Language support: Random symbols appearing in citations?

Thomas A Martin

Hello,

I've noticed that a few of my references are importing to Zotero with random symbols replacing letters. Is this a problem with an unsuported language within Zotero? Or even on my computer itself? It does seem to coincide with references from publications written by foreign authors and it is normally the names where the problem occurs.

Would really appreciate any help with this!

Thanks in advance,

Tom.

ajlyon

Things like this happen with particular sites-- but to troubleshoot and address it, we'll need some example URLs or other example input. In general this should work and does work.

adamsmith

And Zotero has virtually not language limitations - it can do all of UTF-8 which includes Chinese, Arabic etc. the problem is one of encoding, not of a lack of support.
For troubleshooting more specifically what ajlyon says.

Thomas A Martin

Thanks for the replies.

Unfortunately, I didn't notice this issue until I has finished importing my whole library of references.

I'll try and back track and work out which sites or PDFs this problem occurs with and post an update.

Cheers!

Thomas A Martin

OK, I found my first example.

http://onlinelibrary.wiley.com/doi/10.1002/adsc.200303094/abstract;jsessionid=E6E10B8DE5E8E4EF3215407488CCF2A3.d03t02

This is a link to a paper I've cited, when I import it to Zotero the long hyphen between the C and H in C-H activation shows up as an apple! (I'm running on a mac at work, so perhaps this is not as bizarre as it sounds!)

Thomas A Martin

Here is another one.

http://www.sciencedirect.com/science/article/B6TGW-42SR0RS-34/2/f735e9842fa128a2903c6738ae3bface

In this case the first author H. Rabaa (the last 'a' has a kind of hat on it) shows up with a question mark on a black background instead of the last 'a' once imported into Zotero.

Thomas A Martin

This is another example of the first problem with the long hyphen.

http://onlinelibrary.wiley.com/doi/10.1002/anie.199610681/abstract

Thomas A Martin

This time it is the second 'o' in Rosa Lopez, which has a hat too.

http://www.sciencedirect.com/science/article/B6THS-3V9CSX9-4/2/0ca17fe56d29fbb86b711facc0fddd59

Thomas A Martin

And the first 'o' in Ismail Özdemir.

http://www.sciencedirect.com/science/article/B6TGW-43XFW1R-4/2/35f994228b23ef0366457157f5cd2910

Thomas A Martin

I've also noticed one other small gripe, sometimes subscript and superscript is not imported correctly or additional spaces are added. The former is quite problematic for chemical notation!

Here is an example.

http://pubs.acs.org/doi/abs/10.1021/ja980212k

ajlyon

Thanks for finding the examples-- I'll look into these and see what can be done.

ajlyon

The Wiley articles with the apple-hyphens are because Wiley is feeding bad data to the CrossRef database (which powers the DOI service). You can see the broken character here. Please notify Wiley. Note also that the hyphen on the pages is actually an image, not a text symbol.

The items from ScienceDirect work fine for me, although I suppose we could fix the [sigma] and related terms when importing and turn them into the correct letters. The broken letters on your end sounds like they could be a font issue, but those characters should be pretty widely supported.

The ACS site imports fine, although without super- and subscripts, just as in the other sites. Zotero has basic support for super- and subscripts using the <sub>, <sup> notation from HTML, starting in the 2.1 betas; they will show correctly in citations. But since these sites don't include the subscripts and superscripts in their export formats (Zotero uses RIS for ACS and ScienceDirect), it doesn't preserve the formatting. Feel free to encourage the database providers to add the super- and subscript information to their export. One option is using the superscripted and subscripted digits and letters available in Unicode.