching to another group and coming back to the web group it stopped working again. Only quitting and restarting seemed to help. I guess it isn't all that stable yet? Another quirk (in this otherwise very useful feature): when I change the prefs to search only english or french documents, I get Unable to parse bibtex for all records. Lastly: is the UI for this meant to stay different from other searches or will these web scrapping searches be integrated along side others? It would seem more natural to me to access all these in a searches menu, with bookmarks common to all, even if the UI once the search is started is a bit different (three panes vs two panes). Thanks a lot for this great feature, google scholar search is awesome. JiHO --- http://jo.irisson.free.fr/ - 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 Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] View bibtex record from the file content search view
On 2007-November-07 , at 20:37 , Adam R. Maxwell wrote: On Wednesday, November 07, 2007, at 10:38AM, jiho [EMAIL PROTECTED] wrote: I was wondering wether is it possible to open the bibtex edition window of a particular reference from the File Content search view. This view already shows the records titles and double clicking one opens the associated PDF file, so the relation between those two is already known. Well, it's sort-of known. The search index has no idea which item a given file actually belongs to; it's given a title and a file, and maintains that association. When you deselect the file content search, we do a search of your records based on title to figure out which items to select in the standard table view. In this case, could this search be done, while still in the File Content search, and associate file search results to bibtex items? I guess the search should be quite fast and therefore should not impact much the file content search. which is usually quite lengthy compared to other operations. However I don't have a huge bibliography database (500 records) so I may not estimate this correctly. It would be nice to have a bibtex record icon next to the pdf one and click this one to edit the bibtex record. Yeah, I know. There are issues with this, since the only way I can do it is by assuming a 1:1 mapping of file-record. If multiple files link to the same record, results would be non-deterministic. Ah, that's another problem. How is it dealt with now, when the search result contain several files linked to the same record? How are they displayed (which title do they have)? How are the items selected when leaving the File Content search? I tried to test this by associating a folder containing several pdfs to a record (that's the only way I know to associate several PDFs to the same record) but the file content search was unable to search those PDFs apparently. Because of that, we have to come up with a way to displaying multiple records per file, so you click on a file and the publication(s) that link to it would be displayed in some other view. I've been thinking about a table or shallow outline view like a source list on the right side of the window, which displays a summary of the selected records. That would be useful for the web group and the file views, as well as file content search. Each item would show just title when collapsed, but expand to show authors/title/date over multiple lines like Skim's notes table view, but with a disclosure triangle instead of a variable height thumb. Not sure I visualize all this correctly. Anyway, I see a problem with displaying a clickable bibtex icon next to the pdf one only when one file is associated with multiple records, but I don't think this is currently possible in BibDesk (and can't see a sensible use for it). In the opposite case, when several files are associated with the same record, this approach will just mean that several entries in the search results will have the same title and that the bibtex icons next to them will trigger the edition of the same record, but I don't see how that's a problem. A typical use case (well mine actually ;)) is that I often search for all files broadly related to a subject with the (great but necessarily slow) file content search, skim through the highest results and add keywords to the articles that are really well related to the subject in their bibtex records. Then the rest of my organization or searches relies on these keywords (so it is both fast and more accurate). I find this very efficient and the only black spot is that I have to note the names of all interesting records while I skim trough them and then get out of the file content view to add keyword to them. Does it help to know that your selection is preserved when you leave the file content search view by clearing the search field? Well it help a bit but in most cases there are several files I want to edit. In this case, when I leave the search field I end up with a (most likely discontinuous) multiple selection in the main table and I can't edit all record at once (double clicking on one deselects the others). So the solution would be to drag them the sidebar, create a temporary group, edit the records and then delete the group. A bit better indeed, but still not perfect. And this also supposes that I take care of keeping the interesting results in the File Content search selected. Since I need to click other results to open them and that it looses my selection I still have to remember (well write down in this case ;)) the titles of the interesting ones. It would really be easier to access the records directly in the File Content field. I (and hopefully others) would then be able to: search, skim/read, add keywords/annotation to the file, move to the next one, skim/read etc. JiHO --- http
Re: [Bibdesk-users] View bibtex record from the file content search view
On 2007-November-08 , at 12:13 , jiho wrote: On 2007-November-07 , at 20:37 , Adam R. Maxwell wrote: On Wednesday, November 07, 2007, at 10:38AM, jiho [EMAIL PROTECTED] wrote: [...] Because of that, we have to come up with a way to displaying multiple records per file, so you click on a file and the publication(s) that link to it would be displayed in some other view. I've been thinking about a table or shallow outline view like a source list on the right side of the window, which displays a summary of the selected records. That would be useful for the web group and the file views, as well as file content search. Each item would show just title when collapsed, but expand to show authors/title/date over multiple lines like Skim's notes table view, but with a disclosure triangle instead of a variable height thumb. Not sure I visualize all this correctly. Anyway, I see a problem with displaying a clickable bibtex icon next to the pdf one only when one file is associated with multiple records, but I don't think this is currently possible in BibDesk (and can't see a sensible use for it). In the opposite case, when several files are associated with the same record, this approach will just mean that several entries in the search results will have the same title and that the bibtex icons next to them will trigger the edition of the same record, but I don't see how that's a problem. After using BibDesk in this context a bit more, I realized that, if you can relate the PDF to a bibtex record, then displaying the same view in the File Content search that in the other ones would be the most sensible and easy choice. You'll need to enforce the view of the Local-Url column and maybe make the PDF icon a bit bigger to emphasize its importance in these results, but overall keeping the same design than the others means keeping all the same good functionality. Just my 2 cents. JiHO --- http://jo.irisson.free.fr/ - 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 Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
[Bibdesk-users] View bibtex record from the file content search view
Hello everyone, I was wondering wether is it possible to open the bibtex edition window of a particular reference from the File Content search view. This view already shows the records titles and double clicking one opens the associated PDF file, so the relation between those two is already known. It would be nice to have a bibtex record icon next to the pdf one and click this one to edit the bibtex record. A typical use case (well mine actually ;)) is that I often search for all files broadly related to a subject with the (great but necessarily slow) file content search, skim through the highest results and add keywords to the articles that are really well related to the subject in their bibtex records. Then the rest of my organization or searches relies on these keywords (so it is both fast and more accurate). I find this very efficient and the only black spot is that I have to note the names of all interesting records while I skim trough them and then get out of the file content view to add keyword to them. I guess other people could also find many use cases when directly opening the bibtex record would be useful also. Thanks in advance for your help. JiHO --- http://jo.irisson.free.fr/ - 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 Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
[Bibdesk-users] Dublin Core metadata problems
Hello everyone, I am using the New publication from web intensively. I found that many sites I visit apparently offer the publication info in Dublin Core format: BibDesk detects it and asks me what I want to do with it. However, when acceping BibDesk does not do anything with it appart from setting the pub type to misc and sometimes adding some keywords. I don't know wether the data provided by the sites is correctly formated or not so I would like to know what I can do to check and/or see this tested by experts on this list ;) Here are some examples, both of articles in open access so you should be able to test completely: http://compbiol.plosjournals.org/perlserv/?request=cite- builderdoi=10.1371/journal.pcbi.0010057 http://www.int-res.com/abstracts/meps/v347/p285-300/ Thank you in advance. PS: I know I promised a review of the the template editor but I simply did not have the time to write anything. I'll do it ASAP. Meanwhile, the few elements I tested really worked great. JiHO --- http://jo.irisson.free.fr/ - 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 Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] Dublin Core metadata problems
On 2007-November-06 , at 18:33 , Christiaan Hofman wrote: It finds as much data as is available an as we know about. So if it does not fill in much it simply does not find more. It can be that it is data we don't know about, but more probable is that the data is not well formed or is simply not available. Thanks for the answer. Is there a way for me to see this data because in the second case, BibDesk detects dublin core data but does not add _anything_ to the publication info, so that's strange. I guess it is a problem of malformation in the data supplied by the journal and I would like to report it to them. To do so, I should know what to expect and what they give. Thanks in advance. On 6 Nov 2007, at 6:15 PM, jiho wrote: Hello everyone, I am using the New publication from web intensively. I found that many sites I visit apparently offer the publication info in Dublin Core format: BibDesk detects it and asks me what I want to do with it. However, when acceping BibDesk does not do anything with it appart from setting the pub type to misc and sometimes adding some keywords. I don't know wether the data provided by the sites is correctly formated or not so I would like to know what I can do to check and/or see this tested by experts on this list ;) Here are some examples, both of articles in open access so you should be able to test completely: http://compbiol.plosjournals.org/perlserv/?request=cite- builderdoi=10.1371/journal.pcbi.0010057 http://www.int-res.com/abstracts/meps/v347/p285-300/ Thank you in advance. PS: I know I promised a review of the the template editor but I simply did not have the time to write anything. I'll do it ASAP. Meanwhile, the few elements I tested really worked great. JiHO --- http://jo.irisson.free.fr/ -- --- 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 Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users JiHO --- http://jo.irisson.free.fr/ - 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 Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] text editor integration
On 2007-October-10 , at 15:32 , Christiaan Hofman wrote: On 8 Oct 2007, at 2:36 PM, jiho wrote: On 2007-October-08 , at 13:20 , Christiaan Hofman wrote: On 8 Oct 2007, at 12:57 PM, jiho wrote: On 2007-October-08 , at 11:38 , Christiaan Hofman wrote: On 8 Oct 2007, at 9:16 AM, jiho wrote: On 2007-October-08 , at 08:38 , Simon Spiegel wrote: [...] I've created a little template composer app along these lines. It can be downloaded from the Wiki. Please try it out and say what you think. We might at some point include it in BibDesk. That's great... and fast! I cannot fully test it now but I'll try tonight or this week end. Thanks! JiHO --- http://jo.irisson.free.fr/ - 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 Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] what is bibdesk?
On 2007-October-08 , at 18:37 , Maxwell, Adam R wrote: On 10/08/07 09:21, jiho [EMAIL PROTECTED] wrote: On 2007-October-08 , at 18:03 , Simon Spiegel wrote: [...] I also use latex2rtf but simply because there is no real alternative. I also found it has many limitations: - I have found that it does not respect natbib settings for inline citations (at least) Can you provide a sample? I guess I was using an outdated version of latex2rtf (I used the one in MacPorts, with tetex) because I cannot make the svn version fail now. IIRC there was a problem when specifying punctuation other that the default in \bibpunct. latex2rtf used to always output (SomeName, 2000; SomeOtherName, 2004) instead of what was specified in bibpunct. Anyway, that seems fixed now, which is a good thing. - it does not support utf8 coded documents Use the version from latex2rtf project's svn. It supports utf8, and also has some graphics sizing improvements. I see that. I usually rely on utf coded characters for accents only (I try to use latex code for other symbols) and it seems to work fine indeed. That's another great news. - the equations are converted to pictures (in the best cases) Really? One reason I stick with latex2rtf is because it gives editable equations, and doesn't convert them to pictures unless you tell it to. YMMV. Current svn apparently converts latex equations to rtf indeed but, when they get a bit interesting (sums, fractions etc.) I cannot view them in either TextEdit, Pages or even NeoOffice. While thats not surprising for TextEdit and Pages, I would expect NeoOffice to be better. I seems I'll have to install MS Word in the end. Gasp. Thanks for pointing the improvements in the latest version. JiHO --- http://jo.irisson.free.fr/ - 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 Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] Citations include Abstract and/or Notes
On 2007-October-09 , at 18:07 , greg kise wrote: I've been using BibDesk for a while and have a problem that I cannot solve. When I preview a citation that has an abstract or a note, the citation includes the abstract or note like this: [1] G. Mason. Homeschool recruiting: Lessons learned on the journey. Journal of College Admission, pages 2–3, 2004. The article provides the author’s experiences in both working with and recruiting homeschoolers. In this context it is advised that admission officers should help students in the same ways a high school counselor would. This process advantages both parties–students learn the in’s and out’s of admission, while admission officers present their institutions as trustworthy to a potential recruit. Because homeschooled students come from intimate environments, it is important that the recruitment techniques colleges use to reach them be personalized. This is really annoying. This also happens when I insert the citation into Lyx. The only way I've been able to solve this problem is to completely delete the Abstract and/or Note, but this seems like overkill. I looked at the offending references in JabRef (because I can see the code) and I don't see anything obviously wrong with the markup. Also I don't know where the problem is coming from, whether it is BibDesk or LaTeX or who knows. Any ideas about how to solve this without deleting my Abstracts? The fact that the (I assume LaTeX) preview shows the abstract or not depend on the BibTeX style selected in the preferences. You can: - select another style and check what suits you - create a new style with makebst (included with you latex distro somewhere in the bibtex part) and explicitly tell it not to include abstract or notes. check the documentation on where to put the new style and rebuild your tex index to make it available - grab a style from a journal you like there: http:// jo.irisson.free.fr/bstdatabase/ JiHO --- http://jo.irisson.free.fr/ - 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 Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] BibDesk to Connotea to BibDesk
On 2007-October-09 , at 18:32 , P Kishor wrote: In my quest to put up my citations in a place where they can be shared with others, I started exploring Connotea (http://www.connotea.org), an open source online citations manager created by Nature.com and let loose in the wild. Well, I started by importing my BibDesk bib into Connotea. After whirring for about 20 mins, it reported that out of my 98 citations, only 40 had been imported. The primary cause of not importing the bulk of the remaining 58 was missing URI, PMID, ASIN, or DOI. A quick Google-ing explained to me what the heck that meant, and indeed, I didn't have a PubMed Unique Identifier, an Amazon Standard Identification Number, or a Digital Object Identifier let alone a URI from which Connotea could have retrieved the requisite info. I am thankful for BibDesk being lenient and not rejecting my entries, but now I am thinking -- what is it that I can do to make my bibliography more complete, accurate, and reliable. I realize this is not a BibDesk-specific question, but most of you know way more than I do about citations and bibliographies, so I hope you can teach me a few things here. Eventually I want my entire BibDesk to be imported into Connotea (or any other such site that you might suggest as being better) and back again, if required. If the idea is to make your bib file public I can suggest: - exporting it in html from bibdesk and posting this on you personal site - exporting your citations to CiteULike. CuL can import BibTeX and will make less changes than what Connotea apparently does. However it will just take the information in your bib file as is, without completing the missing fields (which Connotea might do apparently, since it requires some identifier to fetch the publication information). From CiteULike you can also import citations from web pages or such and export everything back to bibdesk with no loss of information if you stick to the classic bibtex fields. Linking PDFs is a more delicate mater and I don't remember if I found a solution. CiteULike is really nice but two-way communication with BibDesk is still problematic. I would personally love to see this improved. Anyone motivated ;) Last I remember, CuL guys (well guy actually since it is kind of a one man project) were quite open to such ideas. JiHO --- http://jo.irisson.free.fr/ - 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 Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] what is bibdesk?
On 2007-October-09 , at 19:39 , Ingrid Giffin wrote: On 10/9/07 7:20 AM, Bruce Pourciau [EMAIL PROTECTED] wrote: On Oct 7, 2007, at 9:47 PM, Ingrid Giffin wrote: I tried looking at MacTex (I think it was), but I really don’t need the heavy typesetting capabilities. I ran screaming to Mellel. I use BibDesk for my master publications database because the interface is so nice, although, as I said in another post, I have to use Bookends in between BibDesk and Mellel. Those who do need the typesetting and cross-referencing abilities of LaTeX, but who have run screaming from its complication and coding requirements, might want to look at LyX (http://www.lyx.org/), which is a word processor front end to LaTeX. You work in LyX as you would in any word processor, but when you wish to view the finished product, LyX calls up TeX and has it typeset your document. I use LyX with Bibdesk. The idea sounds good, but the LyX page lists several other required packages. Of the first two, Qt and Xforms, Qt costs several hundred dollars and the Xforms web site is dead. This doesn't look promising No no lyx works out of the box on os x (provided you have a TeX system of course). Just download the dmg and install everything: ftp://ftp.lyx.org/pub/lyx/bin/1.5.2 JiHO --- http://jo.irisson.free.fr/ - 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 Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] request for feature (was Re: BibDesk to Connotea to BibDesk)
On 2007-October-09 , at 22:02 , P Kishor wrote: On 10/9/07, François Briatte [EMAIL PROTECTED] wrote: Sorry, what do you mean by 'my' file? I am not sure what you mean by that question. I am looking at my response and I don't see any reference to my file. I said specifically (quoting myself) -- The error message from CiteULike starts by: I got this error when I tried to parse your file In this case your file is the file from P Kishor he is trying to import. Thanks for the advice. I am now trying citeulike as suggested, but not very successful at this as well. I am getting a lot of errors like so, and then nothing is imported -- everything else is errors reported by CituULike. Also, warnings are normal, especially if you forget to mention the authors or use unconventional types. Yes, I would think so, no? Except, nothing gets imported. My CituULike library is empty. Seems like I am using unconventional types except I am not quite sure what convention to apply to the ones that are erroneous as reported by CiteULike. Besides, as you yourself said, CiteULike should import at least the others (I have around 100 entries, and less than 20 show up in the CiteULike's import error log). But, nothing is imported. I suspect it to be caused by your citekey format. I am not sure CuL can deal with the colon ( : ) in the citekey. Given the error message (lots of colons inside) I guess CuL breaks citekeys at colons and trys to read the rest as a field, until the next colon. Try with another citekey format. The colons there are not very safe anyway. JiHO --- http://jo.irisson.free.fr/ - 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 Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] text editor integration
On 2007-October-08 , at 11:38 , Christiaan Hofman wrote: On 8 Oct 2007, at 9:16 AM, jiho wrote: On 2007-October-08 , at 08:38 , Simon Spiegel wrote: Just out of curiosity, what would help the non-LaTeX users the most? People have mentioned integration with word processors. What would it look like? Basic RTF scanning wouldn't be too hard. A UI for templates would be nice, too. An UI would certainly be a good idea. I don't think that many users would use BibDesk with its current template system even if it had RTF scanning. Toom many users would be scared by it. I'd use it, but I'm certainly not the average user. I also think that the template system needs improvement to deal with more complex styles. For example more options how to deal with multiple authors/editors or with stuff like 'ibid.'. I doubt whether the template mechanism needs to be more complex, it is already quite complex and capable. See also below. For example, it is quite capable to do something like the standard bibtex styles (the Wiki has templates for abbrv.bst and plain.bst). I also think a GUI to build bibliography styles would be very welcome. I suggested this in back in 2005: http://sourceforge.net/mailarchive/message.php? msg_id=9787bb9212cc90ffa59662781198cb8a%40gmail.com [NB: sorry the links to the images are broken, I lost them] and it was considered too big to be in the scope of BibDesk at that time. The context seems to be the same today: people want tighter integration between bibdesk and their word processors and, to do that, a way to scan the document for citations as well as to format the references list is needed. If such a UI is build, I would love it to output rtf templates as well as something more LaTeX related (since apparently most current users of BD use LaTeX). At the time of my first post, I thought about writing .bst files, with BibTeX code, but the fact that it is a new language, and a not very user-friendly one, adds some complexity. Now, if this UI could output files suitable for biblatex, which is much easier to use than bibtex, it would be great. I personally have trouble imagining a workable UI to build templates. The template syntax is quite complex, and the only UIs I can think of would either be able to handle only simple templates, or they would be more bothersome to work with than the raw template itself. I don't know how the other managers (like Bookends and Sente) work with their template UI, but I guess they only use a much simplified syntax (in our terms, the equivalent of only value tags). I don't know the RTF template mechanism well but I was thinking at a UI like this (very quick draft): http://jo.irisson.free.fr/dropbox/bd_styles_ui.png - On the left, a list of publication types, taken from BibDesk - On the right: . a list of fields, which changes according to the pub type, with mandatory bibtex fields in red. The list of fields is also taken from bibdesk (custom fields appear here) . an Inline citations and a References list fields in which tokens can be drag and dropped to select what will be printed in each case. Plus, one can also write in these sections (but I cannot find how to do this in interface builder). Therefore one can write: [Author], [Year]. [Title], In: [Journal] etc. . an option view, with options depending on each token (here some possible options are shown for the [Author] token, which is probably the most complex to print). . a preview of what the output will look like I think such a UI would be quite straightforward to use and could be interfaced to many template mechanisms (BD own templates, biblatex, bibtex etc.) The cumbersome part with such a UI would be to do the templating for all publication types. There are two solutions I can think of: - never start from scratch: BD provides two of three templates to start from, e.g. an author-year one, a numbered one etc. maybe from the three or four classic bibtex styles - define one reference type as a master (as shown here). The modifications done on this master propagates the all other pub types and then one only modifies them. I think a better road for users for whom building a template is too complicated is to provide an (extensive) library of standard citation and bibliography styles. That could be done mostly by users I think. While this could be an efficient solution, I don't think it will be viable if there is not an easy way to produce the templates. I have experience in doing it for latex bibliography styles: http://jo.irisson.free.fr/bstdatabase/ and most styles were collected by me from publishers web pages (actually the large majority comes from elsevier which gives one template for all their journals: 2000 journals in the database). I received approximately 30 submissions of new styles in two years, 10 of which being brand new, the rest of them being
Re: [Bibdesk-users] WoS url
On 2007-September-18 , at 23:20 , Karsten W. Jacobsen wrote: I am new to BibDesk and trying to find out the various features of the program. So far I find it great but I have the following problem. I would like to search and fetch articles from Web of Science. However, my university has a subscription to ISI which requires that I access WoS in a special way with some proxy-stuff in the url. Is it possible to change/set the url used by BibDesk to access WoS or other external databases? I have the same problem. If you want to solve it for yourself quickly, you can change some hardcoded URL in BibDesk source and recompile BibDesk. I looked a bit into this already, search for a message with title: Web of Science search with a special address on BibDesk user list. The only problem will be for BibDesk to be logged in with your proxy, if you need login+password. If you want a more full featured solution, then this would need a preference somewhere, to store the custom address, and a panel to provide login and password. Adam told me that he would be happy to accept a patch that provides this functionality. I did not looked further into this since the stupid proxy I am using does not support Safari (i.e. Apple's WebKit) and this is what BibDesk would be using to log in. I hope you'll be more lucky than I was. Cheers, JiHO --- http://jo.irisson.free.fr/ - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Bibdesk-users mailing list Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] WoS url
On 2007-September-19 , at 18:30 , Adam R. Maxwell wrote: On Wednesday, September 19, 2007, at 09:20AM, Adam M. Goldstein [EMAIL PROTECTED] wrote: I wonder if Camino is any different? It has the same rendering engine as FF, but is built to be more mac-like. It still doesn't use WebKit, so doesn't use shared cookies. On Sep 19, 2007, at 10:49 AM, Karsten W. Jacobsen wrote: I just downloaded and checked Omniweb. The access to WoS worked without any problems, so it seems to be a problem with Safari alone. I suppose the bottom line is that at least I will need to buy Omniweb and adjust the BibDesk source code if I should get BibDesk to fetch WoS references for me. Thanks for your help. We can do some testing before you go to such lengths (I think OmniWeb is worth using anyway, though). It looks like OmniWeb uses its own cookie storage as well, so this might not work at all. Feel free to send me a note off-list, and I'll send you a version that allows changing the URL without modifying source code. My error with safari is also a too many redirects error. I thought this was WebKit related and not Safari related but since it works with OmniWeb it looks like I was wrong... which is great because it means that it may work for Bibdesk then. However, when I try to log in the proxy with bibdesk through the New publication from web panel, I still get the HTTP redirects error. What about you Karsten? As for the special BD version which allows changing the URL, I would be interested also. If this works in the end it would mean that, once logged in with safari or with bibdesk web panel, we should be able to search directly through bibdesk, is that it? I would really be great ;-) Thank you in advance. JiHO --- http://jo.irisson.free.fr/ - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Bibdesk-users mailing list Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] WoS url
On 2007-September-19 , at 20:30 , Adam R. Maxwell wrote: On Wednesday, September 19, 2007, at 11:06AM, jiho [EMAIL PROTECTED] wrote: My error with safari is also a too many redirects error. I thought this was WebKit related and not Safari related but since it works with OmniWeb it looks like I was wrong... which is great because it means that it may work for Bibdesk then. However, when I try to log in the proxy with bibdesk through the New publication from web panel, I still get the HTTP redirects error. So this probably isn't going to work. Does Safari 3 work? No, the problem is still the same. What about you Karsten? As for the special BD version which allows changing the URL, I would be interested also. If this works in the end it would mean that, once logged in with safari or with bibdesk web panel, we should be able to search directly through bibdesk, is that it? I would really be great ;-) I'm not sanguine about this approach, but you can download it from http://bibdesk.sf.net/BibDesk.app.zip. Be warned: this has my development hacks in it for the file view testing. The normal endpoint URL is http://wok-ws.isiknowledge.com/esti/soap/ SearchRetrieve, and you can quit BibDesk and change it in Terminal with `defaults write edu.ucsd.cs.mmccrack.bibdesk BDSKISIEndpoint http://my/url` OK, thank you. But since the redirect problem blocks webkit access I guess this won't work... too bad. JiHO --- http://jo.irisson.free.fr/ - This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse012070mrt/direct/01/ ___ Bibdesk-users mailing list Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] Web of Science search with a special address
On 2007-August-13 , at 19:09 , Christiaan Hofman wrote: On 13 Aug 2007, at 7:01 PM, jiho wrote: [...] now I am trying to compile BibDesk but I get and error because I do not have the Developer/Examples software installed and there is a shell script action requiring it (those examples are big, are they really required?) There shouldn't be any dependency on the examples. Please copy and paste the exact error message from Xcode. Also, be sure that you've set your Xcode preferences to put Build Products in a customized location (Building pref pane), or things will break horribly. I didn't set a custom build location because, from the wiki, I suspected that this was only to avoid paths with spaces in their names and my bibdesk checkout did not have any space in its name. I figured it was not necessary then. This requirement has nothing to do with spaces. It's for xcode to be able to find the dependent build products, such as the frameworks. But I changed it now and I still get the error: /bin/sh -c /Users/jiho/Softwares/bibdesk_build/ OmniFoundation.build/Release/OmniFoundation.build/ Script-4A4E06AC08AA72B10098FF0F.sh Copying/sed to /Users/jiho/Softwares/bibdesk_build/ OmniFoundation.build/Release/OmniFoundation.build/DerivedSources/ OmniFoundation.sdef sdp: /Developer/Examples/Scripting Definitions/NSCoreSuite.sdef: No such file or directory It may be necessary to restart xcode, because it does a lot of caching. I'll try that but /Users/jiho/Softwares/bibdesk_build/ is already my custom selected location so XCode took it in account. JiHO --- http://jo.irisson.free.fr/ - 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 Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
Re: [Bibdesk-users] Web of Science search with a special address
On 2007-August-13 , at 19:12 , jiho wrote: On 2007-August-13 , at 19:09 , Christiaan Hofman wrote: On 13 Aug 2007, at 7:01 PM, jiho wrote: [...] now I am trying to compile BibDesk but I get and error because I do not have the Developer/Examples software installed and there is a shell script action requiring it (those examples are big, are they really required?) There shouldn't be any dependency on the examples. Please copy and paste the exact error message from Xcode. Also, be sure that you've set your Xcode preferences to put Build Products in a customized location (Building pref pane), or things will break horribly. I didn't set a custom build location because, from the wiki, I suspected that this was only to avoid paths with spaces in their names and my bibdesk checkout did not have any space in its name. I figured it was not necessary then. This requirement has nothing to do with spaces. It's for xcode to be able to find the dependent build products, such as the frameworks. But I changed it now and I still get the error: /bin/sh -c /Users/jiho/Softwares/bibdesk_build/ OmniFoundation.build/Release/OmniFoundation.build/ Script-4A4E06AC08AA72B10098FF0F.sh Copying/sed to /Users/jiho/Softwares/bibdesk_build/ OmniFoundation.build/Release/OmniFoundation.build/DerivedSources/ OmniFoundation.sdef sdp: /Developer/Examples/Scripting Definitions/NSCoreSuite.sdef: No such file or directory It may be necessary to restart xcode, because it does a lot of caching. I'll try that but /Users/jiho/Softwares/bibdesk_build/ is already my custom selected location so XCode took it in account. As I suspected, the problem is still there, in building target OmniFoundation (with configuration Release). Since it's Omni stuff I guess there's nothing BibDesk can do about it and I should just install the example software. JiHO --- http://jo.irisson.free.fr/ - 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 Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
[Bibdesk-users] Web of Science search with a special address
Hello everyone, I was thrilled by the announcement for search facilities on ISI Web Of Science in the release notes of 1.3.7. However, my access to the WOS is through a gateway. I.e., I do not access: http://isiknowledge.com/ or whatever the real address is, but I access: http://portal.isiknowledge.com.gate1.inist.fr/ Then I enter a login+password and proceed with my search. This is how all French researchers from the CNRS (french national research agency) access to the WOS so it would be great for us if there was some possibility to work with this in BIbDesk. So I was wondering: - is the search performed from a regular web address or does this use another protocol? - if there is a web address, could there be a pref (hidden possibly) to change it? - if it is not, what is it exactly (z39.50 protocol?)? maybe we can try to have a gateway for this protocol also (there's no reason the CNRS should limit us to web search). Thanks in advance, for all the improvements in Bibdesk and for your answers. NB: sorry if the question already came out on the list but I am unable to search it on sourceforge. I get a time out eveytime I try. JiHO --- http://jo.irisson.free.fr/ - 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 Bibdesk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bibdesk-users
