News export to Kobo with templates

[biffhero]

I subscribe to 120+ news sites, so after four days, there are 500 articles in one directory.

If I want to read articles from today before moving on to yesterday's news, I have to scroll and scroll and scroll. You get the picture. :-)

I fought with templates for a couple of hours today, before giving up and writing a shell script. This is what I wrote:

#!/bin/bash

#
# take care of the news files in the kobo directory
#

NEWSPATH="/media/rob/KOBOeReader/News/"

function findNewsDir() {
    path=${1}
    if [[ -e "${path}" ]]
    then
        REPLY="yes"
    else
        REPLY="no"
    fi
}

function main() {
    # newsDir=findNewsDir ${NEWSPATH}
    findNewsDir ${NEWSPATH}
    if [[ ${REPLY} == "yes" ]]
    then
        echo "Found the news directory"
        echo "It is ${NEWSPATH}"
        echo "getting dates now"
        for date in $(ls "${NEWSPATH}" | rev | cut -f 3- -d "." | cut -f 1 -d "_" | rev | sort -u)
        do
            {
                echo "Date : ${date}"
                if [[ ! -e "${NEWSPATH}${date}" ]]
                then
                    echo "date dir for ${date} does not exist"
                    printf "making directory - %s" "${NEWSPATH}${date}/"
                    mkdir "${NEWSPATH}${date}/"
                fi

                echo "date dir for ${date} exists"
                OLDIFS="${IFS}"
                IFS=$'\n'
                for article in $(ls "${NEWSPATH}" | grep ${date}.kepub.epub$)
                do
                    {
                        printf "article - %s\n" "${article}"
                        echo "I want to move article : \"${article}\""
                        echo "to"
                        echo "${NEWSPATH}${date}/"
                        echo
                        mv "${NEWSPATH}${article}" "${NEWSPATH}${date}/"
                    }
                done
                IFS="${OLDIFS}"
            }
        done
    else
        echo "Cannot find the news directory, exiting"
        exit 1
    fi
}

main

I'm sure there is a better way, but I don't know what it is, and I can't get a template to do it. I have made two columns out of information I found on other posts, or on stackoverflow, or something.

1. news_date : {date:'test($,
strcat(format_date($,'yyyy'), '_',format_date($,'MM'), '_',format_date($,'dd')),
'')'}

2. test_date : {authors:contains(calibre, {#news_date},)}

I figured I would use a device template that had /{#test_date} in it, but that never seemed to work. I'm guessing at this point that there is different code paths for exporting News/ articles and other things.

Thanks,
Rob

[davidfor]

What exactly are you trying to achieve? Are you trying to build a directory structure on the device? If so, that won't help if you are using the Kobo software to read the book. It doesn't use the directory structure in the interface. If you are using KOReader or Plato or something else, it might work.

For reading in the Kobo, you could create a collection for each day. When you have read all the new articles, you can remove the collection.

You could do this with a custom column with the template:

program:
news_date = format_date(raw_field('timestamp'),'yyyy-MM-dd');
contains(field('authors'), 'calibre', news_date, '');

I'm assuming that "calibre" is the author for the news articles. But, you can use whatever other test would be appropriate.

If you do the above, you would add the column the the "Collections columns" in the driver configuration. That should create collections for each day as you send the news articles to the device.

[biffhero]

I'm sorry, I wasn't clear. I am using KOReader, so one of my main interfaces to my collection is the filesystem. I am trying to make a different directory for each days' worth of news that is on the reader.

News/
News/2020-08-19/
News/2020-08-20/
News/2020-08-21/

and so on, and so forth. I run my script after the calibre sync, it creates the directories and moves everything in News/ into the appropriate subdirectory.

Thank you.