Collection of scripts for literature management.
In addition to Perl 5 (with Tk) and a POSIX-compliant shell with coreutils, the following command line tools are required:
parallelzathuraPDF viewerfdfind(also known asfd-findor justfd)pdftotext(notpdf2txt)
Customize PREFIX in the Makefile, if required (~/.local/bin by default). Then, simply run:
make installUsage: mklitentry.sh [FILE]...
Read one or more PDFs and generate a database entry (simple CSV) for each FILE in the following form:
key,tag1;tag2;tagN,/abs/path,unique words from text
If FILE is - or empty, read from STDIN. Output is written to STDOUT.
Usage: mklitdb.sh <DIR>
Create a literature database recursively from all PDFs inside the directory DIR by calling mklitentry.sh in parallel.
Output is written to STDOUT.
Usage: mkiidx.pl <DATABASE>
Generate an inverse index from database file <DATABASE> (previously generated by either mklitentry.sh or mklitdb.sh) in the following form:
term,key1:/absolute/path/1;key2:/absolute/path/2;...
...
Output is written to STDOUT.
Usage: add-to-library.sh <PDF_FILE>
Add a PDF file to the index manually.
Usage: rebuild-index.sh
Rebuild the entire library index.
Usage: bookworm.sh
Watch the currect path recursively and add newly created files to the index.
Usage: iidxlookup.pl <INVERSE_INDEX>
Read an inverse index (CSV file, previously generated by mkiidx.pl) and allow users to search it for all keywords in the query string (AND-search). File paths of matching documents are displayed in a list. When double-clicking a list entry, the absolute path is writen to STDOUT.
Usage litsearch
Wrapper for iidxlookup.pl that opens its output in a default application (defined in litsearchrc).
- use the status bar progress indicator instead of a modal dialog
- calculate
$TF$ or$TF/IDF$ inmklitentry.sh - rank results in
iidxlookup.pl