Importing EndNote libaray, INCLUDING PDF attachments

camngnp223
I'm new to Zotero. I've spent the last 4 hours using the "search" feature of these forums and of the support section of the Zotero site. I have an EndNote library which is full of PDF attachments. I would like to import this library as well as all the attached PDFs into Zotero. It it true that this is impossible to do at the current time? (The library is large and the labour involved with manually re-attaching each of the PDF files from within Zotero really is not-at-all feasible.)
Huge thanks in advance for your advice,
Cameron
«134
  • Tjowens
    Zotero's RIS import does have the ability to import PDFs and Endnote does export links to your PDFs in the RIS files it generates. The problem is that Endnote does not give the full path to the file.

    You should be able to import all the PDFs by using a single find and replace statement.

    You will want to find
    internal-pdf://
    And then replace it with
    file://location/where/endnote/stores/your/pdfs
  • camngnp223
    Excellent: I followed your instructions and the import worked like a charm! Thank you! (Incidentially, giving a relative path didn't work. Giving an absolute directory path worked right away.)
  • kay
    Hi--I'm clearly less computer literate than you guys. Could you please elaborate where exactly I need to look for internal-pdf:// ?

    Thanks so much in advance--
    Kay
  • camngnp223
    In EndNote, you need to export your library using the RIS (Refman) style. It should be exported as a text document. Search thus text document for the string "internal-pdf://" and replace that string with "file://C:/location/where/endnote/stores/your/pdfs". (This type of directory structure is for Windows C: drive. Obviously, for the various flavours of unix, you won't be pointing to a C: drive.)

    I'd also note that, when I was using WordPad to edit the RIS text file exported by EndNote, all of my diacritical characters were getting massacred. If you use a slightly more advanced text editor, everything imports perfectly into Zotero.
  • remi
    Hello. I followed your instructions too. But I've met 2 problems :

    1. Urls of my PDF attachments look like :
    file://Euros/bases_biblio/MyTeam/MyName/PDF/Author2007.pdf
    And when I try to import RIS file, I can read this message :
    "An error occurred while trying to import the selected file. Please ensure that the file is valid and try again."
    [Report ID: 942743707]

    2. When I simplify the URLs by changing files location and beginning :
    file://Euros/bases_biblio/MyTeam/MyName/PDF/Author2007.pdf
    in
    file://D:/Author2007.pdf
    It works but the files have lost their names, they are renamed like "Full Text (PDF)".
    Is it possible to preserve their previous name ?

    Thanks in advance,
    remi
  • Tjowens
    Are the files actually renamed? Or is it just the names as they appear in Zotero. You can check this by clicking the "Show File" button to see the actual file in a sub-directory of the storage directory.
  • remi
    @Tjowens :
    Oups, you're right : it is just the names as they appear in Zotero, thanks.
  • jjagresti
    Just to add one thing. I started endnote several versions ago. At the time, you could "insert object", but not attach a file. This tags the pdf as a "figure". Therefore, all of my PDFs were attached as figures. Although attached files are put into the RIS file, "figures" are not. Therefore I ran into trouble trying to export an RIS to put into zotero.

    Thankfully I found a solution:

    I'm not sure if other versions have this, but X2 allows you to convert figures to attachments. It's in References>Figure>convert figures to file attachments. You can ctrl-A to select all and do this in a batch. Now your RIS file will be imported to zotero with attachments. Saved me hours.

    Now I have no more need for EN. Hooray!
  • ben58
    > Search thus text document for the string
    > "internal-pdf://" and replace that string with
    > "file://C:/location/where/endnote/stores/your/pdfs
    > ". (This type of directory structure is for
    > Windows C: drive. Obviously, for the various
    > flavours of unix, you won't be pointing to a C:
    > drive.)

    Under Mac OS X (which is unix), the only way I could get it to work was copying the PDF folder from the EndNote library to the root ot the startup disk and search-replacing "internal-pdf://" by "file:///PDF/" (important! - 3 slashes).

    But would be nice if Zotero in the next version would guess the meaning of "internal-pdf:" itself.
  • eagutt
    Report ID 240781048: Was trying to import my data from Endnotes 8.0.2 into Zotero, following the instructions above: setting output style in EN to "refman RIS", exporting to text file, then choosing import (I don't have embedded pdfs, so no complications that way.). I got the error message: "An error occurred while trying to import the selected file. Please ensure that the file is valid and try again." A number of entries got imported, but most did not. How can I ensure that the file is valid? Thanks for any help.
  • lemmo
    Thanks for this, my PDFs import from Endnote X1, but a couple of problems:

    Zotero will not import multiple PDFs. A number of my records in Endnote have multiple PDFs embedded, and each PDF has a distinct "internal-pdf://" location. After import, these show up in the URL field in the right Zotero pane, and the middle pane shows "Full Text (PDF)" as an attachment, but Zotero will cannot open them. It says "The attached file could not be found. It may have been moved or deleted outside of Zotero."
    Any solution to this?

    Also, each record now displays the old Endnote file location in the URL field, but my old URL field in Endnote becomes an attached Note in Zotero. Is there any way Zotero can import the URL field in Endnote into the URL field in Zotero at the same time as importing the PDFs?

    Any advice much appreciated, thanks!
  • Tjowens
    You have pointed to something that I don't think will be easy to solve.

    The Zotero URL field is supposed to be a place for a single URL that would be used in citations. For example, when citing a webpage you want to include the URL as part of the reference, clearly you would never want to include the local link to the PDF in a reference.

    Now Endnote uses URL differently, sometimes it is the canonical link to the item, but other times it is just a URL, or list of URLs, that are in some way related to the item.

    Zotero has the ability to support these two different sorts of URLs. The Item URL field is for a single URL used when citing the item. Items can also have attached links to other URLs and or a series of PDFs.

    So the problem is this. How could Zotero figure out which sort of URLs are in the RIS file when the RIS file squishes them together?
  • amacom73
    edited May 26, 2009
    Not sure how technically feasible this is, but what if every imported URL recognized by a Zotero translator is assumed to be the citing URL? Just have the translator mechanism scan all incoming URLs during import. It's imperfect, but would at least mean that URLs of journal sites or other repositories that have translators are correctly assigned as more or less official.
  • lemmo
    Using the Zotero translator is an interesting idea, but wouldn't that slow up the import process? I have 1000+ references to import from Endnote, and I can see it falling over if it had to seek the translators for each one.

    This is how the import process is working now: If my reference in Endnote has a single URL and no attachment, the URL is imported into the Zotero URL field. If my reference in Endnote has a single URL and a single attached PDF, the URL is imported into Zotero as a Note and the PDF link goes into the Zotero URL field. If my reference in Endnote has multiple URLs and/or multiple PDFs then they are not imported.

    The RIS text file from the Endnote export has field labels. This RIS file has the exported URLs listed in a field "UR", and the attached PDF file in a field "N1". So, when importing, why does Zotero effectively move the (now redundant) PDF file location from N1 into the UR field, and in so doing move the URL?

    Whatever, I shall go through my Endnote database and remove multiple URLs in the URL field...
  • noksagt
    N1 is a notes field. Perhaps you mean 'L1'? The translator presently tries to populate the URL field first. This MIGHT be because there are some RIS-generators that use L1 (or L2 or L4) rather than UR for the primary URL? But I don't know. It seems to me that we should follow the RIS spec and only ever put UR into that field.

    If your UR fields appear ahead of your L1 fields, UR should be placed in the Zotero URL field & your L1 should become a PDF attachment.

    The spec allows semicolon separators (I don't think Endnote actually uses these, though) & so we can potentially parse each entry more intelligently too (to extract multiple URLs when given in a single field).
  • lemmo
    Thanks noksagt, yes it's "L1" and not "N1", sorry.

    My L1 fields appear ahead of of UR fields, so is this an issue with the Endnote RIS export, or can the Zotero import switch it to read the UR field first?

    Good news about using semicolons for muliple PDFs, perhaps an option for future releases... Thanks
  • hurlbert
    Hi all,
    I am having trouble getting my Endnote X2 library imported, and it is due to the pointers to PDF files. I have read many of these forum posts about replacing "internal-pdf:", but my PDFs are all stored on a network drive, and so I am wondering if I am just not replacing the new pathname correctly.

    I tried the equivalent of this:
    "file://servername.edu/username/My Library/"
    but the record does not import. If I delete the UR field entirely, the record imports fine. Is it simply a matter of adding a few "/"s or modifying how I specify the path?

    Thanks
  • noksagt
    @lemmo:
    My L1 fields appear ahead of of UR fields, so is this an issue with the Endnote RIS export, or can the Zotero import switch it to read the UR field first?
    In this case, I don't think that Zotero should really be acting this way (regardless of order). However, it may be quicker/easier for you to modify EndNote's RIS exporter than to wait for the Zotero translator to be modified.
  • noksagt
    @hurlbert:
    but my PDFs are all stored on a network drive, and so I am wondering if I am just not replacing the new pathname correctly.
    You have to use a URI that firefox can read from; test these in firefox's location bar. I assume you are on windows because you said "network drive" that has a drive letter mapped to it. If so, the URL would be "file:///X:/path/to/file.pdf". If this assumption is incorrect and this isn't enough of a hint to do, please give more details.
  • hurlbert
    Thanks noksagt.

    It turned out that due to the nature of my server, the URL link only works if it is:
    "\\servername.edu\path\to\file.pdf" with no "file:" and without substituting a drive letter for the servername. It would probably be helpful to others to understand the details of our server set-up such that this is required, but unfortunately I don't know those details. Anyway, this worked fine for me and I was able to import ~1000 records quickly once that was taken care of.

    One other hangup that caused problems that I'll note for other new users is that occasionally I had a "%20" in the URL instead of a space, and these all had to be replaced in order to not cause an error.
  • noksagt
    It would probably be helpful to others to understand the details of our server set-up such that this is required, but unfortunately I don't know those details.
    You use windows file sharing (SMB) & must have not mapped that as a network drive with a letter. (I actually had not realized that Firefox could accept locations without resource types listed.)
  • hillr
    Hi All,

    I'm having some trouble after following the instructions in this thread to change the path that Endnote gives pdfs. After import, the pdf path which was L1 in the RIS format, becomes the URL field in Zotero replacing the URL in the RIS UR field. Is this the correct behavior? Should the pdf not show up in the attachments tab?

    Thanks,
  • noksagt
    edited June 4, 2009
    Please refer to my posts in this thread:
    The URL is currently filled in by the Zotero RIS translator before anything else and the translator should probably be changed, but there is a work-around for now.
  • lemmo
    Thanks noksagt, OK just to make sure I've got this correct:

    I have to go to Endnote, Edit > Output Styles > Open Style Manager, and double click on RefMan (RIS) Export > Bibliography, Templates. Then for each Reference Type I have to go through and move the UR field above the L1 field in the list?

    Also, in the exported RIS file, could I add semi-colons in fields with multiple entries, such as PDFs, keywords,URLs, and Zotero can then import multiple items?

    I'd like to know more about how Zotero imports using RIS, how it assigns the fields and any translation quirks such as importing in a certain order, e.g. UR field before L1.

    This may also help me understand strange translations, e.g. when importing a Conference Paper from Endnote, it becomes a Journal Paper in Zotero, the Conference is imported as Series, conference Dates are imported as a Note, and Abstract also becomes a Note :(

    Is there anywhere I can find whether these issues are being worked on in up-coming versions of Zotero, so I can avoid unnecessary editing work?

    Many thanks :)
  • noksagt
    Then for each Reference Type I have to go through and move the UR field above the L1 field in the list?
    yes
    Also, in the exported RIS file, could I add semi-colons in fields with multiple entries, such as PDFs, keywords,URLs, and Zotero can then import multiple items?
    Not at this time, but the translator should probably be changed for some fields to allow this (where the spec says that a semicolon is valid). For now, you should split these things into multiple fields.
    I'd like to know more about how Zotero imports using RIS, how it assigns the fields and any translation quirks such as importing in a certain order, e.g. UR field before L1.
    The best place to see this is the code:
    https://www.zotero.org/trac/browser/extension/trunk/translators/RIS.js
    This may also help me understand strange translations, e.g. when importing a Conference Paper from Endnote, it becomes a Journal Paper in Zotero, the Conference is imported as Series, conference Dates are imported as a Note, and Abstract also becomes a Note :(
    Much of this might be due to EndNote's export (which has many long-standing bugs). Feel free to share any example record that does not import as expected.
    Is there anywhere I can find whether these issues are being worked on in up-coming versions of Zotero, so I can avoid unnecessary editing work?
    The best way is to search the forums & report any issues that you see. I am unsure of what "unnecessary editing" you are trying to save yourself. You may have to modify EndNote's export style for some bugs that cannot be worked around. You should not have to edit your RIS manually much. If you want to modify zotero, drop a note in the dev group & monitor svn/trac.
  • Peter100
    edited April 29, 2010
    see my new thread discussion here: http://forums.zotero.org/discussion/12476/issues-concerning-duplicates-importexport-ris-pdf-endnote/
  • veronnem
    Hi!
    I'm working on Windows and I tried able to export my Endnote library with PDFs to Zotero. At first, it looked like it worked fine, because I could see my library in Zotero. However, when I clicked the PDF icons (in Zotero), I got the following error message:

    The attached file could not be found.
    It may have been moved or deleted outside of Zotero.

    What did I do wrong?
    Thanks in advance for your help.
  • adamsmith
    this is all on the same computer or did you sync?
    If it's on the same computer, please create an error report immediately after trying to open one of the pdfs.
    http://www.zotero.org/support/reporting_bugs
  • dstillman
    veronnem: And if not the same computer, see Files Not Syncing. In either case, start a new thread.
  • veronnem
    Thanks adamsmith and Dan for answering my post. After trying several times to import my library with PDFs from my PC at work, I just copied the text file that listed the RIS references and also all of my PDFs on my Mac at home. Then I followed the instructions that ben58 outlined above (posted on Apr 5th 2009). Using the 3 slashes was very important to make it work on Mac. I'm glad it worked, in the end.
Sign In or Register to comment.