Instructions for generating the New Abstracts of Interest page.


Greg thornto has written a small script to produce a HTML page of physics abstracts from the preprint server. The only input required is a list of the numbers of the papers to be included. It's all very simple - instructions below. There is also a script to produce an offline presentation version.

UPDATE: 02/08/2017

It seems that it is no longer necessary to use the university proxy server. Select 'No proxy' or equivalent in your browser settings if you haven't already.

Generating the HTML page

NB: This currently only works under Linux.
The script "gen_abs_html" does not work well when running from bragg. Please log in to other machine to run the script.
1. Make a file with a list of the paper numbers, including the section name
   ( astro-ph/ or since 2007 arXiv:)

   e.g.
   --file listing---------------------not part of file
   arXiv:0711.2171    As a recent addition comments are allowed after the paper number.
   astro-ph/0302187   So you can safely annotate the refs.new file.
   astro-ph/0302207   
   hep-ph/0011048     These comments will be stripped out in processing.
   --end file listing-----------------not part of file

   See the current file  /remote/hermes/httpd/htdocs/astrophysics/Abstracts/refs.new 
   for an example.

2. Copy your file to (and overwrite) 
         /remote/hermes/httpd/htdocs/astrophysics/Abstracts/refs.new

3. cd /remote/hermes/httpd/htdocs/astrophysics/Abstracts

4. run the program gen_abs_html. ie ./gen_abs_html
   (The script inserts the name of the person running it, based on `whoami`, into the html.
    This name will be replaced by any arguments given to the script, so if Fred Bloggs is running it on behalf of Jane Smith, 
    ./gen_abs_html Jane Smith 
    will produce the appropriate html.)

5. All done

The script gen_abs_html is very simple (OK, crude) and will probably die or produce garbage on unexpected input, or in other unanticipated circumstances. Let me know ASAP if this occurs.
One problem that may occur is with long authour lists. When the list ends with 'et al. ( xx additional authors not shown)', my abstract extraction code may fail. In this case most of the bad result will be obvious, but 1 reference may be dropped altogether.

Offline presentation version

The offline script is intended to provide a minimal set of files that can be used to present the "New abstracts of interest" in the absence of an internet connection. The script saves the file to the local directory. Running it from a directory on a USB stick provides maximal insurance against network or laptop problems. So, e.g.
After generating the new_abs.html file, as above:
cd /media/usbdisk/abstracts
/remote/hermes/httpd/htdocs/astrophysics/Abstracts/offline

This should copy the PDF files to the current directory and also produce a version of new_abs.html, in that same directory, with the PDF (and only the PDF) link pointing locally.