William was kind enough to help me with the following awk script to convert fields into vCal standard items. It works beautifully in testing, but I cannot get the program I use to output 24h time, only AM and PM. I obviously need a statement to make AM disappear, and PM to add 12 hrs to the number. Do William or any awk gurus have suggestions? Here is the starting point: Title of Event,05/30/2003,11:45 AM,15, 7 Title of Event2,05/30/2003, 9:00 AM,15, 1 The result: BEGIN:VCALENDAR VERSION:1.0 BEGIN:VEVENT SUMMARY:Title of event DTSTART:20030530T114500 DTEND:20030530T120000 CATEGORIES:7 END:VEVENT BEGIN:VEVENT SUMMARY:Title of Event2 DTSTART:20030530T090000 DTEND:20030530T091500 CATEGORIES:1 END:VEVENT END:VCALENDAR The script that works beautifully on military time: #!/usr/bin/awk -f BEGIN { FS = "," OFS = "" print "BEGIN:VCALENDAR\nVERSION:1.0" } { split($2,startdate,"/") month=startdate[1] day=startdate[2] year=startdate[3] split($3,starttime,":") hour=starttime[1] minute=starttime[2] endhour = hour endminute = minute + $4 if ( endminute >= 60 ) { endhour += int(endminute / 60) endminute = endminute % 60 } print "BEGIN:VEVENT" print "SUMMARY:" $1 printf "%s%s%02d%02d%s%02d%02d%02d\n","DTSTART:", year,month,day, "T", hour,minute, "00" printf "%s%s%02d%02d%s%02d%02d%02d\n", "DTEND:", year,month,day, "T", endhour,endminute, "00" print "CATEGORIES:" $5 print "END:VEVENT" } END { print "END:VCALENDAR" } Any thoughts on an awk function that I can add that will use this field --> , 1:00 PM, instead of --> , 13:00, Thanks in advance for any advice..

If there can sometimes be a space before the start time as in your example, it gets trickier. If instead the data will always be like Title of Event,05/30/2003,11:45 AM,15, 7 Title of Event2,05/30/2003,9:00 AM,15, 1 then this should work: #!/usr/bin/awk -f BEGIN { FS = "," OFS = "" print "BEGIN:VCALENDAR\nVERSION:1.0" } { split($2,startdate,"/") month=startdate[1] day=startdate[2] year=startdate[3] split($3,starttime,"[: ]") hour=starttime[1] minute=starttime[2] ampm=starttime[3] if ( ampm == "PM" ) hour += 12 eventlength = $4 endhour = hour endminute = minute + eventlength if ( endminute >= 60 ) { endhour += int(endminute / 60) endminute %= 60 } print "BEGIN:VEVENT" print "SUMMARY:" $1 printf "%s%s%02d%02d%s%02d%02d%s\n","DTSTART:", year, month, day, "T", hour, minute, "00" printf "%s%s%02d%02d%s%02d%02d%s\n","DTEND:",year,month,day,"T",endhour,endminute, "00" print "CATEGORIES:" $5 print "END:VEVENT" } END { print "END:VCALENDAR" }

I'll check with our vendor, but unfortunately this schedule program is very proprietary and does not play well with others. I fully expect him to tell me that it has to be output in this manner, as I know there is a very strict control header, and when the time is less than 4 digit, it puts the space in front. I suspect this will make the script choke. Any way to add an initial statement that strips out certain spaces in certain fields?

Well, the situation is even stickier than I thought. My vendor says that he has to output part of the title field as: Doe, John This is in the same field, but makes the script think this is delimiting to the next field. I've asked him if he can use a different delimiter, allowing us to tweak the script. The time thing is set, unfortunately, and will not left justify. any thoughts??

How about adding: if ( index($3, " ") == 1 ) $3 = substr($3,2) right above the first "split()" in the revised script. That conditionally trims one blank off the beginning of the start time field.

For the problem of the first field potentially containing the delimiter character, it should be possible to have awk check upfront whether field #2 matches [number]/[number]/[number], and if it doesn't, and field #3 does, then perform a different set of assignments. It's not perfect, but at least the script would be making an intelligent guess. I don't have any time to try this out right now unfortunately. Good luck!

Thank you so much. I actually got that one solved by making my software guy give me a file with a delimiter of "^" instead of ",". I changed the delimiter in your awk script. It works like a charm, so nicely I can hardly stand it... I've got a new little problem to solve with the same file, but I'll post it in a different message. awk may not be the answer for this new issue, and I thank you so much for helping me out with this one!!!!