Multi-file parse

[adsabs]

Multi-file parse

[spacemansteve]

From the error message, it appears code tries to open a file whose name is the comma separated list:
"Bibcode '2003A&A...402..531C' is linked to a non-existent file '/proj/ads/fulltext/sources/A+A/backdata/2003/17/aah3724/aah3724.right.html,/proj/ads/fulltext/sources/A+A/backdata/2003/17/aah3724/tableE.1.html,/proj/ads/fulltext/sources/A+A/backdata/2003/17/aah3724/table2.html,/proj/ads/fulltext/sources/A+A/backdata/2003/17/aah3724/table1.html'"

[spacemansteve]

non-existent file error is generated at https://github.com/adsabs/ADSfulltext/blob/master/adsft/checker.py#L235

[spacemansteve]

Currently, there are 11259 bibcodes in fulltext/all.links that list multiple files. This is out of 4823177 bibcodes that have fulltext. When multiple files are listed, it looks like the first file (right.html) contains the text of the paper while the other files hold tables.

[spacemansteve]

One complicating factor, commas can part of a filename:
2008bhgs.confE...1V /proj/ads/fulltext/sources/downloads/cache/POS/pos.sissa.it//archive/conferences/075/001/BHs,%20GR%20and%20Strings_001.pdf POS

Perhaps a comma followed by a '/' is the start of a new filename.

[spacemansteve]

I verified the Solr body field for 2003A&A...402..531C has text from each of the 4 files listed in fulltext/all.links.