Am Donnerstag, 20. Februar 2025, 15:08:27 CET schrieb Richard Owlett: > I wish to extract CSV formatted data from a PDF document. [1] > Page ES-7 has a weekly grocery list for males grouped by age. > I need only the first and last columns. > > Can someone point me in a suitable direction? > > TIA > > [1] https://www.fns.usda.gov/cnpp/thrifty-food-plan-2006 > Table ES-1. Thrifty Food Plan market baskets, quantities of food > purchased for a week, by age-gender group, 2006
Without knowing the content of your pdf file, maybe you can port the pdf file to a text file for example by using "pdftotext". pdftotext [options] <PDF-file> [<text-file>] Then, you could read every line in this text file and filter only lines with a unique word in this line (i.e. "sold") and create a new file with all lines you only need. For example: cat ~/my_file.txt | grep "sold" > my_new_file.txt Now you have this one, you can cat and cut only words you need (see manual of cut). The syntax is similar like: cat `cut --fields 3 5 7` ~/my_new_files.txt > my_target_file.txt This would read linewise and only print the 3rd, 5th and 7th word of the source file. See manual of cut, what options you need. At last, you can edit the my_target_file.txt with any editor and add a separator sign at any space between the words. The space is also a sign like any other and can be exchanged like any other letter. Then you would have a csv file! If you are familiar with these commands, you can write a shell script, which does all in once in a future. Usefull also vor very very big files. Please note: Above might not be the correct syntax!!! My goal was more, to show, which way you could like to go and it maybe not usefull at your special pdf file. Maybe, if my suggestion is usefull, someone more experienced as me can you tell the correct commands. Please take look at the manuals of pdftotext, cat and cut, hope this helps. Best regards Hans
