Open rbellamy opened 7 years ago
artistechnic
commented
6 years ago Noob here. I hope this is the proper place to comment on this. Please let me know if it isn't. I seem to be having the same issue with a ScanSnap iX500: I select ADF Duplex, but only the front side is scanning.
jflesch
commented
6 years ago This is the proper place.
Problem currently is that I don't have any duplex scanner to try it. So at some point I may have to try at random .. :/
What I'm currently considering doing is a special diagnostic program and send it to as many volunteers as possible. The idea is to get detailed informations on as many scanners as possible and attempt scanning on as many scanners and sources as possible (there are other issues with other scanners). I'll try to include scanning from duplex ADF.
artistechnic
commented
6 years ago OK. Thanks.
Don't know if it would help, but if I can provide any useful information please let me know. I know duplex on this scanner works successfully using SANE (with the sane-fujitsu backend) in Simple Scan and gScan2PDF on Ubuntu and Solus OS.
rbellamy
commented
6 years ago I'm also willing to help if you need me to run a debug version and pass along logs.
mdorman
commented
6 years ago @jflesch I would be happy to donate a Fujitsu fi-6130Z that has duplex scanning, if that would help get it enabled.
jflesch
commented
6 years ago @mdorman and I would gladly accept it. You can find my address here below. Also the diagnostic program I was talking about is well under way. I hope to have it ready in the coming weeks.
FLESCH Jerome
28C rue Paul Milleret
25000 BESANCONI had a look. It seems SnapScan S1500 scanners do not support even TWAIN or ISIS. It seems unlikely they support WIA (the interface used by Pyinsane/Paperwork).
Therefore, the question is: Are you sure those scanners support WIA ? Do they work with any other third-party Windows applications ?
artistechnic
commented
6 years ago I didn't know WIA or ISIS existed until you mentioned them. It looks like you are correct that Fujitsu scanners do not support TWAIN). I'm learning here.
I know that recent versions of SANE support the iX500, and I can successfully scan duplex in Linux using Simple Scan and gScan2PDF, both of which use SANE. I think there is a commercial multiplatform product called VueScan that also claims to support the ScanSnap series, though I don't know what sort of driver it uses.
I'm afraid that's all I know. Sorry I can't be more helpful.
February 27, 2018 5:03 AM, "Jerome Flesch" <notifications@github.com (mailto:%22Jerome%20Flesch%22%20notifications@github.com)> wrote: I had a look. It seems SnapScan S1500 scanners do not support even TWAIN or ISIS (http://www.fujitsu.com/global/products/computing/peripheral/scanners/scansnap/discontinued/s1500/s1500.html). It seems unlikely they support WIA (the interface used by Pyinsane/Paperwork).
Therefore, the question is: Are you sure those scanners support WIA ? Do they work with any other third-party applications ?
—You are receiving this because you commented. Reply to this email directly, view it on GitHub (https://github.com/openpaperwork/paperwork/issues/673#issuecomment-368816190), or mute the thread (https://github.com/notifications/unsubscribe-auth/AgFDIlZYVah4B-1mn2q2IOx27c2LmAK2ks5tY9LggaJpZM4PFZO8).
pechfunk
commented
6 years ago My Fujitsu ScanSnap 1300i also defaulted to scanning Simplex. (Linux, SANE)
I found out that there was a problem with the heuristic of choosing the scanner source by matching human-friendly names like "ADF Front". In multiscan/init.py, the call to get_scanner() passes a list of preferred sources: ["ADF", "Feeder"] for SANE to match. However, in the list of sources offered by my S1300i, there are two matches to "ADF", with "ADF Duplex" coming in second. So the multiscan dialog would always pick the simplex source. I helped myself by putting "ADF Duplex" in front of the array, but that means I can't scan simplex any more.
Is there no better way to select scanner sources than by matching human-facing labels? (Honest question, I don't know SANE)
jflesch
commented
6 years ago @pechfunk : Can you send a scan test report to openpaper.work using IronScanner please ? This is the kind of information I'm going to need for the next versions of Paperwork :)
jflesch
commented
6 years ago Same thing for @rbellamy and all the other people that commented on this ticket: For those using Linux, can you send a scan test report using IronScanner please ?
artistechnic
commented
6 years ago Done on my part. I did not realize until I sent the report that, although my iX500 scanned both sides of the page as the settings dictated, the report only included the final page of the scan. I don’t know if that is important, but I include it here in case it may be.
jflesch
commented
6 years ago That's to be expected. I need only one page: I want to check ratio, and make sure the scanners don't add padding to pages (which it does in your case, a lot). I only need the last page for that. All the other information I may need are in the traces.
LS-art
commented
6 years ago @pechfunk: I have a similar issue with a Brother ADS-2100 on Ubuntu 16.04. Which file did you actually edit to "put "ADF Duplex" in front of the array"? Was this "init.py"? Where can I find that file?
pechfunk
commented
6 years ago @jflesch: I tried Ironscanner, but it failed with an undefined-variable error (see report 124).
@LS-art: That was line 170 of paperwork-gtk/src/paperwork/frontend/multiscan/init.py as of 1.2.4. I don't remember which sane-diagnostic program gave me the actual names of my scanner backend's feeding options.
jflesch
commented
6 years ago @pechfunk : I've fixed the issue in IronScanner. However it seems that the problem was that your scanner (driver) immediately told IronScanner that there is no paper at all in the feeder.
pechfunk
commented
6 years ago @jflesch Perhaps I hadn't shoved the sheet hard enough; there's little mechanic feedback. I've submitted two more scan attempts from the latest Ironscanner binary, unfortunately both ending in an AssertionError from the other side of the FIFO.
jflesch
commented
6 years ago Yep, on one side, your scanner seems to indicate an end of page .. after sending 0 pixels. On the other side, there was an assert a little bit too strict in Pyinsane. I've fixed the assert. I am not sure what I can do about your scanner returning a page of 0 pixels ... Can you try again so we see what happens ? Maybe the driver is always sending an empty page first.
pechfunk
commented
6 years ago Latest IronScanner binary (as of this evening): same result: assertion error after 10cm. -- funny, Paperwork itself does scan most of the time.
Report 131 was from ironscanner and pyinsane git master of today. Different error, same result :(
jflesch
commented
6 years ago [epjitsu] read_from_scanner: got GOOD/EOF (65536)
[epjitsu] read_from_scanner: finish rB:454024 len:65536
[epjitsu] sane_read: finish si:0 len:0 max:524288
INFO 2018-04-18 22:00:47,349 ironscanner.main Available lines: (0, 0)len:0 --> There is something wrong here ... :/
pechfunk
commented
6 years ago However, both Paperwork and the little pyinsane API example https://gist.github.com/pechfunk/63ac270acf803bfaffa9ae03fe5747e5 get to read full images from the scanner. I didn't quite follow how Ironscanner does things differently.
I have my ScanSnap S1500 Default source set to
ADF Duplex, but as you can clearly see from the log below, Paperwork is forcingADF Front.