Re: [sane-devel] [pieusb] Fix unused-parameter compiler warnings
Hi Klaus, Your vacation service mentioned you'd be away until 2016-09-13. Have you had a chance to look at this already? I have a patch lined up that simply removes all the unused code. If that's fine with you (or I don't get a reply ;-), I'll just go ahead and push it at the end of September. On 2016-08-28, Olaf Meeuwissen wrote: > Hi Klaus, > > The following functions in pieusb_scancmd.c trigger a pile of unused > parameter warnings: > > sanei_pieusb_cmd_get_exposure_time > sanei_pieusb_cmd_get_halftone_pattern > sanei_pieusb_cmd_get_highlight_shadow > sanei_pieusb_cmd_set_ccd_mask > sanei_pieusb_cmd_set_halftone_pattern > > Upon closer inspection, I noticed that these functions are not called > anywhere in the backend and all but the get-halftone-pattern are stubs. > > I'm trying to get rid of all compiler warnings and wondered how to fix > this. I can completely remove the functions from the code or mark all > unused parameters with a __sane_unused__ on the assumption that they > will be unstubbed and used in the (near?) future. > > What do you suggest? Hope this helps, -- Olaf Meeuwissen, LPIC-2FSF Associate Member since 2004-01-27 GnuPG key: F84A2DD9/B3C0 2F47 EA19 64F4 9F13 F43E B8A4 A88A F84A 2DD9 Support Free Softwarehttps://my.fsf.org/donate Join the Free Software Foundation https://my.fsf.org/join -- sane-devel mailing list: sane-devel@lists.alioth.debian.org http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/sane-devel Unsubscribe: Send mail with subject "unsubscribe your_password" to sane-devel-requ...@lists.alioth.debian.org
Re: [sane-devel] [niash] Fix unused-but-set-variable compiler warning
On 2016-08-28, Olaf Meeuwissen wrote: > Hi Gerhard, > > The startBlackX variable in SimpleClibExt() from niash_core.c is flagged > as set but unused by gcc on my debian-8-full build. I was going to just > remove it but wondered if it ought to be used. I removed it in fb0464d and pushed that just now. > Specifically, I wondered > if it would be needed in the loop that starts at niash_core.c:1301. > >> for (i = 0; i < endBlackY - startBlackY + 1; i++) >>{ >> CircBufferGetLine (iHandle, , abLine, iReversedHead); >> for (j = 0; j < endBlackX; j++) >> { >>bMinR = MIN (abLine[j * 3 + 0], bMinR); >>bMinG = MIN (abLine[j * 3 + 1], bMinG); >>bMinB = MIN (abLine[j * 3 + 2], bMinB); >> } >>} > > Should the initialization of j use startBlackX? For hp scanners the > startBlackX variable is set to 0 but for agfa scanner to 1666. Hope this helps, -- Olaf Meeuwissen, LPIC-2FSF Associate Member since 2004-01-27 GnuPG key: F84A2DD9/B3C0 2F47 EA19 64F4 9F13 F43E B8A4 A88A F84A 2DD9 Support Free Softwarehttps://my.fsf.org/donate Join the Free Software Foundation https://my.fsf.org/join -- sane-devel mailing list: sane-devel@lists.alioth.debian.org http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/sane-devel Unsubscribe: Send mail with subject "unsubscribe your_password" to sane-devel-requ...@lists.alioth.debian.org
Re: [sane-devel] Eventually got a debug log running ET-4500 with backend epson2 over network
Hi Roger, Roger Sewell writes: >> Could you upgrade to 1.0.25? There are some notes on network scanner >> fixes as well as timeout and scanner crash fixes for 1.0.25 that might >> just fix things for you. That is, if you want to drop that non-free >> plugin needed by the imagescan/utsushi backends. > > So I did in fact already try 1.0.25, and got exactly the same > behaviour. > > Given that, do you still need the log from 1.0.25 ? Need is an overstatement ;-) but if you don't mind making one ... One thing though, the epson2 debug output alone is not detailed enough to see what's going wrong. I'd need a network packet capture as well. You can make one with wireshark. If you still feel like making a log, I'll take a look. -- Olaf Meeuwissen, LPIC-2FSF Associate Member since 2004-01-27 GnuPG key: F84A2DD9/B3C0 2F47 EA19 64F4 9F13 F43E B8A4 A88A F84A 2DD9 Support Free Softwarehttps://my.fsf.org/donate Join the Free Software Foundation https://my.fsf.org/join -- sane-devel mailing list: sane-devel@lists.alioth.debian.org http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/sane-devel Unsubscribe: Send mail with subject "unsubscribe your_password" to sane-devel-requ...@lists.alioth.debian.org
Re: [sane-devel] SANE-Backend for Brother PDS-series scanner
James A. Robinson writes: > I appreciate your contacting them about the discrepancy. We'll see what > they say (as you indicate, my own suspicion is that they will say nothing). > > BTW, regarding making sure to read the EULA, I assume you noticed that you > can't actually get to the point where you see what architecture are > supported, nor what the copyrights are, until after you sign the license? Yup, just your "regular" click-through license. Gotta agree to terms of use before you even get to take a look at what it is you can use, but they'll use it against you anyway. -- Olaf Meeuwissen, LPIC-2FSF Associate Member since 2004-01-27 GnuPG key: F84A2DD9/B3C0 2F47 EA19 64F4 9F13 F43E B8A4 A88A F84A 2DD9 Support Free Softwarehttps://my.fsf.org/donate Join the Free Software Foundation https://my.fsf.org/join -- sane-devel mailing list: sane-devel@lists.alioth.debian.org http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/sane-devel Unsubscribe: Send mail with subject "unsubscribe your_password" to sane-devel-requ...@lists.alioth.debian.org
[sane-devel] SANE license interpretation (was Re: SANE-Backend for Brother PDS-series scanner)
Hi Allan, m. allan noah writes: > On Sun, Sep 18, 2016 at 6:25 AM, Olaf Meeuwissen >wrote: >> Hi Allan, >> >> I just dashed off a rather long explation to James and the list. >> >> m. allan noah writes: >> >>> I personally am of the opinion that Brother is in violation of our >>> license. However, our license is not strictly GPL, and the differences >>> were clearly not written by a lawyer. You could argue that we give >>> some space for a company to steal our work, and keep it from their >>> users. >> >> The sane-backends source code contains files that are GPL and some that >> are GPL with an exception. The exception is similar in spirit to what >> the LGPL allows and was, IIRC, added before (or around) the LGPL was >> introduced. As long as they only used GPL with exception code (based on >> library symbols, I they did), there isn't really anything you can object >> to (unless you also object to using LGPL'd code ;-). > > Well, I disagree on that point. I believe we have disagreed on this point in the past ;-) Not trying to start a flame, just would like to arrive at a common point of view for SANE. > In my mind, the key words in the exception are 'an executable'. My thinking on this has been influenced by the libtool documentation which says[1]: libraries are programs with multiple entry points and the GPL FAQ on "aggregation" and "modified versions"[2], which says: If the modules are included in the same executable file, they are definitely combined in one program. If modules are designed to run linked together in a shared address space, that almost surely means combining them into one program. [1] https://www.gnu.org/software/libtool/manual/libtool.html#Libtool-paradigm [2] https://www.gnu.org/licenses/gpl-faq.html#MereAggregation I basically regard libraries and programs (do these cover executables?) as the same thing and draw a border at the process boundary. From that point of view, I don't think that non-free third party backends linking with libsanei violate the "SANE exception". # I would like to arrive at the opposite conclusion, though. # BTW, like all SANE backends, the Brother backend links statically with # libsanei/ so only parts of it are included and used. From what I saw # in terms of library symbols all of those parts are GPL+exception. I # didn't notice any SANE parts outside of sanei/. Whether my interpretation corresponds with the thinking of the people who added that exception, I don't know. Perhaps we should ask? > At the time the exception was added, > many people believed that free OS's were the way of the future, but we > would still be using large numbers of closed-source, shrink-wrapped > apps. The exception seems to have been intended to allow userspace > programs like StarOffice to initiate scanning. This is different from > linking SANE to a library, even if that library acts as a SANE > backend. I think it is a stretch to call Brother's usage in a backend > 'an executable'. Yes, a library contains executable code, but it > cannot be started from the command line without a front-end. If that front-end would be a closed-source, shrink-wrapped app, would Brother be off the hook in your opinion? Does that depend on how the app links with the backend (directly or through libsane-dll)? Approaching things from the opposite end, what about GPL'd front-ends (think scanimage and saned; they carry no exception) using a non-free backend such as Brother's? Would that be okay? Wouldn't that be kind of misleading users of these frontends? They're using a GPL'd app but under the covers they might unwittingly be using non-free backends. >> Personally, I wished that more of the sane-backends code is GPL, making >> it harder for folks to take what we share without sharing back. > > I also wish we could change it, but it is too late. There have been > too many authors over too many years. There is a good chance that a > few of the authors have died. It would be very difficult to identify > entire files which could have their license changed. I know, I've read the LICENSE file ;-) Doesn't stop me from wishing, though. Talking about that file, perhaps we should add a clarification about the use of libsanei. Now, the file only mentions the SANE backends. > When I added sanei_magic, I seriously debated making it GPL. But, I > decided that I could not find a way to convert my backends (which > needed to use it) to the GPL as well. Have you considered making it LGPL rather than GPL+exception? It might be a good idea to suggest that license for *new* code. Just an idea. Hope this helps, -- Olaf Meeuwissen, LPIC-2FSF Associate Member since 2004-01-27 GnuPG key: F84A2DD9/B3C0 2F47 EA19 64F4 9F13 F43E B8A4 A88A F84A 2DD9 Support Free Softwarehttps://my.fsf.org/donate Join the Free Software Foundation https://my.fsf.org/join -- sane-devel mailing list:
Re: [sane-devel] SANE-Backend for Brother PDS-series scanner
I appreciate your contacting them about the discrepancy. We'll see what they say (as you indicate, my own suspicion is that they will say nothing). BTW, regarding making sure to read the EULA, I assume you noticed that you can't actually get to the point where you see what architecture are supported, nor what the copyrights are, until after you sign the license? Jim -- sane-devel mailing list: sane-devel@lists.alioth.debian.org http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/sane-devel Unsubscribe: Send mail with subject "unsubscribe your_password" to sane-devel-requ...@lists.alioth.debian.org
Re: [sane-devel] Eventually got a debug log running ET-4500 with backend epson2 over network
Olaf, > Thanks for the log. Comparing the log with the code I noticed that > you are using sane-backends-1.0.24 (yes, I should have noticed from > your previous mail ;-). The logs hint at a low-level network I/O > error. Thank you... > Could you upgrade to 1.0.25? There are some notes on network scanner > fixes as well as timeout and scanner crash fixes for 1.0.25 that might > just fix things for you. That is, if you want to drop that non-free > plugin needed by the imagescan/utsushi backends. So I did in fact already try 1.0.25, and got exactly the same behaviour. Given that, do you still need the log from 1.0.25 ? Roger. -- sane-devel mailing list: sane-devel@lists.alioth.debian.org http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/sane-devel Unsubscribe: Send mail with subject "unsubscribe your_password" to sane-devel-requ...@lists.alioth.debian.org
Re: [sane-devel] "No: command not found" when building sane-backends
Greetings. On Sunday, September 18, 2016 9:47:32 PM CEST Olaf Meeuwissen wrote: > I confused dhl with a well known mail carrier in configure.ac. On top > of that, I forgot to add the fig2dev check. Fixes have been pushed and > things should be all well on master. Thanks! I've verified that the build now succeeds even when fig2dev and dlh are not present. (Presumably the part of the doc build that depends on them is just skipped.) Regards, Tristan -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Tristan Miller Free Software developer, ferret herder, logologist https://logological.org/ =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- signature.asc Description: This is a digitally signed message part. -- sane-devel mailing list: sane-devel@lists.alioth.debian.org http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/sane-devel Unsubscribe: Send mail with subject "unsubscribe your_password" to sane-devel-requ...@lists.alioth.debian.org
Re: [sane-devel] "No: command not found" when building sane-backends
Hi Tristan, list, Olaf Meeuwissen writes: > Tristan Miller writes: > >> Greetings. >> >> On Saturday, September 17, 2016 3:37:26 PM CEST m. allan noah wrote: >>> Hmm, sounds like fig2dev is missing on your machine, and we mistakenly >>> tested for it and let the string 'no' sneak thru when it was not >>> found. Looks like Olaf worked on this in cc8f99b9 > > I don't think that that's what caused it. I think it was introduced in > 79c62362 in combination with changes in newer versions of the autotools. I was wrong. I confused dhl with a well known mail carrier in configure.ac. On top of that, I forgot to add the fig2dev check. Fixes have been pushed and things should be all well on master. # Unless you `make html`. That target is still potentially broken due # to its sane-html and man-html dependencies. The reason I didn't see these issues in my CI builds is because they don't have any of the other needed utilities installed either. The absence of those other utilities prevented sane-html getting added to the all target. Hope this helps, -- Olaf Meeuwissen, LPIC-2FSF Associate Member since 2004-01-27 GnuPG key: F84A2DD9/B3C0 2F47 EA19 64F4 9F13 F43E B8A4 A88A F84A 2DD9 Support Free Softwarehttps://my.fsf.org/donate Join the Free Software Foundation https://my.fsf.org/join -- sane-devel mailing list: sane-devel@lists.alioth.debian.org http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/sane-devel Unsubscribe: Send mail with subject "unsubscribe your_password" to sane-devel-requ...@lists.alioth.debian.org
Re: [sane-devel] SANE-Backend for Brother PDS-series scanner
On Sun, Sep 18, 2016 at 6:25 AM, Olaf Meeuwissenwrote: > Hi Allan, > > I just dashed off a rather long explation to James and the list. > > m. allan noah writes: > >> I personally am of the opinion that Brother is in violation of our >> license. However, our license is not strictly GPL, and the differences >> were clearly not written by a lawyer. You could argue that we give >> some space for a company to steal our work, and keep it from their >> users. > > The sane-backends source code contains files that are GPL and some that > are GPL with an exception. The exception is similar in spirit to what > the LGPL allows and was, IIRC, added before (or around) the LGPL was > introduced. As long as they only used GPL with exception code (based on > library symbols, I they did), there isn't really anything you can object > to (unless you also object to using LGPL'd code ;-). Well, I disagree on that point. In my mind, the key words in the exception are 'an executable'. At the time the exception was added, many people believed that free OS's were the way of the future, but we would still be using large numbers of closed-source, shrink-wrapped apps. The exception seems to have been intended to allow userspace programs like StarOffice to initiate scanning. This is different from linking SANE to a library, even if that library acts as a SANE backend. I think it is a stretch to call Brother's usage in a backend 'an executable'. Yes, a library contains executable code, but it cannot be started from the command line without a front-end. > > Personally, I wished that more of the sane-backends code is GPL, making > it harder for folks to take what we share without sharing back. I also wish we could change it, but it is too late. There have been too many authors over too many years. There is a good chance that a few of the authors have died. It would be very difficult to identify entire files which could have their license changed. When I added sanei_magic, I seriously debated making it GPL. But, I decided that I could not find a way to convert my backends (which needed to use it) to the GPL as well. allan -- "well, I stand up next to a mountain- and I chop it down with the edge of my hand" -- sane-devel mailing list: sane-devel@lists.alioth.debian.org http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/sane-devel Unsubscribe: Send mail with subject "unsubscribe your_password" to sane-devel-requ...@lists.alioth.debian.org
Re: [sane-devel] Eventually got a debug log running ET-4500 with backend epson2 over network
Hi Roger, Roger Sewell writes: > Olaf, and whoever kindly maintains the epson2 backend: > > Please find attached a debug log of my attempt to run a 2-page scan from > the ADF of an Epson ET-4500 using the epson2 backend over the network. Thanks for the log. Comparing the log with the code I noticed that you are using sane-backends-1.0.24 (yes, I should have noticed from your previous mail ;-). The logs hint at a low-level network I/O error. Could you upgrade to 1.0.25? There are some notes on network scanner fixes as well as timeout and scanner crash fixes for 1.0.25 that might just fix things for you. That is, if you want to drop that non-free plugin needed by the imagescan/utsushi backends. Hope this helps, -- Olaf Meeuwissen, LPIC-2FSF Associate Member since 2004-01-27 GnuPG key: F84A2DD9/B3C0 2F47 EA19 64F4 9F13 F43E B8A4 A88A F84A 2DD9 Support Free Softwarehttps://my.fsf.org/donate Join the Free Software Foundation https://my.fsf.org/join -- sane-devel mailing list: sane-devel@lists.alioth.debian.org http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/sane-devel Unsubscribe: Send mail with subject "unsubscribe your_password" to sane-devel-requ...@lists.alioth.debian.org
Re: [sane-devel] SANE-Backend for Brother PDS-series scanner
Hi Allan, I just dashed off a rather long explation to James and the list. m. allan noah writes: > I personally am of the opinion that Brother is in violation of our > license. However, our license is not strictly GPL, and the differences > were clearly not written by a lawyer. You could argue that we give > some space for a company to steal our work, and keep it from their > users. The sane-backends source code contains files that are GPL and some that are GPL with an exception. The exception is similar in spirit to what the LGPL allows and was, IIRC, added before (or around) the LGPL was introduced. As long as they only used GPL with exception code (based on library symbols, I they did), there isn't really anything you can object to (unless you also object to using LGPL'd code ;-). Personally, I wished that more of the sane-backends code is GPL, making it harder for folks to take what we share without sharing back. > I vote with my money, and don't buy their products. In addition, I'd point out their mistakes, publicly ;-) Hope this helps, -- Olaf Meeuwissen, LPIC-2FSF Associate Member since 2004-01-27 GnuPG key: F84A2DD9/B3C0 2F47 EA19 64F4 9F13 F43E B8A4 A88A F84A 2DD9 Support Free Softwarehttps://my.fsf.org/donate Join the Free Software Foundation https://my.fsf.org/join -- sane-devel mailing list: sane-devel@lists.alioth.debian.org http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/sane-devel Unsubscribe: Send mail with subject "unsubscribe your_password" to sane-devel-requ...@lists.alioth.debian.org
Re: [sane-devel] SANE-Backend for Brother PDS-series scanner
Hi, James A. Robinson writes: > I was wondering whether or not any of the developers here know the > story behind the Brother scanner drivers? They offer a Debian > package that identifies itself as GPL, and that claims it is based > on sane-backends.When I asked them about getting ahold the > source (I want to port it to ARM), I got the stock "thank you for > choosing... unfortunately we cannot support all models..." answer. > > I can't tell if that means that > > (a) they aren't really obligated to release their driver under the > GPL, and mislabeled the Debian package as such. > > (b) they are obligated because they derived it from GPLed code, and > are not following their obligations under that license. > > I did specifically mention the GPL confusion, but they didn't > bother to address that point in their reply. > > The copyright file from the debian binary package > http://support.brother.com/g/b/downloadlist.aspx?c=us=en=pds6000_us_eu_as=128 > says: > > Name : brother-pds-sane > Version : 1.0.0 > Release : 4 > Architecture: amd64 > Install Date: (not installed) > Group : Applications/System > Size : 14276902 > License : GPL > Signature : (none) > Source RPM : brother-pds-sane-1.0.0-4.src.rpm > Build Date : 2015年05月28日 (週四) 16時34分37秒 > Build Host : Brother > Relocations : (not relocatable) > Packager : Brother Industries, Ltd. > Vendor : Brother Industries, Ltd. > Summary : SANE-Backend for Brother PDS-series scanner > Description : > SANE-Backend for Brother PDS-series scanner. It based on sane-backends, so > we need install sane-backends package first. Hmm, before being able to download, I had to agree to a EULA that had this in it: Further, Brother shall have no liability to disclose and/or distribute the source cord[sic] of the Software to User under any circumstances. In no case shall the above license by Brother to modify, alter, translate or otherwise prepare derivative works of the Software be construed as Brother's implied agreement or undertakings to disclose and/or distribute the source cord[sic] of the Software. That flies straight in the face of this from the GPL-v3 preamble: [...] General Public Licenses are designed to make sure that you have the freedom to distribute copies of free software (and charge for them if you wish), that you **receive source code or can get it if you want it**, that you can change the software or use pieces of it in new free programs, and that you know you can do these things. # Emphasis is mine. I've let Brother know via the Content Feedback shown for the download and suggested they fix the discrepancy, preferably by distributing the source code. # Not holding my breath on anything getting actually done, but ... who # knows, they might take notice. BTW, the EULA does grant: [...] User a non-exclusive license: to reproduce and/or distribute (via Internet or in any other manner) the Software. Further, Brother grants User a non-exclusive license to modify, alter, translate or otherwise prepare derivative works of the Software and to reproduce and distribute (via Internet or in any other manner) such modification, alteration, translation or other derivative works for any purpose. That's in line with the GPL but utterly useless without access to the source code. I did a quick check on the contents (btw, everything in usr/lib64 is a copy of what's in usr/local/lib64, bloating the package by ~100%), but all the libraries only seem to link to LGPL'd libraries (or GPL-v3 with the a GCC Runtime Library Exception). This makes me believe that Brother can change the license to whatever it pleases (within the LGPL limitations on debugging for reverse engineering purposes), assuming they own all the other code that went into those libraries. The libsane-brother_pdsseries backend uses some code from sanei/, but all of that is GPL with a special exception. The exception is that, if you link a SANE library with other files to produce an executable, this does not by itself cause the resulting executable to be covered by the GNU General Public License. Your use of that executable is in no way restricted on account of linking the SANE library code into it. This exception does not, however, invalidate any other reasons why the executable file might be covered by the GNU General Public License. which is pretty much what the LGPL does too. # The SANE licensing approach predates the LGPL, IIRC. In short, I'd say that Brother made a very bad-tasting mistake labeling their Debian packages as GPL. Whether on purpose or not, I don't know, but next time make sure to read the EULA before you click agree ;-( Hope this clarifies, -- Olaf Meeuwissen, LPIC-2FSF Associate Member since 2004-01-27 GnuPG key: F84A2DD9/B3C0 2F47 EA19 64F4 9F13 F43E B8A4 A88A F84A 2DD9 Support Free Softwarehttps://my.fsf.org/donate Join the Free Software Foundation