for some files, pdfgrep fails to find existing patterns
When searching a batch of pdf files for the pattern "©.+ISBN" to determine if they are books, pdfgrep failed to find the pattern for some files.
Yet, searching the file with macOS's Preview application showed that the pattern is present in the file. Also, producing a text representation with pdftotext showed that the pattern should have been found.
I expanded the search pattern for pdfgrep to include more text, and saw that in the output produced, one or more characters on the left margin are missing; e.g. for the line beginning with "ISBN 978-0-7506-8210-7", the output shows "SBN 978-0-7506-8210-7". This is for all lines between the beginning and ending parts of the match.
For another pdf file, even more of the leftmost characters of every line were missing, e.g. "Library of Congress Number: 2004111169" showed as " of Congress Number: 2004111169".
To me, those files do not have anything suspect about them.
This is pdfgrep version 2.0.1.
Using poppler version 0.56.0 Using libpcre version 8.41 2017-07-05
Installed with brew on OS X 10.11.6