I don't know of any "ripping" software that will enable you to copy MPG2 files off your playable DVD (although they're stored as Video Object files, or VOBs, they're MPEG2 underneath it all), edit (staying in MPEG2) and then re-author into UFO, BUP and VOB files and re-burn. Trying to stay entirely within MPEG2 is a major pain in the a@$e.
If you play your DVD into a PC via a video capture card, then you've now gone one generation further in the edit because the capture card probably doesn't capture MPEG2 but the (now) de-compressed file. So already it has been through MPEG2 and back again.
If the bit-rate the MPEG was originally encoded with is high (say, video bit-rate above 8Mbits per sec), it may withstand a further generation (MPEG2 back to AVI, editied, then re-rendered back to MPEG2, re-authoring and burning) and still look OK. However, you will be adding the second generation's quantisation-noise to that of the furst, and sometimes this noise may be visible in the picture - depending on how good your display device is and how finicky you are about picture quality.
MPEG2 is heavily compressed, and the process is pretty technical, involving initially a lossless digitisation process of an 8X8 pixel "macroblock", then quantisation and compression of the digitised data using a numerical matrix and a Discrete Cosine Transform function. This compression is lossy, but the losses are in the finest details as far as possible, relying on the inability of the human eye to notice the difference - and the usually mediocre resolution of domestic display devices.
MPEG2 further reduces bandwidth (or filesize) by having three different types of compressed picture frames. The I-frame is a complete, displayable frame of video, entire of itself. The B-frame is a bidirectionally-predictive frame, which contains information as to motion vectors for frames that precede it and follow it (naturally these frames have to be transmitted out-of-normal-time-sequence so the motion vector determinations for frames that should be displayed prior to the B-frame that contains their information can be correctly displayed,) and the P-fame, which contains information about the frame which follows it in normal time-sequence. (I work with this stuff all the time and it still makes my head spin too!)
In satellite transmission, we use a Group Of Pictures (GOP) as follows: IBBPBBPBBPI, so an I-frame occurs only every so often - and if the MPEG2 decoder loses lock, then it has to wait until an I-frame arrives in order to re-lock to the transport stream and begin correctly reproducing piccies again. I-frames are largest in terms of bits, P-frames are smaller and B-frames are also small, thus maximising efficiency and reducing transmission bandwidth requirements for any given video bit-rate.
DVD MPEG2 seems to prefer to use only I and P-frames, allowing DVD MPEG decoders to re-establish frame-sync much more quickly, reducing the visual impact of a glitch. This increases filesizes for DVD Video Object Files.
Forgive the technicalities above, but what I'm trying to show is that repeated trips through the MPEG2 process back to uncompressed video, then back into MPEG2 etc etc etc just erodes picture quality, although the degradation for a second generation may not be noticeable - it would certainly be worth a try. All you'll lose is your time and possibly a single DVD at the end of it all if you're not happy with the result.
The microprocessor power necessary to edit entirely in MPEG2 would be truly massive, and there is the added burden or overhead of correctly reassembling the B and P-frames into correct order to successfully edit in MPEG2...which is almost like de-compressing the video anyway. I don't know of any editing software that works entirely within MPEG2 files, most of the professional stuff prefers to work in uncompressed video, so transitions etc can be previewed instantly instead of needing to be re-rendered into a new MPEG2 group of pictures and re-muxed into the data stream.
Basically, editing in uncompressed files is faster and allows the original material to remain unaltered, as all the editing software us really doing is buildng up an "edit decision list" which is then executed on the files as needed at rendering-time, before the uncompressed video is turned into MPEG2.
Don't know if this is really the answer to your question, but at least it may help you to understand the horrendous things that have to be done to beautiful analogue PAL or NTSC (or SECAM) piccies to crunch them down into DVD files, and why we infinitely prefer to work in uncompressed files in the edit process, which in turn explains our reluctance to re-use camera-tapes. If a production has to be extended, shortened or re-edited, it is by far the best option to have the original camera-tapes and go back to them. They are the best it is ever going to look, and everything after that is a slow trip downhill.
Speaking of "how good it looks", even the best digital handicam of today won't shoot piccies that look any better than those shot on ten-year old broadcast camera. Even an old analogue broadcast camera will have a magnificent Fujinon or Canon lens, built with something between 14 and 18 elements, designed to be superb in resolving-power, tack-sharp, free of distortion and costing probably five times what a digital handicam does anyway. Right at the back of the lens is the best the pix ever look, and a good lens is the greatest ally a cameraman has to get decent pictures - and always will be. I still shoot on Betacam SP (an old analogue broadcast format) which is playable in any digital Betacam deck. Once the pix are on a hard-drive, what tape format they were shot in is immaterial. How good the lens was is still important because that basically determines how sharp the pix will be and that doesn't alter much afterwards until you get to very low bit-rates - far to low to be useable.
Another issue with DVD files is the actualy playable files (Video Objects or VOB files) are seamless, that is one ends precisely where the next begins as far as playback is concerned, so it is essentially seamless when menu-driven DVDs are played. (Each separate menu-item requires a separate VOB file).
Altering the length of any one VOB file within a group will require complete re-writing of all the subsequent VOBs as well. Again, I don't know of any software that does this at the editing stage because VOBs are not editable as such. They are the result of DVD authoring software chopping up an MPEG2 file into chunks to suit the playback requirements of the DVD player. (The largest a VOB can be is 1Gb, so a DVD of any length will have several VOBs.)
Basically, play your DVD back into your PC via your video capture card, add your new material into it, then re-render and burn a new DVD. That's what I would do, if I didn't have original camera-tapes to work with. If I did, then I'd always go straight back to them and start again.
Apologies for a long and technical post, but if a little knowledge is a dangeourous thing, a lot can be positively addictive!